January 2011
NSF10/27/EPADWCTR
EPA/600/R-10/151
Environmental Technology
Verification Report
Removal of Inorganic, Microbial, and
Particulate Contaminants from Secondary
Treated Wastewater
Village Marine Tec.
Expeditionary Unit Water Purifier,
Generation 1
Prepared by
NSF International
Under a Cooperative Agreement with
U.S. Environmental Protection Agency
-------
This page is intentionally blank
-------
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 INORGANIC, MICROBIAL, AND
PARTICULATE CONTAMINANTS FROM SECONDARY
TREATED WASTEWATER
PRODUCT NAME: EXPEDITIONARY UNIT WATER PURIFIER (EUWP),
GENERATION 1
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 evaluated using
secondary wastewater effluent from the Gallup, New Mexico wastewater treatment plant (WWTP).
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 voluntary 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 10/27/EPADWCTR The accompanying notice is an integral part of this verification statement. January 2011
VS-i
-------
PRODUCT DESCRIPTION
The following technology description was provided by the manufacturer for informational purposes only
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 are designed to
operate with a generator and 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. The first pass part
of the RO system has two arrays. One of the arrays is driven by the normal RO feed pump and the other
array is driven by the energy saving device. There is only one array in the second pass part of the RO
system. The EUWP 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,
ferric chloride coagulation pretreatment was used at a dose of 5 mg/L as Fe. There was no post-treatment
chlorination.
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 test was performed at the City of Gallup WWTP at 800 Sweetwater Place, Gallup, New Mexico. The
WWTP treats an average of 3 million gallons per day (MGD) of wastewater with a peak of 5.5 MGD in
the summer. The source water for testing was secondary wastewater effluent prior to chlorination. Initial
characterization samples, which consisted of six grab samples, were collected in May and June of 2006.
Highlights of the source water characterization are presented in Table VS-i. Parameters in the source
water that exceed the EPA's National Primary Drinking Water Regulations (NPDWR) included nitrate,
bromide, gross alpha, and biological components. Secondary drinking water standards were exceeded for
color, sulfate, TDS, surfactants, aluminum, and odor. The source of the city's drinking water is high in
TDS and sulfate with some radioactivity. The rest of the exceedances are caused by municipal use and the
wastewater treatment process. Detailed results of the source water characterization can be found in the
report.
Table VS-i. Source Water Characterization Data
Parameter
Color (color units)
Bromide ( mg/L)
Sulfate (mg/L)
Nitrate (as Nitrogen) ( mg/L)
TDS (mg/L)
Surfactants (mg/L)
Aluminum (ng/L)
Odor (Threshold Odor Number)
Gross Alpha (pCi/L)
Total Coliform (MPN/100 mL)
Fecal Coliform (MPN/100 mL)
Heterotrophic Plate Count (HPC) (CFU/mL)
5/25
35
0.20
320
19.5
1100
0.75
<100
12
9.8
N/A
N/A
N/A
6/01
30
0.20
340
13.8
1100
0.75
N/A
12
0
24,000
5,000
6,600
Background
6/08
75
0.21
330
N/A
1100
0.75
310
17
30
>160,000
140,000
>160,000
Samples,
6/15
40
0.20
340
10.9
1100
0.50
130
17
7.5
70,000
70,000
11,000
2006
6/22
40
0.20
310
10.8
1200
0.50
110
12
1.9
1,600,000
900,000
190,000
6/28
35
0.20
340
8.7
1100
0.75
130
17
16
4,000
<2,000
11,000
NSF 10/27/EPADWCTR
The accompanying notice is an integral part of this verification statement.
VS-ii
January 2011
-------
Methods and Procedures
The EUWP verification test was conducted from July 12 to August 16, 2006 by the U.S. Bureau of
Reclamation (USER), with assistance from the U.S Army Tank-Automotive Research, Development, and
Engineering Center (TARDEC). The test was intended to determine if the EUWP could produce 100,000
gpd of finished water meeting the NPDWR from secondary treated wastewater, based on contaminants
found in the source water during the initial water characterization phase of ETV testing (see Table VS-i).
The testing activities followed a test/quality assurance plan (TQAP) prepared for the project. The TQAP
was developed according to the ETV Protocols EPA/NSFProtocol 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 Paniculate Contaminants - September 2005.
The system was shut down for two days (July 24 and 25, 2006) for RO cleaning and for two days (July 30
and 31, 2006) for UF cleaning. An additional RO cleaning was performed from August 7 to August 8,
when the system was down for approximately 24 hours. The system was in operation on 32 calendar days,
which met the test plan goal for collecting operating data for a minimum of 30 days. The system was
operated as continuously as possible. Shut downs occurred each day to perform a pressure decay test on
the UF system, to calibrate sensors, clean the strainers, etc. The RO system also shut down periodically
for various maintenance activities, or when alarms occurred and shut the system down. When alarms and
shutdown occurred during unattended operation at night, the entire system would remain shut down until
an operator arrived in the morning. Turbidity and conductivity were selected as two key parameters.
Turbidity removal by the system would indicate the ability to remove particulate 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. 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. 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.
Once per week samples were collected from the UF and RO process streams for alkalinity, hardness,
sulfate, total silica, dissolved organic carbon (DOC), TDS, total organic carbon (TOC), total suspended
solids (TSS), ultraviolet light absorbance at 254 nanometers (UV254), dissolved metals, total metals, total
and fecal coliforms, Escherichia coli (E. coli), and HPC. Samples were also collected from the UF system
weekly for color, biological oxygen demand (BOD) and chemical oxygen demand (COD).
VERIFICATION OF PERFORMANCE
Finished Water Quality
The UF system reduced turbidity from a mean of 11.1 Nephelometric Turbidity Units (NTU) in the feed
water to a mean of 0.74 NTU in the UF filtrate as measured by the daily grab samples. The 95%
confidence level shows that filtrate turbidity can be expected to be in the range of 0.62 to 0.86 NTU. The
operators manually recorded in-line turbidity measurements at least once per day. The feed water
turbidity, as recorded from the in-line analyzer, showed a mean value of 8.7 NTU. The UF filtrate in-line
analyzer showed a mean turbidity of 0.69 NTU. Statistics for in-line turbidity measurements were not
calculated for the test because the in-line turbidity data for the process streams was inadvertently erased
for the period July 27 through the end of the test.
The RO permeate had a mean turbidity of 0.15 NTU based on the handheld meter readings. The 95%
confidence interval for the handheld meter results showed an expected range of 0.13 to 0.17 NTU for the
RO permeate. The RO permeate turbidity, as manually recorded from the in-line analyzer, had a mean
value of 0.016 NTU.
NSF 10/27/EPADWCTR The accompanying notice is an integral part of this verification statement. January 2011
VS-iii
-------
The UF system was found to have faulty seals, which is discussed in the verification report. This may
explain why the turbidity reductions by the UF system did not meet the NPDWR of <0.3 NTU 95% of the
time. While the UF system alone did not meet the NPDWR, the RO system which followed it sufficiently
reduced the turbidity to the meet NPDWR. The RO permeate turbidity levels manually recorded from the
in-line meter show that the system did meet the NPDWR of <0.3 NTU 95% of the time, with all values
below 1.0 NTU.
A second turbidity requirement is an action level of 0.15 NTU in the EPA Long Term 2 Enhanced
Surface Water Treatment Rule (LT2ESWTR). The rule states that if the in-line turbidity measurement
exceeds 0.15 NTU over any 15-minute period, the system must be shut down and a direct integrity test
performed. Throughout the period for which in-line turbidity data exists (July 12-27), the RO system
produced permeate with turbidity meeting the LT2ESWTR action level criteria. There were a few single
data points that exceeded 0.15 NTU, but never two readings in a row, which would indicate that the
turbidity did not exceed the action level over an entire 15-minute period. All of the manually recorded
turbidity data was 5 to 10 times lower than the 0.15 NTU action level.
The RO system reduced the dissolved ions in the feed water, as measured by conductivity, by a mean of
99.3%. The mean conductivity in the RO permeate was 11 (iS/cm compared to the mean conductivity in
the RO feed water of 1,600 (iS/cm. The direct measurement of TDS shows that the mean concentration in
the RO permeate was <10 mg/L compared to a mean RO feed water level of 1,100 mg/L. The overall
TDS rejection was 99.5%.
The UF system had no impact on the pH of the water with the feed water having a mean pH of 7.53 and
the filtrate having a mean pH of 7.54. The RO system did lower the pH of the permeate. The pH in the
permeate ranged from 5.38 to 7.30 with a mean of 6.27. The UF and RO systems did not have an effect
on the temperature of the water as it passed through the systems.
After RO treatment, the RO permeate met all primary and secondary drinking water standards measured
during the verification test. The RO unit served as an effective treatment system for removing inorganic
and organic constituents present in the secondary wastewater. To be acceptable for transmission or
drinking, the RO permeate would need stabilization and residual chlorination.
UF and RO Membrane Integrity
Daily pressure decay tests were used to document UF membrane integrity. Turbidity, fecal and total
coliforms, E. coll, and HPC were measured in the UF feed and filtrate as indirect membrane integrity
indicators.
During the test audit, representatives from Koch Membrane Systems, Village MarineTec., NSF, and
USER were present to observe the pressure decay test. During that test the filtrate side of the membranes
was drained and both arrays were simultaneously pressurized to 20 pounds per square inch, gauge (psig).
The feed valve and retentate valves were in their operating positions. The filtrate valves were closed.
After 15 minutes the system had lost 1.5 psig. This rate of pressure decline was acceptable to Koch.
As pressure testing continued, it became apparent that the procedure was not giving an accurate test of the
system. After further inspection of the system, USER realized that the check valve on the feed side and
the long run of piping filled with water on the retentate side would not allow air to escape from the system
at 20 psig. In effect the system was completely closed. Opening a sample port on the feed side remedied
this, but also revealed that the system had lost integrity, as was apparent from the turbidity readings and
biological analysis results that had started arriving by this time.
NSF 10/27/EPADWCTR The accompanying notice is an integral part of this verification statement. January 2011
VS-iv
-------
As discussed above, the UF filtrate turbidity was much higher than expected. None of the remedies of
chemically cleaning the system, cleaning the turbidimeter, and recalibration of the turbidimeter solved the
problem. The leakage was so severe that it was believed to be more than broken fibers. However, the
testing schedule had to be maintained, as the City of Gallup needed the space and the EUWP had to be
off-site by the scheduled end of the test period.
Biological analyses were performed for fecal and total coliform, E. coll, and HPC. Virus counts were
measured for one set of UF feed, filtrate, and RO permeate. The enteric virus results showed
176 MPN/100 mL in the RO feed and <1 MPN in the RO permeate. Coliform species were present in the
feed water in great enough numbers to allow for a log reduction value (LRV) greater than 3 from the UF
filtrate to the RO permeate.
Dye-marker direct integrity tests were performed on the RO system at the start and end of the test period.
The RO membranes rejected the dye at a rate higher than 99%. The rejection rate improved at the end of
the test. These results, supported by the high rejection rate for conductivity, the low turbidity in the
permeate, and the 3 LRV for coliform samples, indicate that the RO membranes maintained integrity
throughout the verification test. Although the UF membrane unit had lost integrity, the subsequent RO
array provided a barrier to microorganisms, turbidity and other contaminants.
UF System Operation
UF process operations data for the test are presented in Table VS-ii. The mean UF operating hours during
the verification test was 14 hours per day. The mean RO operating hours during the verification test was
18 hours per day. The UF operating hours were lower than the RO because the system is designed for the
UF to operate at a higher filtrate flow rate than the RO feed rate to keep the RO feed tank full. Whenever
the RO feed rate tank was at maximum level, the UF was automatically shut down until the RO feed tank
level dropped to the pre-set level to restart the UF system. The intake flow is defined as the source water
pumped into the UF feed water tank. The mean UF feed water flow rate of 250 gallons per minute (gpm)
was slightly below the target feed flow rate of 259 gpm specified for the system. The mean filtrate flow
rate of 229 gpm corresponds to a flow rate of 14.3 gpm for each of the 16 UF membrane modules. The
UF water recovery was 91.6% based on the mean feed water and filtrate flow rates.
The UF system flow rate objective was 200,000 gpd for this test. Based on the mean net filtrate
production of 178,000 gpd over the verification period, the UF system did not achieve the objective. The
reason was that the unit did not operate a sufficient number of hours per day to meet the production goal.
At a mean filtrate flow rate of 229 gpm, and accounting for a backwash volume of 900 gallons every 30
minutes, the UF system would need to operate an average of 17 hours per day to meet the objective. The
UF system operated an average of only 14 hours of per day during the test.
Table VS-ii. UF Operations Productivity Data
Standard 95% Confidence
Parameter Count Mean Median Minimum Maximum Deviation Interval
UF operation (hr/day) 30 14 15 4 20 4.1 +1.5
Intake flow (gpm) 53 281 288 217 301 21.0 +5.65
Feedflow(gpm) 53 250 251 179 314 24.3 +6.55
Filtrate flow (gpm) 53 229 229 154 289 25.0 +6.74
Retentate flow (gpm) 49 24 25 19 30 4.4 +1.2
Backwash flow (gpm) Not measured. 900 gallons per backwash cycle(1); Backwash every 30 minutes
Feed pressure (psig) 53 22 21 16 30 3.9 +1.1
Retentate pressure (psig) 53 19 19 0 28 5.4 +1.5
Filtrate temperature (°F) 54 78 78 76 82 L5 +0.4
(1) Volume not measured. It was provided by the manufacturer.
NSF 10/27/EPADWCTR The accompanying notice is an integral part of this verification statement. January 2011
VS-v
-------
RO System Operation
The RO process operations data are presented in Table VS-iii. The RO system did not achieve the
permeate production of 100,000 gpd claimed in the statement of performance. The mean permeate
production for the 32 calendar days of operation was 78,000 gpd. The mean feed water flows of 107 gpm
for Array 1 and 41 gpm for Array 2 were below the target feed rates established in the test plan (Array 1
target 116 gpm and Array 2 target was 58 gpm). The percent recovery for Array 1 of 50% equaled the
target specification of 50%. The Array 2 percent recovery of 42% was below the target specification of
48%. These recoveries, with the feed water flows, resulted in mean permeate flow rates of 53 gpm for
Array 1 and 17 gpm for Array 2. At these flow rates, the RO unit would need to operate an average of
approximately 24 hours per day to meet the target of 100,000 gpd. The RO unit averaged 18 hours per
day of operation during the test.
It was apparent during the test that the UF treated secondary wastewater was putting a heavier load on the
RO than expected. For this type of application, lower percent recoveries and lower flows were achieved
compared to design specifications for groundwater and seawater. During the last few days of testing the
recovery was set to 40% to protect the system from heavy loading from the WWTP. While this may not
have been necessary, it explains the drop in flows and pressure near the end of the test.
It should be noted that while the RO only achieved approximately 78% of the performance objective for
permeate production, additional operating time each day would have increased the total production. As
noted in the UF system discussion, operators were only present during daylight hours and there was no
coverage over night. Therefore, if an alarm sounded and shutdown the unit, the system remained off-line
until an operator arrived the next morning. While it may not be realistic to operate the RO unit
continuously 24 hours per day for several days, additional operator coverage could increase operating
hours and achieve permeate production closer to the target.
Table VS-iii. RO System Operations Productivity Data
95%
Standard Confidence
Parameter Count Mean Median Minimum Maximum Deviation Interval
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)
54
54
54
54
54
54
54
53
54
54
54
107
53
54
41
17
24
290
197
193
138
20
107
55
53
41
18
23
293
199
195
138
19
104
42
43
32
11
20
222
134
133
91
9
110
64
67
48
22
29
366
263
261
182
42
1.29
5.44
5.52
4.14
2.74
1.70
26.1
24.8
23.6
19.0
5.82
±0.34
±1.45
±1.47
±1.10
±0.73
±0.45
±6.96
±6.67
±6.28
±5.06
+1.55
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.
In-line field meters for particle counts were factory calibrated and certificates were provided as required
in the TQAP. However, incorrect calibration certificate data for bin voltages was entered into the software
program for the particle counters. This resulted in rendering the particle count data inaccurate and not
NSF 10/27/EPADWCTR The accompanying notice is an integral part of this verification statement. January 2011
VS-vi
-------
meeting the Data Quality Objectives. Because of this problem, particle count data could not be used for
documenting system performance for particle count and the data are not included in this report.
Samples were collected for Cryptosporidium and Giardia enumeration, but the analyses did not meet the
QA/QC objectives for the ETV test. Therefore, these data are not included in the verification report.
A complete description of the QA/QC procedures is provided in the verification report.
Original signed by Sally Gutierrez 01/31/11 Original signed by Robert Ferguson 01/17/11
Sally Gutierrez Date Robert Ferguson Date
Director Vice President
National Risk Management Research Laboratory Water Systems
Office of Research and Development NSF International
United States Environmental Protection Agency
NOTICE: Verifications are based on 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 10/27/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 10/27/EPADWCTR The accompanying notice is an integral part of this verification statement. January 2011
VS-vii
-------
This page is intentionally blank
-------
January 2011
Environmental Technology Verification Report
Removal of Inorganic, Microbial, and Particulate Contaminants from
Secondary Treated Wastewater
Village Marine Tec.
Expeditionary Unit Water Purifier, Generation 1
Prepared by:
Michelle Chapman, United Stated Bureau of Reclamation, Denver, CO 80225
Dale Scherger, Scherger Associates, Ann Arbor, MI 48105
Michael Blumenstein and C. Bruce Bartley NSF International, Ann Arbor, MI 48105
Jeffrey Q. Adams, Project Officer, U.S. Environmental Protection Agency, Cincinnati, OH
45268
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
List of Figures vii
List of Tables viii
Appendices ix
Abbreviations and Acronyms x
Acknowledgements xiii
Chapter 1 Introduction 14
1.1 ETV Purpose and Program Operation 14
1.2 Testing Participants and Responsibilities 14
1.2.1 EPA 15
1.2.2 NSF International 15
1.2.3 ONR 16
1.2.4 TARDEC 16
1.2.5 USER 16
1.2.6 Village Marine Tec 17
1.2.7 The City of Gallup, New Mexico 17
1.3 Verification Testing Site 17
Chapter 2 Equipment Capabilities and Description 22
2.1 Equipment Capabilities 22
2.2 General System Description 23
2.3 Concept of Treatment Processes 25
2.3.1 UF Pretreatment/Suspended Solids Filtration 25
2.3.2 RO Desalination 26
2.4 Detailed System Description 26
2.4.1 Raw Water Intake 29
2.4.2 UF System Description 29
2.4.2.1 UF System Operation 31
2.4.2.2 UF Cleaning Procedure 31
2.4.3 RO System 34
2.4.3.1 RO skid statistics 37
2.4.3.2 RO System Operation 37
2.4.3.3 RO Cleaning Procedure 38
2.4.3.4 Pressure Exchanger 41
2.5 General Requirements and Limitations 42
2.6 Waste Generation and Permits 44
2.6.1 UFandROCIP 44
2.6.2 RO Concentrate 45
2.6.3 UF Backwash and Retentate 45
2.6.4 Discharge Permits 45
2.7 Discussion of the Operator Requirements 45
Chapters Methods and Procedures 47
iv
-------
3.1 Quantitative and Qualitative Evaluation Criteria 47
3.2 Key Treated Water Quality and Operational Parameters 47
3.3 Operations and Maintenance 52
3.4 Field Operations 52
3.5 Overview of ETV Testing Plan 52
3.5.1 Task A: Characterization of Feed Water 53
3.5.2 Task B: Equipment Installation, Initial Test Runs and System Integrity Tests.... 53
3.5.3 TaskC: Verification Test 53
3.5.3.1 Task Cl: Membrane Flux and Recovery 53
3.5.3.2 TaskC2: Cleaning Efficiency 53
3.5.3.3 TaskC3: Finished Water Quality 54
3.5.3.4 TaskC4: Membrane Module Integrity 54
3.5.3.5 Task C5: Data Handling Protocol 54
3.5.3.6 Task C6: Quality Assurance and Quality Control 54
3.6 Task A: Characterization ofFeed Water 54
3.7 Task B: Equipment Installation, Initial Test Runs, and Initial System Integrity Tests.. 54
3.8 Task C: Verification Testing 54
3.8.1 Task Cl: Membrane Flux and Operation 54
3.8.1.1 Work Plan 55
3.8.1.2 Evaluation Criteria 55
3.8.1.3 Equations 56
3.8.2 TaskC2: Cleaning Efficiency 60
3.8.2.1 Work Plan 60
3.8.2.2 Evaluation Criteria 61
3.8.3 TaskC3: Finished Water Quality 61
3.8.3.1 Work Plan 61
3.8.3.2 Evaluation Criteria 61
3.8.4 Task C4: Membrane Integrity Testing 61
3.8.4.1 Direct Integrity Testing 62
3.8.4.2 Continuous Indirect Integrity Monitoring 67
3.8.5 TaskCS: Data Handling Protocol 67
3.8.5.1 Work Plan 67
3.8.6 Task C6: Quality Assurance Project PI an 67
3.8.6.1 Experimental Objectives 68
3.8.6.2 Work Plan 68
3.8.6.3 QA/QC Verifications 68
3.8.6.4 Data Correctness 68
3.8.6.5 Operation and Maintenance 76
Chapter 4 Results and Discussion 77
4.1 Introduction 77
4.2 Equipment Installation, Start-up, and Shakedown 77
4.3 Task A: Raw Water Characterization 77
4.4 Task B Initial Test Runs 83
4.5 TaskC: Verification Test 84
4.5.1 Task Cl: Membrane Flux and Operation 84
4.5.1.1 UF Operating Data 85
-------
4.5.1.2 RO System Operational Data 90
4.5.1.3 Power Requirements and Efficiency 95
4.5.2 TaskC2: Cleaning Efficiency 97
4.5.2.1 UF Backwash and Cleaning Frequency and Performance 97
4.5.2.2 RO Cleaning Frequency and Performance 99
4.5.3 TaskC3: Finished Water Quality 101
4.5.3.1 Water Quality Results - Turbidity, Conductivity, pH, and Temperature... 101
4.5.3.2 Other Water Quality Results - UF System 109
4.5.3.3 Other Water Quality Results - RO System 114
4.5.4 Task C4: Membrane Integrity Testing 118
4.5.4.1 UF System - Pressure Hold Test 118
4.5.4.2 RO System - Dye Challenge 119
4.5.4.3 Continuous Indirect Integrity Monitoring 120
4.5.5 TaskC6: Qualitative Evaluations 121
4.5.5.1 Reliability or Susceptibility to Environmental Conditions 121
4.5.5.2 Equipment Safety 122
4.5.5.3 Effect of Operator Knowledge, Skill, and Experience on Results 123
4.5.5.4 Effect of Operator's Technical Knowledge on System Performance and
Robustness of Operation 125
4.5.5.5 Ease of Equipment Operation 125
4.5.5.6 Waste Discharge Requirements 126
4.6 QA/QC 126
4.6.1 Introduction 126
4.6.2 Documentation 126
4.6.3 Quality Audits 126
4.6.4 Test QA/QC Activities 127
4.6.5 Sample Handling 128
4.6.6 Physical and Chemical Analytical Methods QA/QC 128
4.6.6.1 Field Sample Analysis 128
4.6.6.2 Laboratory Methods 129
4.6.7 Documentation 129
4.6.8 Data Review 130
4.6.9 Data Quality Indicators 130
4.6.9.1 Representativeness 130
4.6.9.2 Accuracy 131
4.6.9.3 Precision 131
4.6.9.4 Completeness 131
4.7 References 133
VI
-------
List of Figures
Figure 1-1. Existing WWTP and EUWP location 19
Figure 1 -2. Gallup Wastewater Treatment Plant with EUWP indicated by the circle 20
Figure 1-3. General Layout of EUWP Equipment 21
Figure 2-1. Process component diagram 24
Figure 2-2. Koch UF hollow fiber modules, a single fiber, and the process flow through the
module 25
Figure 2-3. EUWP system process schematic 27
Figure 2-4. Schematic of typical EUWP layout 28
Figure 2-5. Photo of the UF skid 30
Figure 2-6. Piping and instrumentation diagram of UF skid 33
Figure 2-8. Photo of the RO skid membrane vessels 35
Figure 2-9. Vessel arrangement schematic 35
Figure 2-10. Membrane arrangement schematic 36
Figure 2-11. P&ID of RO skid 40
Figure 2-12. PX pressure exchanger 41
Figure 4-1. Plot of UF system flow rates throughout the testing period 86
Figure 4-2. UF system filtrate production through the testing period 86
Figure 4-3. Plot of UF system feed and retentate pressures over the testing period 87
Figure 4-4. Plot of UF system TMP over testing period 87
Figure 4-5. UF system specific flux over testing period 88
Figure 4-6. Loss of specific flux overtime 89
Figure 4-7. UF System Backwash Analysis 90
Figure 4-8. RO system flow rates 92
Figure 4-9. RO system operating pressures 92
Figure 4-10. RO system percent recoveries 93
Figure 4-11. RO system permeate production and feed water volume 93
Figure 4-12. RO system specific flux 94
Figure 4-13. RO and UF power consumption overtime 96
Figure 4-14. RO and UF power requirements per kgal of RO permeate 96
Figure 4-15. RO system energy efficiency calculated from BLIP, WFtP, and energy recovery
based on total feed flow compared to the overall water recovery 97
Figure 4-16. UF feed water turbidity 101
Figure 4-17. UF filtrate and RO permeate turbidity handheld meter 102
Figure 4-18. UF feed and UF filtrate/RO feed turbidity in-line meter 103
Figure 4-19. UF filtrate andRO permeate in-line turbidity readings 104
Figure 4-20. UF filtrate connector and leaking end cap 119
Figure 4-21. Relation between QA and actual productivity 125
vn
-------
List of Tables
Table 2-1. Koch Membrane Systems Targa 10-48-3 5-PMC Cartridge Specifications 29
Table 2-3. RO System Membrane Element Characteristics 36
Table 2-4. RO Skid Statistics 37
Table 2-5. EUWP Site Considerations and Dimensions 42
Table 2-6. Equipment Limitations 43
Table 2-7. Membrane Limitations 44
Table 3-1. Raw Water Quality Sampling Schedule and Analysis Locations 48
Table 3-2. Unregulated Organic Chemicals of Concern Analyzed by Colorado School of Mines
51
Table 3-3. Water Quality and Operational Parameters Measured Online 55
Table 3-4. Key Operating Parameters 55
Table 3-5. Operational Data Plots Appearing in Chapter 4 56
Table 3-6. Properties of FWT Red 25 Liquid Powder Dye 65
Table 3-7. On-Site Analytical Equipment QA Activities 69
Table 3-8. On-Site Data Generation QC Activities 69
Table 3-9. Water Sampling Locations for Water Quality Samples 69
Table 3-10. Analytical Methods for Laboratory Analyses 72
Table 3-12. Completeness Requirements 76
Table 4-1. Background Water Analyses -Severn Trent Results/CSM: General Chemistry 78
Table 4-2. Background Water Analyses -Severn Trent: Dissolved Metals 79
Table 4-3. Background Water Analyses - Severn Trent: Total Metals 79
Table 4-4. Background Water Analyses - Severn Trent: Volatile Organic Compounds 80
Table 4-5. Background Water Analyses - Severn Trent: Semi-Volatile Organic Compounds .. 80
Table 4-6. Background Water Analysis - Week and ACZ 81
Table 4-7. Background Water Analyses - Anatek Labs: Disifection Biproducts and Pesticides 81
Table 4-8. Background Biological Analysis 81
Table 4-9. Background Water Analyses - Colorado School of Mines: Wastewater Contaminants
of Concern 82
Table 4-10. UF Full System Integrity Test Results, July 12,2006 83
Table4-ll. UF Operational Data Statistics 85
Table 4-12. RO System Operational Measurement Statistics 91
Table 4-13. UF System Performance Parameter Values at Key Intervals 98
Table 4-14. Change in UF Performance with Cause and Action Taken 99
Table 4-15. RO System Performance Intervals 100
Table4-16. ROandUF System Cleanings 100
Table 4-17. Summary Statistics for Handheld Turbidity Meter Results 102
Table 4-18. Summary Statistics for In-line Turbidity Meter Manually Recorded Results 104
Table 4-19. Conductivity Results 105
Table 4-20. pH Results 107
Table 4-21. Temperature Results 108
Table 4-22. UF Feed and Filtrate General Water Quality Analysis Ill
Table4-23. Biological Analysis of UF System 113
Table 4-24. UF Retentate and Backwash Analysis 113
Table 4-25. RO Feed, Permeate, and Concentrate - General Chemistry 115
Vlll
-------
Table 4-26. Rejection of Analytes in the RO Feed 117
Table4-27. RO System Mass Balance 117
Table 4-28. Biological Analyses of RO Process Streams 118
Table 4-29. RO Permeate Absorbance after Injection 120
Table4-30. Evaluation of Skill, QA, and Productivity 124
Appendices
Appendix A - Operation and Maintenance Manual
Appendix B - Field Logbooks, Field Log Sheets, Field Calibration Records
Appendix C - Week Laboratory Data Reports
Appendix D - Severn Trent Laboratory Data Reports
Appendix E - Anatek Laboratory Data Reports
Appendix F - CSM Laboratory Data Reports
Appendix G - BioVir Laboratory Data Reports
Appendix H - University of Arizona Laboratory Data Reports
Appendix I - AC Laboratory Data Reports
Appendix J - WaterEye Data
Appendix K - Turbidity Data
Appendix L - Spreadsheets
Appendix M - QA/QC Data Tables
IX
-------
Abbreviations and Acronyms
ANGB Air National Guard Base
bhp brake horsepower
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
DOC dissolved organic carbon
DQO data quality objectives
DWS Drinking Water Systems
EPA United States Environmental Protection Agency
ERI Energy Recovery, Inc.
