Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Insert Client Logo here or delete. Note: If more than two clients logos are to be inserted press ALT CL to use footer
Report Reference: DEFRA10932.05
June 2016
Comparison of Private Water Supply and
Public Water Supply Ultraviolet (UV)
Systems: Final Report
RESTRICTION: This report has the following limited distribution:
External: Defra
Any enquiries relating to this report should be referred to the Project Manager at the
following address:
WRc plc,
Frankland Road, Blagrove,
Swindon, Wiltshire, SN5 8YF
Telephone: + 44 (0) 1793 865000
Website: www.wrcplc.co.uk
Follow Us:
WRc is an Independent Centre
of Excellence for Innovation and
Growth. We bring a shared
purpose of discovering and
delivering new and exciting
solutions that enable our clients
to meet the challenges of the
future. We operate across the
Water, Environment, Gas, Waste
and Resources sectors.
Document History
Version
number
Purpose Issued by Quality Checks
Approved by
Date
V.05 Final report issued to DWI. David Shepherd,
Project Manager
Glenn Dillon June 2016
© Defra 2016 The contents of this document are subject to copyright and all rights are reserved. No part of this document may be reproduced, stored in a retrieval system or transmitted, in any form or by any means electronic, mechanical, photocopying, recording or otherwise, without the prior written consent of Defra.
This document has been produced by WRc plc.
Comparison of Private Water Supply and Public
Water Supply Ultraviolet (UV) Systems: Final
Report
Authors:
Glenn Dillon
Technical Consultant
WATT
Date: June 2016
Report Reference: DEFRA10932.05
Tom Hall
Principal Consultant
WATT
Project Manager: Glenn Dillon
Project No.: 16375-0
David Shepherd
Senior Process Engineer
WATT
Client: Defra
Client Manager: Mick Stanger
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
i
Contents
Glossary and Abbreviations ..................................................................................................... 1
Summary .................................................................................................................................. 2
1. Introduction .................................................................................................................. 6
1.1 Background ................................................................................................................. 6
1.2 Objectives .................................................................................................................... 6
1.3 Report résumé ............................................................................................................. 7
2. UV Technologies in Public Water Supply ................................................................... 8
2.1 Objective ..................................................................................................................... 8
2.2 UV design and operating information provided by water companies for earlier DWI project....................................................................................................... 8
2.3 Further design and operating information provided by water companies ................. 10
2.4 Information provided by UV companies .................................................................... 12
2.5 Further comment from UV Supplier 2 ....................................................................... 15
2.6 Summary ................................................................................................................... 16
3. UV Technologies in Private Water Supply ................................................................ 17
3.1 Objective ................................................................................................................... 17
3.2 UV equipment suppliers ............................................................................................ 17
3.3 Local Authority survey ............................................................................................... 23
3.4 Survey of UV equipment installers ............................................................................ 26
4. Critical differences between UV Technologies used in Public and Private Water Supplies .............................................................................................. 27
4.1 Objective ................................................................................................................... 27
4.2 Comparison of UV systems ....................................................................................... 27
5. Comparison of Validation Criteria for different UV Systems ..................................... 31
5.1 Objective ................................................................................................................... 31
5.2 Current standards...................................................................................................... 31
5.3 Comparison of standards .......................................................................................... 43
5.4 Conclusions ............................................................................................................... 46
6. Review Standards for UV Systems and Identify Validation Criteria suitable for Private Supply ........................................................................................ 49
6.1 Objective ................................................................................................................... 49
6.2 Implications of water quality regulations ................................................................... 49
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
ii
6.3 Requirements for use of UV disinfection for public water supplies ........................... 52
6.4 Potential validation criteria for private supply ............................................................ 52
6.5 Conclusions ............................................................................................................... 57
7. Self-help Leaflet for Households ............................................................................... 58
7.1 Objective ................................................................................................................... 58
