May 2013
NSF13/38/EPADWCTR
EPA/600/R-13/096
Environmental Technology
Verification Report
Inactivation of Microbiological
Contaminants in Drinking Water by
Ultraviolet Technology
NeoTech Aqua Solutions Inc.
Ultraviolet Water Treatment System
NeoTech D438™
Prepared by
NSF International
Under a Cooperative Agreement with
U.S. Environmental Protection Agency
-------
FINAL
Environmental Technology Verification Report
Inactivation of Microbiological Contaminants in Drinking Water by
Ultraviolet Light Technology
NeoTech Aqua Solutions, Inc.
Ultraviolet Water Treatment System
NeoTech D438™
Prepared by:
NSF International
Ann Arbor, Michigan 48105
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
-------
Table of Contents
Title Page i
Notice ii
Table of Contents iv
List of Tables v
List of Figures vi
Abbreviations and Acronyms vii
Verification Statement viii
_Chapter 1 Introduction 1
1.1 ETV Program Purpose and Operation 1
1.2 Purpose of Verification 1
1.3 Verification Test Site 2
1.4 Testing Participants and Responsibilities 2
Chapter 2 Equipment Description 4
2.1 NeoTech General Information 4
2.2 NeoTech UV System Description 4
2.3 NeoTech D438™ Specifications and Information 7
2.4 NeoTech Ultraviolet Treatment System Standard Features 8
Chapters Methods and Procedures 9
3.1 Introduction 9
3.2 UV Sensor Assessment 10
3.3 Headloss Determination 11
3.4 Power Consumption Evaluation 11
3.5 Feed Water Source and Test Rig Setup 11
3.6 Installation of Reactor and Lamp Burn-in 15
3.7 Collimated Beam Bench Scale Testing 15
3.8 Full Scale Testing to Validate UV Dose 19
3.9 Analytical Methods 23
3.10 Full Scale Test QA Controls 25
3.11 Power Measurements 26
3.12 Flow Rate 26
3.13 Evaluation, Documentation and Installation of Reactor 26
Chapter 4 Results and Discussion 28
4.1 Introduction 28
4.2 Sensor Assessment 28
4.3 Collimated Beam Dose Response Data 29
4.4 Development of Dose Response 29
4.5 MS and Operational Flow Test Data 42
4.6 SetLineforREDmeasOf40mJ/cm2 48
4.7 Deriving the Validation Factor and Log Inactivation for Cryptosporidium 49
4.8 Deriving the Validation Factor and Log Inactivation for Giardia 56
4.9 Validated Dose (REDVai) for MS2 as the Target Organism 61
iii
-------
4.10 Water Quality Data 63
4.11 Power Measurement 66
4.12 Headless 66
Chapter 5 Quality Assurance/Quality Control 67
5.1 Introduction 67
5.2 Test Procedure QA/QC 67
5.3 Sample Handling 67
5.4 Chemistry Laboratory QA/QC 67
5.5 Microbiology Laboratory QA/QC 67
5.6 Engineering Lab - Test Rig QA/QC 70
5.7 Documentation 73
5.8 Data Review 73
5.9 Data Quality Indicators 73
Chapter 6 References 76
Attachments
Attachment 1 NeoTech Technical Manual and Documentation
Attachment 2 Sensor and Lamp Information
Attachment 3 NSF Collimated Apparatus® from NSF Std 55® Annex A
Attachment 4 UVT Scans of Influent and Effluent Water at High and Low UVT
List of Tables
Table 2-1. Basic UV Chamber Information 7
Table 2-2. Low Pressure Lamp Information 7
Table 2-3. Lamp Sleeve Information 7
Table 2-4. UV Sensor Information 7
Table 3-1. Test Conditions for Validation 21
Table 3-2. Analytical Methods for Laboratory Analyses 23
Table 4-1. Sensor Assessment Data 29
Table 4-2. Response Data from Collimated Beam Tests at 91% UVT 32
Table 4-3. Response Data from Collimated Beam Tests at 97% UVT 34
Table 4-4. Response Data from Collimated Beam Test at 97% UVT outlier removed 36
Table 4-5. Flow Test Operational Data NeoTech D438™ 43
Table 4-6. Flow Test MS2 Concentrations NeoTech D438™ 44
Table 4-7. MS2 Log Concentration for Influent and Effluent NeoTech D438™ 45
Table 4-8. MS2 Log Inactivation Results NeoTech D438™ 46
Table 4-9. MS2 RED Results NeoTech D438™ 47
Table 4-10. RED Bias Factor for Each Set Point for Cryptosporidium 50
Table 4-11. Uncertainty of the Validation (Uvai) and BRED Factors for Cryptosporidium 53
Table 4-12. Validation Factors and Validated Dose for Cryptosporidium Inactivation 55
Table 4-13. RED Bias Factor for Each Set Point for Giardia 57
Table 4-14. Uncertainty of the Validation (Uvai) and BRED Factors for Giardia 59
iv
-------
Table 4-15. Validation Factors and Validated Dose for Giardia Inactivation 60
Table 4-16. Validation Factor and Validated Dose based on MS2 62
Table 4-17. Temperature and pH Results 64
Table 4-18. Total Chlorine, Free Chlorine, and Turbidity Results 64
Table 4-19. Iron and Manganese Results 65
Table4-17. HPC, Total Coliform, andE. co//Results 66
Table 4-20. Power Measurement Results 67
Table 4-21. Headloss Measurement Results 67
Table 5-1. Stability Results 71
Table 5-2. Trip Blank Results 71
Table 5-3. Flow Meter Calibration 72
Table 5-4. Reactor Control and Reactor Blank MS Results 73
Table 5-5. Completeness Requirements 76
List of Figures
Figure 2-1. NeoTechD438™ 5
Figure 2-2. NeoTech D438™ Basic Dimensions 6
Figure 3-1. Schematic of NSF Test Rig 13
Figure 3-2. Photograph of the Test Unit Setup 14
Figure 4-1. MS2 Collimated Beam Dose versus Log N Day 1 08-02-2012 UVT 91% 38
Figure 4-2. MS2 Collimated Beam Dose versus Log N Day 2 08-03-2012 UVT 97% 39
Figure 4-3. Dose Response Log I versus Dose UVT 91% Day 1 08-02-2012 40
Figure 4-4. Dose Response Log I versus Dose UVT 97% Day 2 08-03-2012 41
Figure 4-5. Set Line for 40 ml/cm2 REDmeas NeoTech D438™ 48
Figure 4-6. Set Line for 3-log Cryptosporidium Inactivation for NeoTech D438™ 54
-------
Abbreviations and Acronyms
A254 Absorbance at 254 nm
ATCC American Type Culture Collection
°C Degrees Celsius
CFU Colony Forming Units
cm Centimeter
DWS Drinking Water Systems
EPA U. S. Environmental Protection Agency
ETV Environmental Technology Verification
°F Degrees Fahrenheit
GP Generic Protocol
gpm Gallons per minute
h Hours
HPC Heterotrophic Plate Count
L Liter
Ibs Pounds
LEVIS Laboratory Information Management System
Log I Log inactivation
LT2ESWTR Long Term 2 Enhanced Surface Water Treatment Rule
m Meter
min Minute
ml Millijoules
mg Milligram
mL Milliliter
MS2 MS2 coliphage ATCC 15597 Bl
NeoTech NeoTech Aqua Solutions, Inc. (formerly Ultraviolet Sciences, Inc.)
NIST National Institute of Standards and Technology
nm Nanometer
NRMRL National Risk Management Research Laboratory
NSF NSF International (formerly known as National Sanitation Foundation)
NTU Nephelometric Turbidity Unit
ORD Office of Research and Development
PFU Plaque Forming Units
psi Pounds per Square Inch
QA Quality Assurance
QC Quality Control
QA/QC Quality Assurance/Quality Control
QAPP Quality Assurance Proj ect Plan
QMP Quality Management Plan
RED Reduction Equivalent Dose
REDmeas Measured Reduction Equivalent Dose - from test runs
REDvai Validated Reduction Equivalent Dose - based on selected pathogen and
uncertainty
RPD Relative Percent Deviation
vi
-------
SM Standard Methods for the Examination of Water and Wastewater
SOP Standard Operating Procedure
TQAP Test /Quality Assurance Plan
IDS Total Dissolved Solids
ISA Tryptic Soy Agar
UVT UV transmittance
TSB Tryptic Soy Broth
ug Microgram
jam Microns
USEPA U. S. Environmental Protection Agency
UVDGM UV Design Guidance Manual 2006
UDR Uncertainty of collimated beam data
Us Uncertainty of sensor
USP Uncertainty of set point
UVAL Uncertainty of validation
vn
-------
Chapter 1
Introduction
1.1 ETV Program Purpose and Operation
The U.S. Environmental Protection Agency (USEPA) has created the Environmental Technology
Verification (ETV) Program to facilitate the deployment of innovative or improved
environmental technologies through performance verification testing 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 that are responsive to the needs of stakeholders;
conducting field or laboratory testing, collecting and analyzing data; and by 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 USEPA has partnered with NSF International (NSF) under the ETV Drinking Water
Systems Center (DWS) to verify performance of drinking water treatment systems that benefit
the public and small communities. It is important to note that verification of the equipment does
not mean the equipment is "certified" by NSF or "accepted" by USEPA. Rather, it recognizes
that the performance of the equipment has been determined and verified by these organizations
under conditions specified in ETV protocols and test plans.
1.2 Purpose of Verification
The purpose of the ETV testing was to validate using the set line approach the ultraviolet light
(UV) dose delivered by the NeoTech Aqua Solutions Inc. (NeoTech) Ultraviolet Water
Treatment System Model D438 (NeoTech D438™) as defined by these regulatory authorities
and their guidelines and regulations:
• Water Supply Committee of the Great Lakes—Upper Mississippi River Board of State
and Provincial Public Health and Environmental Managers otherwise known as The Ten
States Standards 2012;
• The Norwegian Institute of Public Health (NIPH) and its guidelines; and
• The New York Department of Health (NYDOH) and its code.
Another purpose was to use the same data set to calculate the log inactivation of a target
pathogen such as Cryptosporidium and Giardia using the Generic Protocol for Development of
Test / Quality Assurance Plans for Validation of Ultraviolet (UV) Reactors, August 2011
1
-------
11/01/EPADWCTR (GP-2011) which is based on Ultraviolet Design Guidance Manual For the
Long Term 2 Enhanced Surface Water Treatment Rule, Office of Water, US Environmental
Protection Agency, November 2006, EPA 815-R-06-007 (UVDGM-2006).
The setline approach was based on validation testing at three set points (a set point is defined as a
single flow rate and irradiance output that delivers the targeted UV dose). The results of the three
set point tests were used to develop a setline that defines the maximum flow rate - minimum
irradiance output required to ensure the UV dose is achieved. The microorganism used for this
validation test was MS2 coliphage virus (MS2). The target UV dose was a measured Reduction
Equivalent Dose (REDmeas) of >40 ml/cm2. This dose was calculated based on the understanding
of dose calculations used internationally and by the Ten States Standards. The REDmeas was then
adjusted based on the uncertainty of the measurements to calculate a MS2 based validated dose
(REDVai) where the RED bias is set equal to one (1.0) in accordance with the unique approach of
the State of New York. The REDmeas data were also adjusted for uncertainty and the
Cryptosporidium and Giardia RED bias factors from UVDGM-2006 Appendix G. The data
were used to estimate the log inactivation of Cryptosporidium and Giardia so that a regulatory
agency could grant log credits under the Long Term 2 Enhanced Surface Water Treatment Rule
(LT2ESWTR). NeoTech selected flow rates of 150, 250, and 435 gpm as the target flow rates
based on their design for NeoTech D438™.
Based on the results of the three set points, a setline was developed for this unit. During full-
scale commercial operation, Federal regulations require that the UV intensity as measured by the
UV sensor(s) must meet or exceed the validated intensity (irradiance) to ensure delivery of the
required dose. Reactors must be operated within the validated operating conditions for maximum
flow rate - minimum irradiance combinations, UV transmittance (UVT), and lamp status [40
CFR 141.720(d)(2)]. Under the UV setline approach, UVT does not have to be measured
separately. The intensity readings by the sensor take into account changes in the UVT and the
setline establishes the operating conditions over a range of flow rates used during the validation
test.
This verification test did not evaluate cleaning of the lamps or quartz sleeves, nor any other
maintenance and operation.
1.3 Verification Test Site
UV dose validation testing was performed at the NSF Testing Laboratory in Ann Arbor,
Michigan. The NSF laboratory is used for all of the testing activities for NSF certification of
drinking water treatment systems, and pool and spa treatment systems.
1.4 Testing Participants and Responsibilities
The following is a brief description of each of the ETV participants and their roles and
responsibilities.
1.4.1 NSF International
NSF is an independent, not-for-profit organization dedicated to public health and safety, and to
protection of the environment. Founded in 1944 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. The USEPA partnered with NSF to verify the performance of drinking
water treatment systems through the USEPA's ETV Program.
NSF performed all verification testing activities at its Ann Arbor, MI location. NSF prepared the
test/QA plan (TQAP), performed all testing, managed, evaluated, interpreted, and reported on the
data generated by the testing, and reported on the performance of the technology.
Contact: NSF International
789 N. Dixboro Road
Ann Arbor, MI 48105
Phone: 734-769-8010
Contact: Mr. Bruce Bartley, Project Manager
Email: bartley@nsf.org
1.4.2 U.S. Environmental Protection Agency
USEPA, 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.
1.4.3 NeoTech Aqua Solutions, Inc. (formerly Ultraviolet Sciences, Inc.)
NeoTech supplied the UV test unit for testing, required reference sensors, detailed specifications
on the equipment, UV lamps, lamp sleeves, and duty sensors, and written and verbal instructions
for equipment operation. NeoTech also provided logistical and technical support, as needed.
Contact: NeoTech Aqua Solutions, Inc.
5893 Oberlin Drive, Suite 104
San Diego, California 92121
Phone: 1-858-571-6590 or 1-888-718-5040
Email: info@neotechaqua.com
Website: neotechaqua.com
-------
Chapter 2
Equipment Description
2.1 NeoTech General Information
NeoTech, headquartered in San Diego, California, designs and manufactures UV water treatment
systems for disinfection, TOC reduction, and chlorine reduction purposes. NeoTech water
treatment products are designed for industrial and commercial applications; pharmaceutical,
microelectronics, beverage, pools/spas, hospitality, water reclamation, and small municipal
drinking water delivery systems (<10 MGD).
The NeoTech team of scientists and engineers work in collaboration with a group of universities
to develop more efficient UV systems for water purification applications. NeoTech has taken
these innovations and applied them to a new product line of UV reactors.
NeoTech developed a highly reflective UV treatment chamber (US Patent 7,511,281) that
maximizes the use of the UV light emitted by conventional mercury amalgam lamps. NeoTech
has stated the following regarding the new treatment chamber:
The 99.8% reflective surface keeps the UV light inside the treatment chamber, reducing
the amount of light energy necessary to achieve a proper UV dose. This reflective
surface encapsulates the entire flow channel ensuring an even UV dose exists
throughout the treatment chamber. No complicated internal baffling and mixing systems
are necessary. This efficiency gain is unique to the NeoTech product line, resulting in
smaller UV systems with fewer UV lamps to achieve a given UV dose. As a result
NeoTech water treatment systems are more compact in design and require significantly
less power to operate, saving the end user up to 90% in operating costs.
The NeoTech product line of treatment chambers was launched commercially in March 2009
after nearly two years of field trials.
2.2 NeoTech UV System Description
The NeoTech Ultraviolet Water Purification System validated in this test was the largest flow
rated unit, NeoTech D438™. This unit is rated by NeoTech to handle 500 gpm. The system uses
two low-pressure mercury amalgam lamps and one intensity sensor mounted in a stainless steel
flow chamber. Figure 2-1 presents a picture of the system and basic dimensions of the system are
shown in Figure 2-2. NeoTech provided an operating manual (Attachment 1 of this report),
which included additional schematics and drawings with parts and dimensions of the reactor, the
sensors, the lamps and the quartz sleeve placement. NeoTech has also provided additional
information for the UV sensor (spectral data, measuring angle, measuring range, and output
range) and for UV lamps (lamp life, irradiance output, power requirements, aging data, etc.) as
required for the validation test. This information is presented in Attachment 2.
-------
Figure 2-1. NeoTech D438T
M
-------
40.55
1030
36.95
938.6
6.00
152.4
A
3.08
78.2
29.53
750.1
4X .40 10.1 THRU
7.90
200.7
f
7.96
202.2
i
11.40
289.5
Figure 2-2. NeoTech D438™ Basic Dimensions.
-------
2.3 NeoTech D438™ Specifications and Information
NeoTech provided the following information about their UV reactor:
Table 2.1. Basic UV Chamber Information
Manufacturer/Supplier
Type or model
Year of manufacture
Maximum flow rate
Net dry weight
Volume
Electrical power
Operating power consumption
Maximum pressure
Ambient water temperature
Maximum cleaning temperature
Inlet pipe size
NeoTech Aqua Solutions, Inc.
D438
2009
500 gpm
63 pounds
278.38 cu inches
120 VAC, 50/60Hz; 15 A max.
300 watts
150 psi static and transient
35 °F (2 °C) min.; 80 °F (27 °C) max.
Steam sterilize with pure steam up to 257 °F
for up to 90 minutes
3 inches
Table 2.2. Low Pressure Lamp Information
Type
Model
Number of lamps per reactor
UV emission at wavelengths ranging from
240-290 nm
Lamp life
Power supply unit's name, make and serial
numbers
Irradiance @lm
UV output
Operating lamp watts
Lamp current and voltage
Low-pressure
Light Sources, Inc. Model M1-4Y-01
NeoTech Part Number LK 38
2
> 90% irradiance at 253-256 nm
9000 hrs - Aging data in Attachment 2
Fulham WHS or WH7 ballast 120 V;
Power supply by NeoTech,
334 (mW/cm2)
35 W
111 W
1.3 A; 86 V
Table 2.3. UV Lamp Sleeve Information
Type or model
Quartz material
Pressure resistance
Heraeus Quartz;
Supersil310
lOOOkPa
Table 2.4. UV Sensor Information
Type / model
Measuring field angle
Number of sensors per reactor and placement
Signal output range
Measuring range output signal
UVIM-3 - 1660-002
180 degrees with cosine
degrees without cap
corrector cap, 64.7
1
4 - 20 mA
0-160 mW/cm2
-------
UV sensor spectral information has been provided by NeoTech to demonstrate the sensor meets
the basic requirements of the Generic Protocol for Development of Test/Quality Assurance Plans
for Validation of Ultraviolet (UV) Reactors, 7/2010 (GP-2010). The GP-2010 and updated GP-
2011 are based on the EPA's UVDGM-2006. These data are presented in Attachment 2. Sensor
calibration information was provided by NeoTech prior to the start of the test. Sensor calibration
was also checked with reference sensors as part of the test procedures. NeoTech has provided
information on the calibration (NIST traceable) of the equipment used to calibrate the sensors.
These data are provided Attachment 2.
2.4 NeoTech Ultraviolet Treatment System Standard Features
NeoTech has provided the following information on the features of the NeoTech series of
reactors.
Standard Features of all NeoTech systems include:
• 316L and 304 stainless steel construction;
• Ra-15 finish for all wet contact surfaces;
• Low Pressure Amalgam lamps;
• Static Operation up to 1 hour (no flow);
• NIST traceable UV intensity monitor with LED output on control panel;
- provides real-time lamp intensity information;
- 4-20mA signal for data logging of UV intensity;
• High operating pressure (150 psig);
• NEMA 4X Control panel with remote monitoring and control capability;
4-20mA scaled analog output;
- Remote Shut off;
- Alarms for lamp out and power off;
• Sanitary fittings and Viton gaskets;
• UV lamp replacement requires no tools required; and
• UL and CE approved. TUV verified.