ETV Environmental Technology Verification
EUWP Expeditionary Unit Water Purifier
°F degrees Fahrenheit
FRP fiberglass reinforced plastic
ft foot (feet)
gal gallons
gfd gallons per square foot per day
g/mol grams per mole
gpd gallons per day
gpm gallons per minute
h hour
FIPC Heterotrophic plate count
in inch
JP8 j et propellent 8 (j et fuel)
kgal kilogallon
kW kilowatt
kWh kilowatt hour
L liter
Ibs pounds
LLD lower limit of detection
LT2ESWTR Long Term 2 Enhanced Surface Water Treatment Rule
m meter
MDL method detection limit
mg milligram
mL milliliter
MPN most probable number
mS milliSiemens
MWCO molecular weight cutoff
NA not applicable
NBC nuclear, biological, and chemical
-------
NC not calculated
ND non-detect
NDP net driving pressure
NFESC Naval Facilities Engineering Service Center
NIST National Institute of Standards and Technology
NM not measured
nm nanometers
NPDWR National Primary Drinking Water Regulations
NR not recorded
NSF NSF International (previously known as the National Sanitation Foundation)
NSWCCD United States Naval Surface Warfare Center - Carderock Division
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
ORP oxidation reduction potential
pCi picocuries
PE performance evaluation
P&ID piping and instrumentation diagram
ppb parts per billion
ppm parts per million
ppt parts per trillion
PQL practical quantification limit
psi pounds per square inch
psi pounds per square inch, gauge
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 United States Army Tank-Automotive Research, Development, and Engineering
Center
TDS total dissolved solids
TOC total organic carbon
TON total organic nitrogen
TQAP test/quality assurance plan
TQG Tactical Quiet Generator
TSS total suspended solids
TMP trans-membrane pressure
XI
-------
UF ultrafiltration
USER United States Bureau of Reclamation
UV254 ultra violet absorbance at 254 nanometers
VOC volatile organic chemicals
VSS volatile suspended solids
whp water horsepower
jig micrograms
|im micron
xn
-------
Acknowledgements
The U.S. Bureau of Reclamation (USER) was the main field testing organization. USER was
supported by the U.S. Army Tank-Automotive Research, Development, and Engineering Center
(TARDEC). USER and TARDEC 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 Michelle Chapman of USER, Dale Scherger of Scherger
Associates (3017 Rumsey Drive, Ann Arbor, MI 48105), and Michael Blumenstein and C. Bruce
Bartley of the NSF International ETV Drinking Water Systems Center (DWSC). The
verification report was based on the project test/quality assurance plan authored by DWSC,
USER, and TARDEC.
The authors would like to thank Jeff Adams of the EPA Water Quality Division for his assistance
in the ETV process.
The manufacturer of the EUWP was:
Village Marine Tec.
2000 W. 13 5th St.
Gardena, CA 90249
Phone: 310-516-9911
The USER project managers were Ms. Michelle Chapman and Mr. Steve Dundorf. The
engineers responsible for the daily operations of the field test were Saied Delagah, Vanessa
Aguayo, Yuliana Poras, Erik Jorgansen, Andrew Tiffenbach, Ken Yokoyama, Katherine Benko,
and Dan Gonzales. John Walp was instrumental in set up and demobilization. Susan Martella
and Katherine Benko are to be credited for drafting the test plan.
Mark Miller of Naval Facilities Engineering Service Center was very helpful with operational
advice based on his experience with the EUWP Gen 1-2.
USER would like to thank the City of Gallup Mayor Bob Rosebrough and Lance Allgood,
Executive Director of Gallup Joint Utilities for making this project possible. The following
Gallup Water and Wastewater employees were instrumental in helping with site work and
sampling: Gary Munn, Ernest Thompson, Pat Sanchez, Sam Koike, Arcenio Chavez Jr., Herbert
Guillen, Tobias Sandoval, Diann Bowie, Edward Richard Apodaca, Dave A. Guadagnoli, Gregg
Valtierra, and Charles de la Torre.
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.
Xlll
-------
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
(DWS) Center 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 DWS Center 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
secondary wastewater effluent from the Gallup, New Mexico wastewater treatment plant
(WWTP) as the source water for the test.
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
14
-------
(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.
• City of Gallup, New Mexico
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 Agreement No. R-82833301. This verification effort
was supported by the DWS Center 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
15
-------
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
project.
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
The U.S. Army TARDEC provided oversight of EUWP design, construction, and testing.
Contact Information:
US 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 was the FTO for the ETV test and 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. 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
16
-------
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
1.2.7 The City of Gallup, New Mexico
City of Gallup provided a portion of the funding for the wastewater testing phase, the testing
location, and operational assistance. The funding was provided by the City of Gallup through a
grant from the state of New Mexico.
Contact Information:
Lance Allgood
Phone: 505-863-1289
Fax: 505-726-1278
Email: lallgood@ci.gallup.nm.us
1.3 Verification Testing Site
The ETV test of the EUWP Generation 1-1 was performed at the City of Gallup WWTP at 800
Sweetwater Place, Gallup, New Mexico. Gallup is located 140 miles west of Albuquerque.
Gallup is on a high desert plateau at 6,500 ft. The climate is temperate (average low of 16°F,
average high of 87°F) and dry (11 in. of rain per year). In 2005, the population of Gallup was
approximately 20,000 residents. The water chemistry data is presented in Section 4.3.
The City of Gallup WWTP provided secondary treated wastewater for the ETV test. During
most of the operation the secondary effluent is filtered through pressure media filters prior to
chlorination providing tertiary treated effluent. Figure 1-1 presents a flow diagram of the
WWTP and the location of the EUWP Generation 1-1 equipment.
The WWTP treats an average of 3 million gallons per day (MGD) of wastewater with a peak of
5.5 MGD in the summer. The facility was in the midst of expansion to 3.5 MGD during ETV
testing. The WWTP has enough hydraulic capacity to handle the peak loads without significant
change in effluent water quality. Wastewater is discharged into the Rio Puerco of the West, a
dry riverbed.
17
-------
The EUWP Generation 1-1 was situated on a gravel pad at the north side of the facility as shown
in the photograph provided in Figure 1-2. Rio Puerco of the West runs along the north end of the
facility (on left side in Figure 1-2). All water produced during the verification test, including
clean-in-place (CIP) waste, was discharged to the oxidation ditch at the head of the WWTP.
Chemical waste from CIP operations was neutralized prior to discharge to the oxidation ditch.
Figure 1-3 provides a diagram of the general layout of the major components of the EUWP
Generation 1-1 equipment.
18
-------
Raw Influent
Raw Water Pump
Mechanical Bar Screens
Grit Removal
Primary Clarifier
Aeration Basin
Oxidation
Ditch
Secondary
Clarifier
To Rio Puerco
of the West
Chlorine Contact
Basin
EUWP
Pressure Sand
Filter
Chlorine
Contactor
i—I To Golf Course
I—' & Soccer Field
Figure 1-1. Existing WWTP and EUWP location.
19
-------
EUWP Source
Water
All effluent returned to oxidation
ditch at head of WWTP
Figure 1-2. Gallup Wastewater Treatment Plant with EUWP indicated by the circle.
20
-------
TK13
PRODUCT WATER
STORAGE TANK
20.000 GAL
27'-6"x23'-6"
TK2
UF FEED
TANK
3000 GAL
11'-10"
REVERSE
OSMOSIS
SKID
19-104x8-0
ULTRA-FILTRATION
SKID
19'-10j"x8'-0"
STORAGE TRAILER
19'-10}"x8'-0"
TKb
CHEMICAL
WASTE
UF BACK
FLUSH & RO
FEED TANK
3000 GAL
STORAGE TANK
3000 CAL
1T-10"
RO PRODUCT
WATER OUTLET
RO REJECT
WATER OUTLET
UF RAW
WATER INLET
UF REJECT
WATER OUTLET
UF+RO MODE
1ST/2ND PASS
3CAL£ OT FETT
Figure 1-3. General layout of EUWP equipment.
21
-------
Chapter 2
Equipment Capabilities and Description
The EUWP was designed to meet purified water needs in areas with challenging water sources of
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 gallons 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 second 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 while removing as much as
99.7% of dissolved salts and meeting EPA 's National Primary Drinking Water
Regulations (NPDWR) from secondary treated wastewater based on contaminants
found in the source water during the initial water characterization phase of ETV
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
22
-------
Field Water Quality Standards for short-term consumption by healthy adults. However, the
technology used is capable of exceeding the EPA NPDWR.
According to the system designers, 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 TDS 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 the Gallup WWTP.
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.
23
-------
UF Product Tank
RO Skid
—©•
200 urn
UFSkid
Feed Tank
-©—*•
200 urn
UF Hollow Fiber
Membranes
2nd Pass
RO Product Tank
-©—+•
Figure 2-1. Process component diagram.
24
-------
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 psi) membrane process that
separates particulates based on size exclusion. The UF process retains oils, particulate 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.
25
-------
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.
Osmotic pressure can be estimated by the following equation:
x=inRT (2-1)
where:
n = osmotic pressure;
/' = e dissociation constant;
n = number of moles of ions;
R = Universal Gas Constant; and
T = temperature in degrees Kelvin.
A simpler approximation is 1 psi per 100 mg/L TDS.
RO is a moderate to high-pressure (80-1,200 psi) membrane separation process. The membranes
in the EUWP are spiral wound with up to seven modules in a vessel. 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
The system process schematic and detailed layout are shown in Figures 2-3 and 2-4, respectively.
26
-------
Concentrate/
Waste
Figure 2-3. EUWP system process schematic.
27
-------
Figure 2-4. Schematic of typical EUWP layout.
28
-------
2.4.1 Raw Water Intake
The intake strainer was buoyed behind a weir before the WWTP chlorination system and
detention tank. The intake pump draws water from this source to the UF skid where ferric
chloride is injected as a filter aid before the dual 200 jim Amiad strainers. The strainers aid in
mixing the coagulant since there is not enough time to form particles larger than 200 |im
between the injection point and the strainer. The 3,000 gal UF feed tank provides 12 min of
retention time. Ferric chloride is dosed to create a micro-floe under these conditions. Ferric
chloride was used during the ETV test at a dose rate of 5 mg/L as Fe, as described in Section 4.4.
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. A photo of the UF
skid is shown in Figure 2-5. The onion tank in the right foreground of the photograph is for the
UF feed source after the ferric chloride addition and strainer.
Table 2-1. Koch Membrane Systems Targa 10-48-35-PMC Cartridge Specifications
Parameter
Value
Nominal Molecular Weight Cut-off
Max. Recommended Flow (per cartridge)
Maximum Pressure
Maximum Transmembrane Pressure (TMP)
Maximum Backflush TMP
Inner Fiber Diameter
Membrane Area
Cartridge Diameter
Cartridge Length
Maximum Free Chlorine at 25°C
Maximum Total Cholerine Contact
100,000
32.2 gpm
45 psi
30psi
20 psi
0.035 in
554 ft2
10.75 in
48 in
200 mg/1 at 9.5 pH
200,000 ppm hrs cumulative
29
-------
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 psi
30psi
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 kWhr/kilogallon (kgal)
DF2 (Diesel Fuel, Grade 2)
DFA (Diesel Fuel, Arctic Grade)
JP8 (Kerosene type, military jet fuel)
43 gal
Figure 2-5. Photo of the UF skid.
30
-------
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 2-6 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. Ferric chloride was used during the ETV test at a dose rate of 5 mg/L as Fe, as
described in Section 4.4.
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.
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 trans-membrane pressure (TMP) exceeds 35 psi 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 the Gallup WWTP. The following is a basic description of the flow path and functional
description of the UF CIP 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 PLC will automatically
move the pneumatically operated valves to the correct positions.
31
-------
6. Enable heaters to maintain CIP solution to between 96 - 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.
32
-------
Figure 2-6. Piping and instrumentation diagram of UF skid.
33
-------
2.4.3 RO System
The RO skid is shown below in Figures 2-7 and 2-8.
Figure 2-7. Photo of the RO skid.
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 2-9). 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 2-9). 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 2-9).
The second pass RO system consists of a 2—4 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 2-9). The brine from these vessels then feeds one additional four-element vessel
(Vessel 9 in Figure 2-9).
34
-------
Figure 2-8. Photo of the RO skid membrane vessels.
5678
Concentrate
•*! 1 I 2 | 3
5678
1 I 2 | 3 | 4
5678
1 I 2 I 3 I 4
1234
1 234
Numbers indicate pressure vessels
Figure 2-9. Vessel arrangement schematic.
35
-------
The RO design incorporates an internally staged RO element configuration on the first pass
(Figure 2-10). 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.
o
o
0
©
1 *| X1 | X1 | X2 | X2 | — >\ X3 X3
L___________
*| X1 | X1 | X2 | X2 | — >\ X3 X3
l______________
_
^(CPXJ7*I X1 | X1 | X2 | X2 | — >\ X3 X3
^— "^ l______________
X3 X3 |— *•
^^^^^^^^^^jl^^^^^
X3 X3 |— *•
_______J__*
X3 X3 |— *•
_______J__^
X4 X4 X4 X4
X4 X4 X4
X4
Numbers indicate pressure vessels
Figure 2-10. 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 Flow
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
36
-------
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 and 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 2-11 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
pressure pump #6 (P6) supplies pressure for the 1st pass 1st and 2n arrays and the
pressure exchanger #8 (P8) supplies pressure for the 1st pass 3rd array.
37
-------
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
psi 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 water 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
38
-------
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 min.
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.
39
-------
CP4
CHEMICAL
PUMP
SETFOINT
5 PS DECR PB
P6 OFF RO 1ST PASS
HP PUMP SETPOINT
1300 PSI
X——-T »_J_j? H_XI_L_"_
x—=t_g-I!-jr_- L_»_r_!g
T BRINE X 4TX3 "[ X3 f" X3 ]" H "II
GPM: 20-3T I Jl 1 1 J H
PV7
|
pva
1^=-
V59
1ST PASS PERMEATE GPM: 64-102 j-^
2" "" II
X4 ! X4 X4 X4 jf X PERMEATE Vl/ VV J,
PSI: 148-189
2ND PASS FROM P7 -^ ' ' 1
V
p
R
h
X4 X4
316 SS GRP
RO PRODUCT
WATER OUTLET TO
PRODUCT WATER
CHLOfflNATION INLET
GPM:
64-102 (1ST PASS ONLY)
- (W/ 2ND PASS)
50-69
UF+RO 1ST/2ND PASS MODE
, , SETPOINT
[«) 1 PSI INCR
MOV1 DIVERT
tt25 PSI DECR
P9 OFF
Figure 2-11. P&ID of RO skid.
40
-------
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 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 2-12). 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.
In a typical system, the pressurized feed water 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).
Feedwater | | Reject fluid | Liquid piston
Rotation
High-pressure feed water
going to 2nd parallel 1st pass
High-pressure concentrate or
reject fluid from reverse osmosis membranes
B
High pressure
Sealed area
Low pressure
H
Low-pressure feedwater
inlet from brackish supply
pump
PX—Pressure Exchanger™
Low-pressure concentrate or
reject flu id to drain
http://www.energv-recoverv.com/pdf/PX45SPX70SPX90S.pdf
Figure 2-12. PX pressure exchanger.
41
-------
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 for 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 2nd 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.
42
-------
Table 2-6. Equipment Limitations
System
Inlet Pump #1
Strainer
Parameter
Suction head (maximum)
Differential pressure (maximum) before manual
25ft
7 psi
Value
UF
UF Membranes
UF CIP Water
RO
backwash required
Backpressure required for strainer auto flushing
Pretreatment requirements
Feed pressure (maximum)
Ambient temperature range
Water temperature range
Control air pressure
Damaging chemicals
TMP following back flush (maximum) before CIP
required
Pressure surges
Stagnation time (maximum) before preservation
required with 1,000 - 5,000 mg/L sodium bisulfite
(see operations manual for details)
See Table 2.6 for more details.
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 paniculate matter such as
rust, scale, flake sandy, granular material, slurries,
scum, algae, etc.
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)
35 psi
200 um strainer
45 psi
32 - 120°F
34 _104°F
60 psi
Grease, Oil, Silicon
35 psi
Must be minimized by closing and
opening valves slowly
14 days (somewhat temperature
dependent)
<1.0NTU
O.05 mg/L
O.05 mg/L
<0.5 mg/L
ND(1)
ND
<10 mg/L
< saturated at 50°C (122°F)
< saturated
no living or dead material
<3.0
1.5-13
45 psi
15 psi
32°F to 120°F
500 um prior to entering the UF
system
34 - 104°F
5 (membrane dependent)
32°F to 120°F
32°F to 120°F
3% to 95%
- UF treatment
- 200 um strainer on RO skid
200 psi
100 psi
43
-------
Table 2-6 (cont'd). Equipment Limitations
System
Parameter
Value
RO
RO Membranes
2n 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 with 1% sodium bisulfite (see operations
manual for details)
(see Table 2-6 for details)
300 psi
600 revolutions per minute (RPM)
30 psi
1 week (somewhat temperature
dependent)
(1) Non-detect
Table 2-7. Membrane Limitations
8.
I
i
o
-*^
o
eS
S
cS
s
Membrane
a
a.
2 5
a &
S s
w P
M i/5
'•Z
_o
a
U
I
I
i
I
I
O <—
C S
•a a
M
S a
% c
2 S
PH "w
a •=
I 13
•a
o
H
%
sS
fe
sS
CO
TARGA® 10 - 48 - 35 - PMC
FILMTEC™ SW30HRLE-400
FILMTEC™ SW30
XLE-400
FILMTEC™ SW30HR
-12000 (experimental)
AquaPro LE-8040UP(2)
45 104
1,000(1) 113
1,200 113
1,200
600
113
113
200
0.1
O.I
0.1
ND
30
2-11
2-11
2-11
2-11
1-12
1-12
1-12
1-12
15
15
15
20
60
(1) May go up to 1,200 psi 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 and RO CIP
A CIP cycle for the UF or RO system involves filling the 300 gal CIP tank with an acid cleaning
solution followed by a basic solution with chlorine. Each cleaning cycle is followed by a rinse
cycle from the UF filtrate tank. A second base cleaning may be required. The total volume
generated with two cleaning steps using the 300 gal tank plus 200 gal piping/membrane volume
44
-------
each followed by a thorough rinse is approximately 2,500 gal for a total waste volume of
12,500 gal for the whole ETV test. All CIP waste was returned to the head of the WWTP.
2.6.2 RO Concentrate
All product water and concentrate was recombined and returned to the head of the WWTP. A
total of 5.8 million gal was processed and returned to the head of the WWTP during the test.