7.2 Guide to the selection of UV disinfection systems for households ........................... 58
8. Guide for Local Authorities ........................................................................................ 64
8.1 Objective ................................................................................................................... 64
8.2 Guide to the assessment of UV disinfection systems for local authorities................ 64
9. Design of a Pilot Study to evaluate a UV System for Private Water Supplies ..................................................................................................................... 71
9.1 Objective ................................................................................................................... 71
9.2 Test regime ............................................................................................................... 71
9.3 Procedure .................................................................................................................. 72
9.4 UVT measurement .................................................................................................... 73
10. Conclusions ............................................................................................................... 75
11. Recommendations .................................................................................................... 78
References ............................................................................................................................. 79
Appendices
Appendix A UV Technologies in Public Water Supply: Further Design
and Operating Information ....................................................................... 82
Appendix B Biodosimetry ............................................................................................ 88
Appendix C UV sensitivity of micro-organisms ........................................................... 97
Appendix D UV Suppliers .......................................................................................... 100
Appendix E Local authority site visits ....................................................................... 151
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
iii
List of Tables
Table 2.1 Public supplies: Summary of information from previous DWI project ........................................................................................................ 8
Table 2.2 Public supplies: Summary of installations by works size and water type .................................................................................................. 9
Table 2.3 Public supplies: Summary of installations by design dose ........................ 9
Table 2.4 Public supplies: Summary of installations by lamp type ............................ 9
Table 2.5 Public supplies: Summary of installations by supplier ............................. 10
Table 3.1 Summary of UV systems available for small supplies ............................. 18
Table 4.1 Comparison of UV systems for public and private water supplies ................................................................................................... 27
Table 4.2 Typical characteristics of UV mercury vapour lamps (Bolton and Cotton, 2008) .................................................................................... 30
Table 5.1 Summary of standards and guidelines .................................................... 33
Table 5.2 Comparison between UVDGM and ÖNORM validation methodologies ......................................................................................... 45
Table 5.3 Comparison between BSI and NSF/ANSI standards .............................. 47
Table 6.1 PCVs for private water supplies of potential relevance to UV disinfection ............................................................................................... 49
Table B.1 How standards address experimental uncertainties ............................... 92
Table C.1 UV dose (mJ/cm2) for inactivation of protozoa and viruses .................... 97
Table C.2 UV dose (mJ/cm2) for inactivation of spores and bacteria ...................... 98
Table E.1 Summary of site visits to Local Authority ‘A’ ......................................... 154
Table E.2 Summary of site visits to Local Authority ‘B’ ......................................... 159
Table E.3 Summary of site visits to Local Authority ‘C’ ......................................... 164
Table E.4 Summary of site visits to Local Authority ‘D’ ......................................... 168
List of Figures
Figure 6.1 Example correlations between colour and UVT in UK upland raw waters ............................................................................................... 51
Figure 6.2 Relative cost of additional control functionality for small-scale UV devices ..................................................................................... 54
Figure 8.1 Typical treatment flowsheet ..................................................................... 67
Figure B.1 Biodosimetry validation procedure .......................................................... 89
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
1
Glossary and Abbreviations
LP Low pressure (UV lamp).
LPHO Low pressure high output (UV lamp).
MP Medium pressure (UV lamp).
RED Reduction equivalence dose. Common units are mJ/cm2.
Alternative terminology for REF.
REF Reduction equivalence fluence. Common units are mJ/cm2.
Alternative terminology for RED.
UV Ultraviolet.
UVI UV intensity. Common units are mW/cm2.
UVT % UV transmittance through 1 cm of water.
Standards and guidelines applicable to UV systems
Austrian Standards Institute
(ÖNORM)
M 5873-1 2001: Plants for the disinfection of water using UV
radiation – Requirements and testing – Low pressure mercury
lamp plants.
Austrian Standards Institute
(ÖNORM)
M 5873-2 2003: Plants for the disinfection of water using UV
radiation – Requirements and testing – Part 2: Medium
pressure mercury lamp plants.
British Standards Institute (BSI) BS EN 14897:2006 / A1:2007: Water conditioning equipment
inside buildings – Devices using mercury low-pressure UV
radiators – Requirements for performance, safety and testing.
Germany (DVGW) W 294-1 2006: UV devices for the disinfection of water supply –
Part 1: Requirements on the design, function and action.