-------
Chapter 3
Methods and Procedures
3.1 Introduction
The tests followed the procedures described in the Test/Quality Assurance Plan for The
Ultraviolet Sciences Inc. Ultraviolet (UV) Water Purification System Model UVXS438S Reactor,
August 2010 (TQAP). The TQAP was adapted from GP-2010. The GP-2010 and updated GP-
2011 version are based on the EPA's UVDGM-2006. The TQAP is available from NSF upon
request.
The approach used to validate UV reactors is based on biodosimetry, which determines the log
inactivation of a challenge microorganism during full-scale reactor testing for specific operating
conditions of flow rate, UVT, and UV intensity (measured by the duty sensor). A dose-response
equation for the challenge microorganism (MS2 coliphage for this test) is determined using a
collimated beam bench-scale test. The observed log-inactivation values from full-scale testing
are input into the collimated beam derived-UV dose-response equations to estimate a RED. The
RED value is adjusted for uncertainties and biases to produce the validated dose of the reactor
for the specific operating conditions tested.
The methods and procedures were designed to accomplish the primary objective of the validation
test of the NeoTech D438™, which was to develop a set line based on three set points (each set
point is a specific flow rate- UV intensity combination) that would ensure a measured RED
(REDmeas) of at least 40mJ/cm2 based on MS2 as defined by the "Ten States Standards". Test
procedures were also designed so that the REDmeas could be adjusted based on the uncertainty of
the measurements to calculate a MS2 based validated dose (REDvai) in accordance with the
unique approach of the State of New York. The REDmeas data were also adjusted for uncertainty
and the Cryptosporidium and Giardia RED bias factors from the UVDGM-2006 Appendix G.
These validated RED data can be used by states to evaluate applicable log credits under the
LT2ESWTR.
The GP-2010 requires the use of a second less sensitive challenge organism as part of the
validation. T7 was initially included in the ETV UV Generic Protocol in the 2010 version as a
result of research suggesting it could be a surrogate test microorganism with UV sensitivity
similar to the UV sensitivity of Cryptosporidium (Fallen et.al, JAWWA, 99.3, March 2007).
The GP-2010 technical advisory panel had reservations about using any test microorganism other
than MS2 which has an excellent record of quality control response for collimated beam
regression curves (Figure A.I in the UVDGM-2006). The ETV GP-2010 technical advisory
panel opinion was that other test microorganisms simply did not yet have the record of quality
control limits as did MS2.
In 2010 during some initial validation studies, NSF attempted to use T7. The strain referenced
by the JAWWA study (ATCC 11303-B7) was not available through ATCC. In fact, ATCC said
verbally that the strain mentioned was not T7 and was not available. With the counsel of the
EPA, NSF agreed to try using T7 ATCC BAA-1103-B38.
-------
Comments in 2011 on the GP-2010 also found reasons not to specify only T7: ". However, T7
cannot be produced at nearly as high a titer as Tl, so in the validation of high-flow reactors,
replacing all the Tl test conditions with T7 test conditions would consume an unacceptable
volume of raw phage stock." Consequently the GP-2010 technical advisory panel recommended
the use of any organism other than MS2 will be optional and the use of MS2 will be mandatory
for all types of reactors. The use of a challenge organism other than MS2 will be determined by
the consensus of stakeholders.
For the retesting done for this project, NSF chose to only use MS2 based on the concerns raised
about T7 by reviewers and the changes made in the 2011 ETV UV Protocol (GP-2011). Instead,
it was decided to illustrate how MS2 data was being used to satisfy many different regulatory
requirements while using essentially the same data. The basic biodosimetry data was used to
calculate the log inactivation of two target pathogens: Cryptosporidium and Giardia. The data
was also used to calculate the 40mJ/cm dose (REDmeas) requirement found in the "Ten States
Standards" and the NIPH guidelines, and the "validated" dose approach (REDvai) used by the
NYDOH.
UV reactor validation followed these steps:
1. Obtain the technical specifications for the system as provided by NeoTech;
2. Assessment of the UV sensors;
3. Collimated beam laboratory bench scale testing;
4. Full scale reactor testing;
5. Calculate the REDmeas; and
6. Adjust the REDmeas for uncertainty in UV dose and calculate a validated dose
(REDyai) for Cryptosporidium and Giardia to show the log inactivation.
r\
The target UV dosage was a REDmeas of > 40 ml/cm , based on MS2. NeoTech selected flow
rates of 150, 250, and 435 gpm as the target flow rates based on their system design for NeoTech
D438™ and screening and initial data from 2010.
3.2 UV Sensor Assessment
The NeoTech test unit duty sensor was evaluated according to the UV sensor requirements in the
UVDGM-2006 prior to and following the verification testing. All UV intensity sensors (the duty
and two reference sensors) were new sensors. Evidence of calibration of the sensors, traceable to
NIST, was provided by NeoTech.
The validation testing requires confirmation of the duty sensor spectral response to assess
whether the sensors are germicidal (see UVDGM-2006 Glossary for definition of germicidal)
with a defined spectral response of at least 90% between 200 and 300 nm. The technical
specifications of the NeoTech sensor and representation of sensitivity to the germicidal
wavelength were provided by NeoTech and found to meet the requirements. The technical
specifications of the NeoTech UV sensor and representation of sensitivity to the germicidal
wavelength are included in Attachment 2.
10
-------
During validation testing, the duty UV sensor measurement was compared to two reference
sensor measurements to assure the duty sensor was within 10% of the average of the two
reference sensor measurements.
The following steps were used to check the uncertainty of the duty and reference UV sensors.
The sensors were checked before and after the validation testing.
1. Step 1: Water was passed through the reactor at the maximum UVT and the maximum
lamp power setting to be used during validation testing.
2. Step 2: Using two calibrated reference UV sensors, each reference sensor was installed
on the UV reactor at the sensor port. The UV intensity was measured and recorded.
Step 2 was repeated using the duty UV sensor.
3. Step 3: Steps 1 and 2 were repeated at maximum UVT and lamp power decreased to the
minimum level expected to occur during validation testing.
4. Step 4: For a given lamp output and UVT value, the difference between the reference and
duty UV sensor measurements were calculated as follows:
The absolute value of [Sduty/ SAvgRef- 1]
Where:
S duty = Intensity measured by a duty UV sensor,
S Avg Ref = Average UV intensity measured by all the reference UV sensors
in the same UV sensor port with the same UV lamp at the same UV lamp
power.
3.3 Headloss Determination
Headloss through the unit was determined over the range of expected flow rates, in this case
from 150 gpm to 450 gpm. The inlet pressure near the inlet flange and the outlet pressure near
the outlet flange were measured at several flow rates. Measurements were recorded for flow rates
of 150, 250, 350 and 450 gpm. These data are reported Section 4.12.
3.4 Power Consumption Evaluation
The amperage and voltage used by the unit were measured during all reactor test runs.
Power data are presented in Section 4.11.
3.5 Feed Water Source and Test Rig Setup
The water source for this test was City of Ann Arbor, Michigan municipal drinking water. The
water was de-chlorinated using activated carbon and confirmed by laboratory testing. For the
lowered UVT conditions, the chemical Lignosulfonic Acid (LSA) was used to lower the UV
transmittance to a level that achieved a duty sensor reading at the selected UV intensity set point.
11
-------
LS A was added to the supply tank before each set of the lowered UVT runs and was well mixed
using a recirculating pump system. UVT was measured continuously using an in-line UVT meter
(calibrated daily) to confirm that proper UVT was attained.
NSF used a UV test rig and system setup that is designed to conform to the specifications
described in Sections 5.4.3 and 5.4.4 of the UVDGM-2006. Figure 3-1 shows a basic schematic
of the NSF test rig and equipment setup. The schematic is reproduced for informational purposes
and is copyright protected. Figure 3-2 shows a photograph of the actual equipment and piping
setup in the laboratory.
The feed water pump to the test unit was a variable speed pump. Flow rate was controlled by
adjusting the power supplied to the pump and by a control valve. A magnetic water flow meter
was used to monitor flow rate. The meter was calibrated and achieved the required accuracy of +
5%. A chemical feed pump (injector pump) was used to inject MS2 coliphage upstream of an
inline static mixer. The inline mixer ensured sufficient mixing of the microorganism prior to the
influent sampling port, which was located upstream of the 90° elbow installed directly on the
inlet to the unit. The effluent sampling port was located downstream of the test unit, downstream
of a 90° elbow attached directly to the unit outlet and downstream of a second inline static mixer.
This ensured good mixing of the treated water prior to the effluent sampling port. The 90°
elbows prevented stray UV light from exiting the unit.
A power platform that measures amperage, volts, watts, and power factor was used to monitor
power use by the test unit. The unit was wired into the platform and power consumption was
recorded for each test run.
12
-------
Fluid Flow
Dosing Pump
No'.e: A! plumbing is snedule du PVC
Figure 3-1. Schematic of NSF Test Rig
13
-------
Figure 3-2. Photograph of the Test Unit Setup.
14
-------
3.6 Installation of Reactor and Lamp Burn-in
The UV reactor and the reactor inlet and outlet connections were installed at the NSF laboratory
in accordance with the NeoTech installation and assembly instructions. Two 90° elbows, one
upstream and one downstream of the unit, were used in the test rig setup to eliminate stray UV
light. Figure 3-1 shows a schematic of the test rig setup and Figure 3-2 shows a photograph of
the actual equipment setup. The UV lamps were new and therefore the system was operated for
100 hours prior to the start of the tests to provide proper burn-in of the lamps.
There was one duty sensor and two lamps in the NeoTech system. Therefore, the lamp
positioning check requirements (i.e. checking each lamp and placing the lowest output lamp
closest to the sensor) were required for this validation. The sensor readings were basically the
same before and after the lamp positions were switched. The lamp positions that gave the highest
sensor reading were used for all of the test runs. This provided the most conservative approach.
3.7 Collimated Beam Bench Scale Testing
The collimated beam procedure involves placing a sample collected from the test rig and
containing MS2 in a petri dish, and then exposing the sample to collimated UV light for a
predetermined amount of time. The UV dose is calculated using the measured intensity of the
UV light, UV absorbance of the water, and exposure time. The measured concentration of
microorganisms before and after exposure provides the "response," or log inactivation of the
microorganisms from exposure to UV light. Regression analysis of measured log inactivation for
a range of UV doses produces the dose-response curve.
Appendix C of the UVDGM-2006 provides guidance on how to conduct the collimated beam
bench-scale testing and produce a UV dose-response curve. The following sections describe the
details of the collimated beam testing as performed by NSF.
3.7.1 Test Microorganism (Challenge)
MS2 coliphage (ATCC 15597-B1) was used in collimated beam bench scale testing and for the
full-scale reactor dose validation tests.
MS2 coliphage ATCC 15597-B1 is a recommended microorganism for UV lamp validation tests.
Further reasons for selecting this microorganism for UV validation are based on its inter-
laboratory reproducibility (UVDGM-2006), ease of use and culturing, and demonstrated
performance of MS2 in validation testing.
3.7.2 Test Conditions
The collimated beam tests were performed in duplicate at the minimum and maximum UVT test
conditions. For this validation the testing spanned a period of two (2) days for the MS2 test runs,
with Day 1 being the lowered UVT water tests and Day 2 the high UVT water lowered power
tests. Collimated beam tests were run in duplicate on the minimum UVT water (90-91%) on Day
1. Collimated beam tests were run in duplicate on the maximum UVT water (97-98%) on Day 2.
Thus, for this validation test, there are two sets of duplicate collimated beam test data for MS2,
one at low UVT and one at high UVT.
15
-------
For MS2, UV doses covered the range of the targeted RED dose, which in this case is 40mJ/cm2.
UV doses were set at 0, 20, 30, 40, 60 and 80 ml/cm2. The samples are clustered close to the
9 9
40mJ/cm target dose with two doses above and below the target of 40mJ/cm .
The collimated beam radiometers were calibrated to ensure that the measured UV intensity met
the criteria of an uncertainty of 8% or less at a 95% confidence level.
3.7.3 Test Apparatus
NSF uses a collimated beam apparatus that conforms to NSF/ANSI Standard 55c section 7.2.1.2.
Attachment 3 includes a description of the apparatus and is reproduced for informational
purposes and is copyright protected.
3.7.4 Collimated Beam Procedure
NSF collected two (2) one-liter samples from the influent sampling port of the test rig for
collimated beam testing. Each bottle was used for one of the replicates for the collimated beam
test. The MS2 spiked water was collected directly from the test rig each day during the test runs.
The collimated beam test water and microorganism culture were the same as used in the full
scale reactor tests.
NSF microbiological laboratory personnel followed the "Method for Challenge Microorganism
Preparation, Culturing the Challenge Organism and Measuring its Concentration" in Annex A of
NSF/ANSI Standard 55®. Please note that all reproduced portions of NSF/ANSI Standards are
copyright protected.
For collimated beam testing of a water sample containing challenge microorganisms, NSF's
laboratory followed this procedure:
1. Measure the A254 of the sample.
2. Place a known volume from the water sample into a petri dish and add a stir bar.
Measure the water depth in the petri dish.
3. Measure the UV intensity delivered by the collimated beam with no sample present
using a calibrated radiometer using a calibrated UV sensor. The UV sensor is placed at
the same distance from the radiometer as a sample.
4. Calculate the required exposure time to deliver the target UV dose described in the
next section.
5. Block the light from the collimating tube using a shutter or equivalent.
6. Center the petri dish with the water sample under the collimating tube.
7. Remove the light block from the collimating tube and start the timer.
8. When the target exposure time has elapsed, block the light from the collimating tube.
9. Remove the petri dish and collect the sample for measurement of the challenge
microorganism concentration. Analyze immediately or store in the dark at 4 °C (for up
to 6 hours). Multiple dilutions are used to bracket the expected concentration range
(e.g. sample dilutions of 10X, 100X, 1000X). Plate each dilution in triplicate and
16
-------
calculate the average microbial value for the dilution from the three plate replicates
that provide the best colony count.
10. Re-measure the UV intensity and calculate the average of this measurement and the
measurement taken in Step 3. The value should be within 5% of the value measured
in Step 3. If not, recalibrate radiometer and re-start at step 1.
1 1 . Using the equation described in the next section, calculate the UV dose applied to the
sample based on experimental conditions. The calculated experimental dose should
be similar to the planned target dose.
12. Repeat steps 1 through 1 1 for each replicate and target UV dose value. Repeat all
steps for each water test condition replicate.
The UV dose delivered to the sample is calculated using the following equation:
Where:
DCB = UV dose (ml/cm2);
Es = Average UV intensity (measured before and after irradiating the sample)
(mW/cm2);
Pf = Petri Factor (unitless);
R = Reflectance at the air-water interface at 254 nm (unitless);
L = Distance from lamp centerline to suspension surface (cm);
d = Depth of the suspension (cm);
A254= UV absorbance at 254 nm (unitless); and
t = Exposure time (s).
To control for error in the UV dose measurement, the uncertainties of the terms in the UV dose
calculation meet the following criteria:
• Depth of suspension (d)
-------
Concentration of microorganisms in the petri dish prior to UV exposure (No) in
units of plaque forming units (PFU)/mL, and
Concentration of microorganisms in the petri dish after UV exposure (N) in units
ofPFU/mL.
The procedure for developing the UV dose response curves was as follows:
1. For each UV test condition (high or low UVT water) and its replicate and for each day of
testing, log N (PFU/mL) was plotted vs. UV dose (mJ/cm2). A best fit regression line was
determined and a common N0 was identified as the intercept of the curve at UV dose = 0. A
separate equation was developed for each UVT condition (lowest and highest). In this test,
there were two days of testing for MS2, so there were two sets of data.
2. The log inactivation (log I) was calculated for each measured value of N (including zero-
dose) and the common N0 identified in Step 1 using the following equation:
log I = log(No/N)
Where:
No = The common N0 identified in Step 1 (PFU/mL); and
N = Concentration of challenge microorganisms in the petri dish after
exposure to UV light (PFU/mL).
3. The UV dose as a function of log I was plotted for each day of testing, and included water
from both high and low UVT test conditions.
4. Using regression analysis, an equation was derived that best fit the data, forcing the fit
through the origin. The force fit through the origin is used rather than the measured value
of No, because any experimental or analytical error in the measured value is carried to all
the data points, adding an unrelated bias to each measurement. Using the y-intercept of the
curve eliminates error carry through. The regression equation was then used to calculate the
REDmeas for each full scale test sample.
The full set of collimated beam data and all calculations and regression analyses are presented in
Chapter 4.
The regression analysis was used to derive an equation that best fits the data with a force fit
through the origin. Both linear and a quadratic equations were evaluated to determine the best fit
of the data. The regression coefficient, R2, was determined for each trend line and was
considered acceptable if it was 0.9 or greater and for "r" ± 0.95 or greater. The equation
coefficients for each day were also evaluated statistically to determine which terms were
statistically significant based on the P factor. A second order polynomial gave the best fit for the
collimated beam dose response curves.
For this validation a single curve corresponding to one day's worth of full scale reactor testing
was used to calculate REDmeas values for that day. The higher UVT dose response curve was
18
-------
used for the high UVT water day with reduced power and the lower UVT dose response curve
was used for the day when the UVT of the test water was lowered with LSA.
3.7.6 Collimated Beam Data Uncertainty
The collimated beam data was fit to a polynomial regression and the uncertainty of the dose
response equation based on a 95% confidence interval (UDR) was calculated as follows:
UDR = t * [SD/ UV DoseCB] * 100%
Where:
UDR= Uncertainty of the UV dose-response fit at a 95% confidence level;
UV DosecB = UV dose calculated from the UV dose-response curve for the
challenge microorganism;
SD = Standard deviation of the difference between the calculated UV dose
response and the measured value, and
t = t-statistic at a 95% confidence level for a sample size equal to the number of
test condition replicates used to define the dose-response.
The UDR calculations and results are included in Section 4.4
3.8 Full Scale Testing to Validate UV Dose
3.8.1 Test Conditions for UV Intensity Set-Point Approach
The purpose of this testing was to determine a REDmeas dose of >40 ml/cm2 at three set points
that were then used to establish a set line based on the three UV intensity and flow rate pairs.
NeoTech specified the target flow rates (150, 250, 435 gpm) and UV target intensity levels (7.5,
10.0, 13.0 mW/cm2) based on the results of screening test performed at NSF prior to the
validation tests. The intensity targets were based on the expected intensity at UVTs of 91%,
94%, and 97%.
Each set point represents a given flow rate with testing under two conditions, (1) lowered UVT-
maximum power and (2) high UVT-reduced power. The first test condition involved reducing
the UVT until the UV intensity measured by the unit UV sensor equaled the target UV intensity
set point. The second test condition was run with high UVT and with the power reduced until the
unit UV intensity measured by the sensor was equal to the target UV intensity set point. Three
target flow rates - intensity points (150 gpm - 7.5 mW/cm2; 250 gpm - 10.0 mW/cm2; 435 gpm -
13.0 mW/cm2) were tested for the set line. All conditions were performed in duplicate. The
intensity targets were based on expected intensity at UVT's of 91%, 94%, and 97%.
The LT2ESWTR requires validation of UV reactors to determine a log inactivation of
Cryptosporidium or other target pathogens so that States may use the data to grant log credits.