2.6.3 UF Backwash and Retentate
The UF system on the EUWP automatically initiates a backwash every 30 min to remove
contaminants from the membrane surface. Each backwash consists of backflushing the
membrane with UF filtrate for 60 sec followed by a forward flush using UF feed water to remove
the contaminants dislodged from the membranes during the backflush. In addition to the
backwash, the UF system also discharges a continuous stream referred to as retentate. All
backwash and retentate was returned to the head of the WWTP and is included in the total of 5.8
million gal processed during the test period.
2.6.4 Discharge Permits
The City of Gallup has an NPDES permit for the discharge of their wastewater. The permit
limits the increase in TDS between their source water and their discharge to 500 mg/L. If the
product and concentrate flow streams are combined for discharge, the permit requirements are
not exceeded. All product water and concentrate was recombined and returned to the head of the
WWTP.
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.
45
-------
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.
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™. 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.
46
-------
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; and
• Labor hours.
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 on the EPA regulations, and
other water quality parameters of interest. Regulated contaminants and unregulated troublesome
components analyzed are listed in Table 3-1, which also lists the laboratory responsible for
analysis. Contaminants such as pharmaceuticals and hormones have recently emerged as a
concern with municipal wastewater. Pharmaceuticals and hormones compounds monitored are
listed in Table 3-2. They pose at least a public perception problem and at most a health problem,
and deserve consideration for removal. In addition to the analytical water quality parameters, the
Langelier Saturation Index (LSI) was calculated for the RO feed and permeate according to
ASTM Method D 3739.
47
-------
Table 3-1. Raw Water Quality Sampling Schedule and Analysis Locations
Water Quality
Parameter
Type
ted Constituents
1
13
1
1
Dissolved
Gases
1
5
1
•3
I
I
i
•d
1
a
i ^^
•d g
j£JL
i i,
Parameter
pH
Temperature
Conductivity
ORP
Dissolved Organic Carbon (DOC)
Total Organic Carbon (TOC)
UV254
Color
Odor
Biochemical Oxygen Demand (BOD)
Chemical Oxygen Demand (COD)
Total Suspended Solids (TSS)
TDS
Volatile Suspended Solids (VSS)
Alkalinity (total) [as CaCOJ
Carbonate (CO3~2)
Bicarbonate (HCO3~)
Hardness (total) [as CaCOJ
Nitrogen (total)
Nitrogen (total Kjeldahl)
Surfactants
Silica (SiO2) (total and dissolved)
Silt Density Index (SDI)
Ammonia (NH3) (as N)
Soluble Sulfide
Iron (Fe)
Manganese (Mn)
Phosphorous (total) (P)
Aluminum (Al+3)
Boron (B+3)
Calcium (Ca+2)
Magnesium (Mg+2)
Iron (Fe+2)
Manganese (Mn+2)
Nickel (Ni)
Phosphorous (total) (P)
Potassium (K+)
Silver (Ag)
Sodium (Na+)
Strontium (Sr+2)
Zinc (Zn+2)
Bromide (Br~)
Chloride (Cl")
Orthophosphate (PO4"3)
Sulfate (SO4-2)
Sulfide (as S2-)
[3
1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
c/x
X
X
c/x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Analysis Location
J £ § - *
£ 3 8 I S
x q
q q q x q
X X
X
X
X
X
X
Week
x
X
X
X
X
c/x
X
X
c/x
X
X
X
X
X
Week
Week
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
c = performed by calculation
c/x = by calculation and measurement
q = will be performed only as a quality check
STL = Severn Trent Labs
CSM = Colorado School of Mines
ACZ = ACZ Laboratories
BioVir =BioVir Laboratory
Week = Week Laboratory
Anatek = Anatek Laboratory
48
-------
Table 3-1 (cont'd). Raw Water Quality Sampling Schedule and Analysis Locations
Water Qualty
Parameter Type
x
§
1
"S
o
O
I"
1
—
•d
is
1
(S
1
&
§
C-
,0
o
"3 "S-
"o a1
o
•-B
"1
£
dissolved
(0.45 nm)
Radionuclides
DBFs
Biological &
Fouling Potential
Parameter
Antimony (Sb)
Arsenic (As)
Barium (Ba)
Beryillium (Be)
Cadmium (Cd)
Chromium (total) (Cr)
Copper (Cu)
Cyanide (free) (CN~)
Fluoride (F~)
Lead (Pb)
Mercury (inorganic) (Hg)
Nitrate (NO3) (as N)
Nitrite (NO'2) (as N)
Selenium (Se)
Thallium (Tl)
Copper (Cu)
Lead (Pb)
Nitrate (NO3) (as N)
Nitrite (NO'2) (as N)
Radium 226
Radium 228
Gross Alpha (excluding Ra & U)
Beta Particle & Photon Emitters
Uranium (U)
Bromate (BrO3)
Chlorite (CIO'2)
Haloacetic Acids (5 species)
Total Trihalomethanes (4 species)
Turbidity
Crypto sporidium
Giardia
Heterotrophic Plate Count (HPC)
Total Coliforms
Fecal Coliforms
E. coli
•3
•JS
HH
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Analysis Location
3 ^H •£ ^
H 's S§ !2 •§
^
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
ACZ
ACZ
ACZ
ACZ
ACZ
Week
Week
Anatek
Anatek
x
BioVir
BioVir
x BioVir
x BioVir
x BioVir
x BioVir
c = performed by calculation ACZ = ACZ Laboratories
c/x = by calculation and measurement BioVir =BioVir Laboratory
q = will be performed only as a quality check Week = Week Laboratory
STL = Severn Trent Labs Anatek = Anatek Laboratory
CSM = Colorado School of Mines
49
-------
Table 3-1 (cont'd). Raw Water Quality Sampling Schedule and Analysis Locations
Water Qualty
Parameter Type
X
1
1
a
Regulated
X
TS
^w
1
•§
&
0
Parameter
Alachlor
Benzene
Carbofaran
Carbon tetrachloride
Chlordane
Chlorobenzene
2,4-D (2,4-dichlorophenoxyacetic acid)
Dalapon (2,2-Dichloropropionic acid)
l,2-Dibromo-3-chloropropane(DBCP)
o-Dichlorobenzene (1,2-dichlorbenzene)
p-Dichlorobenzene (1,4-dichlorbenzene)
1 ,2-Dichloroethane
1 , 1 -Dichloroethylene
cis-l,2-Dichloroethylene
trans-l,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
Di(2-ethylhexyl) adipate
Di(2-ethylhexyl) phthalate
Diquat
Endothall
Endrin
Ethylbenzene
Glyphosate
Heptachlor
Heptachlor epoxide
Lindane
Methoxychlor
Picloram
Simazine
Styrene
Tetrachloroethylene
Toluene
Toxaphene
2,4,5-TP (Silvex)
1 ,2,4-Tri chlorobenzene
1,1,1 -Trichloroethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
Vinyl chloride
Xylenes (total)
is
|
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Analysis Location
j f § 3 |
— a vi a 5!
vi O O £ O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Anatek
Anatek
X
X
Anatek
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
c = performed by calculation
c/x = by calculation and measurement
q = will be performed only as a quality check
Accuracy reported as "Recovery Limits"
Precision reported as "RPD" or "Relative Percent Difference"
STL = Severn Trent Labs
CSM = Colorado School of Mines
ACZ = ACZ Laboratories
BioVir =BioVir Laboratory
Week = Week Laboratory
Anatek = Anatek Laboratory
50
-------
Table 3-2. Unregulated Organic Chemicals of Concern Analyzed by Colorado School of Mines
Group
Indicator 1:
Hydrophllic
Ionic
Indicator 2:
Hydrophilic
Non-Ionic
Indicator 3:
Hydrophobic
Non-Ionic
Name
Dichlorprop (2-(2,4-dichlorophenoxy) propanoic acid)
Diclofenac
Gemfibrozil
Ibuprofen
Ketoprofen
Mecoprop (2-(2-Methyl-4-chlorophenoxy) propanoic
acid)
Naproxen
Propyphenazone
Caffeine
Primidone
Phenacetine
Tris(l,3-dichloro-2-propyl) phosphate - (TDCPP)
Tris(2-chloroethyl) phosphate - (TCIPP)
Tris(l,3-dichloroisopropyl) phosphate - (TCEP)
Bisphenol A
Carbamazepine
l?p-estradiol
Testosterone
Category 1 Category 2
Herbicide
Pharmaceutical
Pharmaceutical
Pharmaceutical
Pharmaceutical
Herbicide
Pharmaceutical
Pharmaceutical
Pharmaceutical
Pharmaceutical
Pharmaceutical
Flame retardant Carcinogen
Flame retardant Carcinogen
Flame retardant
Industrial Chemical Endocrine disrupter
Pharmaceutical
Natural Hormone Endocrine disrupter
Natural Hormone Endocrine disrupter
Description
Various sources
Anti-inflammatory
Blood lipid regulator
Anti-inflammatory - OTC
Anti-inflammatory
Various sources
Anti-inflammatory
Analgesic
Stimulant - Coffee, soda
Anticonvulsant (seizures)
Analgesic
Often in flexible foams
Plasticizer
Estrogen
EPA
Toxicity Major Concern
Class
Slightly
Public perception
Slightly Public perception
Public perception
Slightly
Health
51
-------
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, USER conducted the testing of the EUWP as described below. USER field
personnel performed field analytical work using field laboratory equipment and procedures for
pH, temperature, conductivity, and turbidity. Six laboratories performed water quality analytical
work for samples not analyzed on site. The laboratories included Severn Trent Labs, Colorado
School of Mines, ACZ Laboratories, BioVir Laboratory, Week Laboratory and Anatek
Laoratory. 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 a 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: Installation, Initial Test Runs, and Initial System Integrity Tests
Task C: Verification Test
Task Cl: Membrane Flux and Recovery
52
-------
Task C2: Cleaning Efficiency
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. As mentioned previously, this ETV test at
Gallup, New Mexico, was specifically performed on wastewater effluent.
3.5.2 Task B: Equipment Installation, Initial Test Runs and System Integrity Tests
The objective of this initial operations task was to evaluate equipment operation and determine
the treatment conditions that resulted 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 Task C: 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 Task C2: 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.
53
-------
3.5.3.3 Task C3: 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.
3.5.3.4 Task C4: 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 Task C5: 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 USER 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. Since sufficient historic data was not available for the feed water source for
the ETV test at Gallup, an initial full water quality analysis of the feed water was performed.
This consisted of six grab samples, one each on the following dates: May 25, June 1, 8, 15, 22
and 28, 2006. The samples were collected between the hours of 12:30pm and 3:30pm. The
initial sampling event also included the full analysis of regulated organic chemicals.
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 included pressure decay testing of the UF membranes and a dye
challenge test of the RO system. See Section 3.8.4.1 for further discussion about this test.
3.8 Task C: Verification Testing
The verification test ran from July 12, 2006 to August 16, 2006. 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 Task Cl: 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
water recovery achieved by the membrane equipment, and the rate of flux decline observed over
54
-------
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
Table 3-3 lists the water quality and operational parameters measured continuously via online
instrumentation. Conductivity, turbidity, and temperature were also verified with manual
measurements. Flows were corroborated with total flow readings or by monitoring fill time on
tanks. Elapsed run time was also recorded daily based on RO high pressure pump and UF feed
pump hours. All continuously measured online data were recorded automatically approximately
every 15 minutes. The set points for key operating parameters are listed in Table 3-4.
Table 3-3. Water Quality and Operational Parameters Measured Online
•a
O
^b
ta
p
Flow FS2
Pressure PS3
Conductivity
Temperature
Turbidity X
Turbidity
(low range)
Particle ,,
x
Count
3
"S "S _.
o> t. ^
-*-> -^. i>
o ^2 w
PS ta ^
ta u. O
P P rt
FS4
PS4 PS5 PS9
Til
x
x
ji* o> cs o> ^*
^2 2s 2 £ w
•
-------
Table 3-5. Operational Data Plots Appearing in Chapter 4
UF Skid RO Skid
Filtrate Production Flow Rates
Flow Rates Percent Recovery
Operating Pressures 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:
.7. =-
S
•/,=% (3-1)
where:
Jt = filtrate flux at time t (gallons per square foot per day (gfd))
Qp = filtrate flow (gpd)
S = membrane surface area (ft2)
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 e-0.0239(r-20)
J,-^- (3-2)
56
-------
where:
Jt = filtrate flux at time t (gfd)
Qp = filtrate flow (gpd)
S = membrane surface area (ft)
T = temperature of the feed water (°C)
Transmembrane Pressure: The pressure across the membrane, equal to the average feed water
pressure on the membrane (average of inlet pressure and outlet pressure) minus the filtrate
(permeate) pressure:
TMP =
(3-3)
where:
TMP = transmembrane pressure (psi)
Pf = inlet pressure to the feed side of the membrane (psi)
PC = outlet pressure on the retentate side of the membrane (psi)
Pp = filtrate pressure on the treated water side of the membrane (psi)
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:
where:
TMP = Transmembrane pressure across the membrane (psi)
Jt = filtrate flux at time t (gfd). Temperature-corrected flux values were employed.
Temperature correction is to 20 °C.
Jtm = specific flux at time t (gfd/psi)
RO System
Permeate: Water produced by the RO membrane process.
Feed Water: Water introduced to the membrane element.
Concentrate: Water rejected by the RO membrane system.
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:
(3-5)
57
-------
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 was used to provide temperature corrections for
specific flux calculations:
Q,
-0.0239-(r-25)
S
(3-6)
where:
Jt (at 25° C) =
Jt = permeate flux at time t (gfd)
Qp = permeate flow (gpd)
S = membrane surface area (ft2)
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:
NDP =
(3-7)
where:
NDP = net driving pressure for solvent transport across the membrane (psi)
Pf = feed water pressure to the feed side of the membrane (psi)
Pc = concentrate pressure on the concentrate side of the membrane (psi)
Pp = permeate pressure on the treated water side of the membrane (psi)
A;r = osmotic pressure (psi)
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 total dissolved solids (TDS)
concentrations on either side of the membrane:
A;r =
(TDSf+TDSc)
2
-TDSr
(3-8)
where:
A;r = osmotic pressure (psi)
TDSf = feed water TDS concentration (mg/L)
58
-------
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 psi 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 psi 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 psi per 100 mg/L TDS gave a much higher osmotic
pressure and the ratio of 0.5 psi per 100 mg/L TDS gave a lower osmotic pressure. It was
determined that the equation for TDS using a factor 0.6 psi 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=~NDP (3"9)
where:
Jtm = specific flux (gfd/psi)
NDP = net driving pressure for solvent transport across the membrane (psi)
Jt = permeate flux at time t (gfd). Temperature-corrected flux values should be
employed.
Water Recovery: The recovery of feed water as permeate water is given as the ratio of permeate
flow to feed water flow:
~Qp
% System Recovery =100 —- (3-10)
where:
Qf = feed water flow to the membrane (gpm)
Qp = permeate flow (gpm)
Loss of Original Specific Flux: The loss of original specific flux is given as the ratio of specific
flux at membrane testing time zero divided by the specific flux at time T, and is calculated using
the following equation:
Percent Loss = 100-1 *_ (3-11)
V Jso)
where:
Jso = specific flux (gfd/psi) at time zero point of membrane testing
Js = specific flux (gfd/psi) at time T of membrane testing
59
-------
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:
'?,-<:.
Percent Solute Rejection = 100 •
Cf
(3-12)
where:
Cf = feed water concentration of specific constituent (mg/L)
Cp = permeate concentration of specific constituent (mg/L)
Break Horse Power:
bhp= ®'P (3-13)
1715•«#•
where:
Q = total feed flow to hydraulic array plus the Energy Recovery, Inc. (ERI) array
P = feed pressure to the hydraulic array
Eff = efficiency of the high pressure pump motor
1715 is a conversion factor.
Water Horse Power:
whp = - (3-14)
^ 3960
where:
Q = total feed flow to hydraulic array plus the ERI array
h = feet of head (pressure)
3960 is a conversion factor.
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 psi, even after a backwash. The manufacturer specified that the RO system be
cleaned when there is a 10 to 15% decrease in normalized permeate flow, 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.
60
-------
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).
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 3-1 were monitored during the testing period. Some
parameters in this table would not normally be measured at some locations (e.g. UF filtrate, RO
feed, total and dissolved silica for the RO permeate), but are in place as quality assurance checks.
In addition to the water quality parameters and manual sample collection listed in Table 3-1, the
following in-line measurements were recorded at 5 to 15 minute intervals (most intervals were
around ten minutes) with the data acquisition software:
• Turbidity readings recorded for UF feed and filtrate (RO feed) and RO permeate;
• Particle count readings recorded for UF feed and filtrate; and
• Conductivity readings recorded for RO filtrate.
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 was 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.
61
-------
3.8.4.1 Direct Integrity Testing
3.8.4.1.1 Pressure Decay Test
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. The pressure decay
test was also performed after each UF system cleaning.
Pressure decay testing was performed on the UF system daily. The product side of the
membranes was drained, both arrays were pressurized to 20 psi with compressed air as shown in
Figure 3-1. The feed valve and retentate valves were left in their operating positions. The
filtrate valves were closed. Pressure indicator P5 was monitored for 15 minutes. A pressure
decline of 0.1 psi/min was determined to be acceptable based on the feed side pipe volume.
In the case that there was a failure of the system, it is necessary to perform a pressure hold test
on individual cartridges using the setup shown in Figure 3-2. The cartridge must be removed
from the unit. The filtrate connectors were plugged on one side and pressurized with compressed
air on the other side. The rate of pressure decline was measured with a gauge attached to the
pressurized line. The cartridge was partly submerged in water, rotating the cartridge periodically
so that any leaking fibers could be detected and plugged.
As described later in Section 4.5.4, the daily pressure hold test did not provide an accurate
reading due to the configuration of the system that was used for the test. The individual cartridge
pressure test, described above, was performed at the end of the ETV test. All cartridges were
found to be in good condition. However, it was found that the end caps did not fit properly which
allowed leakage to occur.
62
-------
Monitor PIS
Figure 3-1. Pressurized Lines for Filtrate Side Pressure Hold Test.
63
-------
Figure 3-2. Individual UF Cartridge Pressure Hold Test.
64
-------
3.8.4.1.2 Dye Challenge Test
The direct integrity testing method employed on the RO system was a dye challenge test.
Florescent Water Tracing (FWT) Red 25 Liquid dye, a formulation of Rodamine WT, was used
as a challenge dye for the RO system. It was obtained from Bright Dyes, Division of Kingscote
Chemicals, 3334 South Tech Blvd., Miamisburg, Ohio 45342. Table 3-6 lists the properties of
the dye.
Table 3-6. Properties of FWT Red 25 Liquid Powder Dye
Property FWT Red 25 Liquid
Detectability of Active ingredient Visual <100 ppb
Maximum absorbance wavelength2 550/588 nm
Appearance clear dark red aqueous solution
NSF Approved maximum concentration for potable water 0.8 ppb
Molecular Weight 479.02 g/mol
Specific Gravity 1.03±0.05 @25°C
Viscosity 3 1.3 centipoises (cps)
pH 8.7±0.05 @25°C
Coverage of Product One Pint Liquid
Light Visual 31,250 gallons
Strong Visual 3,125 gallons
1 In deionized water in 100 mL flask. Actual detectability and coverage in the field will vary with specific water
condition.
2 No significant change in fluorescence between 6 and 11 pH.
3 Measured on a Brookfield viscometer, Model LV, UL adapter, 60 rpm @25°C.
To determine the concentration that would be required to detect a 99% or greater reduction in
concentration between the feed and product during the challenge, first an absorption response
curve and then a concentration response curve were developed. Figures 3-3 and 3-4 show the
absorbance of a concentration of 2 mL/L of dye solution and the change in absorbance as the dye
was diluted from 2 mL/L to 0.005 mL/L.
The FWT Red absorbance is saturated above 1 mL/L. Absorbance at 1 mL/L is 2.922 while the
lowest that can be detected reproducibly is 0.005 mL/L with an absorbance of 0.005. Therefore,
the maximum detectable log removal was 2.77 logic. A solution was prepared in the RO feed
tank to attain an absorbance of approximately 2.9. After monitoring the concentrate until the dye
was detected, samples of the product and concentrate were collected at one minute intervals for
ten minutes. Absorbance was measured at a wavelength of 560 nm. The concentrate sample was
diluted with RO permeate and concentration calculated from absorbance.
65
-------
O.vJ
•2
? EL
O
2 9
n
.Q
^
o
£ <( EL
^ 1 .O
-\
n 'i
n
» * * '
*
•
490 500 510 520 530 540 550 560 570 580
Wavelength (nm)
Figure 3-3. Absorbance of 2 mL/L FWT Red Liquid Dye with Maximum Absorbance at
Wavelength 560 nm.
E
c
o
<£>
IO
0)
o
c
re
£1
O
V)
y = 3.0817x
R2 = 0.9917
0.0
0.5
1.0 1.5
Concentration (mL/L)
2.0
2.5
Figure 3-4. Absorbance has a linear correlation to concentration below ImL/L.
66
-------
3.8.4.2 Continuous Indirect Integrity Monitoring
The continuous indirect integrity monitoring method used on both the UF and RO membrane
systems was turbidity monitoring. Turbidity was monitored continuously on the UF filtrate
(RO feed) and RO permeate using a Hach 1720E Low Range Process Turbidimeter with sclOO
controller. The UF feed was monitored continuously with the Hach FilterTrak 660™ sc Laser
Nephelometer. Turbidity was also measured twice daily from each process stream using a Hach
21 OOP Portable Turbidimeter. Data was downloaded by WaterEye and also saved locally on the
controllers. As a backup, WaterEye readings were manually documented twice per day.
In addition to turbidity monitoring, 2200 PCX Particle Counters were installed on the UF feed,
filtrate, and RO permeate. Data was written to a dedicated computer at five to 15 minute
intervals with most of the data collected at ten minute intervals.
Results of the direct integrity tests and indirect integrity monitoring are presented in Chapter 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 USER provided sufficient and reliable data; and 2)
develop an effective and accurate statistical analysis of the data.
3.5.5.7 Work Plan
The EUWP test system was equipped with a computer monitoring system. Some of the required
measurements (see Table 3-3) were recorded automatically by the automated system. The
remaining required measurements were recorded by hand by the field operator on-site.
All field activities were documented. Field documentation included field logbooks, photographs,
field data sheets, and chain-of-custody forms. USER was responsible for maintaining all field
documentation. Field notes were kept in bound logbooks with each page numbered sequentially.
Field data sheets were used to record all operating data as backup and check of data recorded
inline by WaterEye.
The database for the project was set up in the form of custom-designed spreadsheets.
Spreadsheets containing the operational and water quality data, including calculations, were
developed by USER. Following data entry, 100% of the data in the spreadsheets was checked
against the numbers on the field log sheets or laboratory analysis outputs.
Chain-of-custody forms accompanied all samples shipped to the analytical laboratory. Copies of
field sheets and chain-of-custody forms are included in the Appendix.
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.
67
-------
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.
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.
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.
68
-------
Table 3-7. On-Site Analytical Equipment QA Activities
Equipment
Action Required
Flowmeters - electronic
Turbidimeter - in-line (1720E)
Initial 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
Conductivity meter - portable
Volumetrically checked flow
Verified with portable turbidimeter
3-point calibration (4,7,10)
Volumetrically checked flow
Volumetrically checked flow
Calibrated at 2 points
Weekly
Rotameters
UF filtrate flow
Particle counter - in-line
Temperature - portable
Turbidimeter - portable
Inspected for buildup of algae, salt, etc.
Verified volumetrically
Cleaned sensors
Verified calibration with NIST-certified precision
thermometer
Calibrated using <0.1, 20, 100, and 800 NTU
standards
Every Two
Weeks
Flowmeters - electronic
Verified calibration volumetrically
Prior to
Test
Tubing
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
3.8.6.4.1.1 Sampling Locations
Sampling locations are detailed in Table 3-9.
69
-------
Table 3-9. Water Sampling Locations for Water Quality Samples
Sample Stream Location
Raw Grab sample from feed pipe side tap, upstream of any pre-
UF chemical addition, or source water near intake
UF Filtrate UF filtrate sample tap
UF Retentate and Backwash (combined) Outfall
RO Feed V90 - Valve immediately after RO strainer
RO 1st Pass Permeate V58 - only available during 1st Pass only operation
RO 1st Pass Concentrate V91 - Temporary valve off the RO unit
RO 2nd Pass Permeate V61
RO 2nd Pass Concentrate V91 - Temporary valve off the RO unit
3.8.6.4.1.2 On-Site Analytical Methods
The analytical methods for on-site monitoring of raw and treated water quality are described in
the following sections.
EH
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 compared to the
readings from the hand-held turbidimeter daily. In addition to calibration, the lens was cleaned
before each calibration using lint-free paper to prevent any particle or microbiological build-up
that could produce inaccurate readings. If the comparison suggested inaccurate readings, the in-
line turbidimeter was recalibrated. A volumetric check on the sample flow was performed daily.
70
-------
The LED readout was also recorded in the logbook to ensure a back-up record of the in-line data
acquisition system.
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.
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 jim to 750 jim in up to 32 user-defined bins. The
particle counter manufacturer provided calibration certificates documenting that the inline
particle sensors meet these criteria. The particle counter manufacturer provided the methods for
demonstration of coincidence error.
3.8.6.4.1.3 Sample Collection, Shipment, and Storage for Laboratory Analyses
Samples were collected in bottles prepared and shipped to the test site by the laboratories
identified in Table 3-1. 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 during storage and shipment to the
laboratories. Chain of custody-like forms accompanied all samples.
3.8.6.4.1.4 Laboratory Analytical Methods
A comprehensive list of laboratory analytical methods used can be found in Table 3-10. 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.
71
-------
Table 3-10. Analytical Methods for Laboratory Analyses
Parameter Analytical Method
Bicarbonate EPA 310.1
Carbonate EPA 310.1
BOD SM(1)5210B
COD EPA 410.4
Color EPA 110.2
Ammonia EPA 350.1
Bromide EPA 300.0
Bromate EPA 300.1
Chloride EPA 300.0
Chlorite EPA 300.1
Odor EPA 140.1
DOC EPA 415.1
UV254 SM5910-B
Sulfate EPA 300.0
Fluoride EPA 300.0
Free Cyanide SM 4500-CN-I
Hardness (total) EPA 130.2
Nitrate EPA 300.0
Nitrite EPA 300.0
Conductivity EPA 120.1
Alkalinity EPA 310.1
TDS EPA 160.1
TOC EPA 415.1
TSS EPA 160.2
VSS EPA 160.4
Dissolved Ortho-Phosphate EPA 300.0
Dissolved Sulfide EPA 376.1 and SM 4500 S2
Surfactants Special Method
Total Kjeldahl Nitrogen EPA 351.2
Total Nitrogen EPA 300.0
Aluminum SW8466010B
Boron SW8466010B
Calcium SW8466010B
Iron SW8466010B
Potassium SW8466010B
Magnesium SW8466010B
Manganese SW8466010B
Sodium SW8466010B
Silica SW 846 6010B
Strontium SW 846 601 OB
Zinc SW8466010B
Phosphorous SW8466010B
Silver SW8466010B
Copper SW8466010B
Nickel SW8466010B
Lead SW8466010B
Antimony SW 846 601 OB
Arsenic SW8466010B
Barium SW8466010B
Beryllium SW8466010B
Cadmium SW8466010B
Chromium (total) SW8466010B
72
-------
Table 3-10. Analytical Methods for Laboratory Analyses (continued)
Parameter
Analytical Method
Selenium
Thallium
1,2-Dibromoethane (EDB)
Benzene
Carbon Tetrachloride
Chlorobenzene
l,2-Dibromo-3-chloro propane (DBCP)
1,2-Dichlorobenzene
1,4-Dichlorobenzene
1,2-Dichloroethane
cisl,2-Dichloroethylene
trans 1,2-Dichloroethylene
1,1-Dichloroethylene
1,2-Dichloropropane
Ethylbenzene
Methyl Chloride
Styrene
Tetrachloroethylene
Toluene
1,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1,1,2 -Trichloroethane
Trichlorethylene
Vinyl Chloride
Xylenes (total)
Lindane
Chlordane
Endrin
Heptachlor
Heptachlor Epoxide
Methoxychlor
Toxaphene
Aroclor 1016
Aroclor 1221
Aroclor 1232
Aroclor 1242
Aroclor 1248
Aroclor 1254
Aroclor 1260
Simazine
2,4-D
Dalapon
Picloram
2,4,5-TP (Silvex)
Carbofuran
N-Nitrosodimethylamine (NDMA)
Gross Alpha/Beta
Radium 226
Radium 228
Uranium
Chloroform
Bromodichloromethane
SW8466010B
SW8466010B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8260B
SW 846 8081A
SW 846 8081A
SW 846 8081A
SW 846 8081A
SW 846 8081A
SW 846 8081A
SW 846 8081A
SW 846 8082
SW 846 8082
SW 846 8082
SW 846 8082
SW 846 8082
SW 846 8082
SW 846 8082
SW 846 8141 A
SW8468151A
SW8468151A
SW8468151A
SW8468151A
SW 846 8321A
EPA 1625M
EPA 900.0
EPA 903.1
EPA 904.0
EPA 200.8
EPA 524.2
EPA 524.2
73
-------
Table 3-10. Analytical Methods for Laboratory Analyses (continued)
Parameter Analytical Method
Chlorodibromomethane EPA 524.2
Bromoform EPA 524.2
Monochloroacetic Acid SM 6251 B
Dichloroacetic Acid SM 6251 B
Trichloroacetic Acid SM 6251 B
Monobromoacetic Acid SM 6251 B
Dibromoacetic Acid SM 6251 B
Endothall EPA 548.1
Diquat EPA 549.2
Glyphosate EPA 547
Total Coliform SM9221B
Fecal Coliform SM9221E
HPC SM9216A
Ibuprofen (2)
Diclofenac (2)
Ketoprofen (2)
Naproxen (2)
Gemfibrozil (2)
Propylphenazone (2)
Mecoprop (2)
Diclorprop (2)
Primidone (2)
Phenacetine (2)
Caffeine (2)
Tris(l,3-dichloro-2-propyl)phosphate (TDCPP) (2)
Tris(2-chloroethyl)-phosphate (TCIPP) (2)
Tris(2-chloroisopropyl)- phosphate (TCEP) (2)
l?p-estradiol (3)
Testosterone (3)
Carbamazepine (2)
Bisphenol A (2)
(1) SM= Standard Methods for the Examination of Water and Wastewater
(2) Method described in Reddersen and Heberer (2003)
(3) Method described in Mansell and Drewes (2004)
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-u. (0/2) x (S/Vn)] (3-13)
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
74
-------
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 (SA/n)] (3-14)
3.8.6.4.3 A ccuracy
The accuracy of on-site analytical equipment was periodically verified according to the schedule
in Table 3-5. 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. Accuracy of analytical readings
was measured through the use of spiked samples and laboratory control samples. Accuracy also
incorporates calibration procedures and use of certified standards to ensure the calibration curves
and references for analysis are near the "true value."
Recoveries for spiked samples are calculated in the following manner:
WO*(SSR- SR)
% Recovery =
SA (3-15)
where:
SSR = spiked sample result
SR = sample result
SA = spike amount added
Recoveries for laboratory control samples are calculated as follows:
100 • (Found Concentration)
% Recovery =
True Concentration (3-16)
3.8.6.4.4 Precision
Precision refers to the degree of mutual agreement among individual measurements and provides
an estimate of random error. To quantify precision, the relative percent difference (RPD) of
duplicate analyses can be calculated. Every sampling event with samples shipped to the lab
included 10% sample duplicates. RPD was measured by use of the following equation:
:200 (3-17)
where:
Sl = sample analysis result; and
S2 = sample duplicate analysis result.
RPD should be less than 30%.
75
-------
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 (3-18)
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 3-12
presents the completeness requirements based on the sampling frequency spelled out in the
test/QA plan.
Table 3-12. Completeness Requirements
Number of Samples per Parameter and/or Method Percent Completeness
0-10 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.
76
-------
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 flows 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.
QA/QC information, as described by the QAPP in the PSTP for this verification test, is presented
at the end of the chapter.
4.2 Equipment Installation, Start-up, and Shakedown
The equipment installation, start-up and shakedown tests took place between June 28, 2006, and
the beginning of the official ETV test on July 12, 2006. During this period all sensors were
calibrated, communications were established with the particle counters and turbidimeters, and
several programming issues were resolved. Handheld analyzers were calibrated and checked;
colorimetric methods were tested; and the intake screen was installed. It was determined that
ferric chloride coagulation would be necessary to keep the UF system running smoothly. Jar
tests were performed to estimate the necessary dose rate for this water source.
Background sampling continued during the initial test runs, giving the operators experience in
sampling, packaging, and shipping the water samples. A pressure hold test for UF system, and
an initial dye test on the RO system were performed during this time.
4.3 Task A: Raw Water Characterization
The objective of this task was to determine the chemical, biological, and physical characteristics
of the feed water. As described in the Section 1.5, the feed (raw) water for this ETV test is
treated municipal wastewater collected prior to chlorination. Wastewater analysis data for the
background samples are listed in Tables 4-1 through 4-9. Parameters that exceed primary
drinking water standards included nitrate, haloacetic acids, bromate, gross alpha, and biological
components. Secondary drinking water standards are exceeded for color, sulfate, TDS,
surfactants, aluminum, and odor. The source of the city's drinking water is high in TDS and
sulfate with some radioactivity. The rest of the exceedances are caused by municipal use and the
wastewater treatment process.
Additional contaminants of concern found in the feed water (treated municipal wastewater),
listed in Table 4-9, include:
• Naproxen - a nonsteroidal anti-inflammatory drug (NSAID) found in Aleve®;
• Gemfibrozil - a lipid regulating agent used to treat heart disease and high cholesterol;
• Primidone - an anticonvulsant used to treat epilepsy and neuralgia;
• Carbamazepine - also an anticonvulsant;
• Tris(l,3-dichloro-3-propyl)phosphate (TDCPP), Tris(2-chloroethyl)phosphate (TCIPP),
tris(2-chloroisopropyl)phosphate(TCEP) - components of fire retardant;
77
-------
• 17p-estradiol - female hormone;
• Testosterone - male hormone; and
• Bisphenol A - a degradation product of plastic and potential endocrine disrupter.
Table 4-1. Background Water Analyses -Severn Trent Results/CSM: General Chemistry
Parameter
pH
Bicarbonate
Carbonate
COD
Color
Ammonia
Ammonia (CSM)
Bromide
Chloride
DOC
UV254 absorbance
Sulfate
Fluoride
Free Cyanide
Hardness (total, as CaCO3)
Nitrate
Nitrite
Conductivity
Conductivity (CSM)
Alkalinity (as CaCO3)
TDS
TOC
TSS
Volatile Suspended Solids
Dissolved Ortho-Phosphate
Dissolved Sulfide
Surfactants
Total Kjeldahl Nitrogen
Total Nitrogen
Nitrate (as N)
Units
-
mg/L
mg/L
mg/L
Color Units
mg/L
mg/L
mg/L
mg/L
mg/L
(cm'1)
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
|j,S/cm
|j,S/cm
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
Reporting
Limit
0.1
5
5
20
5
0.10
0.10
0.20
30
1.0
-
50
0.5
0.01
5
1
0.5
2.0
2.0
5.0
10
1.0
4.0
4.0
0.5
1.0
0.10
0.50
0.50
0.50
5/25/06
7.8
260
<5
20
35
0.12
0.48
O.20
110JQ
6.9J
0.1745
320Q
1.4
NM
63
13Q
0.5
1700
1672
260
1100
7.4
10
7.6
3.1
<1.0
0.75
0.84
14
19.5
6/01/06
7.8
270
<5
38
30
<0.10
NM
O.20
110Q
8.6J
NM
340JQ
1.4
0.01
44
11Q
0.5
1700
NM
270
1100
22
7.6
5.2
4.4
<1.0
0.75
2.2
14
13.8
Date
6/08/06 6/15/06
7.9
300
<5
40
75
O.10
0.69
0.21
100JQ
13.0
0.1700
330JQ
1.3J
0.01
47
10
0.5
1700
1653
300
1100
12
21
14
0.5
<1.0
0.75
3.1
14
NM
8.0
280
<5
26
40
O.10
0.56
O.20
110Q
7.7
0.1819
340Q
1.3
0.01
54
10Q
0.5
1700
1642
280
1100
8.1
11
8
2.4
<1.0
0.50
2.2
13
10.9
6/22/06
8.0
290
<5
34
40
O.10
0.45
0.20
88
6.9
0.1759
310Q
1.3
0.01
59
8.5
0.5
1700
1636
290
1200
6.9
15
11J
2.6
<1.0
0.50
2.3
11
10.8
6/28/06
8.0
270
<5
32
35
O.ll1
0.37
O.20
100Q
7.1
0.1825
340Q
1.0
0.01
44
11Q
0.5
1700
1659
270
1100
7.1
<4.0
<4.0
3.4
<1.0
0.75
1.6
13
8.7
J - Method blank contamination. The associated method blank contains the target analyte at a
reportable level.
Q - Elevated reporting limit. The reporting limit is elevated due to high analyte levels.
NM - Not measured.
78
-------
Table 4-2. Background Water Analyses -Severn Trent: Dissolved Metals
Dissolved Metals
Aluminum
Boron
Calcium
Iron
Potassium
Magnesium
Manganese
Sodium
Silica
Strontium
Zinc
Phosphorus (total)
Silver
Copper
Nickel
Lead
Reporting
Limit ((J.g/L)
100
100
200
100
3,000
200
10
1,000
1,100
10
20
3,000
10
15
40
9.0
5/25/06
(HS/L)
<100
320
14,000 (J)
<100
15,000
2,800
<10
400,000
22,000
170
62 (J)
3,000
<10
<15
<40
<9.0
6/1/06
(ng/L)
NM
320
13,000
<100
15,000 (L)
3,000
19
380,000
18,000
140
60
4,400
<10
<15
<40
<9.0
6/8/06
(HS/L)
310
310
13,000
140
13,000 (L)
2,300
13
360,000 (J)
19,000
160
53 (J)
<3,000
<10
<15
<40
<9.0
6/15/06
(HS/L)
130
320
14,000 (J)
<100
13,000
3,000
<10
380,000 (J)
17,000
170
90
<3,000
<10
<15
<40
<9.0
6/22/06
(HS/L)
110
320
12,000 (J)
<100
13,000
3,000
<10
370,000
20,000
170
62 (J)
<3,000
<10
<15
<40
<9.0
6/28/06
(HS/L)
130
300
12,000 (J)
<100
15,000
2,900
<10
340,000
21,000
150
61 (J)
3,200
<10
<15
<40
<9.0
NM - Not measured.
J - Method blank contamination. The associated method blank contains the target analyte at a
reportable level.
L - Serial dilution of a digestate in the analytical batch indicated that physical and chemical
interferences are present.
Table 4-3. Background Water Analyses - Severn Trent: Total Metals
Total Metals
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium
(total)
Copper
Iron
Lead
Manganese
Mercury
Phosphorus
Selenium
Silica
Thallium
Reporting
Limit (ng/L)
10
15
10
5
5
10
15
100
9
10
0.2
3,000
15
1,100
15
5/25/06
<10
<15
<10
<5
<5
<15
150
<9
11
0.2
3,000
<15
21,000
<15
6/1/06
<10
<15
<10
<5
<5
<15
130
<9
30
0.2
4,400
<15
21,000
<15
6/8/06
<10
<15
10 (J,L)
<5
<5
<15
170 (J)
<9
17
0.2
<3,000
<15
21,000
<15
6/15/06
<10
<15
<10
<5
<5
<15
140
<9
12
0.2
<3,000
<15
15,000
<15
6/22/06
<10
<15
<10
<5
<5
<15
<100
<9
<10
0.2
<3,000
<15
19,000
<15
6/28/06
<10
<15
<10
<5
<5
<15
<100
<9
<10
0.2
3,300
<15
19,000
<15
J - Method blank contamination. The associated method blank contains the target
analyte at a reportable level.
L - Serial dilution of a digestate in the analytical batch indicated that physical and
chemical interferences are present.
79
-------
Table 4-4. Background Water Analyses - Severn Trent: Volatile Organic Compounds
Volatiles
Reporting
Limit ((J.g/L)
5/28/06 6/1/06 6/8/06 6/15/06 6/22/06 6/28/06
1,2-Dibromoethane (EDB)
Benzene
Carbon Tetrachloride
Chlorobenzene
l,2-Dibromo-3-chloro propane (DBCP)
1,2-Dichlorobenzene
1,4-Dichlorobenzene
1,2 -Dichloroethane
cis 1,2 -Dichloroethylene
trans 1,2-Dichloroethylene
1,1 -Dichloroethylene
1,2 -Dichloropropane
Ethylbenzene
Methyl Chloride
Styrene
Tetrachloroethylene
Toluene
1,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1,1,2-Trichloroethane
Trichlorethylene
Vinyl Chloride
Xylenes (total)
.0
.0
.0
.0
5.0
.0
.0
.0
.0
.0
.0
.0
.0
5.0
.0
.0
.0
.0
.0
.0
.0
.0
2.0
<5.0 <5.0 <5.0 <5.0 <5.0 <5.0
<5.0 <5.0 <5.0 <5.0 <5.0 <5.0
<2.0 <2.0 <2.0 <2.0 <2.0 <2.0
Table 4-5. Background Water Analyses - Severn Trent: Semi-Volatile Organic Compounds
GC Semi Volatiles
Lindane
Chlordane
Endrin
Heptachlor
Heptachlor Epoxide
Methoxychlor
Toxaphene
Aroclor 1016
Aroclor 1221
Aroclor 1232
Aroclor 1242
Aroclor 1248
Aroclor 1254
Aroclor 1260
Simazine
2,4-D
Dalapon
Picloram
2,4,5 -TP (Silvex)
Carbofuran
Reporting
Limit ((J.g/L)
0.05
0.5
0.05
0.05
0.05
0.1
2.5
1.0
1.0
1.0
1.0
1.0
1.0
1.0
10.0
4.0
2.0
0.1
1.0
1.0
6/1/06
O.05
O.5
O.05
0.05
0.05
0.1
<2.5
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<10.0
<4.0
<2.0
O.I
-------
Table 4-6. Background Water Analysis - Week and ACZ
Analyte
Bromate
Chlorite
"o Ammonia as N
y£ Odor
Soluble Sulfide
NDMA
Gross Alpha
LLD(1)
Gross Beta
LLD
£j Radium 226
< LLD
Radium 228
LLD
Uranium
Units
ug/L
ug/L
mg/L
TON
mg/L
ng/L
pCi/L
pCi/L
pCi/L
pCi/L
mg/L
Reporting Limit
25
50
0.1
1
0.1
2
MDL(2) = 0.0001
PQL(3) = 0.0005
5/25/06
<25
<50
NM
12
O.I
<2
9.8
4.4
18
6.3
0.09
0.22
0.34
0.75
0.0001
6/1/06
<25
<50
O.I
12
O.I
6.8
0
4.9
11
7.6
0.06
0.221
1.6
0.86
0.0001
6/8/06
<25
<50
NM
17
O.I
<2
30
4.4
11
6.3
0.16
0.45
1.1
0.84
0.0002
6/15/06
<25
<50
O.I
17
O.I
<2
7.5
4.4
14
6.3
0.1
0.35
0.15
1.6
0.0001
6/22/06
<25
<50
0.1
12
O.I
<2
1.9
4.7
12
7.2
0.16
0.64
0.86
0.62
0.0002
6/28/06
<25
<50
0.11
17
O.I
<2
16
4.8
29
7
0.05
0.42
1.6
0.67
0.0001
(1) LLD - lower limit of detection
(2) MDL - method detection limit
(3) PQL - practical quantitation limit
NM - Not analyzed.
Table 4-7. Background Water Analyses - Anatek Labs: Disifection Biproducts and
Pesticides
Analyte
Chloroform
Bromodichloromethane
Chlorodibromomethane
Bromoform
Monochloroacetic Acid
Dichloroacetic Acid
Trichloroacetic Acid
Monobromoacetic Acid
Dibromoacetic Acid
Endothall
Diquat
Glyphosate
MDL (ng/L)
0.05
0.05
0.05
0.05
0.6
0.5
0.4
0.7
0.4
1.9
0.2
3.3
5/25/06
(Ug/L)
0.46
0.05
O.05
O.05
O.6
0.6
1
O.7
O.4
<1.9
O.2
<3.3
6/1/06
(Ug/L)
0.39
0.05
O.05
O.05
O.6
0.8
0.7
O.7
O.4
<1.9
O.2
<3.3
6/8/06
(Ug/L)
0.33
0.05
O.05
O.05
O.6
0.5
0.6
O.7
O.4
<1.9
O.2
<3.3
6/15/06
(Ug/L)
0.32
0.05
O.05
O.05
O.6
0.5
0.6
O.7
O.4
<1.9
O.2
<3.3
6/22/06
(Ug/L)
0.52
0.08
O.05
O.05
O.6
0.4
0.6
O.7
O.4
<1.9
O.2
<3.3
6/28/06
(Ug/L)
0.5
0.06
O.05
O.05
O.6
0.6
0.9
O.7
O.4
<1.9
O.2
<3.3
Table 4-8. Background Biological Analysis
Analyte
Total Coliforms
Fecal Coliforms
HPC
Units
MPN/100 mL
MPN/100 mL
CFU/mL
6/1/06
24,000
5,000
6,600
6/8/06
>160,000
140,000
>160,000
6/15/06
70,000
70,000
11,000
6/22/06
1,600,000
900,000
190,000
6/28/06
4,000
<2,000
11,000
81
-------
Table 4-9. Background Water Analyses - Colorado School of Mines: Wastewater
Contaminants of Concern
Detection
limit 05/25/06 06/01/06 06/08/06 06/15/06 06/22/06 06/28/06
Samples (ng/L) (ng/L) (ng/L) (ng/L) (ng/L) (ng/L) (ng/L)
Indicator 1:
Hydrophilic, ionic
muicaior ^:
Hydrophilic,
non-ionic
a i .a
a g a
2 § .2
Ibuprofen
Diclofenac
Ketoprofen
Naproxen
Gemfibrozil
Propylphenazone
Mecoprop
Diclorprop
Primidone
Phenacetine
Caffeine
TDCPP
TCIPP
TCEP
l?p-estradiol
Testosterone
Carbamazepine
Bisphenol A
4
1
2
1
2
20
2
1
1
40
40
50
50
50
0.4
0.5
2
5
<4
<1
<2
168
143
<20
<2
<1
287
<40
<40
1110
989
1055
0.7
1
552
462
<4
NA
<2
65
35
<20
<2
<1
277
<40
<40
1185
558
1040
0.8
<0.5
444
<5
<4
<1
<2
40
<2
<20
<2
<1
286
<40
<40
760
520
480
0.8
<0.5
<2
<5
<4
<1
<2
177
NA
<20
<2
<1
195
<40
<40
1012
1008
995
1
0.6
<2
<5
<4
<1
<2
127
<2
<20
<2
<1
283
<40
<40
651
988
800
ND
<0.5
<2
<5
<4
<1
<2
99
<2
<20
<2
<1
341
<40
<40
888
840
537
1
0.8
NA
<5
82
-------
4.4 Task B Initial Test Runs
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 was considered
a shakedown testing period and was completed before the main tasks.
Initial equipment checks, UF integrity tests, required calibration checks, and initial test runs took
place between June 28, 2006 and the beginning of the official ETV test on July 12, 2006. 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 June 21, 2006. This test was a low
pressure test demonstrating that the system held a pressure of 2.4 psi for greater than five (5)
minutes. A second higher pressure test was performed on July 12, 2006. These results are
presented in Table 4-10. The pressure loss was only 1.5 psi over a 15-minute period (0.1
psi/min). Later, during the verification test it was discovered that the UF pressure decay data was
incorrect. Section 4.5.4.1 presents additional information on UF integrity testing.
Table 4-10. UF Full System Integrity Test Results, July 12, 2006
Time
(min)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Pressure
(psi)
20.0
19.8
19.7
19.7
19.5
19.5
19.3
19.2
19.2
19.1
18.9
18.9
18.8
18.7
18.7
18.5
Pressure Change
(psi/min)
NA
0.2
0.1
0.0
0.2
0.0
0.2
0.1
0.0
0.1
0.2
0.0
0.1
0.1
0.0
0.2
A dye test to check the RO system was performed on July 10, 2006. A liquid tracing dye stock
solution was prepared and then used to develop an absorbance calibration curve by serial dilution
of the stock solution. RO feed water with dye added was pumped to the RO system and the RO
permeate checked for absorbance. The RO permeate had low absorbance and dye rejection was
83
-------
>99.4% both arrays. Data from the initial dye test and the dye test at the end of the verification
test are presented in Section 4.5.4.2.