Therefore, in addition to determining the setline to achieve a minimum REDmeas of 40 ml/cm2,
additional calculations were run to show an example of determining the log inactivation of
Cryptosporidium and Giardia by adjusting the REDmeas for uncertainty and RED bias.
19
-------
A reactor control test (MS2 injection with the lamp off) was run at the low flow rate (150 gpm)
and with high UVT water, which demonstrated that there was no inactivation of MS2 with the
lamps off. A reactor blank was also run on each day of testing. The reactor blank was run with
no phage injection at the low flow rate with the lamps at full power to demonstrate the testing
system was low in MS2 concentration and other microorganisms. Reactor blank and control
samples were collected in triplicate at the influent and effluent sampling locations and submitted
for MS2 analyses.
Trip blanks were prepared and analyzed for each day of testing. The microbiology laboratory
took two samples from the challenge solution prepared for one of the test runs. The first sample
remained in the microbiology laboratory and the second sample traveled with challenge solution
to the engineering laboratory and then was returned with the samples collected from the test run.
Both samples were analyzed for MS2 and the results were compared to determine any change
that might have occurred during transport of the samples. As with stability testing, trip blanks are
important when samples must be shipped or carried long distance with the inherent holding time
before delivery to the lab. At NSF, the test rig and laboratory are in the same building and the
trip is "down the hall". Therefore, travel related impacts are of less concern, but trip blanks were
run as part of the QC plan for these tests.
Table 3-1 shows a summary of the test conditions that were run for the validation test. A Sample
and Analysis Management Program was also prepared and was provided to the NSF engineering
and microbiology laboratories for use during the testing and for setting up the sample and
analysis in the NSF sample management system.
Five sets of samples were collected at the influent and effluent sample ports for MS2 analysis
during each test condition and it's duplicate. The delivered dose was calculated for each of the
five samples and then the average of the five results was calculated to determine an average
delivered dose (REDmeas).
Flow rate, intensity, and UVT data (from the NSF in-line UVT monitor) were collected at each
of the five sample collection times for all test runs. These data were averaged to determine the
average flow rate, UVT, and intensity for each test condition and it's duplicate.
In addition, samples for pH, turbidity, temperature, total and residual chlorine, E coli, and
heterotrophic plate count (HPC) were collected at the influent and effluent sample ports once
during each test run. Samples for iron (Fe) and manganese (Mn) analyses were collected once
during each test run at the influent sample port to provide additional basic water quality data.
Samples were also collected at the influent and effluent for UVT analysis by the chemistry
laboratory bench scale spectrophotometer to confirm the in-line UVT measurements.
20
-------
Table 3-1. Test Conditions for Validation
Validation Test
Condition 1
Condition 2
Condition 3
(reactor control)
Condition 4
(reactor blank)
Flow Rate
150 gpm
250 gpm
435 gpm
150 gpm
250 gpm
435 gpm
150 gpm
150 gpm
UVT (%)
91%
94%
97%
> 97%
> 97%
>97%
>97%
Daily Source water
- ether high or low
UVT
Lamp Power
Maximum
Lowered to
achieved intensity
from Condition 1
Turned off
Full Power
Intensity
Sensor Reading
Record actual
reading
Set to equal
Condition 1 by
lowering lamp
power
Not applicable
Record
Condition 1 and 2 performed in duplicate
Reactor blanks run for each day of testing
UVT scan of feed water with and without UVT adjustment
Trip blanks and method blanks run for each day of testing
3.8.2 Preparation of the Challenge Microorganisms
The challenge microorganism (MS2) used to validate the UV reactor was cultured and analyzed
by NSF's microbiology laboratory as specified in Standard Methods for the Examination of
Water and Wastewater. NSF microbiological laboratory personnel followed the method for
"Culture of challenge microorganism" in Annex A of NSF/ANSI Standard 55.
1 9
Propagation resulted in a highly concentrated stock solution (approximately 1.0x10 PFU/mL)
of essentially monodispersed phage whose UV dose-response follows second-order kinetics with
minimal tailing. Over the range of RED values demonstrated during validation testing, the mean
UV dose-response of the MS2 phage stock solution was within the 95-percent prediction interval
of the mean response in Figure A.I in Appendix A of the UVDGM-2006. Over a UV dose range
of 0 to 120 mJ/cm2, the prediction intervals of the data shown in Appendix A of the UVDGM-
2006 are represented by the following equations"
Upper Bound: log/= -1.4x10 *UVDose2 + 7.6x10 2 *UVDose
Lower Bound: log/ = -9.6xlO~5 *UVDose2 + 4.5xlO~2 *UVDose
City of Ann Arbor tap water was filtered using activated carbon to remove any residual chlorine
(confirmed by chemical analysis for total chlorine of the test water), organic surfactants and
dissolved organic chemicals that may be UV absorbers. The filtered challenge water was then
tested for the following parameters and found acceptable if the result is non-detectable or as
otherwise indicated below:
21
-------
• Total Chlorine;
• Free Chlorine;
• UV254 ;
• UVT > 95%;
• Total Iron;
• Total Manganese;
• Turbidity < 0.3 Nephelometric Turbidity Units (NTU);
• Total coliform (<1 CFU/lOOmL); and
• Heterotrophic Plate Count (<100 CFU/mL).
3.8.3 Testing - Measuring UV Dose
During full-scale reactor testing, the reactor was operated at each of the test conditions for flow
rate, UVT, and lamp power as described in section 3.6.1. The following steps were taken to
assure meeting data quality objectives:
1. Steady-state conditions were confirmed before injecting the challenge microorganism.
Confirmation of steady state involved monitoring the flow rate, UV sensor measurements
and the UVT to assure the test water and reactor meet the test conditions. After typically
3-5 minutes of operation and confirmation that UVT, sensor readings, and flow rate were
steady, the injection pump was started and steady state conditions were demonstrated by
waiting until the injection pump was at a steady flow rate based on measurements of
weight loss of solution over 15-second time intervals. In all cases, sampling did not start
until at least two minutes after the injection pump was started.
2. MS2 was injected into the feed water flow upstream of the reactor to achieve a
concentration greater than IxlO5 PFU/mL so that a minimum of a 4-log inactivation
could be measured during the runs.
3. Sample taps remained open over the duration of the test.
4. Samples were collected in accordance with standards of good practice as defined by
Standard Methods Section 9060.
5. Five (5) sample pairs were collected during approximately ten to twelve minutes of
continuous flow at steady conditions. Each set of influent and effluent grab samples were
collected as close in time as possible. The five sets of samples were spread out over the
10 to 12-minute continuous flow run.
6. Samples for assessing the challenge microorganism concentrations in the influent and
effluent were collected in 125 mL bottles.
7. Samples were collected in bottles that have been cleaned and sterilized by the NSF
laboratory.
8. Collected samples were delivered directly to the microbiological lab located in the same
building after each sampling period. Sample analysis was generally started immediately,
but could be stored in the dark and analysis started a few of hours later. All MS2 analysis
was started within 4-6 hours of the time the samples were collected.
22
-------
The following measurements and recordings were taken during each test run:
1. The flow rate through the reactor, UV sensor reading and on-line UVT measurements
were recorded when each sample was collected during each run, yielding a minimum of
five measurements for each test run.
2. Water chemistry and other microbiological grab samples were collected once per test
condition after one of the challenge organism samples were collected. Samples for
temperature, pH, E. coli, and HPC were collected at the influent and effluent locations,
and samples for iron, manganese, turbidity and residual chlorine were collected at the
influent location.
3. A sample for UVT was collected and measured by a UV spectrophotometer for each
influent sample and at least one effluent sample;
4. A sample of the influent and effluent water was collected at the beginning of each test
day and a UVT scan performed over the range of 200 to 400 nm, and
5. The electrical power consumed by system was recorded.
Chapter 4 describes the calculations and presents the data for determining the REDn
validated dose (REDyai) at a each set point.
3.9 Analytical Methods
All laboratory analytical methods for water quality parameters are listed in Table 3-2.
Table 3-2. Analytical Methods for Laboratory Analyses
and the
Parameter
Temperature
pH
E. coli 1 Total
Coliform
Iron
Manganese
Turbidity
MS2
Absorbance UV254
Residual chlorine
HPC
Method
SM(2) 2550
SM 4500-H+B
SM 9223
EPA 200.7
EPA 200.8
SM2130
Top Agar
Overlay
SM5910B
SM 4500-C1 D
SM9215B
NSF
Reporting
Limit
-
1CFU
/lOOmL
20 ug/L
1 ug/L
0.1 NTU
1 PFU/mL
NA
0.05 mg/L
1 CFU/mL
Lab Accuracy
(% Recovery)
-
±.1SU
of buffer
-
70-130
70-130
95-105
-
60-140
90-110
-
Lab
Precision
(%RPD (1))
-
±0.1 SU
-
10%
10%
-
-
<20
<10%
-
Hold
Time
-
(3)
24
hours
180
days
180
days
(3)
24
hours
2 days
(3)
24
hours
Sample
Container
-
NA
500 mL
plastic
125 mL
polyethylene
125 mL
polyethylene
NA
125 mL
plastic
1 L plastic
NA
125 mL
plastic
Sample
Preservation
-
None
1% Tween 80
Nitric acid
Nitric acid
None
1% Tween 80
None
None
1% Tween 80
(1) RPD = Relative Percent Deviation
(2) SM = Standard Methods
(3) Immediate analysis required
23
-------
3.9.1 Sample Processing, and Enumeration of MS2:
MS2 sample processing and enumeration followed the procedures used in NSF/ANSI Standard
55.
3.9.2 Percent UVT Measurements:
The percent UVT for laboratory measurements was calculated from A254. The equation for UVT
using A254 is:
UVT(%)= 100* 10 -A254
The on-line UVT analyzer provided immediate data throughout all test runs. The on-line
analyzer was calibrated every day of operation. A primary standard was used at a minimum on
the first day of testing. Daily calibration was performed on all test days using a certified
secondary standard. Before the start of each day's testing a sample was taken to the laboratory
and analyzed for direct comparison with the on-line analyzer to ensure the data were comparable.
All UVT measurements used a 1-cm path length and are reported on a 1-cm path length basis.
Spectrophotometer measurements of A254 were verified using NIST-traceable potassium
dichromate UV absorbance standards and holmium oxide UV wavelength standards. The UV
Spectrophotometer internal QA/QC procedures outlined in the UVDGM-2006 were used to
verify calibration. UV absorbance of solutions used to zero the Spectrophotometer were verified
using reagent grade organic-free water certified by the supplier to have zero UV absorbance.
The measurement uncertainty of the Spectrophotometer must be 10% or less. To achieve this
goal, the following procedures were used:
1. Verify that the Spectrophotometer reads the wavelength to within the accuracy of a
holmium oxide standard (typically ± 0.2 nm at a 95-percent confidence level),
2. Verify that the Spectrophotometer reads A254 within the accuracy of a dichromate
standard (e.g., 0.281 ± 0.005 at 257 nm with a 20 mg/L standard), and
3. Verify that the water used to zero the instrument has an A254 value that is within 0.002
cm"1 of a certified zero absorbance solution.
3.9.3 Analytical QA/QC Procedures
Accuracy and precision of sample analyses were ensured through the following measures:
• pH - Three-point calibration (4, 7, and 10) of the pH meter was conducted daily using
traceable buffers. The accuracy of the calibration was checked daily with a pH 8.00
buffer. The pH readings for the buffer were within 10% of its true value. The precision
of the meter was checked daily using duplicate synthetic drinking water samples. The
difference of the duplicate samples was within ±0.1 SU.
24
-------
• Temperature - The thermometer used to give the reportable data had a scale marked for
every 0.1°C. The thermometer is calibrated yearly using a Hart Scientific Dry Well
Calibrator Model 9105.
• Total chlorine - The calibration of the chlorine meter was checked daily using a DI water
sample (blank), and three QC standards. The measured QC standard values were within
10% of their true values. The precision of the meter was checked daily by duplicate
analysis of synthetic drinking water samples. The RPD of the duplicate samples was less
than 10%.
• Turbidity - The turbidimeter was calibrated as needed according to the manufacturer's
instructions with formazin standards. Accuracy was checked daily with a secondary
Gelex standard. The calibration check provided readings within 5% of the true value.
The precision of the meter was checked daily by duplicate analysis of synthetic drinking
water samples. The RPD of the duplicate samples was less than 10% or had a difference
of less than or equal to 0.1 NTU at low turbidity levels.
3.9.4 Sample Handling
All samples were labeled with unique identification numbers. These identification numbers were
entered into the NSF Laboratory Information Management System (LEVIS), and were used on the
NSF lab reports for the tests. All challenge organism samples were stored in the dark at 4 + 2 °C
and processed for analysis within 4-6 hours.
3.10 Full Scale Test QA Controls
The following quality-control samples and tests for full-scale reactor testing were performed:
• Reactor controls - Influent and effluent water samples were collected with the reactor
UV lamps in the reactor turned off. The change in log concentration from influent to
effluent should correspond to no more than 0.2 logic.
• Reactor blanks - Influent and effluent water samples were collected with no addition of
challenge microorganisms to the flow passing through the reactor. Blanks were collected
once on each day of testing. The reactor blank is acceptable when the MS2 concentration
is less than 0.2 logio.
• Trip controls - Trip controls were collected to monitor any change in challenge
microorganisms during transport to the laboratory (in the same building).
• Method blanks - A sample bottle of sterilized reagent grade water was analyzed
following the challenge microorganism assay procedure. The concentration of challenge
microorganism with the method blank was non-detectable.
• Stability samples - Influent and effluent samples at low and high UVT were collected
prior to the introduction of MS2. These samples were used to assess the stability of the
challenge microorganism concentration and its UV dose-response over the time period
from sample collection to completion of challenge microorganism assay. The challenge
microorganism was added to achieve a concentration of approximately IxlO3 PFU/L in
the samples containing test water at the lowest and highest UVT. A sample was analyzed
immediately (called time 0) and then 4 h, 8 h and 24 h after time 0. All analyses were
performed in triplicate. While stability samples were performed during the test, they are
25
-------
not directly applicable in this case as all sample analyses for MS2 were started within a 4
to 6 hours of collection.
3.11 Power Measurements
The voltmeter and ammeter used to measure UV equipment voltage and amperage had traceable
evidence of being in calibration (e.g., have a tag showing that it was calibrated). Calibrations are
performed at least yearly and all power equipment was calibrated within the past year.
3.12 Flow Rate
During validation testing, the QC goal was that the accuracy of flow rate measurements should
be within ± 5% of the true value. Flow meter accuracy was verified by monitoring the draw
down volume in the supply tanks over time. The supply tanks have been calibrated using the
catch and weigh technique. The flow meter was within 1.66% of the true value. Flow meter
calibration data are presented in Section 5.6.
3.13 Evaluation, Documentation and Installation of Reactor
NeoTech provided technical information on the NeoTech D438™ and basic information on the
UV lamps, sensor, and related equipment. An operating manual was provided. Additional
information on the lamp output (confirmation of spectral output) was provided prior to the start
of the validation test. All documentation and equipment data was reviewed prior to the start of
testing. The following documentation was reviewed and found to conform to the GP-2011 and
UVDGM-2006 requirements:
Reactor Specifications:
• Technical description of the reactor's UV dose-monitoring strategy, including the use of
sensors, signal processing, and calculations (if applicable);
• Dimensions and placement of all wetted components (e.g., lamps, sleeves, UV sensors,
baffles, and cleaning mechanisms) within the UV reactor;
• A technical description of lamp placement within the sleeve; and
• Specifications for the UV sensor port indicating all dimensions and tolerances that impact
the positioning of the sensor relative to the lamps.
Lamp specifications:
• Technical description;
• Lamp manufacturer and product number;
• Electrical power rating;
• Electrode-to-electrode length; and
• Spectral output of the lamps (specified for 5 nm intervals or less over a wavelength range
that includes the germicidal range of 250 - 280 nm and the response range of the UV
sensors).
Lamp sleeve specifications:
• Technical description including sleeve dimensions;
• Materials of construction; and
• UVT at 254 nm for the medium pressure (MP) lamp with germicidal sensors.
Specifications for the reference and the duty UV sensors:
26
-------
• Manufacturer and product number; and
• Technical description including external dimensions.
Sensor measurement properties:
• Working range;
• Spectral and angular response;
• Linearity;
• Calibration factor;
• Temperature stability; and
• Long-term stability.
Installation and operation documentation:
• Flow rate and pressure rating of the reactor;
• Assembly and installation instructions;
• Electrical requirements, including required line frequency, voltage, amperage, and power;
and
• Operation and maintenance manual including cleaning procedures, required spare parts,
and safety requirements.
27
-------
Chapter 4
Results and Discussion
4.1 Introduction
The validation tests to demonstrate a minimum REDmeas of 40 ml/cm were run on August 2 and
3, 2012. The first day of testing was dedicated to the test conditions and duplicate runs where the
UVT of the feed water was lowered to the target levels (91%, 94%, and 97%) and the lamps
were operated at full power. The second day of testing was dedicated to the test conditions and
duplicates where highest UVT feed water (97% target) was used and the lamp power was
reduced to achieve the target intensity level. The test conditions and detail on the test rig setup,
sampling procedures, and unit operation have been described in Chapter 3 Methods and
Procedures.
All tests were conducted at the NSF laboratory in Ann Arbor MI and all analyses were
performed by the NSF microbiological and chemistry laboratories at this location.
4.2 Sensor Assessment
The same duty sensor was used for monitoring intensity (irradiance) for all test runs. This sensor
measured the intensity from the two low pressure lamps in the unit. The control panel provided
direct readings of intensity in mW/cm . This direct reading was based on converting the 4-20 mA
output signal to intensity based on the calibration set by NeoTech.
The duty sensor was compared against two reference sensors to demonstrate that the duty sensor
was within 10% of the average of the two reference sensors. This evaluation was conducted
before and after the validation test runs, using the procedure described in Section 3.2. Table 4-1
presents the results of the sensor assessment. These data demonstrate that the duty sensor was
within 10% of the average of the two reference sensors. The two reference sensors showed a
variance of 2.9% at 100% power and 1.3% and 6.8% at reduced power.
28
-------
Table 4.1. Sensor Assessment Data
Sensor
Reference #1
2012-08
Reference #2
2012-09
Average of
Reference Sensor
Duty Sensor
2012-10
Deviation of Duty
Sensor from
Reference average
Intensity at
100% Power
Before testing
90.5% UVT
(mW/cm2)
8.6
9.1
8.85
8.2
7.3
Intensity at
100% Power
After testing
98.2% UVT
(mW/cm2)
17.0
18.0
17.5
18.4
5.1
Intensity with
Power Reduced
Before testing
90.5% UVT
(mW/cm2)
3.4
3.9
3.65
3.9
6.8
Intensity with
Power Reduced
After testing
98.2% UVT
(mW/cm2)
7.8
8.0
7.9
8.6
8.8
4.3 Collimated Beam Dose Response Data
Collimated Beam dose response data were generated for each of the test days for MS2 in
accordance with the procedures described in Section 3.5. On test Day 1, collimated beam tests
for MS2 were run on the minimum UVT water (91%). On test Day 2, collimated beam tests were
run on maximum UVT water (97%). All collimated beam tests were performed in duplicate.
UV doses for the MS2 tests covered the range of the targeted RED dose, which in this case was
40mJ/cm2. UV doses were set at 0, 20, 30, 40, 60 and 80 mJ/cm2.
The collimated beam samples were collected directly from the test rig during the normal testing
runs. Two 1 L bottles of the seeded influent water (MS2 was injected into the influent water
during a test run) were collected to provide the two samples for duplicate analyses. Using this
approach, the dose response data reflect the identical conditions to the biodosimetric flow tests
for sample matrix, UVT, and MS2 concentration. The collimated beam samples were irradiated
on the same day as sample collection, and were plated in triplicate along with the flow test
samples. Therefore, analytical conditions for the dose response data were also identical to those
for the flow test samples.
The collimated beam results for MS2 are presented in Tables 4-2 and 4-3. These data were
calculated as the average of the three individual results obtained at each dose level.
4.4 Development of Dose Response
The development of the UV dose response curves for MS2 for use with flow test data to establish
the REDmeas is a three step process.
29
-------
1. For each MS2 collimated beam test and its replicate for each day of testing, the log N
(PFU/mL) was plotted vs. UV dose (mJ/cm2). Figures 4-1 and 4-2 show the curves for
each day.
2. A separate equation (second order polynomial) was developed for each UVT condition
(low and high). In this test there were two days of testing for MS2, so there were two sets
of data. A common N0 was identified for each data set as the intercept of the curve at UV
dose = 0.
3. The log inactivation (log I) was then calculated for each day for each measured value of
N (including zero-dose) and the common N0 identified in Step 1 using the following
equation:
log I = log(No/N)
Where:
No = The common N0 identified in Step 1 (PFU/mL); and
N = Concentration of challenge microorganisms in the petri dish after
exposure to UV light (PFU/mL).
Tables 4-2 and 4-3 show the calculated values for MS2 log inactivation.
Finally, the UV dose as a function of log I was plotted for each day of the testing. Figures 4-3
through 4-4 show the curves for dose as a function of log I. Using regression analysis, an
equation was derived that best fit the data, forcing the fit through the origin. The equation used
for the dose response curve was a second order polynomial, which is the most common for MS2
collimated beam data. The regression equation was then used to calculate the REDmeas for each
full scale flow test sample. REDmeas calculations and full scale data is presented in the Section
4.5.
A Grubb's test was run to determine if any replicates should be omitted from the development of
the dose response curve. The Grubb's test results show that there was one outlier for replicate 2
for the high UVT water. The calculations were rerun after removing the outlier and that data was
used for the subsequent determination of dose response and the RED calculations. Table 4-4
shows the data with the outlier removed. Both sets of data are presented for informational
purposes. There is very little difference in the dose response curve coefficients between the data
set with the outlier included when compared to the one with the outlier removed.
A summary of the statistics for uncertainty for the collimated beam dose response data is
presented at the end of Tables 4-2, 4-3 and 4-4. The uncertainty (UDR) of the collimated beam
test was slightly higher than 30% at 1 log inactivation for the low UVT water (32.58%). The UDR
for the high UVT water was 12.25% once the outlier was removed (19.35% with the outlier
included). At 2-log inactivation (dose of approximately 40 mJ/cm2 RED) the UDR was 15.31%
and 5.53%. The uncertainty of the collimated beam results for Day 1 is greater than 30%.
Therefore, the uncertainty calculations for the Cryptosporidium and Giardia log inactivation
calculations presented in Sections 4.7 and 4.8 include the highest UDR value (32.58%) applied to
30
-------
all of the validation factors for the set points. The highest UDR is also used for calculating the
validation factor for calculating the REDVai based on MS2 as the target organism, as shown in
Section 4.9.
Figures 4-3 through 4-4 show the results of the UDR calculations plotted on the dose response
curve. Also shown in Figures 4-3 and 4-4 are the QC limits for MS2 taken from the UVDGM-
2006. The results show that the MS2 dose response curves are within the boundaries established
forMS2.
The polynomial equation coefficients for each day (Aug 2 and 3, 2012) were evaluated
statistically to determine which terms were statistically significant based on the P factor. All
coefficients were found to be significant except for the x term for the August 2, 2012 day of
testing. The P factor was 0.0539, just above the 0.05 significance test cutoff. When P is greater
than 0.05, it indicates the dose-response relationship could be linear rather than second order.
Both linear and polynomial regressions were evaluated for this data set. The polynomial equation
was selected for the REDmeas calculations because the R2 coefficient was slightly better (0.996
versus 0.994), the standard error was slightly less (3.356 versus 3.755) and the other set of
collimated beam data was a second order polynomial equation. Also the polynomial equation
gave the more conservative (lower) results for the REDmeas. If the linear relationship is used, the
REDmeas is approximately 6% higher than REDmeas calculated using the second order polynomial
equation.
-------
Table 4-2. UV Dose - Response Data from Collimated Beam Tests at 91% UVT
UVT
(%)
90.6
Rep
1
2
Target UV
Dose
(mJ/cm2)
0
20
30
40
60
80
0
20
30
40
60
80
Actual
UV
Dose
0.00
20.82
31.25
41.54
62.27
82.56
0.00
20.73
31.12
41.37
62.05
82.74
UV
Dose2
0
433
977
1726
3878
6816
0
430
968
1711
3850
6846
Avg (PFU/mL)
1,820,000
140,000
80,300
18,200
6,970
707
2,190,000
146,000
84,700
23,600
7,030
873
Avg Log(PFU/mL)
6.26
5.15
4.90
4.26
3.84
2.85
6.34
5.16
4.93
4.37
3.85
2.94
Log I
-0.02
1.09
1.33
1.98
2.40
3.39
-0.10
1.07
1.31
1.87
2.39
3.30
Log I2
0.000
1.194
1.780
3.916
5.739
11.488
0.010
1.155
1.719
3.482
5.721
10.876
Predicted
Dose
-0.43
24.05
29.81
46.00
57.09
85.49
-2.07
23.62
29.25
43.09
56.99
82.75
Avg:
SD:
n:
P:
t (95%):
Residual
(mJ/cm2)
0.4
-3.2
1.4
-4.5
5.2
-2.9
2.1
-2.9
1.9
-1.7
5.1
0.0
0.07
3.20
12
0.05
2.228
G
0.1
1.0
0.4
1.4
1.6
0.9
0.6
0.9
0.6
0.6
1.6
0.0
Outlier?
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
Grubb's Statistic
32
-------
Table 4-2. (continued)
Uncertainty of Dose-Response (UDR)
Log I
0.001
0.25
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
3.39
Dose
0.0
5.2
10.6
21.9
33.9
46.6
59.9
74.0
88.8
104.3
85.5
t
2.23
2.23
2.23
2.23
2.23
2.23
2.23
2.23
2.23
2.23
SD
3.20
3.20
3.20
3.20
3.20
3.20
3.20
3.20
3.20
3.20
UDR (%)
136.89
67.31
32.58
21.05
15.31
11.89
9.63
8.03
6.83
8.34
DL
(mJ/cm2/Log T)
20.48
20.83
21.18
21.88
22.58
23.28
23.98
24.68
25.38
26.08
25.22
t - student t test factor SD - standard deviation
Regression Statistics
Multiple R
R Square
Adjusted R Square
Standard Error
Observations
0.99796
0.995924
0.895516
3.356386
12
ANOVA
Regression
Residual
Total
df
2
10
12
ss
27522.43
112.6533
27635.09
MS
13761.22
11.26533
F
1221.555
Significance F
1.0994 1E-11
Intercept
X Variable 1
X Variable 2
Coefficients
0
20.48157
1.398414
Standard Error
1.751277
0.640225
tStat
11.695219
2.1842525
P-value
3.72E-07
0.053857
Lower 95%
16.57948
-0.0281
Upper 95%
24.38365
2.824924
Lower
95.0%
16.57948
-0.0281
Upper 95.0%
24.38365272
2.824924411
33
-------
Table 4-3. UV Dose - Response Data from Collimated Beam Tests at 97% UVT
UVT
(%)
97.7
Rep
1
2
Target
UV Dose
(mJ/cm2)
0
20
30
40
60
80
0
20
30
40
60
80
Actual
UV
Dose
0.00
20.79
31.09
41.34
61.63
81.83
0.00
20.67
30.85
41.05
61.40
81.78
UV
Dose2
0
432
967
1709
3798
6696
0
427
952
1685
3770
6688
Avg
(PFU/mL
)
3,770,000
173,000
74,300
19,900
3,070
577
3,470,000
128,000
70,000
20,100
2,980
500
Avg
Log(PFU/mL
)
6.58
5.24
4.87
4.30
3.49
2.76
6.54
5.11
4.85
4.30
3.47
2.70
Log I
-0.06
1.28
1.64
2.21
3.03
3.75
-0.03
1.41
1.67
2.21
3.04
3.81
Log I2
0.004
1.626
2.696
4.903
9.156
14.077
0.001
1.976
2.782
4.884
9.235
14.547
Predicte
dDose
-0.90
21.49
28.83
41.31
61.17
81.07
-0.39
24.05
29.37
41.21
61.50
82.87
Avg:
SD:
n:
P:
t (95%):
Grubb's Test for O
P:
t (90%):
Grubb's Statistic
(GCRIT):
Residual
(mJ/cm2)
0.9
-0.7
2.3
0.0
0.5
0.8
0.4
-3.4
1.5
-0.2
-0.1
-1.1
0.07
1.42
12
0.05
2.228
utliers
0.10
3.691
2.412
G
0.6
0.5
1.5
0.0
0.3
0.5
0.2
2.4
1.0
0.2
0.1
0.8
Outlier?
OK
OK
OK
OK
OK
OK
OK
OUTLIER
OK
OK
OK
OK
34
-------
Table 4-3. (continued)
Uncertainty of Dose-Response (UDR)
Regression Statistics
Multiple R
R Square
Adjusted R
Square
Standard Error
Observations
0.999591
0.999183
0.899101
1.488914
12
Log I
0.001
0.25
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
3.81
Dose
0.0
3.7
7.7
16.3
25.9
36.5
48.0
60.5
73.9
88.3
82.9
t
2.23
2.23
2.23
2.23
2.23
2.23
2.23
2.23
2.23
2.23
SD
1.42
1.42
1.42
1.42
1.42
1.42
1.42
1.42
1.42
1.42
UDR (%)
84.88
41.11
19.35
12.18
8.66
6.58
5.22
4.27
3.58
3.81
DL (mJ/cm2/Log
I)
14.41
14.89
15.37
16.33
17.28
18.24
19.20
20.16
21.12
22.08
21.73
t - student t test factor SD - standard deviation
ANOVA
Regression
Residual
Total
df
2
10
12
ss
27102.05
22.16864
27124.22
MS
13551.02
2.216864
F
6112.7
Significance F
7.94E-15
Intercept
X Variable 1
X Variable 2
Coefficients
0
14.40581
1.919398
Standard Error
0.708567
0.22588
tStat
20.33092
8.497439
P-value
1.83E-09
6.92E-06
Lower 95%
12.82703
1.416107
Upper 95%
15.9846
2.422689
Lower 95.0%
12.82703
1.416107
Upper 95.0%
15.9846
2.422689
35
-------
Table 4-4. UV Dose - Response Data from Collimated Beam Tests at 97% UVT with outlier removed
UVT
(%)
97.7
Rep
1
2
Target
UV Dose
(mJ/cm2)
0
20
30
40
60
80
0
30
40
60
80
Actual
UV
Dose
0.00
20.79
31.09
41.34
61.63
81.83
0.00
30.85
41.05
61.40
81.78
UV
Dose2
0
432
967
1709
3798
6696
0
952
1685
3770
6688
Avg
(PFU/mL
)
3,770,000
173,000
74,300
19,900
3,070
577
3,470,000
70,000
20,100
2,980
500
Avg
Log(PFU/mL
)
6.58
5.24
4.87
4.30
3.49
2.76
6.54
4.85
4.30
3.47
2.70
Log I
-0.03
1.30
1.67
2.24
3.06
3.78
0.00
1.70
2.24
3.07
3.84
Log I2
0.001
1.702
2.794
5.034
9.335
14.298
0.000
2.881
5.014
9.414
14.773
Predicte
d Dose
-0.49
22.16
29.48
41.85
61.39
80.84
0.03
30.02
41.75
61.72
82.59
Avg:
SD:
n:
P:
t (95%):
Grubb's Test for O
P:
t (90%):
Grubb's Statistic
(GCRIT):
Residual
(mJ/cm2)
0.5
-1.4
1.6
-0.5
0.2
1.0
0.0
0.8
-0.7
-0.3
-0.8
0.04
0.89
11
0.05
2.262
utliers
0.10
3.751
2.355
G
0.5
1.6
1.8
0.6
0.2
1.1
0.1
0.9
0.8
0.4
1.0
Outlier?
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
OK
36
-------
Table 4-4. (continued)
Uncertainty of Dose-Response (UDR)
Log I
0.001
0.25
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
3.84
Dose
0.0
3.8
7.8
16.4
26.0
36.4
47.8
60.0
73.1
87.1
82.6
t
2.26
2.26
2.26
2.26
2.26
2.26
2.26
2.26
2.26
2.26
SD
0.89
0.89
0.89
0.89
0.89
0.89
0.89
0.89
0.89
0.89
UDR (%)
53.32
25.90
12.25
7.75
5.53
4.22
3.36
2.76
2.31
2.44
DL
(mJ/cm2/Log I)
14.68
15.12
15.56
16.45
17.33
18.22
19.11
19.99
20.88
21.77
21.49
t - student t test factor SD - standard deviation
Regression Statistics
Multiple R
R Square
Adjusted R Square
Standard Error
Observations
0.999851
0.999702
0.888558
0.940025
11
ANOVA
Regression
Residual
Total
df
2
9
11
SS
26689.01
7.952815
26696.97
MS
13344.51
0.883646
F
15101.64
Significance F
4.91681E-15
Intercept
X Variable 1
X Variable 2
Coefficients
0
14.67514
1.772698
Standard Error
0.480973
0.150326
tStat
30.511383
11.792378
P-value
2.14E-10
8.93E-07
Lower 95%
13.5871
1.432638
Upper 95%
15.76317
2.112759
Lower 95%
13.5871
1.432638
Upper 95%
15.76317461
2.112758854
37
-------
Log N as a function of UV Dose
y = 0.0001 x2-0.048x + 6.2389
R2 = 0.9871
z
o
0 10 20 30 40 50 60 70 80 90
UV Dose (mJ/cm2)
Figure 4-1. MS2 Collimated Beam Dose versus Log N Day 1 08-2-2012 UVT 91%.
38
-------
Log N as a function of UV Dose
O)
o
y = 0.0002x2-0.0615x +6.5425
R2 = 0.9988
10 20 30 40 50 60 70 80
UV Dose (mJ/cm2)
90
Figure 4-2. MS2 Collimated Beam Dose versus Log N Day 2 08-03-2012 97% UVT (outlier
removed).
39
-------
Dose-Response Curve
MS2
100
so -
60
IN
E
{j
-^
_§
o>
a 40
20
1.3984x2 + 20.482x
150
120
90
60
30
0123456
Log Inactivation
CB Data UVDGM QC Limits Udr(%) Dose-Response Curve
Figure 4-3. Dose response - log I versus dose -UVT 91% Day 1 08-02-2012.
40
-------
100
80 -
Dose-Response Curve
MS2
= 1.7727x2 + 14.675x
150
120
30
123456
Log Inactivation
CB Data UVDGM QC Limits Udr{%) Dose-Response Curve
Figure 4-4. Dose response - log I versus dose -UVT 97% Day 2 08-03-2012 with outlier
removed.
41
-------
4.5 MS and Operational Flow Test Data
The operational data for each test run (flow rate, UVT, and UV sensor intensity measurements)
are presented in Table 4-5. Flow rate, UVT and intensity were recorded when each sample was
collected, thus providing five data points for each test run. These values were then used to obtain
an average flow rate, UVT and intensity for each test run. The first influent and effluent samples
for MS2 were taken simultaneously beginning after approximately 2 to 3 minutes of steady state
operation. Subsequent influent and effluent samples were collected simultaneously after an
additional 2 to 3 minutes of operation, yielding five sets of samples over a 10 to 12 minute
period. The MS2 concentrations measured during the flow tests are presented in Table 4-6.
For each test condition replicate (i.e., each of the three influent and effluent samples), the log
inactivation (log I) was calculated using the following equation:
log I = log (No /N)
Where:
No = challenge microorganism concentration in influent sample (PFU/mL); and
N = Challenge microorganism concentration in corresponding effluent sample
(PFU/mL).
Table 4-7 shows the log concentration for the influent (N0) and effluent (N) samples. Table 4-8
presents the log I values for each sample.
The calculated REDmeas results in ml/cm2 are shown in Table 4-9. For each test condition the
REDmeas for each replicate was determined using the measured log inactivation (log I) and the
collimated beam test UV dose-response curve. The collimated beam regression line, based on the
replicate collimated beam tests for each day of testing, was used for this calculation. Figures 4-3
and 4-4 present the dose response curves and the equations that were used to calculate the
The replicate REDmeas values were averaged to produce one REDmeas for each test condition and
it's duplicate. All of the flow tests at 150, 250, and 435 gpm, with feed water at 91%, 94%, and
97% UVT or the equivalent reduced power respectively, achieved a minimum REDmeas of 40
ml/cm .