It was expected that ferric chloride coagulation would be necessary to keep the UF system
running smoothly. Field screening tests were performed to identify the initial dose rate for this
water source. Tests at ferric chloride feed rates of approximately 1.5 mg/L (as Fe) and 2.5 mg/L
(as Fe) indicated by visual observation that the higher dose provided better floe formation and
clearer water. As system shakedown proceeded, the target feed rate was increased to 5.0 mg/L
(as Fe) for the start of the test.
4.5 Task C: Verification Test
The verification test was started on July 12, 2006 and ran until August 16, 2006, covering 36
calendar days. The system was shut down for two days (July 24 and 25, 2006) for RO cleaning
and for two days (July 30 and 31, 2006) for UF cleaning. An additional RO cleaning was
performed from August 7 to August 8, when the system was down for approximately 24 hours.
The system was in operation on 32 calendar days, which met the test plan goal for collecting
operating data for a minimum of 30 days. The EUWP was operated as continuously as possible.
Shut downs occurred each day to perform the pressure hold test on the UF system, to calibrate
sensors, clean the strainers, etc. Operators were on site only during the day light periods. 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 mean RO operating hours during
the verification test was 18 hours per day. The mean UF operating hours during the verification
test was 14 hours per day with a median of 15 hours. The UF operating hours were lower than
the RO as the system is designed for the UF to operate at a higher filtrate flow rate than the RO
feed rate in order to keep the RO feed tank full. Whenever the RO feed rate tank is at maximum
level, the UF is automatically shut down until the RO feed tank level drops to the pre-set level to
restart the UF system.
4.5.1 Task Cl: Membrane Flux and Operation
The purpose of this task was to evaluate membrane system performance during operation. The
objectives of this task were to demonstrate the appropriate operational conditions for the
membrane equipment, the feed water recovery achieved by the membrane equipment, and the
rate of flux decline observed over extended membrane operation.
Operational data were collected and on-site water quality measurements were made twice per
day on most days of operation. Occasionally, only one set of measurements were obtained due to
maintenance activities limiting the time available for operators to collect operating data. The data
were summarized for presentation and discussion in this Section. The complete data set can be
found in Appendix L.
84
-------
4.5.1.1 UF Operating Data
4.5.1.1.1 UFflow Rate, Filtration Production, IMP Results, and Specific Flux Results
The UF operational statistics are presented in Table 4-11. The UF skid does not have a filtrate
flow meter or filtrate pressure gauge. Therefore, the total filtrate flow rate was calculated as the
UF feed water flow rate minus the UF retentate flow rate. The intake flow is the intake from the
source water into the UF feed water 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 4-11. UF Operational Data Statistics
Parameter
UF Operation per day (hr)
Intake Flow (gpm)
Feed Flow (gpm)
Filtrate Flow (gpm)
Retentate Flow (gpm)
Backwash Flow (gpm)
Feed Pressure (psi)
Retentate Pressure (psi)
Filtrate Temperature (°F)
Count Mean Median
30
53
53
53
49
14 15
281 288
250 251
229 229
24 25
95%
Confidence
Standard Interval
Minimum Maximum Deviation (CI)
4
217
179
154
19
900 gallons per backwash cycle0 };
53
53
54
22 21
19 19
78 78
16
0
76
20
301
314
289
30
4.1
21.0
24.3
25.0
4.4
Backwash every 30
30
28
82
3.9
5.4
1.5
+1.5
+5.65
+6.55
+6.74
+1.2
minutes
±1.1
+1.5
+0.4
(1) Volume not measured. It was provided by the manufacturer.
The mean UF feed water flow rate of 250 gpm was slightly below the target feed flow rate of
259 gpm specified for the system (See Table 3.5 Section 3.7). The mean filtrate flow rate of
229 gpm corresponds to a flow rate of 14.3 gpm for each of the 16 UF membrane modules. The
UF water recovery was 91.6% based on the mean feed water and filtrate flow rates.
Figure 4-1 shows the UF system flow rates over the duration of the verification test. The
retentate flow rate remained steady throughout the test. The feed water flow rate and filtrate flow
rates dropped as the intake strainers and UF membranes became fouled with solids and TMP
increased. Manual adjustment of the flow control valve was made to hold the feed water and
filtrate flows as steady as possible. The increase in flow rates on July 18 occurred after the
strainer on the feed line was cleaned, which provided more feed line pressure and flow to the UF
system.
85
-------
350
300
IT 25°
o.
S.
o
200
150
100
50
0
RO Cleaning UF Cleaning RO Cleaning
A A A A A A A
7/12/06 7/17/06 7/22/06 7/27/06 8/1/06 8/6/06 8/11/06 8/16/06
Date
Figure 4-1. Plot of UF system flow rates throughout the testing period.
UF filtrate production was tracked using the RO feed totalizer plus the number of backwash
cycles performed (900 gallons of UF filtrate used per backwash). The total UF filtrate volume
produced was 5,693 kgal, which gives an average total production rate of 178 kgal per day. This
daily production rate calculation excludes the four RO and UF cleaning days when the UF was
not operated, but includes all other operating days. The net UF filtrate production, which also
equals the RO feed water volume, was 5,089 kgal. Figure 4-2 shows the cumulative total and net
filtrate production for the UF system over the duration of the verification test.
6000
5000
4000
3000
•a
g
fe
2000
1000
0
RO Feed Production A UF Total Production
7/12/2006 7/17/2006 7/22/2006 7/27/2006 8/1/2006 8/6/2006 8/11/2006 8/16/2006
Date
Figure 4-2. UF system filtrate production through the testing period.
86
-------
Figure 4-3 shows the feed and retentate pressures during the test and Figure 4-4 shows the
calculated IMP results. These figures show the impact of solids build up on the UF membranes
during operation.
35
30
25
5ft
320
s
10
5
0
RO Cleaning UF Cleaning RO Cleanin]
- Feed Pressure —•— Retentate Pressure (psig)
7/12/06 7/17/06 7/22/06 7/27/06 8/1/06 8/6/06 8/11/06 8/16/06
Date
Figure 4-3. Plot of UF system feed and retentate pressures over the testing period.
35
30
25
'SB
1 20
H 15
10
RO Cleaning UF Cleaning RO Cleaning
o
7/12/06 7/17/06 7/22/06 7/27/06 8/1/06 8/6/06 8/11/06 8/16/06
Date
Figure 4-4. Plot of UF system TMP over testing period.
87
-------
Water production for UF membrane systems is typically expressed in terms of flux (gfd) or
specific flux (gfd/psi) adjusted to a standard temperature of 20 °C. The use of specific flux
allows for comparison of filtrate production between various types of membranes and provides
data for determining the number of membrane modules and pressure needed to produce the
desired volume of water. The impact of temperature is accounted for by adjusting the data to a
standard temperature. Figure 4-5 shows the specific flux calculated for the UF system during the
test. The impact of solids buildup on the system between July 12 and July 29 is clear. As
described further in Section 4.2, the CIP in late July was successful, as the specific flux was
restored to a level equal to the start of the test.
2.50
2.00
o. 1.50
1
J2 1.00
c.
V)
0.50
RO Cleaning UF Cleaning RO Cleaning
7/12/06 7/17/06 7/22/06 7/27/06 8/1/06 8/6/06 8/11/06 8/16/06
Date
Figure 4-5. UF system specific flux over testing period.
4.5.1.1.2 UF Flow Rate, Filtrate Production, IMP, and Specific Flux Discussion
The EUWP stated that the net UF production design rate is 250,000 gpd (not including backwash
water) and the performance objective stated for this verification (Section 2.8) was 200,000 gpd.
Based on the mean net filtrate production of 178,000 gpd over the verification period, the UF
system did not achieve the specified design production rate or the verification performance
objective. The reason the specified production rate was not achieved was that the unit did not
operate a sufficient number of hours per day to meet the production goal. At a mean filtrate flow
rate of 229 gpm and accounting for a backwash volume of 900 gallons every 30 minutes, the unit
would need to operate an average of 21.3 hours per day to achieve a net filtrate production of
250,000 gpd and 17 hours per day to meet the verification performance objective of 200,000 gpd.
The UF system operated an average of 14 hours per day and the RO unit operated an average of
18.1 hours per day.
88
-------
The primary reason the UF system did not achieve the design production rate is that the UF
system automatically shutdown anytime the RO system feed water tank was full. The test was
designed to verify the entire system with both UF and RO in operation. Since the RO system has
less production capacity than the UF system, the UF system did not need to meet the stand alone
design specification of 250,000 gpd or performance target of 200,000 gpd. The UF produced
sufficient water to meet the RO system water requirements.
Occasionally, UF system downtime for maintenance and integrity testing, resulted in the RO
system being shutdown as well. This was due to the limited UF filtrate storage capacity (RO feed
water storage tank). With more storage capacity for UF filtrate, the UF system would have been
able to meet the feed requirements for the RO system during the daily integrity testing and
maintenance periods.
During the verification test, the operators were only on site during daytime periods. Therefore,
any time there was an alarm during the unattended hours, which shutdown either the UF or the
RO system, the units would remain shut down until an operator arrived the next morning. This
situation further reduced the operating hours of the UF system.
The UF system operated with an average filtrate flux of 36.5 gfd (temperature adjusted
normalized flux of 31.8 gfd). As shown in Figure 4-5, the specific flux started for the clean
system at 2.04 gfd/psi and dropped to 0.93 gfd/psi prior to cleaning. Figure 4-6 shows the loss
(or gain) of specific flux over the duration of the verification test. The loss (or gain) of 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 a
given day's specific flux to the start of the test. As can be seen, there was a steady loss of
specific flux before the first cleaning on July 30. The cleaning was successful and resulted in the
UF system having a specific flux after the cleaning similar to the start of the test.
u
01
o.
VI
St.
o
20
10
0
-10
-20
-30
-40
-50
-60
-70
RO Cleaning UF Cleaning RO Cleaning
7
7/12/06 7/17/06 7/22/06 7/27/06 8/1/06 8/6/06 8/11/06 8/16/06
Date
Figure 4-6. Loss of specific flux over time.
89
-------
Figure 4-7 shows the backwash record during the testing period. The UF unit is programmed to
perform a 205 second backwash cycle each 30 minutes if the Amiad strainers are not
backwashing. The timer restarts after the backwash is completed. Based on the UF hour meter
data, the backwash counter data, and the expected frequency of a backwash occurring every
33.42 minutes, it is apparent that the system was only achieving an average of 34% of the
backwashes. The UF system was not achieving the backwash rate anticipated, because at the start
of testing the Amiad strainers were attempting to backwash much more often to keep up with the
high solids loading from the WWTP. This prevented as much as 68% of the scheduled
backwashes. The PLC was set to skip UF backwashes if the Amiad strainer was backwashing.
Declines in the UF performance during the first two weeks of testing appear to be caused by the
increase in the time between backwashes. The backpressure on the Amiad Strainers was
increased from 40 to 70 psi on August 6* on instruction from Village Marine. This improved the
backwash frequency to 35% of scheduled, enabling a more consistent performance.
'S
+^
^ £
S3 M
PH g
vi >
1 «
•S PQ
« -O
it
| w
-^
O
-------
Table 4-12. RO System Operational Measurement Statistics
Parameter
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 (psi)
Array 1 Concentrate Pressure (psi)
Array 2 Feed Pressure (psi)
Array 2 Concentrate Pressure (psi)
Array 1 and 2 Combined Permeate
Pressure (psi)
Count
54
54
54
54
54
54
54
53
54
54
54
Mean
107
53
54
41
17
24
290
197
193
138
20
Median
107
55
53
41
18
23
293
199
195
138
19
Minimum
104
42
43
32
11
20
222
134
133
91
9
Maximum
110
64
67
48
22
29
366
263
261
182
42
Standard
Deviation
1.29
5.44
5.52
4.14
2.74
1.70
26.1
24.8
23.6
19.0
5.82
95% CI
±0.34
±1.45
±1.47
±1.10
±0.73
±0.45
±6.96
±6.67
±6.28
±5.06
+1.55
Figure 4-8 shows the daily flow rates for permeate and concentrate for both arrays. Figure 4-9
shows the feed water and concentrate pressures for both arrays. The RO system showed a
decrease in permeate flow rate for both arrays during the first two weeks of the verification test.
The feed water pressure to array 1 remained steady during this time, but the concentrate pressure
on array 1, which also impacts the feed pressure to array 2, decreased during these first two
weeks. Following the RO cleaning, flows and pressures returned to conditions similar to the start
of the test and remained steady until near the end of the test, when operators reduced the
recovery rate by lowering the feed pressure. This was done to reduce the loading to the RO. Feed
flow and pressure were increased again on the last day of the test.
The concentrate pressure from Array 1 was used by the energy conservation device to provide
the feed water pressure for Array 2. This energy saving device eliminated the need for a high
pressure pump for the Array 2 flow rate, which was approximately 32% of the Array 1 flow rate.
Without the energy saving device, additional pumping capacity and the associated energy use
would be required. The energy saving device achieved array 2 feed pressures that were similar to
the Array 1 concentrate pressures throughout the test. Based on the permeate flow rate from
Array 2 representing 24% of the RO water production (mean feed flow rate of 17 gpm out of a
mean 70 gpm total), it can be roughly estimated that the energy conservation device saved 25%
of the energy that would have been required if all the permeate was produced by high pressure
pumps.
91
-------
C.
80
70
60
50
40
30
20
10
RO Cleaning UF Cleaning
RO Cleaning
-Array 1 Perm. Flow(gpm) —A—Array 1 Cone Flow(gpm) —B—Array 2 Perm. Flow(gpm) —A—Array 2 Cone Flow(gpm)
o
7/12/06 7/16/06 7/20/06 7/24/06 7/28/06 8/1/06 8/5/06 8/9/06 8/13/06
Date
Figure 4-8. RO system flow rates.
400
350
300
250
200
150
100
50
0
RO Cleaning UF Cleaning RO Cleanu
-Array 1 Feed Pressure —A—Array 2 Feed Pressure —•—Array 1 Concentrate Pressure —B—Array 2 Concentrate Pressure
7/12/2006 7/19/2006 7/26/2006 8/2/2006
Date
8/9/2006
8/16/2006
Figure 4-9. RO system operating pressures.
Figure 4-10 shows the percent recoveries achieved by the RO system. Recoveries, calculated as
the permeate flow rate divided by the feed water flow rate, were consistent throughout the test.
The mean percent recovery for Array 1 was 50% with a median of 51%. The mean recovery for
Array 2 was 42% with a median of 43%. As expected, the recoveries for Array 2 were lower than
for Array 1, as Array 2 operates at a lower feed water pressure.
92
-------
70%
60%
Array 1 Recovery
Array 2 Recovery
7/12/06
7/19/06
7/26/06 8/2/06
Date
8/9/06
8/16/06
Figure 4-10. RO system percent recoveries.
Figure 4-11 shows the RO permeate production during the test and the total volume of feed water
pumped to the RO system. The RO produced 2,419 kgal of permeate from a total RO feed water
of 5,089 kgal. This yields an overall recovery of 48%.
RO Permeate Production —A— RO Feed Water Total Volume
7/12/06 7/17/06 7/22/06
7/27/06 8/1/06
Date
8/6/06
8/11/06 8/16/06
Figure 4-11. RO system permeate production and feed water volume.
93
-------
A common method of evaluating RO membrane performance is to calculate the specific flux,
which is based on the permeate flux and the 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. 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) = 0.6014x(conductivity)
The permeate flux was also adjusted for temperature to 25 °C, as is the convention.
(4-1)
Figure 4-12 shows the specific flux for the two RO system arrays based on NDP and adjusted to
a temperature of 25 °C. The decrease in the specific flux over the first two weeks of the test,
further indicates that the RO membranes were being fouled over time. After cleaning on July 24
to July 25, the specific flux remained steady for the next two weeks and then began to decline
once again near the end of the test. Based on the pattern established over the verification test
period, it would appear that the RO system would require cleaning every two to three weeks in
this type of application. If the leakage problem with UF system was resolved, then it would be
expected that the RO cleaning frequency would be reduced.
0.12
Specific Flux Array 1
Specific Flux Array 2
0.00
7/12/06 7/17/06 7/22/06 7/27/06 8/1/06 8/6/06 8/11/06 8/16/06
Date
Figure 4-12. RO system specific flux.
94
-------
4.5.1.2.2 RO Flow Rate, Pressures, and Specific Flux Discussion
The RO system did not achieve the permeate production of 100,000 gpd claimed in the statement
of performance. The mean permeate production for the 32 calendar days of operation was
78,000 gpd. The mean feed water flows of 107 gpm for Array 1 and 41 gpm for Array 2 were
below the target feed rates established in the test plan (Array 1 target 116 gpm and Array 2 target
was 58 gpm). The percent recovery for Array 1 of 50% equaled the target specification of 50%.
The Array 2 percent recovery of 42% was below the target specification of 48%. These
recoveries in conjunction with the feed water flows resulted in mean permeate flow rates of 53
gpm for Array 1 and 17 gpm for Array 2. At these flow rates, the RO unit would need to operate
an average of approximately 24 hours per day to meet the claimed target of 100,000 gpd. The
RO unit averaged 18 hours per day of operation during the test.
It was apparent during the test that the UF treated secondary wastewater was putting a heavier
load on the RO than initially expected. For this type of application, it appears that lower percent
recoveries and lower flows were achieved as compared to design specifications based on
groundwater and seawater. During the last few days of testing the recovery was set to 40% to
protect the system from heavy loading from and poor performance of the WWT. While this may
not have been necessary, it explains the drop in flows and pressure near the end of the test.
It should be noted that while the RO only achieved approximately 78% of the performance
objective for permeate production, additional operating time each day would have increased the
total production. As noted in the UF system discussion, operators were only present during
daylight hours and there was no coverage over night. Therefore, if an alarm sounded and
shutdown the unit, the system remained off-line until an operator arrived the next morning.
While it is not realistic to operate the RO unit continuously 24 hrs per day for several days,
additional operator coverage could increase operating hours and achieve permeate production
closer to the specified target.
4.5.1.3 Power Requirements and Efficiency
Figures 4-13 and 4-14 show the UF and RO system power requirements per hour and per volume
of water produced from the UF and RO. The values are comparable to those for water recycled
using RO presented by the Affordable Desalination Collaboration (ADC News Release May 4,
2006).
The efficiency of the high pressure RO pump motor is labeled as 92%. There are a variety of
ways to verify actual performance, one is to calculate the brake horse power (bhp) and/or water
horse power (whp), calculate the theoretical energy needed, and compare to the actual energy
used. The efficiency of the ERI device was calculated as the ratio of the Array 1 concentrate
pressure to the Array 2 feed pressure. The results of all three methods are shown in Figure 4-15.
Considering that the system is designed to operate at 800 - 1000 psi, rather than 300 psi, the
pump and energy recovery device did very well. Efficiencies greater than 1 are possible due to
the additional work by the ERI device.
95
-------
LTF Power kWh=38.025x
R2 = 0.9995
0 50 100 150 200 250 300
Hours of Operation
Figure 4-13. RO and UF power consumption over time.
350
400
450
30000
Total Power = 6.0399 Whr/gal RO Permeate
RO Power = 3.7753 Whr/gal RO Permeate
UF Power =1.401 Whr/gal RO Permeate
0 1000 2000 3000 4000 5000 6000
kgal Produced
Figure 4-14. RO and UF power requirements per kgal of RO permeate.
96
-------
e
UJ
1 A
1 "3
1 9
i i
i n
n Q
n 8
07
n f.
u.o
n -N -
A
n LJ - O "
u- n n _, ._ _
n p i-i '-H '-' A LJ L
tA - A n A p D A- .n^n :-. w
^teoCDoQjC^bfl^3 u^*'o *fi'^4«^<* a^A4^'fe&)CD£S
AA* .ja A AD Ojg
A *- • ft& . A .AA,
A DA A A A AA
A
OBHP AWHP
A O ERI • Recovery
on0/
y\j /o
8n%
?n°/
/U/o
A no/
H-U/o
i no/
lU/o
no/
o
u
7/12
7/17
7/22
7/27 8/1
Date
8/6
8/11
8/16
Figure 4-15. RO system energy efficiency calculated from bhp, whp, and energy recovery
based on total feed flow compared to the overall water recovery.
4.5.2 Task C2: Cleaning Efficiency
An important aspect of membrane operation is the ability to achieve long run times between
chemical cleanings (maintain up 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.
As shown in Figure 4-7 and discussed in Section 4.5.1.1.2, the automatic backwash system was
only operating between 30 and 35% of the time. This was due to frequent backwashing of the
Amiad filter ahead of the UF system. The PLC was programmed to not allow a UF system
backwash to occur if the Amiad strainer was in backwash mode. Adjustments were made to the
pressure setting on the Amiad strainer during the test to reduce the time it was in backwash mode
and increase the number of successful UF system backwashes.
97
-------
The UF system was expected to require CIP chemical cleaning about every 15 to 30 days. The
UF system was only cleaned once, after 15 calendar days of operation and then ran without
additional cleaning for 16 days until the end of the test. There were indications that the UF would
need cleaning again at the end of the test, as TMP was increasing.
The CIP for the UF system started on July 30 and continued through July 31, 2006. The specific
flux had dropped from 2.04 gpd/psi to 0.93 gpd/psi and the TMP had increased to 31 psi. The UF
system was still producing filtrate at an acceptable rate for overall system operation, but the TMP
had reached the target cleaning level of 30 psi and the specific flux had dropped by more than
50%. The operators were also informed that the wastewater facility had not activated a secondary
clarifier at increased flows, and that the secondary effluent was elevated in TSS. Based on these
facts, the CIP was initiated.
As shown in Figure 4-6 and Table 4-13, the CIP was successful in restoring the specific flux and
lowering the TMP of the UF system. The initial specific flux started at 2.04 gfd/psi and was
restored to 2.18 gfd/psi after the cleaning. TMP started at 18 psi, had increased to 31 psi, and was
restored to 16 psi after the cleaning.
Table 4-13. UF System Performance Parameter Values at Key Intervals
Date Range Specific Flux (gfd/psi) Transmembrane pressure (psi)
7/12 2~0418
7/17 1.17 21
7/19 1.81 19
7/29 0.93 31
8/1 2.18 16
8/16 1.42 22
Table 4-14 provides a summary of the performance parameters for the UF system and also a
history of the strainer and UF cleaning based on the operator logs and operating data.
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. Citric acid and sodium
hydroxide were added to water in the cleaning solution tank to make a pH 3 or pH 11 cleaning
solution. 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-
14 shows the amount of each chemical that was used for the cleaning. The CIP mixing tank
contained 270 to 300 gal. Each bank of modules was circulated with the each solution for 20 to
30 minutes. The membranes were then soaked overnight with the high pH solution.
98
-------
Table 4-14. Change in UF Performance with Cause and Action Taken
Date Range
7/13-7/17
Change in Specific
Flux
-42%
Change in
Transmembrane
pressure
+16%
Action
Cleaned &
Adjusted
Chemical Usage
7/17-7/19
7/19-7/29
8/1
8/1 - 8/15
-11%
-54%
+7%
-30%
+6%
+72%
-11%
+22%
backpressure on
Amiad strainer
Improved Strainer
performance
6.5 Ibs. Citric Acid
@40°C/2.3L
Cleaned UF System NaOH pH 14 plus
11.6 L bleach,
39°C
After Cleaning
End of Testing
4.5.2.2 RO Cleaning Frequency and Performance
The RO system had initially been expected to operate for 30 or more days before cleaning was
required. However, the operators were noticing difficulty maintaining the flows and recoveries
on the RO system. On July 24, the Gallup wastewater plant needed to perform maintenance
activities that interrupted the availability of treated effluent to the EUWP, so it was decided to
perform a RO CIP. The cleaning began on July 25 and was completed the afternoon of July 26,
2006.