42
-------
Table 4-5. NeoTech D438™ Flow Test Operational Data
Test Condition
Lowered UVT Full Power
Lowered UVT-Full Power Dup
High UVT - Lowered Power
High UVT - Lowered Power Dup
Lowered UVT Full Power
Lowered UVT-Full Power Dup
High UVT - Lowered Power
High UVT - Lowered Power Dup
Lowered UVT Full Power
Lowered UVT-Full Power Dup
High UVT - Lowered Power
High UVT - Lowered Power Dup
Test
Day
1
1
2
2
1
1
2
2
1
1
2
2
Run
2
o
5
10
11
4
5
12
13
6
7
14
15
UVT
(%)
91
91
98
98
94
94
98
98
97
96
98
98
Flow
(gpm)
151
151
155
154
251
251
252
252
436
436
434
435
Intensity
(mW/cm2)
7.9
7.9
7.9
7.9
10.0
10.0
10.4
10.4
13.0
13.0
13.1
13.2
43
-------
Table 4-6. Flow Tests MS2 Concentrations NeoTech D438™
Test Condition
Lowered UVT 91%
Full Power
Lowered UVT 91%
Full Power Dup
High UVT - Lowered
Power
High UVT - Lowered
Power Dup
Lowered UVT 94%
Full Power
Lowered UVT 94%
Full Power Dup
High UVT -Lowered
Power
High UVT -Lowered
Power Dup
Lowered UVT 97%
Full Power
Lowered UVT 97%
Full Power Dup
High UVT -Lowered
Power
High UVT -Lowered
Power Dup
Run
2
o
J
10
11
4
5
12
13
6
7
14
15
Influent (PFU/mL)
Repl
3.53E+05
8.40E+05
2.36E+06
1.07E+06
1.13E+06
1.75E+06
9.57E+05
1.81E+06
3.47E+06
4.37E+06
3.18E+06
2.97E+06
Rep 2
4.53E+05
5.60E+05
1.18E+06
1.55E+06
8.73E+05
1.59E+06
9.60E+05
1.79E+06
1.56E+06
5.60E+06
2.05E+06
2.90E+06
Rep 3
4.27E+05
7.90E+05
1.33E+06
9.10E+05
NA
5.80E+05
2.26E+06
2.24E+06
1.98E+06
2.00E+06
1.94E+06
3.81E+06
Rep 4
4.27E+05
7.37E+05
1.15E+06
9.00E+05
8.97E+05
6.20E+05
1.59E+06
2.56E+06
2.00E+06
1.78E+06
1.57E+06
2.86E+06
Rep 5
3.17E+05
8.53E+05
8.57E+05
8.07E+05
8.17E+05
1.04E+06
2.15E+06
2.04E+06
2.23E+06
2.31E+06
2.49E+06
3.38E+06
Effluent (PFU/mL)
Repl
1.24E+03
4.37E+03
1.40E+03
6.47E+02
3.88E+03
4.20E+03
6.27E+03
3.37E+03
6.33E+03
2.28E+04
9.43E+03
1.08E+04
Rep 2
1.12E+03
3.69E+03
1.07E+03
6.23E+02
3.72E+03
3.92E+03
4.83E+03
4.10E+03
6.17E+03
1.18E+04
1.25E+04
1.50E+04
Rep 3
1.10E+03
4.22E+03
1.19E+03
8.13E+02
NA
4.00E+03
5.30E+03
4.79E+03
5.33E+03
1.42E+04
5.60E+03
7.30E+03
Rep 4
8.87E+02
3.08E+03
1.51E+03
8.67E+02
3.76E+03
4.33E+03
6.97E+03
3.13E+03
1.02E+04
1.63E+04
6.40E+03
1.17E+04
Rep 5
7.80E+02
3.11E+03
1.31E+03
1.57E+03
4.64E+03
4.43E+03
7.50E+03
5.37E+03
7.50E+03
1.53E+04
1.13E+04
7.40E+03
NA- samples not analyzed
44
-------
Table 4-7. MS2 Log Concentration for Influent and Effluent Samples NeoTech D438™
Test Condition
Lowered UVT 91% Full Power
Lowered UVT 91% Full Power Dup
High UVT - Lowered Power
High UVT - Lowered Power Dup
Lowered UVT 94% Full Power
Lowered UVT 94% Full Power Dup
High UVT - Lowered Power
High UVT - Lowered Power Dup
Lowered UVT 97% Full Power
Lowered UVT 97% Full Power Dup
High UVT - Lowered Power
High UVT - Lowered Power Dup
Run
2
o
J
10
11
4
5
12
13
6
7
14
15
Log Influent Concentration
Repl
5.55
5.92
6.37
6.03
6.05
6.24
5.98
6.26
6.54
6.64
6.50
6.47
Rep 2
5.66
5.75
6.07
6.19
5.94
6.20
5.98
6.25
6.19
6.75
6.31
6.46
Rep 3
5.63
5.90
6.12
5.96
NA
5.76
6.35
6.35
6.30
6.30
6.29
6.58
Rep 4
5.63
5.87
6.06
5.95
5.95
5.79
6.20
6.41
6.30
6.25
6.20
6.46
Rep 5
5.50
5.93
5.93
5.91
5.91
6.02
6.33
6.31
6.35
6.36
6.40
6.53
Log Effluent Concentration
Repl
3.09
3.64
3.15
2.81
3.59
3.62
3.80
3.53
3.80
4.36
3.97
4.03
Rep 2
3.05
3.57
3.03
2.79
3.57
3.59
3.68
3.61
3.79
4.07
4.10
4.18
Rep 3
3.04
3.63
3.08
2.91
NA
3.60
3.72
3.68
3.73
4.15
3.75
3.86
Rep 4
2.95
3.49
3.18
2.94
3.58
3.64
3.84
3.50
4.01
4.21
3.81
4.07
Rep 5
2.89
3.49
3.12
3.20
3.67
3.65
3.88
3.73
3.88
4.18
4.05
3.87
NA- samples not analyzed
45
-------
Table 4-8. MS2 Log Inactivation Results NeoTech D438™
Test Condition
Lowered UVT 91% Full Power
Lowered UVT 91% Full Power Dup
High UVT - Lowered Power
High UVT - Lowered Power Dup
Lowered UVT 94% Full Power
Lowered UVT 94% Full Power Dup
High UVT - Lowered Power
High UVT - Lowered Power Dup
Lowered UVT 97% Full Power
Lowered UVT 97% Full Power Dup
High UVT - Lowered Power
High UVT - Lowered Power Dup
Test
Run#
2
3
10
11
4
5
12
13
6
7
14
15
Test
Day
1
1
2
2
1
1
2
2
1
1
2
2
UVT
(%)
90.6
90.6
98.1
98.1
93.9
93.9
98.2
98.2
96.5
96.4
98.1
98.2
Flow
(gpm)
151.1
151.3
154.8
154.1
251.4
250.8
251.8
251.6
436.0
435.5
434.2
435.8
Intensity
(mW/cm2)
7.9
7.9
7.9
7.9
10.0
10.0
10.4
10.4
13.0
13.0
13.1
13.2
Log I
Repl
2.45
2.28
3.23
3.22
2.46
2.62
2.18
2.73
2.74
2.28
2.53
2.44
Rep 2
2.61
2.18
3.04
3.40
2.37
2.61
2.30
2.64
2.40
2.68
2.21
2.29
Rep 3
2.59
2.27
3.05
3.05
NA
2.16
2.63
2.67
2.57
2.15
2.54
2.72
Rep 4
2.68
2.38
2.88
3.02
2.38
2.16
2.36
2.91
2.29
2.04
2.39
2.39
Rep 5
2.61
2.44
2.82
2.71
2.25
2.37
2.46
2.58
2.47
2.18
2.34
2.66
NA- samples not analyzed
46
-------
Table 4-9. MS2 RED Results NeoTech D438™
Test Condition
Lowered UVT 91%
Lowered UVT 91% Dup
Lowered Power
Lowered Power Dup
Lowered UVT 94%
Lowered UVT 94% Dup
Lowered Power
Lowered Power Dup
Lowered UVT 97%
Lowered UVT 97% Dup
Lowered Power
Lowered Power Dup
Test
Run
2
3
10
11
4
5
12
13
6
7
14
15
Test
Day
1
1
2
2
1
1
2
2
1
1
2
2
UVT
(%)
90.6
90.6
98.1
98.1
93.9
93.9
98.2
98.2
96.5
96.4
98.1
98.2
Flow
(gpm)
151
151
155
154
251
251
252
252
436
436
434
436
Intensity
(mW/cm2)
7.9
7.9
7.9
7.9
10.0
10.0
10.4
10.4
13.0
13.0
13.1
13.2
RED (mJ/cm2)
Repl
58.69
54.07
65.81
65.59
58.96
63.26
40.50
53.28
66.59
54.04
48.43
46.35
Rep 2
62.90
51.33
61.06
70.28
56.41
62.93
43.09
51.10
57.29
64.83
41.20
42.82
Rep 3
62.40
53.76
61.21
61.22
NA
50.80
50.85
51.82
61.87
50.47
48.70
52.97
Rep 4
65.01
56.64
57.01
60.39
56.60
50.66
44.46
57.78
54.30
47.56
45.19
45.16
Rep 5
62.96
58.25
55.37
52.81
53.05
56.41
46.77
49.65
59.21
51.27
44.12
51.57
Average
62.39
54.81
60.09
62.06
56.26
56.81
45.13
52.73
59.85
53.63
45.53
47.77
SD(RED)
2.30
2.69
4.08
6.50
2.43
6.19
3.92
3.11
4.67
6.67
3.13
4.33
USP
10.21
13.63
18.85
29.08
13.76
30.23
24.11
16.40
21.66
34.54
19.12
25.14
NA- not analyzed
47
-------
4.6 Set Line for REDmeas of 40 mJ/cm
The three set point conditions selected for this validation all achieved a minimum REDmeas of 40
mJ/cm2, which was the target minimum REDmeas for developing the set line. Figure 4-5 shows
the set line. The unit is validated for a minimum REDmeas of 40 mJ/cm2 for any flow rate -
intensity combination above and to the left of the set line. The maximum flow rate demonstrated
was 434 gpm. A UV system cannot operate above the highest validated flow rate and claim a 40
mJ/cm2 REDmeas. The lowest intensity demonstrating a RED of 40 mJ/cm2 was 7.9 mW/cm2. A
UV system cannot operate below the lowest validated irradiance and claim a REDmeas of 40
mJ/cm .
Set Point 1-151 gpm; 7.9 mW/cm2
Set Point 2-251 gpm; 10.4 mW/cm2
Set Point 3 - 434 gpm; 13.2 mW/cm2
18
O
!/>
2
0
50 100 150 200 250 300
Flow Rate (gpm)
350
400
450
500
Figure 4-5. Set Line for 40 mJ/cm REDmeas - NeoTech D438™.
48
-------
4.7 Deriving the Validation Factor and Log Inactivation for Cryptosporidium
4.7.1 Validation Factor Definition
Several uncertainties and biases are involved in using experimental testing to define a validated
dose and validated operating conditions such as challenge microorganism UV sensitivity, and
sensor placement or variability. The validation factor (VF) for Cryptosporidium was determined
quantitatively to account for key areas of uncertainty and variability. The equation for the VF is
shown below.
W = BRED x[l+(UVai/100)]
where:
VF = Validation Factor;
BRED = RED bias factor;
Uvai = Uncertainty of validation expressed as a percentage.
The data used for the VF calculations and final results are presented in the following section.
4.7.2 RED Bias (BRED)
The RED bias factor (BRED) is a correction factor that accounts for the difference between the
UV sensitivity of a selected target pathogen and the UV sensitivity of the challenge
microorganism (MS2). If the challenge microorganism is more resistant (less sensitive) to UV
light than the target pathogen, the RED measured during the validation will be greater than the
than the RED that would be measured for the target pathogen. In this case the RED bias would
be greater than 1.0. If the challenge microorganism is less resistant (more sensitive) to UV light
than the target pathogen, then RED measured by the validation will be less than the RED that
would be measured for the target pathogen.
A target pathogen must be selected to calculate the RED bias factor. For this test, the target
pathogen Cryptosporidium was selected for use in presenting an example calculation of RED
bias as it is a common pathogen that is evaluated for drinking water applications.
Cryptosporidium was also selected because the EPA's LT2ESWTR requires UV reactors be
validated to demonstrate a log inactivation for Cryptosporidium. A target of 3-log inactivation of
Cryptosporidium was selected as water utilities in the highest risk category or "bin" may need
this maximum level of inactivation. The RED bias tables in Appendix G of the UVDGM-2006
were used for determining the RED bias. The RED bias is determined from the Tables based on
the sensitivity calculated for each test run replicate at a given set point (test condition) and the
UVT of the water. Sensitivity is calculated as:
9 _
Sensitivity (ml/cm per log I) = RED/ Log I
Per the GP-2011 and UVDGM-2006, the sensitivity is calculated for each test replicate (five per
test run, 20 samples total per set point). The highest BRED value found among the replicates at a
given set point is then selected for the BRED value for use in the VF calculation per the UVDGM-
2006 requirement. Table 4-10 shows the data for the replicates at each set point. The highest
49
-------
RED bias at each set point is used in the validation factor calculations shown later in Section
4.7.4.
Table 4-10. RED Bias Factor for Each Set Point for Cryptosporidium
Sample
Number
2-1
2-2
2-3
2-4
2-5
3-1
3-2
3-3
3-4
3-5
4-1
4-2
4-3
4-4
4-5
5-1
5-2
5-3
5-4
5-5
6-1
6-2
6-3
6-4
6-5
7-1
7-2
7-3
7-4
7-5
10-1
10-1
10-1
10-1
10-1
11-1
11-2
11-3
11-4
11-5
12-1
12-2
Test
Run
2
2
2
2
2
3
3
3
3
3
4
4
4
4
4
5
5
5
5
5
6
6
6
6
6
7
7
7
7
7
10
10
10
10
10
11
11
11
11
11
12
12
UVT
%
90.6
90.6
90.6
90.6
90.6
90.6
90.6
90.6
90.6
90.6
93.9
93.9
93.9
93.9
93.9
93.9
93.9
93.9
93.9
93.9
96.5
96.5
96.5
96.5
96.5
96.4
96.4
96.4
96.4
96.4
98.1
98.1
98.1
98.1
98.1
98.1
98.1
98.1
98.1
98.1
98.2
98.2
Sensitivity mJ/cm2 per Log I
RED
58.69
62.90
62.40
65.01
62.96
54.07
51.33
53.76
56.64
58.25
58.96
56.41
NA
56.60
53.05
63.26
62.93
50.80
50.66
56.41
66.59
57.29
61.87
54.30
59.21
54.04
64.83
50.47
47.56
51.27
65.81
61.06
61.21
57.01
55.37
65.59
70.28
61.22
60.39
52.81
40.50
43.09
Log I
2.45
2.61
2.59
2.68
2.61
2.28
2.18
2.27
2.38
2.44
2.46
2.37
NA
2.38
2.25
2.62
2.61
2.16
2.16
2.37
2.74
2.40
2.57
2.29
2.47
2.28
2.68
2.15
2.04
2.18
3.23
3.04
3.05
2.88
2.82
3.22
3.40
3.05
3.02
2.71
2.18
2.30
Sensitivity
23.91
24.13
24.10
24.23
24.13
23.68
23.53
23.66
23.81
23.89
23.93
23.80
NA
23.81
23.62
24.15
24.13
23.50
23.50
23.80
24.31
23.84
24.08
23.69
23.94
23.67
24.22
23.49
23.33
23.53
20.40
20.07
20.08
19.78
19.67
20.38
20.69
20.08
20.02
19.48
18.55
18.75
BRED 4-log
Crypto
1.66
1.70
1.70
1.70
1.70
1.66
1.66
1.66
1.66
1.66
1.66
1.66
NA
1.66
1.66
1.70
1.70
1.66
1.66
1.66
1.40
1.38
1.40
1.38
1.38
1.38
1.40
1.38
1.38
1.38
1.18
1.18
1.18
1.17
1.17
1.18
1.18
1.18
1.18
1.17
1.17
1.17
BRED 3.5-log
Crypto
1.79
1.83
1.83
1.83
1.83
1.79
1.79
1.79
1.79
1.79
1.79
1.79
NA
1.79
1.79
1.83
1.83
1.79
1.79
1.79
1.43
1.42
1.43
1.42
1.42
1.42
1.43
1.42
1.42
1.42
1.20
1.20
1.20
1.19
1.19
1.20
1.20
1.20
1.20
1.19
1.19
1.19
BRED 3.0 log
Crypto
1.82
1.85
1.85
1.85
1.85
1.82
1.82
1.82
1.82
1.82
1.82
1.82
NA
1.82
1.82
1.85
1.85
1.82
1.82
1.82
1.42
1.41
1.42
1.41
1.41
1.41
1.42
1.41
1.41
1.41
1.20
1.20
1.20
1.19
1.19
1.20
1.20
1.20
1.20
1.19
1.19
1.19
50
-------
Sample
Number
12-3
12-4
12-5
13-1
13-2
13-3
13-4
13-5
14-1
14-2
14-3
14-4
14-5
15-1
15-2
15-3
15-4
15-5
Test
Run
12
12
12
13
13
13
13
13
14
14
14
14
14
15
15
15
15
15
Maximum BRED
UVT
%
98.2
98.2
98.2
98.2
98.2
98.2
98.2
98.2
98.1
98.1
98.1
98.1
98.1
98.2
98.2
98.2
98.2
98.2
Sensitivity mJ/cm2 per Log I
RED
50.85
44.46
46.77
53.28
51.10
51.82
57.78
49.65
48.43
41.20
48.70
45.19
44.12
46.35
42.82
52.97
45.16
51.57
Log I
2.63
2.36
2.46
2.73
2.64
2.67
2.91
2.58
2.53
2.21
2.54
2.39
2.34
2.44
2.29
2.72
2.39
2.66
Sensitivity
19.34
18.86
19.03
19.51
19.36
19.41
19.84
19.25
19.16
18.60
19.18
18.91
18.83
19.00
18.73
19.49
18.91
19.39
Set Point 151 gpm - 7.9 mW/cm2
Set Point 251 gpm - 10.4mW/cm2
Set Point 434 gpm - 13.2 mW/cm2
BRED 4-log
Crypto
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.70
1.70
1.40
BRED 3.5-log
Crypto
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.83
1.83
1.43
BRED 3.0 log
Crypto
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.85
1.85
1.42
NA - sample not analyzed so RED and bias not determined
Uncertainty of Validation
The uncertainty of validation (Uvai) addresses many sources of experimental uncertainty. As the
critical source of uncertainty, such as the sensor readings, or the fit of the dose-response curve, is
unknown in advance of the validation testing, the USEPA developed a decision tree to assist in
establishing Uvai. Figure 5.4 of the UVDGM-2006, which is specific to a UV intensity set point
approach, was used to determine Uvai in calculating the validated dose. Per Figure 5.4 in EPA's
UVDGM-2006, any of the following equations may be used to establish the Uvai:
Uvai = (USP2 +UDR2)1/2
TT
UDR )
Where:
Us = Uncertainty of sensor value, expressed as a fraction;
UDR = Uncertainty of the fit of the dose-response curve;
USP = Uncertainty of set-point; and
Uvai = Uncertainty of the validation.
The QC objective for the duty sensor is that the measurements with the duty sensor should be
<10% of the average of two or more reference sensors. If this objective is met, then it eliminates
51
-------
the need to calculate the Us factor per the GP-2011 and UVDGM-2006, Section 5.4.4. The
sensor met the 10% requirement, as shown in Table 4-1, therefore, Us is not used in determining
the uncertainty of validation.
The UVDGM-2006 also shows the formula and calculations for UDR in Appendix C Section C4.
The equation is:
UDR = t * [SD/ UV DoseCB] * 100%
Where:
UDR= Uncertainty of the UV dose-response fit at a 95% confidence level;
UV DosecB = UV dose calculated from the UV dose-response curve for the
challenge microorganism;
SD = Standard deviation of the difference between the calculated UV dose
response and the measured value; and
t = t-statistic at a 95% confidence level for a sample size equal to the number of
test condition replicates used to define the dose-response.
The UDR results are shown in Tables 4-2 and 4-4 for the low and high UVT waters (32.58% and
12.25%, respectively). Since the UDR was > 30% at the UV dose corresponding to 1-log
inactivation of the challenge organism, the uncertainty of the dose response (UDR) is included in
the calculation of uncertainty.
The uncertainty in the set point value (Usp) is based on a prediction interval at a 95% confidence
level using the following procedure:
1. Calculate the average and standard deviation of REDmeas values for each test condition
2. Calculate the uncertainty of the set point REDmeas using:
USP = [(t x SDRED) / (REDmeas)] x 100%
Where:
REDmeaS = Average REDmeas value measured for each test condition;
SDRED = Standard deviation of the REDmeas values measured for each test
condition; and
t = t-statistic for a 95% confidence level defined as a function of the number of
replicate samples, in this case 5 replicates were used for testing yielding a t value
of 2.776 (n-1 = 4), except run 4 which had four valid replicates so the t value is
3.182).
3. Select the highest USP from the replicates at each set point for calculating the VF.
52
-------
The USP results based on the REDmeas and standard deviation are shown in Table 4-9. The highest
USP for each set point is 29.08% (151 gpm set point), 30.23% (251 gpm set point), and 34.54%
(434 gpm set point).
The uncertainty of the validation is then calculated using the highest UDR (32.58%) from the
2012 dose response data and the highest USP for each test condition using the equation:
Uval = (USP2 +UDR2)1/2
Table 4-11 shows the Uyai values used for determining the uncertainty of the validation at each
set point.
Table 4-11. Uncertainty of the Validation (Uyai) and BRED Values for Cryptosporidium
Set Point
151 gpm- 7.9mW/cm2
251 gpm- 10.4mW/cm2
434gpm-13.2mW/cm2
Max
UDR
(%)
32.58
32.58
32.58
Max
USP
(%)
29.08
30.23
34.54
UVa,
(%)
43.67
44.44
47.48
Max
BRED
4.0-log
1.70
1.70
1.40
3.5-log
1.83
1.83
1.43
3.0-log
1.85
1.85
1.42
4.7.4 Validated Dose and Set Line for Cryptosporidium
After establishing the Uyai and the RED bias as described above, the VF is calculated using the
equation:
W = BRED X[l+(UVai/100)]
Where:
VF = Validation Factor;
BRED = RED bias factor for Cryptosporidium: and
Uvai = Uncertainty of validation expressed as a percentage.
The validated dose is then calculated as follows:
Validated dose (REDVai) = REDmeas / VF
Table 4-12 shows the calculated VF for various Cryptosporidium log inactivation levels (3.0,
3.5, and 4.0 log inactivation).
Table 4-12 shows the REDyai for Cryptosporidium for each test run using the validation factors
for the various Cryptosporidium log inactivation levels. Table 4-12 also shows the Validated
Dose for each set point and a comparison to the dose required for various levels of inactivation
of Cryptosporidium. As can be seen, the tests for the first and third set points (151 and 434 gpm)
show a validated dose for Cryptosporidium that would achieve a minimum of 4.0 log
inactivation. The second set point (middle point) at 251 gpm showed a minimum of 3.5 log
inactivation.
53
-------
The NeoTech D438™ achieved a minimum of 3.5 log inactivation for the low power runs at 251
gpm at 10.4 mW/cm2 and a 4.0-log inactivation for all of the test runs at the set points at 151
gpm - 7.9 mW/cm2 and 434 gpm - 13.2 mW/cm2. Figure 4-6 shows the operating conditions
based on three set points that achieved the 3-log inactivation oi Cryptosporidium.
The NeoTech D438™ achieved a minimum 3.0-log inactivation for Cryptosporidium over the
range of flow (151 to 434 gpm) and intensity (7.9 to 13.2 mW/cm2) as shown in the set line in
Figure 4-14.
The set points in Figure 4-6 are:
Set Point 1-151 gpm; 7.9 mW/cm2
Set Point 2-251 gpm; 10.4 mW/cm2
Set Point 3 - 434 gpm; 13.2 mW/cm2
0 50 100 150 200 250 300 350 400 450 500
Flow Rate (gpm)
Figure 4-6. Set Line for Minimum 3-log Cryptosporidium Inactivation for NeoTech D438™.
54
-------
Table 4-12. Validation Factors and Validated Dose (REDyai) for Cryptosporidium
Condition
Lowered UVT
Lowered UVT Dup
Lowered Power
Lowered Power Dup
Lowered UVT
Lowered UVT Dup
Lowered Power
Lowered Power Dup
Lowered UVT
Lowered UVT Dup
Lowered Power
Lowered Power Dup
Run
#
2
3
10
11
4
5
12
13
6
7
14
15
Flow
Rate
(gpm)
151.1
151.3
154.8
154.1
251.4
250.8
251.8
251.6
436.0
435.5
434.2
435.8
Intensity
(mW/cm2)
7.9
7.9
7.9
7.9
10.0
10.0
10.4
10.4
13.0
13.0
13.1
13.2
Validation Factor
4.0-log
2.44
2.44
2.44
2.44
2.46
2.46
2.46
2.46
2.06
2.06
2.06
2.06
3.5-log
2.63
2.63
2.63
2.63
2.64
2.64
2.64
2.64
2.11
2.11
2.11
2.11
3.0-log
2.66
2.66
2.66
2.66
2.67
2.67
2.67
2.67
2.09
2.09
2.09
2.09
REDmeas
(mJ/cm2)
62.39
54.81
60.09
62.06
56.26
56.81
45.13
52.73
59.85
53.63
45.53
47.77
REDval
4-log
(mJ/cm2)
22(1)
25.5
22.4
24.6
25.4
22.9
23.1
18.4
21.5
29.0
26.0
22.1
23.1
3.5-log
(mJ/cm2)
15(1)
23.7
20.8
22.9
23.6
21.3
21.5
17.1
19.9
28.4
25.4
21.6
22.7
3.0-log
(mJ/cm2)
12(1)
23.5
20.6
22.6
23.3
21.1
21.3
16.9
19.7
28.6
25.6
21.7
22.8
(1) Required dose for log inactivation validation per the UVDGM-2006 Appendix G.
55
-------
4.8 Deriving the Validation Factor and Log Inactivation for Giardia
4.8.1 Validation Factor Definition
As described earlier in Section 4.7.1 on Cryptosporidium, several uncertainties and biases are
involved in using experimental testing to define a validated dose and validated operating
conditions such as challenge microorganism UV sensitivity, and sensor placement or variability.
The VF for Giardia was determined quantitatively to account for key areas of uncertainty and
variability. The equation for the VF is shown below.
W = BRED x[l+(UVai/100)]
where:
VF = Validation Factor;
BRED = RED bias factor;
Uvai = Uncertainty of validation expressed as a percentage.
The data used for the VF calculations and final results are presented below.
4.8.2 RED Bias (BRED)
The RED bias factor (BRED) is a correction factor that accounts for the difference between the
UV sensitivity of a selected target pathogen and the UV sensitivity of the challenge
microorganism (MS2). As described in Section 4.7.2, a target pathogen must be selected to
calculate the RED bias factor. In addition to the target pathogen Cryptosporidium described
previously, Giardia was also selected for evaluation. The RED bias tables for Giardia in
Appendix G of the UVDGM-2006 were used for determining the RED bias. The RED bias is
determined from the Tables based on the sensitivity calculated for each test run replicate at a
given set point (test condition) and the UVT of the water. Sensitivity is calculated as:
9 _
Sensitivity (ml/cm per log I) = RED/ Log I
Per the GP-2011 and UVDGM-2006, the sensitivity is calculated for each test replicate (five per
test run, 20 samples total per set point). The highest BRED value found among the replicates at a
given set point is then selected for the BRED value for use in the VF calculation per the UVDGM-
2006 requirement. Table 4-13 shows the data for the replicates at each set point. The highest
RED bias at each set point is used in the validation factor calculations shown later in Section
4.8.4.
56
-------
Table 4-13. RED Bias Factor for Each Set Point for Giardia
Sample
Number
2-1
2-2
2-3
2-4
2-5
3-1
3-2
o o
J-J
3-4
3-5
4-1
4-2
4-3
4-4
4-5
5-1
5-2
5-3
5-4
5-5
6-1
6-2
6-3
6-4
6-5
7-1
7-2
7-3
7-4
7-5
10-1
10-1
10-1
10-1
10-1
11-1
11-2
11-3
11-4
11-5
12-1
12-2
12-3
12-4
12-5
Test
Run
2
2
2
2
2
3
3
3
3
3
4
4
4
4
4
5
5
5
5
5
6
6
6
6
6
7
7
7
7
7
10
10
10
10
10
11
11
11
11
11
12
12
12
12
12
UVT
(%)
90.6
90.6
90.6
90.6
90.6
90.6
90.6
90.6
90.6
90.6
93.9
93.9
93.9
93.9
93.9
93.9
93.9
93.9
93.9
93.9
96.5
96.5
96.5
96.5
96.5
96.4
96.4
96.4
96.4
96.4
98.
98.
98.
98.
98.
98.
98.
98.
98.
98.
98.2
98.2
98.2
98.2
98.2
Sensitivity mJ/cm2
RED
58.69
62.90
62.40
65.01
62.96
54.07
51.33
53.76
56.64
58.25
58.96
56.41
NA
56.60
53.05
63.26
62.93
50.80
50.66
56.41
66.59
57.29
61.87
54.30
59.21
54.04
64.83
50.47
47.56
51.27
65.81
61.06
61.21
57.01
55.37
65.59
70.28
61.22
60.39
52.81
40.50
43.09
50.85
44.46
46.77
Log I
2.45
2.61
2.59
2.68
2.61
2.28
2.18
2.27
2.38
2.44
2.46
2.37
NA
2.38
2.25
2.62
2.61
2.16
2.16
2.37
2.74
2.40
2.57
2.29
2.47
2.28
2.68
2.15
2.04
2.18
3.23
3.04
3.05
2.88
2.82
3.22
3.40
3.05
3.02
2.71
2.18
2.30
2.63
2.36
2.46
per Log I
Sensitivity
23.91
24.13
24.10
24.23
24.13
23.68
23.53
23.66
23.81
23.89
23.93
23.80
NA
23.81
23.62
24.15
24.13
23.50
23.50
23.80
24.31
23.84
24.08
23.69
23.94
23.67
24.22
23.49
23.33
23.53
20.40
20.07
20.08
19.78
19.67
20.38
20.69
20.08
20.02
19.48
18.55
18.75
19.34
18.86
19.03
BRED
4.0-log
Giardia
1.61
1.66
1.66
1.66
1.66
1.61
1.61
1.61
1.61
1.61
1.61
1.61
NA
1.61
1.61
1.66
1.66
1.66
1.66
1.66
1.38
1.36
1.38
1.36
1.36
1.36
1.38
1.36
1.36
1.36
1.17
1.17
1.17
1.16
1.16
1.17
1.17
1.17
1.17
1.41
1.16
1.16
1.16
1.16
1.16
BRED
3.5-log
Giardia
1.79
1.83
1.83
1.83
1.83
1.79
1.79
1.79
1.79
1.79
1.79
1.79
NA
1.79
1.79
1.83
1.83
1.83
1.83
1.83
1.43
1.42
1.43
1.42
1.42
1.42
1.43
1.42
1.42
1.42
1.2
1.2
1.2
1.19
1.19
1.2
1.2
1.2
1.2
1.19
1.19
1.19
1.19
1.19
1.19
BRED
3.0-log
Giardia
.85
.89
.89
.89
.89
.85
.85
.85
.85
.85
.85
.85
NA
.85
.85
.89
.89
.89
.89
.89
.44
.42
.44
.42
.42
.42
.44
.42
.42
.42
.21
.21
.21
1.2
1.2
1.21
1.21
1.21
1.21
1.2
1.20
1.20
1.20
1.20
1.20
57
-------
Sample
Number
13-1
13-2
13-3
13-4
13-5
14-1
14-2
14-3
14-4
14-5
15-1
15-2
15-3
15-4
15-5
Test
Run
13
13
13
13
13
14
14
14
14
14
15
15
15
15
15
Maximum BRED
UVT
(%)
98.2
98.2
98.2
98.2
98.2
98.
98.
98.
98.
98.
98.2
98.2
98.2
98.2
98.2
Sensitivity mJ/cm2
RED
53.28
51.10
51.82
57.78
49.65
48.43
41.20
48.70
45.19
44.12
46.35
42.82
52.97
45.16
51.57
Log I
2.73
2.64
2.67
2.91
2.58
2.53
2.21
2.54
2.39
2.34
2.44
2.29
2.72
2.39
2.66
per Log I
Sensitivity
19.51
19.36
19.41
19.84
19.25
19.16
18.60
19.18
18.91
18.83
19.00
18.73
19.49
18.91
19.39
Set Point 151 gpm - 7.9 mW/cm2
Set Point 251 gpm - 10.4 mW/cm2
Set Point 434 gpm - 13.2 mW/cm2
BRED
4.0-log
Giardia
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.66
1.66
1.38
BRED
3.5-log
Giardia
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.83
1.83
1.43
BRED
3.0-log
Giardia
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.89
.89
.44
NA - sample not analyzed so RED and bias not determined
4.8.3 Uncertainty of Validation
As described in Section 4.7.3, the uncertainty of validation (Uvai) addresses many sources of
experimental uncertainty. The same approach to uncertainty calculations described in Section
4.7.3 apply to calculating the VF for Giardia and in fact the uncertainty values are the same.
Please refer to Section 4.7.3 for the equations and discussion of the various uncertainty factors
that are used for determining Uyai at each set point.
The USP results based on the REDmeas and standard deviation are shown in Table 4-9. The highest
USP for each set point is 29.08% (151 gpm set point), 30.23% (251 gpm set point), and 34.54%
(434 gpm set point).
The uncertainty of the validation is then calculated using the highest UDR (32.58%) from the dose
response data and the highest USP for each test condition using the equation:
Uval = (USP2 +UDR2)1/2
Table 4-14 shows the Uyai values used for determining the uncertainty of the validation at each
set point.
58
-------
Table 4-14. Uncertainty of the Validation (Uyai) and BRED Values for Giardia
Set Point
151 gpm- 7.9mW/cm2
251gpm- 10.4mW/cm2
434gpm-13.2mW/cm2
Max
UDR
0%
32.58
32.58
32.58
Max
USP
(%)
29.08
30.23
34.54
UVa,
%
43.67
44.44
47.48
Max
BRED
4.0-log
1.66
1.66
1.38
3.5-log
1.83
1.83
1.43
3.0-log
1.89
1.89
1.44
4.8.4 Validated Dose for Giardia
After establishing the Uvai and the RED bias as described above, the VF is calculated using the
equation:
W = BRED x[l+(UVai/100)]
Where:
VF = Validation Factor;
BRED = RED bias factor for Giardia (see Table 4-13); and
Uvai = Uncertainty of validation expressed as a percentage.
The validated dose is then calculated as follows:
Validated dose (REDVai) = REDmeas / VF
Table 4-15 shows the calculated VF for various Giardia log inactivation levels (3.0, 3.5, and 4.0
log inactivation).
Table 4-15 shows the REDyai for Giardia for each test run using the validation factors for the
various Giardia log inactivation levels. Table 4-15 also shows the Validated Dose for each set
point and a comparison to the dose required for various levels of inactivation of Giardia. As can
be seen, the tests for the first and third set points (151 and 434 gpm) show a validated dose for
Giardia that would achieve a minimum of 4.0 log inactivation. The second set point (middle
point) at 251 gpm showed a minimum of 3.5 log inactivation.
The NeoTech D438™ achieved a minimum of 3.5 log inactivation for the low power runs at 251
gpm at 10.4 mW/cm and a 4.0 log inactivation for all of the test runs at the set points at 151
gpm - 7.9 mW/cm2 and 434 gpm - 13.2 mW/cm2.
59
-------
Table 4-15 Validation Factors and Validated Dose (REDyai) for Giardia
Condition
Lowered UVT
Lowered UVT Dup
Lowered Power
Lowered Power Dup
Lowered UVT
Lowered UVT Dup
Lowered Power
Lowered Power Dup
Lowered UVT
Lowered UVT Dup
Lowered Power
Lowered Power Dup
Run
#
2
3
10
11
4
5
12
13
6
7
14
15
Flow
Rate
(gpm)
151.1
151.3
154.8
154.1
251.4
250.8
251.8
251.6
436.0
435.5
434.2
435.8
Intensity
(mW/cm2)
7.9
7.9
7.9
7.9
10.0
10.0
10.4
10.4
13.0
13.0
13.1
13.2
Validation Factor
4.0-log
2.38
2.38
2.38
2.38
2.40
2.40
2.40
2.40
1.99
1.99
1.99
1.99
3.5-log
2.63
2.63
2.63
2.63
2.64
2.64
2.64
2.64
2.07
2.07
2.07
2.07
3.0-log
2.72
2.72
2.72
2.72
2.73
2.73
2.73
2.73
2.08
2.08
2.08
2.08
REDmeas
(mJ/cm2)
62.39
54.81
60.09
62.06
56.26
56.81
45.13
52.73
59.85
53.63
45.53
47.77
REDval
4.0-log
(mJ/cm2)
22(1)
26.2
23.0
25.2
26.0
23.5
23.7
18.8
22.0
30.0
26.9
22.8
24.0
3.5-log
(mJ/cm2)
15(1)
23.7
20.8
22.9
23.6
21.3
21.5
17.1
19.9
29.0
26.0
22.0
23.1
3.0-log
(mJ/cm2)
12(1)
23.0
20.2
22.1
22.9
20.6
20.8
16.5
19.3
28.8
25.8
21.9
23.0
(1) Required dose for log inactivation validation per the UVDGM-2006 Appendix G.
60
-------
4.9 Validated Dose (REDVai) for MS2 as the Target Organism
Some regulatory agencies, such as the NYDOH, have established a standard for spray parks and
other applications based on a validated dose (REDVai) of 40 ml/cm2 based on MS2. The
calculation of the validation factor for a validated dose based on MS2 is performed using BRED
set equal to 1.0. For MS2 validated dose calculations, BRED is set equal to 1.0 because the
pathogen selected, namely MS2, is the same as the test organism, so there is no bias correction.
Therefore, the validation factor will not vary by the log inactivation level.
The Uvai is calculated in the same manner as described in Section 4.7.3.
The VF for evaluating validated dose (REDVai) based on MS2 is calculated using the same
formula as for other pathogens as follows:
W = BRED x[l+(UVai/100)]
Where:
VF = Validation Factor;
BRED = RED bias factor (set equal 1.0); and
Uvai = Uncertainty of validation expressed as a percentage.
The validated dose is then calculated as follows:
Validated dose (REDVai) = REDobserved / VF
Table 4-16 shows the REDyai based on MS2 for each test run.
Using the VF calculated for each set point, the REDVai based on MS2 was calculated for each
test run. Only some of the test runs achieved a 40 mJ/cm2 validated dose based on MS2. The
lowest flow rate condition (151 gpm) showed that three of the four test runs achieved a validated
dose of 40 mJ/cm . One run was at 38 mJ/cm , just below the REDyai target of 40 mJ/cm based
on MS2. A main influence on the validated dose for the lower flow tests is the UDR being just
above the 30% level at 32.58%. If the UDR had been less than or equal to 30%, the validation
factor would be significantly lower. In that case, all of the 151 gpm set point test runs would be >
40 mJ/cm and three of the four 251 gm set point test runs would be above 40 mJ/cm .
61
-------
Table 4-16. Validation Factors and Validated Dose (REDVai) based on MS2
Condition
Lowered UVT
Lowered UVT Dup
Lowered Power
Lowered Power Dup
Lowered UVT
Lowered UVT Dup
Lowered Power
Lowered Power Dup
Lowered UVT
Lowered UVT Dup
Lowered Power
Lowered Power Dup
Run
2
o
3
10
11
4
5
12
13
6
7
14
15
Flow
Rate
(gpm)
151
151
155
154
251
251
252
252
436
436
434
436
Intensity
(mW/cm2)
7.9
7.9
7.9
7.9
10.0
10.0
10.4
10.4
13.0
13.0
13.1
13.2
Validation
Factor
(1)
1.44
1.44
1.44
1.44
1.44
1.44
1.44
1.44
1.47
1.47
1.47
1.47
REDmeas
(mJ/cm2)
62.39
54.81
60.09
62.06
56.26
56.81
45.13
52.73
59.85
53.63
45.53
47.77
REDva,
(mJ/cm2)
43.4
38.2
41.8
43.2
38.9
39.3
31.2
36.5
40.6
36.4
30.9
32.4
(1) BRED equal to 1.0, therefore all log inactivation levels have the same validation factor.
62
-------
4.10 Water Quality Data
Samples were collected for general water quality characterization. Influent and effluent samples
were collected during each flow test run and analyzed for temperature, pH, total chlorine, and
free chlorine. An influent sample was collected from each flow test run and analyzed for
turbidity, iron, and manganese.
An influent and effluent sample from each test run was also collected and analyzed for total
coliform, E. coli, and HPC.
The general chemistry and microbiological results are presented in Tables 4-17 through 4-20.
63
-------
Table 4-17. Temperature and pH Results
Test
Blank
Low UVT (91%) Full power
Low UVT (91%) Full power Dup
Low UVT (94%) Full Power
Low UVT (94%) Full Power Dup
Low UVT (97%) Full power
Low UVT (97%) Full power Dup
Blank
Reactor Control
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Run#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Temperature
(°F)
Influent
72.3
72.4
72.5
72.5
72.3
73.3
72.6
72.0
71.8
71.4
71.3
72.1
72.0
72.2
72.4
Effluent
72.2
72.3
72.3
72.5
72.4
73.2
72.6
72.0
71.8
71.4
71.4
72.1
72.0
72.0
72.4
pH
(S.U.)