The RO cleaning was performed using citric acid for a low pH cleaning and MemClean
detergent cleaner for an alkaline cleaning. Citric acid was added to the 300 gallon CIP tank to
achieve a pH of 3.03. The acid solution was circulated through the RO and then it was allowed to
soak for approximately one hour. The ending pH was 4.07. A total of 7 kg of MemClean
detergent was then added directly into the CIP tank. The system was circulated for a total of 18
minutes and then allowed to soak overnight (15 hours). The ending pH was 11.14. After the
overnight soak, the RO was circulated for 30 minutes and then flushed with permeate. Citric acid
(10 ounces) was then added to the CIP tank and circulated through the system to neutralize the
unit. Finally, the system was then flushed with permeate for 1.5 hours and readied for return to
operation.
Table 4-15 shows the specific flux results for Arrays 1 and 2 before and after the CIP procedure.
The CIP restored Array 1 from a specific flux of 0.057 gfd/psi before the cleaning to 0.063
gfd/psi, which is a recovery of 98%. For Array 2, the CIP restored the membranes to a specific
flux of 0.064 gfd/psi from 0.060 gfd/psi, also yielding a recovery of 98%.
99
-------
Table 4-15. RO System Performance Intervals
Pressure Difference
Specific Flux (Feed to
(gfd/psi) Concentrate) (psi)
Date Array 1 Array 2 Array 1 Array 2 Event Chemical Usage
7/12
7/23
7/23
7/24
7/26
8/2
8/7
8/8
8/15
0.064
0.058
0.059
0.057
0.063
0.057
0.058
0.059
0.049
0.065
0.063
0.068
0.060
0.064
0.058
0.060
0.060
0.045
50.5
130.5
129.5
126.5
81.5
91
124
86.5
116.5
40.5
67.5
40.5
38.5
37
54
58.5
53
60.5
Baseline
Before Product Flush
(@59% Recovery)
After Product Flush
RO Cleaning
After Cleaning
After UF Cleaning
RO Cleaning
After Cleaning
Final Value
1.2 kg Citric/
7 kg MemClean
6.3 kg MemClean
The RO was cleaned with detergent only on August 7 to 8, 2006. The system was soaked
overnight. As shown in Table 4-15, this cleaning did not change the specified flux. However, the
pressure differential between the feed pressure and the concentrate pressure did decrease and
improve operation.
The operators noted that the RO system cleaning appeared to be more effective for Array 1 than
for Array 2, the ERI array. It is not possible to clean the two arrays separately. The RO
forwarding pump (P5) is used for RO cleaning, not the high pressure pump (P6). The maximum
flow possible is 25-30 gallons per minute. This is not enough for a good cleaning cycle as is
possible with the UF system that uses its production pump for the cleaning cycle. Also the
cleaning tank thermometer is in the recycle line for the UF system, not in the tank. It was moved
to provide an operating temperature reading. Unfortunately, this action disabled heating
capability when cleaning the RO system.
Table 4-16 shows additional information on the chemicals use and time for the CIP.
Table 4-16. RO and UF System Cleanings
Date
7/25/06
7/25/06
8/1/06
8/1/06
8/7/06
System
RO
RO
UF
UF
RO
Chemical
Citric 1.1 8 kg
MemClean 7 kg
Citric 2.95 kg
Chlorine 11.6L and
NaOH 2.28 L
MemClean 6.3 kg
Temp (°C)
21.6-30.1
37
37-40
37.8-40.3
23.2-24.2
pH
3.03-4.07
11.3
2.92-3.06
9.9-10.73
10.2-11.46
Flow
(gpm)
N/A
N/A
516-564
520-566
N/A
Duration
(hr:min)
N/A
17:15*
1:20
2:10
15:05*
* System left to soak overnight. Flow data are not available during cleaning cycle.
100
-------
4.5.3 TaskCS: Finished Water Quality
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 USEPA National
Drinking Water Regulations. 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 particulate 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 measured with in-line meters in
the EUWP and were measured with portable equipment on site. In addition, pH and temperature
were measured on site. Other water quality parameters were monitored by collecting samples on
a weekly basis.
Samples were also collected for bacteriological analyses. Data for the bacteriological samples
(total coliform, fecal coliform, E. coli, and HPC) are also presented in this section.
4.5.3.1 Water Quality Results - Turbidity, Conductivity, pH, and Temperature
Figures 4-16 and 4-17 present the grab sample turbidity results for the UF feed, UF filtrate, and
RO permeate over the duration of the test. Table 4-17 shows a summary of the daily turbidity
results for the grab samples taken during the verification test. Based on the grab samples, the UF
system reduced turbidity from a mean of 11.1 NTU in the feed water to a mean of 0.74 NTU in
the UF filtrate. The 95% confidence level shows that filtrate turbidity can be expected to be in
the range of 0.62 to 0.86 NTU. As discussed Section 4.5.1, the UF system was found to have
faulty seals, which may explain the lower than expected reductions of contaminants.
£3
H
=
H
tu
£3
7/12/06 7/17/06 7/22/06
Figure 4-16. UF feed water turbidity.
7/27/06 8/1/06
Date
8/6/06
8/11/06
8/16/06
101
-------
2.5
3
2
1.5
0.5
-UF Filtrate (NTU) Handheld
-RO Permeate (NTU) Handheld
o
7/12/2006 7/17/2006 7/22/2006 7/27/2006 8/1/2006 8/6/2006 8/11/2006 8/16/2006
Date
Figure 4-17. UF filtrate and RO permeate turbidity handheld meter.
Table 4-17. Summary Statistics for Handheld Turbidity Meter Results
Parameter
Mean:
Median:
Minimum:
Maximum:
Count:
Std. Dev.:
95% CI:
UF Feed
(NTU)
11.1
8.55
3.7
41
52
8.6
2.3
UF Filtrate
(NTU)
0.74
0.58
0.22
2.3
51
0.44
0.12
RO Feed
(NTU)
0.80
0.58
0.11
3.7
52
0.62
0.17
RO Permeate
(NTU)
0.15
0.12
0.02
0.44
51
0.09
0.02
The RO permeate had a mean turbidity of 0.15 NTU based on the handheld meter readings. The
95% confidence interval for the handheld meter results showed expected ranges of 0.13 to 0.17
NTU for the RO permeate. The RO permeate turbidity levels based on the handheld meter
results, did not quite meet the National Primary Drinking Water Regulation (NPDWR)
(<0.3 NTU 95% of the time and all values below 1.0 NTU). All results were less than 1.0 NTU,
but three results out of 51 data points recorded were above 0.3 NTU, giving 94% of the data
being less than 0.3 NTU. As described below, the handheld meter turbidity results for the UF
feed and filtrate were similar to the in-line turbidimeters results, however, the in-line results for
the RO permeate showed much lower turbidity levels. The in-line meters were more sensitive
and had a lower detection limit. Additional discussion of the turbidity detection limits and
sensitivity is presented in the QC Section 4.6.
The operators manually recorded in-line turbidity measurements at least once per day. The feed
water turbidity, as recorded from the in-line analyzer, showed a mean value of 8.7 NTU with a
median of 7.5 NTU. The UF filtrate in-line analyzer recorded readings showed a mean turbidity
of 0.69 NTU with a median value of 0.53 NTU. The RO permeate turbidity, as manually
102
-------
recorded from the in-line analyzer had a mean value of 0.016 NTU and a median value of
0.015 NTU. Figure 4-18 shows UF feed and UF filtrate in-line turbidity readings. Note that there
are two y-axes (different scales) in the figure, one for the feed and one for the filtrate. Figure 4-
19 shows the RO feed and permeate in-line analyzer results. It should be noted that only the
manually recorded UF filtrate, RO feed and RO permeate in-line analyzer results are shown in
the figures after July 27. Unfortunately, the in line turbidity data for these process streams was
inadvertently erased for the period July 27 through the end of the test. Table 4-18 shows the
summary statistics for the UF feed, UF filtrate, and RO permeate in-line turbidity readings, as
recorded on a daily basis by the operators.
40
35
0
» UF Feed
• UF Filtrate
n UF Filtrate Manual Reading
6000
-- 5000
-- 4000
3000
H
E
=
H
•a
O>
O>
tu
O
eg
-- 2000 -g
tu
tu
-- 1000
7/7 7/12 7/17 7/22 7/27 8/1 8/6 8/11 8/16 8/21
Date
Figure 4-18. UF feed and UF filtrate/RO feed turbidity in-line meter.
103
-------
m P rr 11 n in mn i n i n n~i
o UF Filtrate
» UF Filtrate Manual Reading
• RO Permeate
D RO Permeate Manual Reading
500
450
400
350 H
300 >.
.±
250 2
H
4*
i
200 S
150
100
50
E
BH
i
81
So
811
8/16
821
Figure 4-19. UF filtrate and RO permeate in-line turbidity readings.
Table 4-18. Summary Statistics for In-line Turbidity Meter Manually Recorded Results
Date
Mean:
Median:
Minimum:
Maximum:
Count:
Std. Dev.:
95% CI:
UF Feed
(NTU)
8.8
7.5
0.62
39
52
7.5
2.0
UF Filtrate
(NTU)
0.69
0.53
0.026
4.0
48
0.69
0.20
RO Permeate
(NTU)
0.016
0.015
0.013
0.029
51
0.003
0.001
While the UF system did not achieve the results expected, the RO system handled the increased
loading and did serve as the ultimate barrier to ensure low turbidity product water was produced.
The RO permeate turbidity levels based on the manually recorded in-line meter results show that
the system did meet the National Primary Drinking Water Regulation (NPDWR) (<0.3 NTU
95% of the time and all values below 1.0 NTU).
The Long-Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR) states that if the
filtrate turbidity exceeds 0.15 NTU over any 15-minute period, the system must be shut down
and a direct integrity test performed. The EUWP was not set up to be compliant with the
LT2ESWTR, as the in-line turbidity meters are not tied to an automatic system shutdown if the
turbidity level exceeds 0.15 NTU for any 15-minute period. The in-line turbidity data was logged
onto a laptop computer, and the computer is not connected to the EUWP for the purpose of
shutting down the system. The EUWP was designed and built before LT2ESWTR and did not
have the necessary control equipment to be compliant with the LT2 monitoring requirements.
104
-------
However, the RO system produced permeate with turbidity below the LT2ESWTR action level
of 0.15 NTU most of the time. However, there are few data points recorded from the in-line
meters (during the period July 12-27) when single data points exceeded the 0.15 NTU action
level. It is not possible to determine if the action level was exceeded after the CIP was performed
on July 25-26, as the 15-minute increment data for the remainder of the test was erased. All of
the manually recorded data (once or twice per day) were 5 to 10 times lower than the 0.15 NTU
action level. This would suggest that the RO permeate did meet the LT2ESWTR turbidity
requirements, but it cannot be confirmed, because the computer file with the continuous (15-
minute increment) data was erased.
The RO system reduced the dissolved ions in the feed water, as measured by conductivity. The
mean conductivity in the RO permeate was 14.2 |iS/cm compared to the mean conductivity in
the RO feed water of 1,726 jiS/cm. The RO unit reduced the conductivity by a mean value of
99.2%. Table 4-19 shows the conductivity results for the UF and RO systems, and the summary
statistics for the verification test. The direct measurement of TDS, presented later in Table 4-27,
shows that the mean TDS concentration in the RO permeate was 5.0 mg/L compared to the mean
RO feed water TDS of 1,113 mg/L. The overall TDS rejection was 99.6%.
Table 4-19. Conductivity Results
Date
7/12/06
7/12/06
7/13/06
7/13/06
7/14/06
7/15/06
7/16/06
7/16/06
7/17/06
7/18/06
7/18/06
7/19/06
7/19/06
7/20/06
7/20/06
7/21/06
7/22/06
7/22/06
7/23/06
7/26/06
7/27/06
7/27/06
7/28/06
7/29/06
7/29/06
8/1/06
UF Feed
(uS/cm)
1733
1726
1747
1739
1779
1729
1729
1714
1663
1926
1927
1952
1768
1787
1760
1860
1850
1705
1700
1741
1682
1672
1669
1677
1667
1665
UF Filtrate
(uS/cm)
1722
1726
1742
1751
1767
1723
1705
1718
1664
nr
1915
1924
1658
1777
1755
1860
1780
1702
1695
1653
1680
1666
1665
1675
1670
1661
RO Feed
(uS/cm)
1719
1729
1745
1755
1769
1728
1714
1711
1658
1857
1918
1932
1758
1774
1760
1850
1773
1715
1694
1647
1676
1669
1664
1663
1667
nr
RO Permeate
(uS/cm)
9.34
9.81
9.2
9.81
9.41
9.1
9.22
9.51
9.47
8.55
73.69
69.13
11.48
10.25
11.15
22
39.4
21.72
10.5
15.4
10.5
10.12
10.4
10.55
10.01
nr
RO Concentrate
(uS/cm)
3637
3640
3650
3684
3717
3626
3655
3740
3512
3277
3921
3973
3749
3754
3690
3830
3180
2800
2879
2704
2760
3226
3159
3172
3207
nr
RO%
Conductivity
Red.
99.5
99.4
99.5
99.4
99.5
99.5
99.5
99.4
99.4
99.6
96.2
96.5
99.4
99.4
99.4
98.8
97.9
98.7
99.4
99.1
99.4
99.4
99.4
99.4
99.4
nc
105
-------
Table 4-19. Conductivity Results
Date
8/2/06
8/3/06
8/3/06
8/4/06
8/4/06
8/5/06
8/5/06
8/6/06
8/6/06
8/7/06
8/8/06
8/9/06
8/9/06
8/10/06
8/11/06
8/11/06
8/12/06
8/12/06
8/13/06
8/13/06
8/14/06
8/14/06
8/15/06
8/15/06
8/16/06
8/16/06
Mean:
Median:
Minimum:
Maximum:
Count:
Std. Dev.:
95% CI:
UF Feed
(uS/cm)
1771
1734
1721
1739
1752
1717
1710
1721
1708
1660
1700
1730
1727
1716
1709
1707
1677
1626
1718
1798
1717
1691
1732
1715
1414
1706
1729
1720
1414
1952
52
79.7
±21.7
UF Filtrate
(uS/cm)
1725
1736
1726
1770
1750
1723
1714
1716
1702
1654
1702
1722
1725
1685
1704
1711
1667
1627
1708
1788
1705
1696
1721
1715
1716
1710
1721
1715
1627
1924
51
57.7
±15.85
RO Feed
(uS/cm)
1720
1725
1721
1754
1751
1721
1711
1715
1706
1656
1699
1721
1719
1688
1710
1713
1672
1631
1701
1766
1706
1693
1723
1715
1715
1712
1726
1715
1631
1932
51
59.4
±16.3
RO Permeate
(uS/cm)
10.85
10.85
10.6
10.62
14.75
10.04
9.75
9.37
11.02
9.08
11.15
9.96
10.96
10.85
10.02
10.87
9.34
9.87
16.73
21.58
10.2
11.05
10.2
12.33
10.94
10.72
14.2
10.5
8.55
73.7
51
12.7
±3.48
RO Concentrate
(uS/cm)
3246
3316
3280
3355
3320
3288
3266
3230
3194
3092
3107
3311
3363
3276
3342
3368
3248
3108
3431
3513
2968
3015
3035
3053
3021
3368
3338
3288
2704
3973
51
300
±82.4
RO%
Conductivity
Red.
99.4
99.4
99.4
99.4
99.2
99.4
99.4
99.5
99.4
99.5
99.3
99.4
99.4
99.4
99.4
99.4
99.4
99.4
99.0
98.8
99.4
99.3
99.4
99.3
99.2
99.4
99.2
99.4
96.2
99.6
51
0.65
±0.18
nr - not recorded
nc - not calculated
Tables 4-20 and 4-21 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 water having a mean
pH of 7.53 (median 7.55) and the filtrate having a mean pH of 7.54 (median 7.55). The RO
system did lower the pH of the permeate, with a mean of 6.27 (median 6.14), and a range of 5.38
to 7.30.
The UF and RO systems also did not have an effect on the temperature of the water as it passed
through the EUWP. The feed water temperature ranged from 23.9 °C to 28.9 °C, with a mean of
106
-------
26.6 °C. The mean temperature of the RO permeate was 26.8 °C, with a range of 20.7 °C to 31.3
°C. Temperature variation and impact on membrane 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 4.5.1.1.1 and 4.5.1.2.1 The temperature data in Table 4-21 served as the basis for the
temperature adjustment calculations.
Table 4-20. pH Results
Date
7/12/06
7/12/06
7/13/06
7/13/06
7/14/06
7/15/06
7/16/06
7/16/06
7/17/06
7/18/06
7/18/06
7/19/06
7/19/06
7/20/06
7/20/06
7/21/06
7/22/06
7/22/06
7/23/06
7/26/06
7/27/06
7/27/06
7/28/06
7/29/06
7/29/06
8/1/06
8/2/06
8/3/06
8/3/06
8/4/06
8/4/06
8/5/06
8/5/06
8/6/06
8/6/06
8/7/06
8/8/06
8/9/06
8/9/06
UF Feed
7.94
7.84
7.49
7.67
7.77
8.5
7.57
8.09
7.57
7.58
7.4
7.23
7.45
7.62
7.31
7.75
7.8
7.68
7.71
7.46
7.6
7.6
7.66
7.61
7.58
7.47
7.40
7.56
7.54
7.45
6.89
7.68
7.73
7.61
7.42
7.58
7.41
7.4
7.51
UF Filtrate
7.93
7.84
7.47
7.45
7.6
8.38
7.62
8.01
8.14
nr
7.4
7.2
7.38
7.6
7.36
7.72
7.8
7.62
7.45
7.64
7.55
7.61
7.55
7.58
7.6
7.59
7.68
7.51
7.44
7.65
6.96
7.53
7.9
7.75
7.56
7.67
7.58
7.55
7.38
RO Feed at
Strainer
7.77
7.85
7.45
7.62
7.41
8.33
7.46
7.66
7.68
7.4
7.4
7.27
7.3
7.52
7.31
7.44
7.6
7.43
7.6
7.43
7.6
7.49
7.55
7.42
7.33
7.08
7.48
7.51
7.45
7.43
6.88
7.59
7.72
7.7
7.37
7.63
7.51
7.47
7.43
RO 1st Pass
Permeate
6.6
7.24
5.87
6.83
6.98
6.85
6.64
7.21
7.24
6.62
6.75
6.52
5.79
5.82
5.67
6.07
6.2
6.15
6.05
6.19
6.3
6.55
6.21
6.2
6.0
Nr
6.0
7.3
5.92
6.08
5.55
6.02
6.59
6.14
5.95
6.07
6.01
5.46
6.06
RO Concentrate
8.06
8.02
7.73
7.83
7.9
8.49
7.66
8.04
8.26
7.4
7.69
7.34
7.55
7.75
7.55
7.8
7.8
7.65
7.75
7.65
7.72
7.56
7.72
7.73
7.62
nr
7.69
7.75
7.65
7.64
7.14
7.8
7.86
7.87
7.63
7.8
7.67
7.7
7.64
107
-------
Table 4-20. pH Results
Date
8/10/06
8/11/06
8/11/06
8/12/06
8/12/06
8/13/06
8/13/06
8/14/06
8/14/06
8/15/06
8/15/06
8/16/06
8/16/06
Mean:
Median:
Minimum:
Maximum:
Count:
Std. Dev.:
95% CI:
UF Feed
7.42
7.49
7.41
7.44
7.47
7.15
5.99
7.56
7.34
7.69
7.46
7.54
7.47
7.53
7.55
5.99
8.5
52
0.32
±0.087
UF Filtrate
7.44
7.52
7.39
7.49
7.44
7.22
6.13
7.45
7.39
7.43
7.43
7.38
7.4
7.54
7.55
6.13
8.38
51
0.31
±0.084
RO Feed at
Strainer
7.37
7.46
7.34
7.42
7.37
7.09
6.17
7.26
7.3
7.36
7.29
7.49
7.34
7.44
7.44
6.17
8.33
52
0.28
±0.075
RO 1st Pass
Permeate
6.01
5.52
6.28
6.1
6.04
5.75
5.38
6.55
6.13
7.21
5.82
6.75
6.61
6.27
6.14
5.38
7.3
51
0.49
±0.135
RO Concentrate
7.61
7.65
7.47
7.61
7.56
7.31
6.46
7.44
7.52
7.58
7.55
7.61
7.58
7.67
7.65
6.46
8.49
51
0.28
±0.077
nr - not recorded
Table 4-21.
Date
7/12/06
7/12/06
7/13/06
7/13/06
7/14/06
7/15/06
7/16/06
7/16/06
7/17/06
7/18/06
7/18/06
7/19/06
7/19/06
7/20/06
7/20/06
7/21/06
7/22/06
7/22/06
7/23/06
7/26/06
7/27/06
7/27/06
Temperature
UF Feed
(°C)
26.5
27.7
25.8
27.8
25.5
26.9
25.1
28.1
28.9
27.2
26.7
25.2
26.4
25.3
28
25.2
26.5
27
27.8
27
26.2
28.2
Results
UF Filtrate
(°C)
26.9
28
25.8
27.1
26
26.6
26.1
28.2
28.4
nr
26.6
25.4
26.3
26.3
28.2
24.8
26.7
26.8
27.1
26.6
26.2
28
RO Feed
(°C)
27.0
27.6
25.5
26.8
26.1
26.5
25.7
27.8
28.5
27.9
26.1
25.9
26.7
25.9
28.1
24.9
26.6
26.7
27.3
26.9
26.4
28.2
RO Permeate
(°C)
26.7
28.3
25.5
27
26.4
27.1
26.7
28.1
31.3
28
27
26
26.6
26.2
28.9
25.8
27.3
26.2
27.6
26
26.7
28.4
RO Concentrate
(°C)
27.4
28.2
26.2
26.9
26.4
27.1
26.7
28.2
28.8
28.2
27.1
26.5
27
26.4
28.4
25.4
26.6
27.3
27.7
26.9
27
28.5
108
-------
Table 4-21. Temperature Results
Date
7/28/06
7/29/06
7/29/06
8/1/06
8/2/06
8/3/06
8/3/06
8/4/06
8/4/06
8/5/06
8/5/06
8/6/06
8/6/06
8/7/06
8/8/06
8/9/06
8/9/06
8/10/06
8/11/06
8/11/06
8/12/06
8/12/06
8/13/06
8/13/06
8/14/06
8/14/06
8/15/06
8/15/06
8/16/06
8/16/06
Mean:
Median:
Minimum:
Maximum:
Count:
Std. Dev.:
95% CI:
UF Feed
(°C)
26.9
26.3
27.2
26.0
27.7
26.7
26.8
26.3
26.6
26
27.5
25.7
25.8
25.1
28.1
25.2
27.4
27.8
25.3
28.7
25.1
26.8
23.9
27.5
24.8
28.1
24.6
27.1
25.6
27.3
26.6
26.7
23.9
28.9
52
1.15
±0.31
UF Filtrate
(°C)
26.9
25.8
26.4
25.9
27.7
26.1
27.7
24.9
26.2
25.1
27.5
26.2
25.9
25.7
27.4
25.4
27.5
27.9
26
28.3
25.2
26.5
24.5
27.4
24.9
27.7
25.2
27.1
25.1
27.1
26.5
26.5
24.5
28.4
51
1.04
±0.29
RO Feed
(°C)
27.3
26
26.8
26.2
27.7
26.7
27
26
26.7
25.7
27.7
26.5
26.2
25.9
27.7
26.1
27.7
28.2
26
28.3
25.6
26.9
24.6
27.9
24.8
28
25.1
27.3
25.5
27.2
26.7
26.7
24.6
28.5
52
0.98
±0.27
RO Permeate
(°C)
27
26.6
27.2
nr
29.8
26.9
27.6
25.7
26.7
25.1
27.9
25.9
26.2
25.5
28.2
25.5
27.5
28.4
24.4
28.6
25.4
26.9
20.7
28.3
23.2
28.3
25.2
27.5
24.3
27.7
26.8
26.9
20.7
31.3
51
1.67
±0.46
RO Concentrate
(°C)
27.6
26.8
27.4
nr
28.1
27.2
27.4
26.5
27.2
26.6
28.8
27
26.4
26.4
28.1
26.7
28.1
28.6
26.6
28.8
26
27.4
25.1
28.4
25.9
28.4
25.8
27.5
23.3
27.9
27.2
27.1
23.3
28.8
51
1.06
±0.29
nr - not recorded
4.5.3.2 Other Water Quality Results - UF System
UF feed and filtrate water general water quality statistics are shown in Table 4-22. Parameters
that improved significantly by UF treatment, based on the 95% confidence intervals, are BOD,
DOC, TOC, TSS, VSS, and iron.