Influent
7.39
7.42
7.41
7.52
7.57
8.28
8.43
8.37
8.36
8.36
8.35
8.48
8.51
8.60
8.64
Effluent
7.41
7.38
7.39
7.52
7.55
8.28
7.47
8.35
8.37
8.37
8.37
8.48
8.53
8.58
7.64
Table 4-18. Total Chlorine, Free Chlorine and Turbidity Results
Test
Blank
Low UVT (91%) Full power
Low UVT (91%) Full power Dup
Low UVT (94%) Full Power
Low UVT (94%) Full Power Dup
Low UVT (97%) Full power
Low UVT (97%) Full power Dup
Blank
Reactor Control
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Run#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Total
Chlorine
(mg/L)
Influent
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.02
0.02
0.02
0.02
Free
Chlorine
(mg/L)
Influent
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
Turbidity
(NTU)
Influent
0.45
0.41
0.44
0.36
0.34
0.23
0.23
0.22
0.28
0.18
0.19
0.17
0.21
0.14
0.20
64
-------
Table 4-19. Iron and Manganese Results
Test
Blank
Low UVT (91%) Full power
Low UVT (91%) Full power Dup
Low UVT (94%) Full Power
Low UVT (94%) Full Power Dup
Low UVT (97%) Full power
Low UVT (97%) Full power Dup
Blank
Reactor Control
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Run#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Iron
(mg/L)
Influent
0.03
0.05
0.05
O.02
0.02
O.02
0.72
O.02
O.02
O.02
O.02
O.02
O.02
O.02
<0.02
Manganese
(mg/L)
Influent
O.001
O.001
O.001
O.001
O.001
O.001
0.004
O.001
O.001
O.001
O.001
O.001
0.001
O.001
O.001
UVT(1)
(%)
Influent
91
91
91
93
93
95
95
96
96
96
96
96
96
96
96
Effluent
91
90
90
93
92
95
95
96
96
96
96
96
96
96
96
(l)UVT on grab samples, measured in laboratory after tests; In- line UVT meter used for flow test results
Table 4-20. HPC, Total Coliform and K coli Results
Test
Blank
Low UVT (91%) Full power
Low UVT (91%) Full power Dup
Low UVT (94%) Full Power
Low UVT (94%) Full Power Dup
Low UVT (97%) Full power
Low UVT (97%) Full power Dup
Blank
Reactor Control
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Run#
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
Total Coliform
(MPN/lOOmL)
Influent
2
8.66E+02
5.94E+01
3.17E+01
6.30E+00
1.60E+01
4.10E+01
<1
1.05E+03
1.57E+02
7.76E+01
1.34E+01
5.20E+00
4.10E+00
1.46E+01
Effluent
<1
<1
<1
<1
<1
<1
<1
<1
1.05E+03
7.76E+01
<1
<1
<1
<1
<1
E. coli
(MPN/lOOmL)
Influent
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
Effluent
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
HPC
(CFU/mL)
Influent
4.76E+03
7.17E+03
3.99E+03
2.80E+03
3.11E+03
1.34E+03
4.65E+02
1.61E+03
3.62E+03
1.75E+03
2.82E+03
9.45E+02
3.70E+02
4.65E+02
7.20E+02
Effluent
3.95E+01
2.35E+00
2.10E+01
1.33E+02
2.60E+01
1.28E+02
4.15E+01
1.15E+02
4.20E+03
4.80E+01
1.85E+01
6.10E+01
1.50E+01
4.70E+01
4.65E+01
65
-------
4.11 Power Measurement
A power monitoring platform was connected to the unit. This monitoring platform provided
continuous readout of the volts, amperage, and watts being used by the unit under each test
condition. Volts, amperes, and wattage were recorded during each flow test. The power factor
was also recorded. Table 4-21 presents the power measurements taken during the flow tests.
Table 4-21. Power Measurement Results
Test
Blank
Low UVT (91%) Full power
Low UVT (91%) Full power Dup
Low UVT (94%) Full Power
Low UVT (94%) Full Power Dup
Low UVT (97%) Full power
Low UVT (97%) Full power Dup
Blank
Reactor Control
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Reduced Power - High UVT
Reduced Power - High UVT Dup
Run#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Volts
(volts)
117
117
116
117
117
117
117
118
119
117
117
116
NR
116
116
Amperage
(amps)
3.0
3.0
3.0
3.0
3.0
3.0
3.0
2.0
0.0
2.0
2.0
3.0
NR
3.5
3.5
Watts
(watts)
340
340
340
340
340
340
340
230
0.0
230
230
340
NR
390
390
Power Factor
0.97
0.96
0.96
0.97
0.96
0.96
0.97
0.99
—
0.98
0.98
0.97
NR
0.96
0.96
NR-not reported because the data were not recorded on log sheet
4.12 Headloss
Headloss was measured over the flow range of 100 to 450 gpm. Pressure at the inlet and outlet of
the reactor was measured at several flow rates as shown in Table 4-22.
Table 4-22. Headloss Measurement Results
Flow Rate
100
150
250
350
450
Inlet (psi)
1.59
1.92
2.87
4.22
6.16
Outlet (psi)
1.20
1.29
1.52
1.88
2.35
Headloss (psi)
0.39
0.63
1.35
2.34
3.81
66
-------
Chapter 5
Quality Assurance/Quality Control
5.1 Introduction
An important aspect of verification testing is the QA/QC procedures and requirements. Careful
adherence to the procedures ensures that the data presented in this report is of sound quality,
defensible, and representative of the equipment performance. The primary areas of evaluation
were representativeness, accuracy, precision, and completeness.
Because this ETV was conducted at the NSF testing lab, all laboratory activities were conducted
in accordance with the provisions of the NSF International Laboratories Quality Assurance
Manual.
5.2 Test Procedure QA/QC
NSF testing laboratory staff conducted the tests by following a USEPA-approved TQAP created
specifically for this verification. NSF QA Department staff performed an audit during testing to
ensure the proper procedures were followed. The audit yielded no significant findings.
5.3 Sample Handling
All samples analyzed by the NSF Chemistry and Microbiology Laboratories were labeled with
unique identification numbers. All samples were analyzed within allowable holding times.
5.4 Chemistry Laboratory QA/QC
The calibrations of all analytical instruments and the analyses of all parameters complied with
the QA/QC provisions of the NSF International Laboratories Quality Assurance Manual.
The NSF QA/QC requirements are all compliant with those given in the USEPA method or
Standard Method for the parameter. Also, every analytical method has an NSF standard
operating procedure.
The bench top UV spectrophotometer was calibrated with Holmium Oxide with each batch of
samples analyzed. The peaks at 241.1, 287.6 and 293.5 nm all passed QC criteria for peak
intensity and peak nm location. Bichromate standards were also run with each batch of samples
and found to be within 102% of the true value.
5.5 Microbiology Laboratory QA/QC
5.5.1 Growth Media Positive Controls
All media were checked for sterility and positive growth response when prepared and when used
for microorganism enumeration. The media was discarded if growth occurred on the sterility
check media, or if there was an absence of growth in the positive response check.
67
-------
5.5.2 Negative Controls
For each sample batch processed, an unused membrane filter and a blank with 100 mL of
buffered, sterilized dilution water was filtered through the membrane were also placed onto the
appropriate media and incubated with the samples as negative controls. No growth was observed
on any blanks.
5.5.3 Collimated Beam Apparatus and QA/QC
The petri dish factor was determined for the collimated beam apparatus prior to the start of the
test program. Radiometers were calibrated and checked in accordance with operating procedure
and UVDGM-2006 requirements. These procedures and data were reviewed as part of the NSF
QA department review of the microbiological laboratory data.
NSF received reviewer comments about the collimated beam data in the Draft EPA ETV NSF
UV reports in November of 2011. They identified two issues related to collimated beam data for
NSF to investigate. The two issues were a high degree of uncertainty with replicates in the
collimated beam data and the data trending at or below the lower 95% confidence interval for
MS2 UV sensitivity. The initial investigation revealed no systematic error. However, further
investigation revealed a miscommunication between the company that calibrated NSF's
radiometers and NSF. It was learned that the radiometers were calibrated at one of two settings
and that NSF used the setting that the radiometer was not calibrated at. Hence the MS2 had
received about 25% more UV dose than estimated by the collimated beam data. Therefore, the
calculated REDmeas at each set point was lower than the actual dose delivered by the unit. NSF
determined that the best action was to retest all previous tested units. That testing was done in
the summer of 2012 and is reported in this report. All previous data is not reported herein as it
was deemed to be biased.
The factors used in the collimated test shown below were evaluated against the protocol
requirements and found to meet the QC objectives. The length (distance from the lamp centerline
to the suspension) and the depth of suspension were fixed parameters. These measurements were
made multiple times at the "fixed mark" on the collimated beam apparatus to estimate the
precision of the measurements. The time was checked based on a stop watch with minimal
uncertainty. The petri dish factor was measured four times prior to the start of the test.
Absorbance uncertainty is based on spectrophotometer precision, as is the related reflectance
factor. The average intensity is measured for every collimated beam test, as it is required that
intensity be measured before and after each test.
To control for error in the UV dose measurement, the uncertainties of the terms in the UV dose
calculation met the following criteria:
Estimated Required
Depth of suspension (d) <5% <10%
Average incident irradiance (Es) 2.5% <8%
Petri Factor (Pf) 2.1% <5%
L/(d + L) 0.7% <1%
Time(t) 1.6% <5 %
(l-10-ad)/ad 1.2%
68
-------
The collimated beam test procedures and key test parameters (radiometer intensity, petri dish
factor, exposure time, suspension depth, distance from lamp to suspension surface, and
absorbance/reflectance) were thoroughly reviewed for all days of testing.
The collimated beam data, as presented in Tables 4-3 and 4-4 show that on Day 2 of testing with
the high UVT water, there was one outlier in replicate 2 for the collimated beam results. The
outlier, as determined by the Grubb statistics presented in Table 4-3 was eliminated from the
dose response curve and equation. Both sets of data, with and without the outlier are presented in
Tables 4-3 and 4-4. As can be seen there was very little difference in the dose response
relationship between the data set with the outlier included and with it removed.
The dose response curve for the collimated beam data were within the QC boundaries set in the
GP-2011 and UVDGM-2006. The boundaries are shown on Figures 4-3 and 4-4 along with all of
the collimated beam dose response data. The low UVT water collimated beam data did have two
data points that were on the upper boundary line and the curve was on the high side of the
midpoint. However, the dose response curve was within the established QC boundaries.
Stability tests for MS2 are normally performed to show that the phage does not change during
holding times when samples are shipped from the test site to the laboratory and/or held in the
laboratory prior to analysis. For these tests, the test rig was located in the same building as the
microbiology laboratory. Samples were delivered to the laboratory after each test run and the
laboratory ran the samples within 4 to 6 hours of sample collection. Therefore, stability samples
were not necessarily needed for these test runs, as the holding time was very short. However,
stability tests were run for the times 0, 4, 8, 24 hours for informational purposes. Table 5-1
shows the stability test results.
Trip blanks are also normally performed to show that the stock phage solution does not change
during shipment to and from the test site. The phage stock solution was delivered from the
microbiology before each test run and the samples were returned to the laboratory after each test
run. Therefore trip blanks were not required for these tests, as all stock solution and test samples
were received from and delivered to the microbiology laboratory before/after each test run. No
shipping or long holding times were required. However, trip blanks were performed once each
day to demonstrate that no significant change was occurring during transport and handling of the
samples. Table 5-2 shows the trip blank results.
69
-------
Table 5-1. Stability Results
Influent 0 Hour
Influent 4 Hour
Influent 8 Hour
Influent 24 Hour
Effluent 0 Hour
Effluent 4 Hour
Effluent 8 Hour
Effluent 24 Hour
Low UVT Water
MS2
PFU/mL
1.07E+05
1.20E+05
1.24E+05
4.80E+04
4.10E+04
5.50E+04
5.30E+05
4.60E+05
4.50E+05
4.60E+05
5.10E+05
4.40E+05
9.40E+02
8.90E+02
9.60E+02
1.83E+03
8.50E+02
6.00E+02
4.90E+03
2.43E+03
3.10E+03
5.20E+03
3.70E+03
4.20E+03
Average
1.17E+05
4.80E+04
4.80E+05
4.70E+05
9.30E+02
1.09E+03
3.48E+03
4.37E+03
logio
5.07
4.68
5.68
5.67
2.97
3.04
3.54
3.64
Influent 0 Hour
Influent 4 Hour
Influent 8 Hour
Influent 24 Hour
Effluent 0 Hour
Effluent 4 Hour
Effluent 8 Hour
Effluent 24 Hour
High UVT Water
MS2
PFU/mL
1.20E+05
1.21E+05
1.14E+05
6.30E+04
8.30E+04
8.70E+04
5.70E+05
5.40E+05
4.40E+05
7.10E+05
6.80E+05
5.40E+05
1.06E+03
8.80E+02
8.70E+02
4.20E+02
5.10E+02
5.20E+02
2.56E+03
2.04E+03
2.23E+03
6.60E+03
6.50E+03
4.10E+03
Average
1.18E+05
7.77E+04
5.17E+05
6.43E+05
9.37E+02
4.83E+02
2.28E+03
5.73E+03
loglO
5.07
4.89
5.71
5.81
2.97
2.68
3.36
3.76
Table 5-2. Trip Blank Results
Held in Micro Lab
Travel to Test Setup
Difference
Day 1 (08-02-12)
MS2
PFU/mL
6.07E+08
5.40E+08
Log™
8.78
8.73
0.05
Day 2 (08-03-12)
MS2
PFU/mL
7.15E+08
7.40E+08
Log™
8.85
8.87
0.02
5.6 Engineering Lab - Test Rig QA/QC
The flow meter for the test rig is part of the NSF tank, pump, and flow control system used for
UV testing and other tests in the engineering laboratory. The flow meter is calibrated by the NSF
70
-------
QA staff at least annually. Calibration is performed by measuring the draw down volume from
the calibrated feed tank over time. The tank was calibrated by filling with measured volumes of
water and the corresponding depth measured. In addition to the annual calibration, the flow
meter was calibrated after the test runs were completed. Calibration was performed at 150, 250,
350, and 450 gpm covering the range of flow rates tested. The calibration of the flow meter
showed accuracy of <1.66% (average of 1.45% over the flow range) and easily met the
requirement of ± 5%. Table 5-3 presents the flow meter calibration data.
Table 5-3. Flow Meter Calibration Data
Flow Meter Reading
(gpm)
453.5
346.0
254.0
150.3
Average
Actual Calculated
Flow Rate based on tank
level change and time
(gpm)
460.1
351.8
256.0
153.2
Deviation
(%)
1.44
1.66
0.78
1.91
1.45
Reactor control and reactor blanks were performed as part of the validation. One reactor control,
with MS2 coliphage injection, and the lamps off, was performed to demonstrate that the MS2
concentration was not changing as the seeded water passed though the reactor. Reactor blanks
were collected for each day of testing to demonstrate that the system was not accumulating or
being contaminated with MS2 at levels that would interfere with the test.
Table 5-4 presents the results of the reactor control and reactor blanks. The reactor control had
an average influent concentration of 5.74 logic and an average effluent concentration of 5.76
logic showing an increase of 0.02 logic through the system with lamps off. This meets the criteria
of less than a 0.2 logic change through the unit with lamps turned off.
During the first day of testing, the reactor blanks showed no detectable MS2 (<1 PFU/ml)
present in either the influent or effluent samples. On Day 2, the influent showed some MS2
present at 20 to 23 PFU/mL (1.30 to 1.36 logic) which is above the objective concentration of
less than 0.2 logic. The effluent samples were clean (<1 PFU/mL). The presence of some MS2 in
the influent did not directly impact the data analysis, as the amount of MS2 was small compared
the influent concentration based on an injection rate of > 5.0 logic.
The results for the blank samples for FtPC, total coliform, and E. coli were presented in Table 4-
20.
71
-------
Table 5-4. Reactor Control and Reactor Blank MS2 Results
Test
Condition
Reactor
Blank
Reactor
Blank
Reactor
Control
Test
Day
1
2
2
Test
Run
1
8
9
UVT
(%)
91
98
98
Flow
(gpm)
150
150
154
Intensity
(mW/cm2)
8.0
7.9
0.0
Influent (PFU/mL)
Repl
<1
23
6.50E+05
Rep 2
<1
22
5.50E+05
Rep 3
<1
20
5.53E+05
Effluent (PFU/mL)
Repl
<1
<1
8.40E+05
Rep 2
<1
<1
4.67E+05
Rep 3
<1
<1
5.47E+05
Test
Condition
Reactor
Blank
Reactor
Blank
Reactor
Control
Test
Day
1
2
2
Test
Run
1
8
9
UVT
(%)
91
98
98
Flow
(gpm)
150
150
154
Intensity
(mW/cm2)
8.0
7.9
0.0
Influent logio
Rep 1
0.00
1.36
5.81
Rep 2
0.00
1.34
5.74
Rep 3
0.00
1.30
5.74
Ave
0.00
1.33
5.76
Effluent Iog10
Rep 1
0.00
0.00
5.92
Rep 2
0.00
0.00
5.67
Rep 3
0.00
0.00
5.74
Ave
0.00
0.00
5.78
72
-------
5.7 Documentation
All laboratory activities were documented using specially prepared laboratory bench sheets and
NSF laboratory reports. Data from laboratory reports were entered into Microsoft™ Excel®
spreadsheets. These spreadsheets were used to calculate the means and logic inactivations. One
hundred percent of the data entered into the spreadsheets was checked by a reviewer to confirm
all data and calculations were correct.
5.8 Data Review
NSF QA/QC staff reviewed the raw data records for compliance with QA/QC requirements. As
required in the ETV Quality Management Plan, NSF ETV staff checked at least 10% of the data
in the NSF laboratory reports against the lab bench sheets.
5.9 Data Quality Indicators
The quality of data generated for this ETV verification is established through four indicators of
data quality: representativeness, accuracy, precision, and completeness.
5.9.1 Representativeness
Representativeness is a qualitative term that expresses "the degree to which data accurately and
precisely represent a characteristic of a population, parameter variations at a sampling point, a
process condition, or an environmental condition." Representativeness was ensuredby consistent
execution of the test protocol for each challenge, 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.
5.9.2 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, where
applicable, and certified standards during calibration of an instrument.
The following equation was used to calculate percent recovery:
Percent Recovery = 100 X [(Xknown - XmeasuredyXknovm]
where:
= known concentration of the measured parameter
= measured concentration of parameter
Accuracy of the bench top chlorine, pH, and turbidity meters was checked daily during the
calibration procedures using certified check standards. The in-line UVT monitor was calibrated
daily with both a purchased UVT standard and with DI water at 99.9% UVT, before the flow
tests.
73
-------
The NSF Laboratory Quality Assurance Manual establishes the frequency of spike sample
analyses at 10% of the samples analyzed for chemical analyses. Laboratory control samples are
also run at a frequency of 10%. The recovery limits specified for the parameters in this
verification, excluding microbiological analyses, were 70-130% for laboratory-fortified (spiked)
samples and 85-115% for laboratory control samples. The NSF QA department reviewed the
laboratory records and found that all recoveries were within the prescribed QC requirements.
Calibration requirements were also achieved for all analyses.
5.9.3 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 iron and
manganese measurement. At least one out of every ten samples for pH, total chlorine, free
chlorine, temperature, and turbidity was analyzed in duplicate as part of the daily calibration
process. Precision of duplicate analyses was measured by use of the following equation to
calculate Relative Percent Deviation (RPD):
RPD =
:200
where:
Sl = sample analysis result; and
S2 = sample duplicate analysis result.
Acceptable analytical precision for the verification test was set at an RPD of 30%. Field
duplicates were collected at a frequency of one out of every 10 samples for each parameter, to
incorporate both sampling and analytical variation to measure overall precision against this
objective. In addition, the NSF Laboratory also conducted laboratory duplicate measurements at
10% frequency of samples analyzed. The laboratory precision for the methods selected was
tighter than the 30% overall requirement, generally set at 20% based on the standard NSF
Chemistry Laboratory method performance.
All RPD were within NSF's established allowable limits for each parameter.
5.9.4 Completeness
Completeness is the proportion of valid, acceptable data generated using each method as
compared to the requirements of the TQAP plan. The completeness objective for data generated
during verification testing is based on the number of samples collected and analyzed for each
parameter and/or method, as presented in Table 5-5.
74
-------
Table 5-5. Completeness Requirements.
Number of Samples per Parameter and/or Method
0-10
11-50
>50
Percent Completeness
80%
90%
95%
Completeness is defined as follows for all measurements:
%C = (V/T)x 100
where:
%C = percent completeness;
V = number of measurements judged valid; and
T = total number of measurements.
One hundred percent completeness was achieved for all aspects of this validation except for the
MS2 run samples, power measurements, and collimated beam analyses. Two samples (one
influent and one effluent - test run 4) were not analyzed out of a total of 120 samples scheduled
during the main test runs. The completeness is calculated as 98.3% complete for the MS2 test run
samples. One power measurement was not recorded out of a total of 15 scheduled measurements
for a completeness of 93.3%. There was one outlier result for replicate 2 from the collimated
beam tests on the high UVT water. A total of 24 samples were scheduled and 23 used for the
dose response relationship. Therefore, completeness was 95.8%. All planned testing activities
were conducted as scheduled.
75
-------
Chapter 6 References
1. Test/Quality Assurance Plan for The Ultraviolet Sciences Inc. Ultraviolet (UV) Water
Purification System Model UVS438s-500 Reactor, August 2010.
2. Generic Protocol for Development of Test/Quality Assurance Plans for Validation of
Ultraviolet (UV) Reactors, NSF International, 7/2010.
3. Generic Protocol for Development of Test / Quality Assurance Plans for Validation of
Ultraviolet (UV) Reactors, 10/01/EPADWCTR, August 2011.
4. Ultraviolet Disinfection Guidance Manual For the Long Term 2 Enhanced surface Water
Treatment Rule, Office of Water, US Environmental Protection Agency, EPA 815-R-06-
007, November 2006.
5. German Association for Gas and Water (DVGW) Technical Standard
Work Sheet W 294-1,2,3 (June 2006).
6. Austrian Standards, ONORM M5873-1, Plants for the disinfection of water using
ultraviolet radiation, Requirements and testing, Low pressure mercury lamp plants (
March, 2001).
7. APHA, AWWA, and WEF (1999). Standard Methods for the Examination of Water and
Wastewater, 20th Edition.
8. NSF International (2007). NSF/ANSI Standard 55 - Ultraviolet Microbiological Water
Treatment Systems.
9. NSF International (2011). NSF/ANSI Standard 50 - Equipment for Swimming Pools,
Spas, Hot Tubs and Other Recreational Water Facilities.
10. Water Report 113, Safe, Sufficient and Good Potable Water Offshore: A guideline to
design and operation of offshore potable water systems. 2nd edition. By Eyvind
Andersen and Bj0rn E. L0fsgaard, Norwegian Institute of Public Health, November 2009.
11. Recommended Standards For Water Works, Policies for the Review and Approval of
Plans and Specifications for Public Water Supplies, 2012 Edition, A Report of the Water
Supply Committee of the Great Lakes—Upper Mississippi River Board of State and
Provincial Public Health and Environmental Managers.
76
-------
Attachment 1
NeoTech D438™ Technical Manual and Specifications
Contact Mr. Bruce Bartley at 734-769-5148 or bartley@nsf.org for a copy of this document.
This attachment is a large pdf file.
-------
Attachment 2
Additional Sensor and Lamp Information
Contact Mr. Bruce Bartley at 734-769-5148 or bartley@nsf.org for a copy of this document.
This attachment is a pdf file.
-------
Attachment 3
NSF Collimated Beam Apparatus
-------
Description of the Collimated Beam Apparatus
(NSF Standard 55 7.2.1.2)
An apparatus shall be assembled in which a small stirred sample can be irradiated in a nearly collimated
beam. A radiometer meeting specification in 7.2.1.2.1 can then be used to measure the incident
irradiance (E0).
A low-pressure mercury vapor UV lamp shall be wired to a ballast and a voltage regulator (figure 2). A
solution contained in a small dish equal to or smaller in diameter than that of the collimated tube shall be
used. The solution shall be 1 cm deep. E0 shall be measured at the surface of the liquid by removing the
dish and stirrer and placing the radiometer at the corresponding position from which the dish was
removed. The UV irradiance at each point of the surface shall be within ± 5% of the average irradiance
across the solution surface.
shield '
jport
nd *-
I 1
UV lamp
collimating
tube
*+ — petri dish
o
-* — magnetic stirrer
NOTE 1 - The collimating tubes shall be a minimum of 53 cm (21 in) in length and the interior shall be painted
flat black.
NOTE 2 - The support stand, if used, shall be adjustable to raise or lower the collimating tube to the surface
of the petri dish.
NOTE 3 - The petri dish shall be set so the surface of the liquid is at the same level as the radiometer.
NOTE 4 - Measurement of the UV dose must be done at the same point at which the petri dish surface is ex-
posed.
Figure 2 - Collimated beam apparatus
7.2.1.2.1 Radiometer specifications
A radiometer with the following specification shall be used:
- linearity: ± 0.5%;
-------
- spectral response: visible-blind detector with narrow band-pass filter centered at 254 nm, full
width at half maximum = 20 nm or less;
- spatial response: cosine response ± 5%;
- calibration: Radiometer calibration (including optics, transducer and electronics) shall be
traceable to the National Institute of Standards and Technology (NIST) or another national standards
laboratory. Calibration shall be performed annually or at the intervals specified by manufacturer,
whichever is more frequent;
Eave=0.98
- uncertainty: The calibration documentation provided with each radiometer (including optics,
transducer, and electronics) shall include both calibration uncertainties (transfer uncertainty to
customer) and the uncertainty associated with the calibration standard. The NIST (or other national
laboratory) uncertainty is added the transfer uncertainty to customer to yield total uncertainty; and
- maximum total uncertainty: ± 9 % at 254 nm.
-------
Method for Challenge Microorganism Preparation, Culturing the Challenge Organism and
Measuring its Concentration
(NSF Standard 55 Annex A)
A.1 Summary
MS-2 Coliphage is used as the biological surrogate to determine the average UV dose output of UV water
treatment systems. The methods that are used for suspension preparation, titration, and analysis of the
challenge organisms for use in the sensitivity calibration and testing are presented in this annex.
A.2 Equipment
- autoclave;
- radiometer (International light IL-700);
- UV collimating beam apparatus and 254 nm photo detector;
- incubator, 35 ± 1 °C (95 ± 1 °F);
- refrigerator, 5 ± 3 °C (41 ± 3 °F);
- water bath 50 ± 1 °C (122 ±1 °F);
- freezer;
- microwave;
- vortex mixer;
- UV-vis spectrophotometer;
- pH meter;
- hemocytometer;
- Colony Counter; and
- centrifuge.
A.3 Microorganisms
All organisms shall be obtained from ATCC.
- MS-2 Coliphage (ATCC # 15597-BI); and
- Escherichia coli host strain (ATCC # 15597).
A.4 Supplies
- Petri dishes, 20 x 60 mm and 15 x 100 mm: sterile;
- pipettes, 1 ml and 10 ml, sterile;
- sterile centrifuge tubes, 10 ml and 50 ml;
- sample bottles, 125 ml sterile screw cap;
- test tubes, 16 x 125 mm;
- sterile inoculating loop;
- sterile filtration apparatus;
- sterile 0.22 urn polycarbonate membrane filters;
- Whatman #1 filter;
- chlorine detection kit; and
- disposable sterile 250 ml polypropylene container.
-------
A. 5 Reagents
- Sterile buffered dilution water (SBDW). This shall be prepared according to the Standard Methods
for the Examination of Water and Wastewater (dilution water: buffered water);
- Phosphate buffer saline (PBS). A stock solution shall be prepared by dissolving 80 g sodium
chloride (NaCI), 2 g potassium dihydrogen phosphate (KH2PO4), 29 g hydrated disodium hydrogen
phosphate (Na2HPO4-12H2O), and 2 g potassium chloride (KCI) in water to a final volume of 1 L. A
working solution shall be prepared from the stock solution by diluting 1 volume of the stock with 9
volumes of water. The pH shall be adjusted using a pH meter to 7.4 with 0.1 N HCI or 0.1 N NaOH
before use;
- Ethylenediaminetetraacetic acid (EDTA), Sigma # ED2SS; and
- Lysozyme, Boehringer Mannheim, #1 243004. Store at 2 to 8 °C (35 to 46 °F).
A.6 Safety precautions and hazards
A.6.1 Steam sterilized samples and equipment shall be handled with protective gloves when being
removed from the autoclave.
A.6.2 Cryogenic culture vials shall be handled with cryoprotective gloves.
A.6.3 Ultraviolet light shall be used to expose the organism during calibration. This light can result in
skin cancer and retinal damage; hence personnel must be protected from exposure.
A.6.4 All microbiological samples and contaminated test supplies shall be steam sterilized to 121 ± 1 °C
(250 ±1 °F) at 15 psi fora minimum of 20 min prior to being discarded.
A.7 Growth medium
NOTE 1 - Common bacteriological media may be purchased from bacteriological medium manufacturers and
prepared according to the manufacturer's instructions.
NOTE 2 - The quality of the growth media shall be monitored by examining growth promotion and sterility prior
to use.
A.7.2 Formula to be used when MS-2 Coliphage is chosen for microbiological agent
A.7.2.1 TSB (Tryptic Soy Broth)
Ingredient
tryptone
soytone
dextrose
sodium chloride
dipotassium phosphate
Dl water
pH
Amount
1.7g
0.3 g
0.25 g
0.5 g
0.25 g
100mL
7.3 ±0.2
-------
TSB shall be dissolved by boiling and adjusted to final pH. 8-mL aliquots shall be dispensed into 16 x 150
mm test tubes. TSB shall be autoclaved at 121 ±1 °C (250 ±1 °F) at 15 psi for 20 min. Cooled broth shall
be stored at 5 ± 1 °C (41 ± 1 °F).
A.7.2.2 1.5% TSA (Tryptic Soy Agar)
Ingredient
tryptone
soytone
sodium chloride
bacto-agar
Dl water
PH
Amount
7.5 g
2.5 g
2.5 g
7.5 g
500 ml
7.3 + 0.2
TSA shall be dissolved by boiling, adjusted to final pH, and autoclaved at 121 ±1 °C (250 ±1 °F) at 15
psi for 20 min. Tempered media shall be poured into sterile petri dishes. Agar plates shall be stored at 5 ±
1 °C (41 ± 1 °F). Plates shall be allowed to come to room temperature before use.
A.7.2.3 Phage top agar 1% ISA (Tryptic Soy Agar)
Ingredient
tryptone
soytone
sodium chloride
agar
Dl Water
PH
Amount
7.5 g
2.5 g
2.5 g
5.0 g
500 ml
7.3 ±0.2
TSA shall be dissolved by boiling, adjusted to final pH, and autoclaved at 121 ±1 °C (250 ±1 °F) at 15
psi for 20 min. Agar shall be stored at 5 ± 3 °C (41 ±1 °F). On the day of testing, the TSA shall be
liquefied and placed in the 45 ± 1 °C (113 ± 1 °F) water bath. The MS-2 Coliphage top agar shall be
maintained at 45 ± 1 °C (113 ± 1 °F) to prevent agar solidification.
A.8 Culture of challenge organisms
A.8.2 MS-2 Coliphage
A.8.2.1 Stock culture preparation of MS-2 Coliphage
NOTE - This section describes the propagation and harvesting methods for stock suspensions of MS-2
Coliphage for use as a challenge suspension for low flow (< 1 gpm) water treatment units. If units possessing a
flow rate greater than 1 gpm are to be tested, the stock preparation procedure may have to be repeated multiple
times to achieve the required volume of MS-2 Coliphage. This method should also be repeated when cryogenic
stocks are low.
a) One day prior to preparation of MS-2 Coliphage stock, a cryogenically frozen E. coli host strain
shall be thawed. One TSB tube shall be inoculated with 0.1 ml of the stock suspension. The stock
suspension shall be incubated at 35 ± 1°C (95 ±1 °F) for 18 ± 2 h.
b) On the day of preparing MS-2 Coliphage stock, 1% TSA shall be liquefied and the media shall be
tempered in a 45 ± 1 °C (113 ± 1 °F) water bath. 1.5% TSA plates shall be room temperature prior to
use.
-------
c) Serial dilutions of MS-2 Coliphage suspension (10~1 to 10~12) shall be made using sterile PBS. 10~5
to 10~12 dilutions shall be plated in triplicate on 1.5% ISA plates. In a sterile tube, 1 ml of diluted MS-
2 Coliphage shall be transferred. Then 0.1 ml of E. coli host shall be added quickly to ~ 5 ml of
melted 1% ISA. The inoculum and media shall be vortexed and poured on ISA plates. The plates
shall be rocked to spread inoculum evenly. After the 1% TSA layer has solidified, the plates shall be
inverted and incubated at 35 ± 1°C (95 ±1 °F) for 18 ± 2 h.
d) Plates shall be selected that show complete lysis of host cells by the MS-2 Coliphage. The
surface of each plate shallbe flooded with 3 ml of TSB. The 1% TSA layer shall be gently removed
using a cell scraper. The contents shall be poured into two sterile 50 ml centrifuge tubes and the total
volume brought to 40 ml with TSB. 0.2 g EDTA and 0.026 g lysozyme shall be added to each tube.
The centrifuge tubes shall be incubated at room temperature for 2 h, mixing every 15 min.
e) After the 2 h incubation, the tubes shall be centrifuged at 9280 xg for 5 min, or 2320 xg for 20
min, at 20 ± 1 °C (68 ± 1 °F). The resulting supernatant shall be removed while avoiding the pellet. A
sterile 47-mm filtration assembly shall be aseptically constructed using a 0.22-|j,m polycarbonate filter.
The filter shall be pretreated with 10 ml of TSB broth just prior to the filtration to minimize MS-2
Coliphage adsorption to the filter. The supernatant shall be filtered.
f) For long-term storage (greater than 28 d), V10 volume of sterile glycerol shall be added to
suspension, dispensed into 1 ml and 3 ml aliquots in cryovials, and stored at -70° ± 1 °C (-94 ± 1
°F).
g) The MS-2 Coliphage suspension shall be titrated as in A.8.2.2. The concentration of MS-2
Coliphage should be 10™ to 1012 PFU/mL
A.8.2.2 Enumeration of MS-2 Coliphage plaques
a) A cryogenically frozen E. coli host strain shall be thawed. One TSB tube shall be inoculated with
0.1 ml of the stock suspension. The TSB tube shall be incubated at 35 ± 1 °C (95 ±1 °F) for 18 ± 2
h.
b) 1% TSA shall be liquefied and the media shall be tempered in a 45 ± 1 °C (113 ± 1 °F) water
bath. 1.5% TSA plates shall be room temperature prior to use.
c) Serial dilutions of MS-2 Coliphage suspension (10~1 to 10~12) shall be made using sterile PBS. 10~7
to 10~12 dilutions shall be plated in triplicate on 1.5% TSA plates. In a sterile tube, 1 ml of diluted MS-
2 Coliphage shall be transferred. Then 0.1 ml of E. coli host shall be added quickly to ~ 5 ml of
melted 1% TSA. The inoculum and media shall be vortexed and poured on TSA plates. The plates
shall be rocked to spread inoculum evenly. After the 1% TSA layer has solidified, the plates shall be
inverted and incubated at 35 ± 1°C (95 ±1 °F) for 18 ± 2 h.
d) After incubation, plates containing 20 - 200 distinct plaque forming units (PFU) shall be
enumerated using a Colony Counter. The MS-2 Coliphage suspension titer shall be calculated by
multiplying the number of PFU obtained by the inverse of the dilution factor. The concentration of MS-
2 Coliphage should be 1010 to 1012 PFU/mL.
A.9 Drinking water treatment unit challenge organism suspension preparation
A.9.1 Determination of the concentration of challenge organism
This determination will be based upon the unit flow rates, injection feed pump rate, suspension density,
and the final challenge organism concentration for the unit challenge. The suspension shall be of
adequate volume to deliver the challenge organism to two complete on/off cycles at each sample point.
-------
Example:
unit flow rate: 1.0 gpm; duplicate units tested so total of 2.0 gpm (7560 mL/min);
injection rate: 10 mL/min;
suspension density: 1 x 109/ml_;
final concentration: 7.0 x 104/ml_; and
on/off cycle: 10 min /10 min (20 min on for two complete cycles).
a) To challenge for 20 min at two 10 min intervals, a total of 200 ml of suspension is needed to
challenge 151,200 ml of water (7560 min x 20 min):
(7.0 x 104/ml_)(151,200 ml) = (injection feed conc.)(200 ml); and
injection feed concentration = 5.3x107/ml_.
b) To prepare this from the stock suspension, combine:
(200 ml_)(5.3 x 107/ml_) = (ml of suspension density)(1.0 x 109/ml_);
ml of suspension density = 10.6 ml; and
10.6 ml of suspension to 189.4 ml of PBS.
Once suspension has been made, the suspension shall be mixed using a magnetic stirrer.
A 10-mL aliquot shall be removed from the challenge suspension and set aside for density
verification according to Standard Methods for the Examination of Water and Wastewater.
A. 10 Analysis of influent and effluent samples
A.10.2 Enumeration of MS-2 Coliphage plaques
a) Serial dilutions of the influent and effluent samples (10° to 10~5) shall be made using sterile PBS.
10° to 10~5 dilutions shall be plated in duplicate on 1.5% TSA plates. In a sterile tube, 1 ml of diluted
MS-2 Coliphage shall be transferred. Then 0.1 ml of E. coli host shall be added quickly to ~ 5 ml of
melted 1% TSA. The inoculum and media shall be vortexed and poured on TSA plates. The plates
shall be rocked to spread inoculum evenly. After the 1% TSA layer has solidified, the plates shall be
inverted and incubated at 35 ± 1°C (95 ±1 °F) for 18 ± 2 h.
b) After incubation, plates containing 20 - 200 distinct plaque forming units (PFU) shall be
enumerated using a Colony Counter. The MS-2 Coliphage suspension titer shall be calculated by
multiplying the number of PFU obtained by the inverse of the dilution factor. Results shall be
expressed as the number of PFU/mL.
A.11 Challenge verification
After the appropriate incubation period for MS-2 Coliphage, the plaques shall be counted on all of the
density determination plates. The mean number of microorganisms per milliliter for plates with 25 to 250
colonies/plaques shall be calculated. This shall verify that the challenge organism was present in the
challenge test water at the optimum concentration before being added to test apparatus.
-------
Attachment 4
UVT Scans of Influent and Effluent Water at high and low UVT
Contact Mr. Bruce Bartley at 734-769-5148 or bartley@nsf.org for a copy of this document.
This is a large pdf file.
------- |