Table 4-23 presents the biological analyses results for the UF system. Since the UF
interconnectors were leaking, the removals across the UF system are not what would be
109
-------
expected. In most cases, there was less than one log removal of the various bacteriological
indicators analyzed.
UF retentate TOC and TSS, and backwash TSS statistics are listed in Table 4-24. The variability
of backwash TSS is due to the natural variation in TSS loading over time during a backwash. It
is difficult to get a representative sample from the large backwash flows from this system.
110
-------
Table 4-22. UF Feed and Filtrate General Water Quality Analysis
i
s
^
•£
1
JS
—
o3
1
W
Parameter
pH (pH Units)
Bicarbonate
Carbonate
BOD
COD
Color (Color Units)
Ammonia
Bromide
Chloride
DOC
Sulfate
Free Cyanide
Hardness
Conductivity (|aS/cm)
Total Alkalinity
TDS
TOC
TSS
vss
Count
8
8
8
8
8(i)
8
8
8(3)
8
8
8
8
8
8
8
8
8
8(3)
gd)
(1) Count and statistics include three
Mean
7.8
250
<5
6.2
24
71
0.10
0.21
100
7.3
330
O.01
45
1,700
250
1,100
7.1
14
7.7
UF Feed
Standard
Deviation
0.1
9
NA
1.8
7.7
50
NA
0.02
6.6
0.7
16
NA
6.0
46
9.3
35
0.2
9.9
5.2
95th Percentile
Upper Lower
7.8
256
NA
7.5
29
106
NA
0.22
105
7.8
341
NA
49
1,732
256
1,125
7.3
21
11.3
7.8
244
NA
4.9
19
36
NA
0.20
96
6.8
319
NA
41
1,668
244
1,076
6.9
7.1
4.1
Count
8
8
8
8
8(2)
8
8
8(3)
8
8
8
8
8
8
8
8
8
8
Mean
7.8
240
<5
<2.0
20
34
0.10
0.21
110
4.9
370
44
1,700
240
1,100
6.6
<4.0
<4.0
UF Filtrate
Standard
Deviation
0.1
12
NA
NA
7.3
5.6
NA
0.02
6.7
2.7
103
5.2
76
9.2
46
0.2
NA
NA
95th Percentile
Upper Lower
7.8
248
NA
NA
25
38
NA
0.22
115
6.8
442
48
1,752
246
1,132
6.8
NA
NA
7.8
232
NA
NA
15
30
NA
0.20
105
3.0
299
40
1,648
234
1,068
6.4
NA
NA
estimated results below the reporting limit.
(2) Count and statistics include four estimated results below the reporting limit.
(3) Count and statistics include one estimated result below the
reporting limit.
111
-------
Table 4-22 (cont'd). UF Feed and Filtrate General Water Quality Analysis
Parameter
3
55.
M
^
"3
"8
•a
o
"3
£
a
— i a ' — v
^ ^ f^
"^ -^ ^v
O OJ 6*
Aluminum
Boron
Calcium
Iron
Potassium
Magnesium
Manganese
Sodium
Silica
Strontium
Zinc
Iron
Manganese
Phosphorus
Silica
Count
7
7
7
70)
7
7
7(D
7
7
6
7
7
7
7
7
Mean
<100
310
13,000
110
14,000
3,000
8.5
350,000
20,000
150
67
1,600
17
4,400
21,000
UF Feed
Standard
Deviation
NA
14
690
64
787
315
4.4
13,452
900
26
29
1,498
9.2
872
900
95th Percentile
Upper
NA
320
13,511
157
14,583
3,233
12
359,965
20,667
171
89
2,710
24
5,046
21,667
Lower
NA
300
12,489
63
13,417
2,767
5.2
340,035
19,334
130
45
490
10
3,754
20,334
Count
6
7
7
7
7
7
6
7
7
7
7
7
7
7
7
Mean
<100
310
13,000
<100
14,000
3,100
13
350,000
20,000
140
52
240
14
3,800
21,000
UF Filtrate
Standard
Deviation
NA
19
976
NA
900
364
3.9
7559
1,134
16
5.9
179
5.1
931
1,380
95th Percentile
Upper
NA
324
13,723
NA
14,667
3,370
16
355,600
20,840
152
56
373
18
4,490
22,022
Lower
NA
296
12,277
NA
13,334
2,831
10
344,400
19,160
128
48
107
10
3,111
19,978
(1) Count and statistics include three estimated results below the reporting limit.
112
-------
Table 4-23. Biological Analysis of UF System
Fecal Coliform (MPN/100 mL)
Date
07/13/06
07/17/06
07/18/06
07/20/06
07/26/06
07/27/06
08/02/06
08/03/06
08/07/06
08/08/06
08/10/06
08/14/06
08/15/06
08/16/06
UF Feed
5.0E+05
4.0E+04
4.0E+04
2.0E+04
2.0E+04
2.0E+04
2.6E+05
2.3E+05
<2.0E+04
4.0E+04
5.0E+05
1.1E+05
4.0E+04
UF Filtrate
1.6E+04
8.0E+04
2.3E+03
5.0E+03
2.2E+03
4.0E+03
7.0E+03
1.6E+04
2.3E+03
3.0E+03
2.4E+04
UF Retentate
1.6E+04
3.0E+04
1.6E+05
2.4E+04
1.6E+05
UF Backwash
2.0E+04
2.0E+04
1.2E+05
Date
07/13/06
07/20/06
07/26/06
08/02/06
08/03/06
08/10/06
08/16/06
UF Feed
3.0E+05
2.0E+04
2.0E+04
1.4E+05
4.0E+04
<2.0E+04
E. coll
UF Filtrate
3.0E+03
2.0E+04
2.0E+03
1.4E+03
4.0E+04
<2.0E+04
(MPN/100 mL)
UF Retentate
1.6E+04
1.7E+04
1.6E+05
2.4E+04
1.3E+04
UF Backwash
<2.0E+04
2.0E+04
<2.0E+04
Date
07/13/06
07/20/06
07/26/06
07/27/06
08/02/06
08/03/06
08/10/06
08/16/06
UF Feed
4.0E+07
9.3E+06
5.3E+06
7.2E+06
3.3E+07
1.3E+07
2.0E+07
HPC
UF Filtrate
1.2E+07
3.8E+06
1.3E+06
6.8E+05
2.1E+07
6.2E+06
1.5E+07
(CFU/100 mL)
UF Retentate
9.4E+06
9.6E+06
1.9E+07
1.8E+07
2.0E+07
UF Backwash
2.1E+07
1.5E+07
5.5E+07
Total Coliform (MPN/100 mL)
Date
07/13/06
07/17/06
07/18/06
07/20/06
07/26/06
07/27/06
08/02/06
08/03/06
08/07/06
08/08/06
08/10/06
08/14/06
08/15/06
08/16/06
UF Feed
1.6E+06
8.0E+03
4.0E+04
1.3E+05
8.0E+04
4.0E+04
2.2E+06
7.0E+05
8.0E+04
8.0E+04
3.0E+04
1.7E+05
5.0E+05
UF Filtrate
1.6E+04
2.3E+03
2.3E+03
3.0E+04
1.4E+04
5.0E+03
3.0E+04
2.4E+04
1.3E+06
1.4E+04
9.0E+04
UF Retentate
1.6E+04
3.0E+04
1.6E+05
1.6E+05
1.6E+05
UF Backwash
1.7E+05
1.1E+05
3.3E+05
Table 4-24. UF Retentate and Backwash Analysis
113
-------
UF Retentate TOC
UF Retentate TSS
UF Backwash TSS
Count
6
6
6
Mean
7.38
18.2
31
Standard
Deviation
0.26
5.4
29
Upper
7.59
22.5
54.4
95% CI
Lower
7.17
13.9
8.0
4.5.3.3 Other Water Quality Results - RO System
A summary of the water quality results and statistics for the RO system feed water, concentrate,
and permeate are reported in Table 4-25. Most constituents were reduced in the permeate to
below the reporting limit. Table 4-26 shows the rejection achieved for various water quality
parameters, based on the mean concentrations. Some contaminants showed low rejection rates
due to low concentrations in the RO feed water, and higher reporting limits resulting from the
wastewater matrix.
After RO treatment, the final production water (RO permeate) met all primary and secondary
drinking water standards. The RO unit served as the an effective treatment system for removing
inorganic and organic constituents present in the secondary wastewater, based on meeting the
objective of achieving a treated water that met primary and secondary drinking water standards.
To be acceptable for transmission or drinking, the RO permeate would need stabilization and
residual chlorination.
The LSI of the concentrate was 0.5 with 28% saturation of silica, 3% saturation of calcium
sulfate, and 2% for strontium sulfate. Antiscalant is recommended. In this test, Nalco
Permatreat PC-191 was used at a dose of 3.0 mg/L. RO permeate requires stabilization with 7
mg/L sodium bicarbonate, 7 mg/L sodium carbonate, and 7 mg/L of calcium hydroxide to attain
an LSI of zero with a pH of 9.5.
Mass balances for selected inorganic constituents (Na, Ca, Mg, SO/i, HCOs, Cl, and TDS) were
calculated to evaluate possible buildup of salts in the membranes. The results are listed in Table
4-27. Inorganic mass balances indicate that outgoing salts were on average 8% higher
concentration than the incoming salts. Therefore, there does not appear to have been a
significant accumulation of salts in the RO system.
The results for the biological analysis of UF filtrate (RO Feed), RO permeate, and concentrate
are listed in Table 4-28. Biological analyses were performed for fecal and total coliforms, E.
co//', and UPC. Enteric virus counts were measured for one set of UF feed, UF filtrate, and RO
permeate samples. The enteric virus results showed 176 MPN/100 mL in the RO feed and <1
MPN/100 mL in the RO permeate. Coliform species were present in the feed water in great
enough numbers to allow for a log reduction value greater than 3 from the UF filtrate to the RO
permeate.
114
-------
Table 4-25. RO Feed, Permeate, and Concentrate - General Chemistry
Parameter
(mg/L)
pH (pH units)
Bicarbonate
Carbonate
Color (Color
Units)
Ammonia
Bromide
Chloride
DOC
Sulfate
Fluoride
Free Cyanide
Hardness
Nitrate
Nitrite
Conductivity
(|4,S/cm)
Total Alkalinity
TDS
TOC
TSS
Count Mean
8 7.7
8 250
8 <5
8 37
8 0.10
8(2) 0.21
8 110
8 6.6
8 340
8 1.1
8 <0.01
8 69
8 18
8 <0.5
8 1,600
8 250
8 1,100
8 6.6
8 <4.0
RO Feed
Std.
Dev.
0.0
12
NA
5.3
NA
0.02
5.2
0.2
27
0.3
NA
69.0
1.8
NA
52
12
35
0.3
NA
95th Percentile
Upper Lower
NA
258
NA
41
NA
0.22
114
6.7
359
1.3
NA
117
19
NA
1,636
258
1,125
6.8
NA
NA
242
NA
33
NA
0.20
106
6.5
321
0.9
NA
21
17
NA
1,564
242
1,076
6.4
NA
RO
Count Mean
5 6.0
<5 NA
5 <5
5 <5
5 0.10
5 0.20
5 <30
5 <1.0
5 <50
5 O.5
5 O.01
5 <5
5 <1
5 O.5
5(2) 11
5 <5.0
5 <10
5 <1.0
5 <4.0
Permeate
Std. 95th Percentile
Dev. Upper Lower
0.1
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
5.4
NA
NA
NA
NA
6.1
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
16
NA
NA
NA
NA
5.9
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
5.1
NA
NA
NA
NA
RO
Count Mean
6 8.0
6 500
6 <5
6 83
6(1) 0.13
6 0.42
6 220
6 13
6 650
6 2.7
6 O.01
6 84
6 37
6 O.5
6 3,300
6 500
6 2,300
6 14
6 <4.0
Concentrate
Std. 95th Percentile
Dev. Upper Lower
0.1 8.1
60 548
NA NA
24 102
0.05 0.17
0.10 0.50
25 240
3.6 16
140 762
1.2 3.7
NA NA
22.2 102
5.9 42
NA NA
288 3,530
57 546
248 2,499
1.1 15
NA NA
7.9
452
NA
64
0.09
0.34
200
10
538
1.7
NA
66
32
NA
3,070
454
2,101
13
NA
(1) Count and statistics include two estimated results below the reporting limit.
(2) Count and statistics include one estimated result below the reporting limit.
115
-------
Table 4-25 (cont'd). RO Feed, Permeate, and Concentrate Inorganic Analysis
—
i
•a
>
1
a
__ a,
"3 "3
Parameter
(Mg/L)
Aluminum
Boron
Calcium
Iron
Potassium
Magnesium
Manganese
Sodium
Silica
Strontium
Zinc
Phosphorus
Lead
Silica
Count
5
7
7
7
7
7
7
7
7
7
7
7
6
6
Avg.
<100
300
12.71
<100
14,000
3,000
n(D
350,000
20,000
140
54
3,400
<100
20,000
RO Feed
Std 95th Percentile
Dev Upper
NA NA
17 313
1.38 13.74
NA NA
900 14,667
264 3,195
6.3 16
7,559 355,600
756 20,560
13 149
6.0 58
785 3,982
NA NA
983 20,787
Lower
NA
287
11.69
NA
13,334
2,805
6.4
344,400
19,440
131
50
2,819
NA
19,213
RO Permeate
Count Avg.
5 <100
5(2) 110
5 0.075
5 <100
5 <3,000
5 <200
5 <10
5 2,200
5 <1,100
5 0.10
5 <20
5 <3,000
5 <100
5 <1,100
Std 95th Percentile
Dev Upper
NA NA
9.4 118
0.025 0.095
NA NA
NA NA
NA NA
NA NA
261 2,429
NA NA
NA NA
NA NA
NA NA
NA NA
NA NA
Lower
NA
102
0.055
NA
NA
NA
NA
1,971
NA
NA
NA
NA
NA
NA
RO
Concentrate
Std 95th Percentile
Count Avg.
5 <100
6 520
6 26.50
6 200
6 30,000
6 6,200
6 22
6 730,000
6 42,000
6 300
6 110
6 7,800
6 <100
6 40,000
Dev Upper
NA NA
57 565
3.94 29.65
95 276
4,622 33,699
973 6,978
9.3 29
82,624 796,112
5,269 46,216
36 329
16 123
2,112 9,490
NA NA
3,933 43,147
Lower
NA
475
23.35
124
26,301
5,422
15
663,889
37,784
272
98
6,110
NA
36,853
116
-------
Table 4-26. Rejection of Analytes in the RO Feed
Analyte
Conductivity
DOC
TOC
Alkalinity
Hardness
TDS
Silica (SiO2)
Manganese
Phosphorous
Percent Rejection*
99.3
>92.4
>92.4
>99.0
>96.4
99.5
>97.3
»54.5
>55.9
Analyte
Boron
Calcium
Magnesium
Potassium
Sodium
Zinc
Bromide
Chloride
Sulfate
Percent Rejection*
63.3
>99.2
>96.7
>96.2
99.4
>81.5
>52.4
>86.4
>92.6
* Components at or below the method detection limit indicate a rejection >X based on one half the
method detection limit.
Table 4-27. RO System Mass Balance
Analyte
Sodium
(mg/L)
Calcium
(mg/L)
Magnesium
(mg/L)
Sulfate
(mg/L)
Bicarbonate
(mg/L)
Chloride
(mg/L)
TDS
(mg/L)
Process
Stream
Feed
Concentrate
Permeate
MB
Feed
Concentrate
Permeate*
MB
Feed
Concentrate
Permeate*
MB
Feed
Concentrated
Permeate*
MB
Feed
Concentrate
Permeate*
MB
Feed
Concentrate
Permeate
MB
Feed
Concentrate
Permeate*
MB
7/19/2006
360
800
2
1.11
13
28
0.1
1.07
3.1
6.9
0.1
1.12
310
710
25
1.18
250
560
2.5
1.12
110
250
15
1.20
1100
2600
5
1.18
7/26/2006
350
590
2.4
1.05
13
21
0.1
1.00
3.1
5.2
0.1
1.05
340
610
25
1.14
240
420
2.5
1.09
100
180
15
1.17
1100
1900
5
1.07
Date
8/2/2006
350
690
2.2
1.02
10
23
0.1
1.19
2.4
4.7
0.1
1.03
360
720
25
1.07
250
490
2.5
1.02
100
210
15
1.16
1100
2200
5
1.04
8/9/2006
350
740
2.6
1.11
13
27
0.1
1.09
3.2
6.7
0.1
1.11
320
680
25
1.15
230
450
2.5
1.03
100
220
15
1.22
1100
2300
5
1.10
8/16/2006
340
770
2.0
1.16
12
28
0.1
1.19
3.0
6.9
0.1
1.19
390
390
25
0.54
240
480
2.5
1.02
110
220
15
1.09
1100
2200
5
1.02
*Permeate values were not detected and are assumed to be one half the reporting limits.
^ Concentrate value is that reported by the lab. Its true value should be at least 750 mg/L for mass balance.
MB - Mass Balance -1.00 would indicate a "perfect" mass balance.
117
-------
Table 4-28. Biological Analyses of RO Process Streams
Date
07/13/06
07/26/06
08/02/06
08/10/06
08/14/06
08/15/06
08/16/06
Fecal Coliform (MPN/100 mL)
RO RO
RO Feed Permeate Concentrate
5.0E+05 <2
2.2E+03* <2
1.1E+04 <2
<2
5.0E+05* <2
1.1E+05* <2
4.0E+04* <2
1.6E+03
2.4E+00
1.6E+04
1.3E+03
3.0E+03
E Coll (MPN/100 mL)
RO RO
RO Feed Permeate Concentrate
3.0E+05
2.0E+04*
5.0E+03
2.0E+04*
<2 1.6E+03
<2 8.0E+02
<2 9.0E+03
<2
1.3E+03
<2 2.4E+03
: - Sample from UF Filtrate - which was feed for RO.
Date
07/13/06
07/26/06
08/02/06
08/03/09
08/10/06
08/14/06
08/15/06
08/16/06
RO Feed
4.0E+07*
5.3E+06*
8.9E+06
1.3E+07*
2.0E+07*
HPC (CFU/L)
RO
Permeate
2.0E+05
l.OE+04
2.1E+06
5.5E+03
1.8E+04
RO
Concentrate
1.4E+07
3.0E+07
1.1E+07
3.3E+06
5.8E+06
Total
RO Feed
1.6E+06*
8.0E+04*
9.0E+04
3.0E+06*
1.7E+05*
5.0E+05
Coliform (MPN/100 mL)
RO RO
Permeate Concentrate
7
<2
2
<2
<2
<2
2
1.6E+03
1.6E+04
1.6E+04
9.0E+03
9.0E+03
* - Sample from UF Filtrate - which was feed for RO.
4.5.4 Task C4: Membrane Integrity Testing
4.5.4.1 UF System - Pressure Hold Test
Pressure hold testing was performed on the UF system each day. During the test audit
representatives from Koch Membrane Systems, Village Marine, Inc., NSF, and USER were
present to observe the pressure hold test procedures. During that test the product side of the
membranes was drained and both arrays were simultaneously pressurized to 20 psi. The feed
valve and retentate valves were in their operating positions. The filtrate valves were closed.
After 15 minutes the system had lost 1.5 psi. This rate of pressure decline was acceptable to
Koch Membrane Systems.
As the verification test progressed, it became apparent that the pressure hold procedure being
used was not providing an accurate evaluation of the UF system. After further inspection, it was
discovered that the check valve on the feed side and the long run of piping filled with water on
the retentate side would not allow air to escape from the system at 20 psi. In effect the system
was completely closed. Opening a sample port on the feed side remedied the situation, but also
revealed that the system was not intact, as was apparent from the turbidity readings and
biological analysis results that had started arriving by this time.
Since the ETV test objectives require both the UF and RO systems for effective treatment, the
testing was continued with the RO system as the ultimate barrier. After the completion of the
118
-------
test period, the UF cartridges were removed for individual cartridge pressure hold tests as
described in Section 3.8. All the cartridges passed. However while re-installing the cartridges
with new o-rings, the operators discovered that the end caps did not fit tightly over the filtrate
connectors, see Figure 4-20. At this point the future tests were postponed while Village Marine
constructed new interconnectors and endcaps.
Village Marine determined that the endcaps were 0.04 inches larger than the design
specifications. The end caps on the EUWP are made from nylon according to Koch Membrane
System's specifications. When the system was built, Koch did not sell seawater compatible
endcaps. Apparently the nylon endcaps and interconnectors deformed over time with exposure
to chlorine and sun light. However, Koch Membrane Systems has a seawater compatible endcap
assembly made from Noryl under development at this time. It is recommended that these new
endcaps be procured for both EUWP systems if they are put into commercial production.
Figure 4-20. UF filtrate connector and leaking end cap.
4.5.4.2 RO System - Dye Challenge
Dye tests were performed on the RO system at the start and end of the test period. For the
second dye test there was only enough dye left for a six minute test. Table 4-29 gives the results
for each sample before and after the test period. As can be seen the RO membranes rejected the
dye at a rate of higher than 99%. The rejection rate actually improved at the end of the test.
These results, supported by the high rejection rate for conductivity, the low turbidity in the
permeate, and the 3 log reduction calculated from the bacteriological samples, indicate that the
RO membranes maintained integrity throughout the verification test.
119
-------
Table 4-29. RO Permeate Absorbance after Injection
Minutes after
injection
1
2
3
4
5
6
71
8
9
10
Feed (mean)
Concentrate
1:1 Dilution
Permeate
Sample 1
0.018
0.014
0.012
0.015
0.013
0.013
0.002
0.001
0.000
0.003
3.068
2.92
Gallup Start Up
Rejection
Sample 1
99.41%
99.54%
99.61%
99.51%
99.58%
99.58%
99.93%
99.97%
>99.99%
99.90%
Permeate
Sample 2
0.020
0.012
0.013
0.014
0.016
0.014
0.003
0.000
0.001
0.004
Rejection
Sample 2
99.35%
99.61%
99.58%
99.54%
99.48%
99.54%
99.90%
>99.99%
99.97%
99.87%
Gallup End Point
Permeate
Sample 1
0.004
0.004
0.007
0.003
0.001
0.002
2.41
2.358
Rejection
99.8%
99.8%
99.7%
99.8%
>99.9%
99.9%
1 Cleaned the product vials and applied silicone oil. The powder had adhered to the outside of the vials giving a
higher reading than expected but still over the 99% rejection level.
4.5.4.3 Continuous Indirect Integrity Monitoring
Turbidity data is presented previously in Figures 4-18 and 4-19, and Table 4-18. RO permeate
turbidity was well below 50 mNTU except for regular brief excursions to 50-60 mNTU during
sampling. However, these excursions are an artifact of sampling. Sampling disrupts the flow to
the turbidimeter.
Due to operator error, the on-line turbidity records were erased after July 27 for RO permeate
and UF filtrate. The readings that were recorded on the data sheets are included in Figures 4-18
and 4-19.
UF filtrate turbidity was much higher than expected. None of the remedies of cleaning the
system, cleaning the turbidimeter, and recalibration did anything to solve the problem. After the
bacteria counts started coming in, it became apparent that the UF system had significant integrity
problems. The UF filtrate water quality was so poor that USER believed it had to be worse than
just broken fibers. However, the schedule with the City of Gallup had to be maintained, as the
City needed the space and the EUWP had to be off site by the scheduled end of this test period.
Testing continued with reliance on the RO system as the ultimate barrier.
Particle counts for the UF filtrate were also high, with only about 50-75% retention of particles
in all size ranges. Hach technical assistance was sought, but their technical representative could
not explain the erratic and poor results. When the testing was complete, and the calibration data
was verified, it was discovered that the particle counter software had been set up with erroneous
calibration data. This means that none of the particle data is meaningful. Even if some of the
ranges roughly correspond to configured ranges, there can be no confidence in any of the data.
NSF agreed that since the test was based on the whole system, it was acceptable to disregard the
erroneous particle count data and use turbidity and biological analysis as criteria for success.
120
-------
4.5.5 Task C6: Qualitative Evaluations
The qualitative evaluation is based on events and observations. Information here should be taken
as advice for those who would operate this equipment or design similar equipment.
4.5.5.1 Reliability or Susceptibility to Environmental Conditions
4.5.5.1.1 Protective Covers
One of the key requirements of this equipment is transportability. It is designed to be picked up
and moved around. Before transport to Gallup, the unit was fitted with brand new, custom-
made, Envelop® Protective Covers from Shield Technologies. When the equipment arrived it
already had two tears in the covers from wind during transport. These tarps should be covered
during transportation. They are mainly for protection while in storage and during operation.
4.5.5.1.2 PLC
The GE Fanuc PLC lost its memory during transportation. This resulted in a delay of four weeks
while Village Marine attempted to replace it. They finally sent their programmer to correct
issues resulting from loading the wrong program. Remote high speed telecommunication access
to the PLC was not available at this remote site. If remote access could have been achieved, it
would have quickened trouble shooting and software updates. Otherwise, it is critical to have the
programming software and a programmer to participate in deployment of the system.
4.5.5.1.3 Intake Strainer
The intake strainer required frequent cleaning, sometimes daily. Although this is indicative of
poor source water, it is also the effect of a relatively high velocity going through the screen.
Larger surface area intake screens may be required to reduce cleaning frequency or in areas
where cleaning is not practical.
4.5.5.1.4 Amiad Strainers
The Amiad Strainers were very difficult to extract from their mounting near the top of the UF
system. It was necessary to remove them for cleaning every one to two days. They need to be
mounted in a more accessible location. Also, the UF is disabled from backwashing while the
Amiad is backwashing, and the Amiad strainers were backwashing much more often than
assumed in the system design. As a result there was a loss of approximately 65% of all
scheduled backwashes for the UF system, thereby decreasing their efficiency.
4.5.5.1.5 UF System Hoses
Several times during the backwash cycle hoses flew off the bottom of the UF cartridges. Hoses
are secured with hose clamps on hose barb fittings. The barbs on the EUWP Gen 1-1 are not
long enough for two hose clamps. The EUWP Gen 1-2 unit has longer hose barbs to
accommodate two hose clamps. Both units should be equipped with the longer barbs.
121
-------
4.5.5.1.6 UF System Interconnections
The filtrate tubes were too small or the caps were too large. This resulted in the failure of the UF
system. Village Marine is redesigning the interconnections to prevent mixing of feed water with
the filtrate.
4.5.5.1.7 Cleaning
The EUWP must be re-configured to accomplish cleaning either system. The longer time for the
first cleaning of the RO and UF reflects the need to find more plumbing parts to get RO permeate
to the Cleaning tank. The PLC did not allow operation of one of the pumps required for water
transfer if it was not feeding the RO system. This required moving the pressure sensor on the
outlet of the RO feed pump (P6) from the RO skid to the cleaning system line. This type of
reconfiguration should not be standard procedure. The PLC program should be modified to
include an RO cleaning cycle that disables the check on the P6 outlet pressure sensor.
4.5.5.1.8 Chemical Feed Pumps
The chemical feed pumps did not meter accurately. There was no correlation between pump
speed, stroke length, and volume delivered. The FTO mixed the chemicals to match the observed
delivery rate at an intermediate pump speed and stroke length setting. The coagulant chemical
injection line was repeatedly blown off the injection fitting. The fitting was not the appropriate
type for rigid 3/8" plastic tubing. This was later moved to the feed side of the pump to alleviate
the problem, but this introduced suction issues where at certain pump settings, there would be
free flow of coagulant. The pumps need to be replaced and/or properly installed to provide
adequate backpressure.
4.5.5.1.9 Flow Measurement
There is not enough redundancy in flow measurement. The UF filtrate is not measured. The
retentate flow is measured with a rotometer which appears to be as much as 30% low according
to bucket and stop watch measurements. The filtrate flow was estimated using the pressure
indicators for onion tank level. However the relation between pressure and volume depends on
tank dimension measurements at various heights. Volume was derived from modeling sections
of the tank. There was a 9-12% difference from the calculated flow in one instance and 2-5% in
another.
4.5.5.1.10 Pumps
All the water pumps were 100% reliable. There were no pump failures during the test period.
4.5.5.1.11RO System
The RO system ran extremely well for the challenge it received during this test.
4.5.5.2 Equipment Safety
There were no safety incidents during the test period.
122
-------
4.5.5.3 Effect of Operator Knowledge, Skill, and Experience on Results
In addition to familiarity with the pilot process and data requirements, desirable skills for
operators are:
• Positive attitude;
• Self confidence;
• Instrumentation skills for troubleshooting and calibration;
• Plumbing skills;
• Familiarity with electrical systems;
• Attention to detail; and
• Neat handwriting.
Successful operations teams own these skills among their members. The Gallup ETV test was
carried out by a total of eleven operators who have been rated on a scale from 1-10 based on past
experience with pilot systems, membrane processes, and general skills such as those listed above.
With the exception of a couple of travel day disconnects, there were two operators on site each
day. The minimum combined score for any team was 4 during a three day overlap of relatively
in-experienced operators. The maximum score was 10 for two days at the start of testing. The
average score was 7.1.
Teams were assigned a Quality Assurance (QA) Rating based on their completion of the several
quality assurance activities that were required, such as writing down the operating conditions,
field measurements, duplicate field measurements, completion of the daily, weekly, and
bi-weekly calibrations and maintenance checks and, if there was a cleaning, whether the
chemical usage was recorded either in the log book or on a cleaning record form. Each activity
earned one point. Operator teams also got a point for recording activities in the log book. Actual
daily production of RO permeate was used as a dependant parameter. Point score was divided
by the number possible for the day resulting in a perfect QA Rating of 1. Table 4-30 lists
operator codes, skill ratings, quality assurance scores, and the RO permeate production of the
day.
There was no correlation between skill level and productivity or the number of quality assurance
measures completed. This was a very challenging test for all of the operators. The UF system
struggled with the waste water effluent, especially during the Fourth of July holiday when the
tourist population was high. Maintenance was required for the Amiad filters, the hoses, and/or
cleaning, most every day of the test, in addition to the extensive sampling program, and quality
assurance activities. Most days, the operators were on site for 10-15 hours. All the operators
dealt with the problems of the day in a professional manner.
Performance at any one period is dependent on the previous performance of all parts of the
process. The responsibility for flux decline and the need for cleaning cannot be assigned to the
current operator. The performance of the WWTP played a major role but the only true indicator
available is the performance of the RO system. Even that is the combined result of the poor
performance of both the WWTP and the UF system.
There was a fairly strong correlation between the QA score and productivity as shown in
Figure 4-21. However, this may be an indication of the correlation between smooth operation
123
-------
and productivity which is indirectly the result of the operator's knowledge and skill. The wide
range of productivity for a perfect score of 1 demonstrates that even conscientious operators have
days when they need to shut the system down for maintenance.
Table 4-30. Evaluation of Skill, QA, and Productivity
Operator Team
A/B
A/B
B/C
B/C
B/C
B/D
B/D
B/D
B/D
D/E
D/E
D/E
E/F
E/F
E/F
E/F
F/G
F/G
F/G
F/G
G/H
G/H
G/H
H/I
H/I
H/I
H/I
I/J
I/J
I/J
I/J
I/J
J/B
J/B
J/B
B/C/K
B/C/K
Skill Rating
10
10
9
9
9
9
9
9
9
5
5
5
5
5
5
5
7
7
7
7
4
4
4
5
5
5
5
8
8
8
8
8
9
9
9
9
9
QA Rating
1.00
1.00
0.63
0.62
1.00
0.67
0.91
1.00
1.00
0.67
1.00
0.78
0.15
0.20
0.67
1.00
1.00
1.00
0.35
0.11
0.59
0.67
.00
.00
.00
.00
.00
0.50
0.95
0.78
.00
.00
.00
.00
.00
.00
0.00
kgal/day produced
42
121
74
84
169
83
157
162
167
86
169
168
0
0
72
144
72
68
0
0
59
58
118
120
119
120
61
59
121
124
126
128
126
126
128
129
0
Activity
Audit/Sampling
Sampling
RO System
Cleaning/ Sampling
Calibrations
UF System
Cleaning
Sampling
RO System
Cleaning/Sampling
Sampling
Dye Test/Cleaning
124
-------
180
160
0.4 0.6 0.8
Quality Assurance Score
i.o
1.2
Figure 4-21. Relation between QA and actual productivity.
4.5.5.4 Effect of Operator's Technical Knowledge on System Performance and Robustness of
Operation
The more an operator knows a system, the easier it will be to troubleshoot any problems. All the
operators had access to the user manual for the system. All but two had prior experience
operating the system. It is vital that operators understand the principles of the process they are
operating so that they can detect changes - and know the probable cause and consequences of the
change.
During the test period, changes in performance were related to the ability of the equipment to
perform with poorly treated wastewater effluent as the feed. The first UF system decline in
performance corresponds to a week of discussions with Village Marine about the efficiency of
the UF backwash and how the Amiad Strainers were preventing UF backwash initiation.
Unfortunately it took three weeks to get the backpressure on the Amiads increased high enough
to allow more consistent UF performance. A thorough knowledge of the system would have
brought the solution to light in a timelier manner. However, since the Amiads were only recently
moved to their current position, this information would not be in the user manual.
4.5.5.5 Ease of Equipment Operation
The EUWP is very easy to operate. The touch screen control system shows what is on, flow
rates, temperature, and pressures. All the pumps can be started from the control screen.
Cleaning is not easy. However if additional hoses were set in place and an extra pump was
available to move water from the RO permeate tank, it would be simpler. The changes
mentioned above need to be incorporated into the RO cleaning algorithm.
125
-------
4.5.5.6 Waste Discharge Requirements
The design waste discharge from the EUWP is comprised of the UF system retentate, backwash,
RO concentrate, and water for cleaning. Theoretically, the discharge should be the UF feed flow
minus the RO permeate, or 3.46 million gallons for the 35.2 days of data or on average 97.8 kgal
per day.
4.6 QA/QC
4.6.1 Introduction
An important aspect of verification testing is the QA/QC procedures and requirements. As
described in Task 7 of the methods and procedures (Section 3.9.9) and the QAPP in the PSTP
prepared for this ETV test, 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.
4.6.2 Documentation
The field technicians recorded on-site data and calculations in a field logbook and on specially
prepared field log sheets. The operating logbook included calibration records for the field
equipment used for on-site analyses. Copies of the logbook, 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 DWS Center staff checked 100% of the data entered into the spreadsheets to confirm the
information was correct. The spreadsheets are presented in Appendix L.
Samples collected and delivered to the contract laboratories for analysis were tracked using
chain-of-custody forms. Each sample was assigned a location name, date, and time of collection.
The laboratories reported the analytical results in laboratory reports. These reports were received
and reviewed by Bureau of Reclamation staff. These laboratory data were entered into the data
spreadsheets, corrected, and verified in the same manner as the field data. Lab reports are
presented in Appendices C through I.
4.6.3 Quality Audits
Representatives from NSF performed an audit of the QA plan at the start of the testing period on
July 12-13, 2006. The audit focused on review of the field procedures, including the collection of
operating data and performance of on-site analytical methods. The TQAP requirements and
QAPP were used as the basis for the audit. The NSF representatives prepared an audit report. All
deficiencies were corrected immediately.
The NSF QA Department reviewed the contract laboratory and field analytical results for
adherence to the QA requirements for precision and accuracy detailed in the project QAPP and
for compliance with the laboratory quality assurance requirements. The laboratory raw data
126
-------
records (run logs, bench sheets, calibrations records, etc.) are maintained at NSF and are
available for review.
4.6.4 Test QA/QC Activities
The USER staff conducted the field monitoring, measurements, and sample collection and
handling in accordance with the USEPA-approved TQAP created specifically for this
verification. The testing laboratory staff conducted the chemical and microbiological analyses by
following the TQAP. NSF QA Department staff and representatives performed audits during
testing to ensure the proper procedures were followed.
Table 4-31 lists a summary of QC activities that were performed in the field and Table 4-32 lists
a summary of QA activities performed by field and project staff.
Table 4-31. Quality Assurance Activities
Equipment
Action Required
Initial Flowmeters - electronic
Turbidimeter - online (1720E)
Turbidimeter - online (FilterTrak)
Particle counter - online
Verify calibration volumetrically
Provide factory calibration certificate
Provide factory calibration certificate
Provide factory calibration certificate
Daily
Chemical Feed Pump
Turbidimeter - online
Turbidimeter - portable
Particle Counters - online
Myron pH meter
UF System
Particle counter - online
Conductivity meter - portable
Volumetrically check flow
Verify with portable turbidimeter
Volumetrically check flow
Volumetrically check flow
Calibrate - 3 point (4,7,10)
Pressure hold test
Clean sensors used to monitor feed
Calibrate - 2 points with certified conductivity
buffers (required monthly)
Weekly Rotameters
UF filtrate flow
Temperature - portable
Turbidimeter - portable
Tubing
On-Line Pressure/flow indicators
Check for algae & verify volumetrically
Verify volumetrically
Verify calibration with NIST certified precision
thermometer (required by ETV only)
Calibrate using <0.1, 20, 100, and 800 NTU
standards (required quarterly)
Check for algae and leaks
Verify calibration
127
-------
Table 4-32. Quality Control Activities
Item Action Required
Daily Data Review system performance data since previous day
„, ., _. ^ Qualitative online data review & review field and Lab data
Weekly Data v, ., ,,
when available.
4.6.5 Sample Handling
All samples analyzed by the laboratories were labeled with unique identification numbers. These
identification numbers appear in the laboratory reports for the tests. All samples for chemical
analytes were analyzed within required holding times.
Some microbiological samples for fecal coliform and HPC were not analyzed within 24 hours. It
was necessary to ship all microbiological samples off site for analysis. Shipment was by
overnight carrier for next- day delivery. In most cases samples collected in the afternoon arrived
the next morning and the analysis was started within the 24 hour holding time. However, in a
few cases, by the time the samples arrived at the laboratory and the test was started, that actual
time was slightly more than 24 hours. All tests were started within a few hours of arrival at the
laboratory. The exceedance of holding times by a couple of hours should not have a major
impact on test results. It should be noted that fecal coliform has an 8 hour holding time
requirement for compliance samples for drinking water, but has a 24 hour holding time for other
data use. The 24 hour holding time was used for this ETV test as it was not a compliance test.
4.6.6 Physical and Chemical Analytical Methods QA/QC
4.6.6.1 Field Sample Analysis
Bench top field instruments that measured turbidity, pH, temperature and specific conductance
(conductivity) were calibrated in accordance with the data quality objectives (DQO) in the
TQAP. Procedures followed USEPA methods.
In-line field meters for particle counts and turbidity measurements were factory calibrated, and
certificates were provided as required in the TQAP. However, the incorrect calibration certificate
data for bin voltages was entered into the software program for the particle counters. This
resulted in rendering the particle count data inaccurate and not meeting the DQO. Because of this
problem, particle count data could not be used for documenting system performance for particle
count and the data are not included in this report.
Turbidity was measured with two different approaches during the test. The in-line turbidity
meters provided continuous data. Grab samples for turbidity were also collected either once or
twice per day and analyzed using a field turbidimeter. The data presented in Section 4.5.3.1
showed that the feed water and UF filtrate results were similar between the methods. However,
the RO permeate results from the in-line meter were an order of magnitude lower than the results
from the handheld turbidimeter (mean of 0.15 NTU for grab samples versus a mean of 0.016
NTU for in-line measurements). It would be expected that the RO permeate would have a very
low turbidity. The difference in the results is due to the better sensitivity and lower detection
limit of the in-line meter. A reasonable detection limit for a standard turbidimeter with grab
128
-------
samples is typically in the 0.05 to 0.10 NTU range. This lower detection limit assumes the use of
a very clean cuvette with no fogging of the glass due to temperature changes in the sample water.
In the measurement of low turbidity levels, such as at 0.10 NTU, error trends are normally biased
positive due to effects from such sources as bubbles, contamination and sample cell
imperfections. While good results can be obtained, it is very difficult in field conditions to
achieve reproducible results much below 0.10 NTU limit. All of the RO permeate readings were
very close to the 0.1 to 0.2 NTU, with a 95% confidence level of 0.13 to 0.17 NTU (see
Table 4-17)
The in-line meter used for the RO permeate was Hach FilterTrak 660™ laser nephelometer
designed specifically to measure very low turbidity levels. The unit specifications show that it
can detect changes in turbidity as low as 0.0005 NTU and has a rated limit of detection of 0.0004
NTU. The typical operating range is 0.001 to 5.0 NTU. Thus, this unit measures turbidity by
passing a steady flow of water through the unit, which reduces problems of fogging due to water
temperature, presence of air bubbles, and similar problems encountered when using grab samples
and field or laboratory turbidimeters. It can be expected that a in-line laser unit, such as the one
installed in the RO permeate line, properly calibrated and maintained, should provide more
accurate data at the low turbidity levels expected in the RO permeate. Based on the evaluation of
the procedures and equipment used for the turbidity measurements of the RO permeate, it is
judged that the in-line meter results are of good quality and appropriate for the measurement of
the RO permeates. Therefore, the lower results obtained from this unit should be the data used
for evaluating the turbidity of the RO permeate water.
4.6.6.2 Laboratory Methods
All of the analytical methods used by the contract laboratories were EPA methods or Standard
Methods. However, most of the EPA methods referenced by the laboratories, particularly for the
background organics work and some of the general water quality measurements, were methods
from EPA SW-846 and not the approved EPA methods for drinking water. The PSTP did not
require that the methods being used for this technology evaluation test (this was not a drinking
water compliance test) be EPA drinking water methods. The PSTP did require that the
laboratories be certified laboratories (NELAC, State, or similar certification) and that proper
methods for water/wastewater be used. In this case, the EPA SW-846 methods are very similar to
the drinking water methods, and are published by EPA as appropriate for water/wastewater type
testing. These methods have QA/QC procedures that met the requirements of the PSTP and
QAPP. Review of the methods found that they were appropriate for this work and produced
results that met the quality objectives.
Review of the microbiological results show that the proper procedures and QA/QC were
followed for the fecal coliform, E. coli, total coliform, and HPC analyses. However, the
Cryptosporidium and Giardia analyses did not meet the QA/QC objectives for the ETV test.
Therefore, these data are not included in the report.
4.6.7 Documentation
The contract laboratories documented their activities using their prepared laboratory bench
sheets and standard laboratory reports. Data laboratory reports were entered into Excel
spreadsheets. These spreadsheets were used to calculate mean, median, and confidence intervals
129
-------
for feeds and filtrates when sufficient numbers of sample results were available (generally 8 or
more). One hundred percent of the data entered into the spreadsheets was checked by a reviewer
to confirm all data and calculations were correct.
4.6.8 Data Review
NSF QA/QC staff reviewed the data records for compliance with QA/QC requirements. NSF
ETV staff checked at least 10% of the data in the laboratory reports against the Excel®
spreadsheets.
4.6.9 Data Quality Indicators
The quality of data generated for this ETV was established through four indicators of data
quality: representativeness, accuracy, precision, and completeness.
4.6.9.1 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 a secondary
wastewater, representing a possible application (highly contaminated surface water) for the
EUWP during deployment. Two other ETV reports considered the EUWP performance when
using sea water and normal surface water (lake water) as its feed.
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.
4.6.9.1.1 Sampling Locations
Samples were collected from the points listed in Table 3-9 for all analyses both by labs and by
operators in the field.
4.6.9.1.2 Timing of Sample Collection
Automatic data acquisition of PLC inputs occurred every 15 minutes, which was adequate to
catch the UF system at various stages of backwash and forward operation. Turbidity data was
recorded by the turbidimeters every five minutes. Data was also recorded manually at the start
and end of the workday. Weekly samples were collected on Wednesdays until July 13 when the
batch of samples arrived too cold for analysis. Biological samples must be above zero degree C,
however the lab maintains that the water temperature was below zero. Sampling was repeated on
July 18 and thereafter sampling was performed on Tuesdays. Sampling was begun at
approximately 1:00 pm to allow for the minimum transit time for samples with 24 hour hold
times. Field measurements were performed twice per day in the morning and evening as other
tasks allowed.
130
-------
4.6.9.2 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, certified QC standards
and/or matrix spikes were run with each batch of samples. Every sampling event with samples
shipped to the lab included 10% sample duplicates, travel blanks, method blanks, matrix spikes
(MS), and matrix spike duplicates (MSD) provided by the labs. For acceptable analytical
accuracy, the recoveries must be within control limits for each analyte, where control limits are
defined as the mean recovery plus or minus 3x the standard deviation. Recovery of matrix spike
samples, duplicates and laboratory control samples are reported in Appendix M.
The percent recoveries of all matrix spikes and standards were within the allowable limits for all
analytical methods.
For physical and chemical analyses performed in the field, certified QC standards (performance
evaluation, or PE) for pH and turbidity were run once during the testing period. The reported
values for pH and turbidity were within the acceptable range for the PE samples.
4.6.9.3 Precision
Precision refers to the degree of mutual agreement among individual measurements and provides
an estimate of random error. One sample per batch was analyzed in duplicate for the NSF
Laboratory measurements. For field measurements, one process stream was analyzed in
duplicate every day. Precision of duplicate analyses was measured through RPD.
The duplicate analysis RPD calculations are presented in Appendix M. All RPDs were within
the allowable limit of 30% for each parameter with the following exceptions:
• 15 of the 48 turbidity measurements using grab samples and the field turbidimeter
exceeded the 30% RPD. As discussed in Section 4.6.6.1, the field test is more susceptible
to air bubbles, fogging, and other issues than the in-line measurements.
4.6.9.4 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 actual number of samples collected and analyzed for
each parameter and/or method compared to the test plan requirements.
All planned water chemistry samples were collected and analyzed.
Initially total and fecal coliform were scheduled to be collected daily, Monday through
Thursday, for each week of testing for the UF feed and filtrate. Full sets of feed and filtrate were
obtained for 11 of 17 possible sampling days yielding a completeness of 65%. FtPC andE1. coli
were scheduled for weekly analysis and were analyzed on 4 of the 5 weeks during the test for
80% completeness. As discussed earlier, the Cryptosporidium and Giardia analyses did not meet
the QA/QC requirements, thus were not valid.
131
-------
The field parameters, pH, temperature, conductivity, and turbidity were scheduled for twice daily
analysis for a 30 day test yielding a projected 60 results. The actual number of results collected
varied from a minimum of 48 UF filtrate turbidity measurements to a maximum of 52 UF feed
water samples for pH, temperature, conductivity and turbidity. This yields a completeness of 80
to 87%.
132
-------
4.7 References
EPA/NSF Protocol for Equipment Verification Testing for Removal of Inorganic Constituents -
April 2002.
EPA/NSF Protocol for Equipment Verification Testing for Physical Removal of Microbiological
and Paniculate Contaminants - September 2005.
Chan, R. 2002. "Fouling Mechanisms in the Membrane Filtration of Single and Binary Protein
Solutions". Doctorate Thesis School of Chemical Engineering and Industrial Chemistry,
The University of New South Wales. September 2002.
Drewes, J.E, G. Amy, P. Xu. 2005. "Rejection of Wastewater-Derived Micropollutants in High-
Pressure Membrane Applications Leading to Indirect Potable Reuse" WateReuse
Foundation Report WRF-02-001.
ASTM D3739. 2006. "Standard Practice for Calculation and Adjustment of the Langelier
Saturation Index for Reverse Osmosis" in ASTM Standards Volume 11.02, Water (II).
ADC News Release, May 4, 2006. "Affordable Desalination Sets low Energy Record." In
numerous sources such as Filtration & Separation May, 2006.
Reddersen, K.; Heberer, T. 2003. Multi-compound methods for the detection of pharmaceutical
residues in various waters applying solid phase extraction (SPE) and gas chromatography
with mass spectrometric (GC-MS) detection. J. Sep. Sci., 26, 1443-1450.
Mansell, J. and Drewes, J. E. 2004.. Fate of Steroidal Hormones during Soil Aquifer Treatment.
Ground Water Monitoring and Remediation, 24(2), 94-101.
133
-------
Appendices
Contact Bruce Bartley at bartley@nsf.org or 734-769-5148 for copies of
appendices.
------- |