Germany (DVGW) W 294-2 2006: UV devices for the disinfection of water supply –
Part 2: Tests of design, function and disinfection effectiveness.
Germany (DVGW) W 294-3 2006: UV devices for the disinfection of water supply –
Part 3: Sensors for the photometric monitoring of UV
disinfection: Tests and calibration.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
2
Summary
i Reasons
Regulatory sampling of drinking water shows greater than 99% compliance with
microbiological standards on public supplies but considerably lower compliance on private
supplies. One factor affecting private supplies is believed to be inappropriate implementation
of some UV systems.
This study has highlighted the critical differences between UV technologies used on public
and private supplies, and established the suitability and performance of the most common UV
system(s) used on private supplies.
ii Objectives
Establish the range of UV technologies used on public and private supplies in England
and Wales, and establish the critical differences in functionality and application.
Review international standards for UV treatment systems to compare validation criteria
and identify which criteria would demonstrate suitability for use in private supplies.
Produce simple guidance for householders and local authorities to help in the selection
and assessment of UV systems used in private supplies.
iii Benefits
This study has highlighted some major difficulties associated with the implementation of UV
disinfection for private supplies. Addressing these difficulties will increase the reliability and
performance of such systems.
iv Conclusions
UV technologies in public water supply
UV disinfection is widely used in public water supply, with most installations <10 Ml/d but also
larger installations >100 Ml/d.
Design is usually based on detailed feed water quality data, with a dose of 40 mJ/cm2 or
higher for the majority of units. Dose validation according to ÖNORM, DVGW or USEPA is
becoming the norm.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
3
Monitoring and control is usually based on measurement of UV intensity; UVT is also
measured and sometimes used for control. Feed water turbidity is monitored according to
Regulation 26 requirements.
UV lamps are of the MP or LPHO type, with cleaning and replacement carried out routinely at
supplier defined intervals. UV intensity monitors are routinely calibrated.
UV technologies in private water supply
UV disinfection used in private water supply is mostly <10 m3/d (often much smaller); the
larger units are usually installed at commercial premises rather than domestic.
Design may be based on limited feed water quality data, with pre-treatment specified to deal
with poorer feed water quality. UV dose is typically 30 mJ/cm2 for domestic units and
40 mJ/cm2 for larger commercial units. Little, if any, biodosimetric dose validation; some larger
suppliers may carry out microbial challenge testing or hydraulic and UV intensity modelling.
Limited monitoring and control, particularly for domestic units, with control usually based on
maximum flow rate and specified UVT of the feed water. No measurement of turbidity or UVT;
some of larger commercial units may include UV intensity monitors which provide a shutdown
rather than a control capability.
UV lamps are of the LP type, with cleaning and lamp replacement carried out annually
(typically) by installers under service agreements in many cases; some simple systems may
be serviced by owners.
Key findings from site visits to private supplies incorporating UV disinfection
There was a general lack of understanding amongst users regarding the treatment of their
private supplies. This was compounded by the lack of information provided by equipment
providers/installers.
There was no indication that UV equipment had been selected correctly for the flow or water
quality.
Smaller private supplies and SDDs incorporated simple treatment, typically particulate
filtration and/or UV disinfection. Some larger commercial private supplies incorporated more
complex treatment systems.
UV equipment was generally serviced by specialist companies, plumbers or the users, with
quartz sleeves cleaned at intervals between 2-12 months and lamps changed around every
12 months; the frequency of maintenance of other equipment and replacement of cartridge
filters was less clear. Maintenance logs are generally not kept by users.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
4
Monitoring and control of UV equipment is inadequate. Failure of a UV lamp may go
undetected for some time because a lack of a prominent alarm, and will generally not prevent
flow and the possibility of the consumption of non-disinfected water.
The potential for contamination of stored UV-treated water may not be well understood by
users.
There is currently no licensing or approved contractor scheme applicable to the installation of
equipment for private water supplies.
Review and comparison of standards and validation criteria for UV systems
UK (BSI) and international standards (USEPA (UVDGM), ÖNORM, DVGW, NWRI/WRF and
NSF/ANSI) have been reviewed and compared.
The USEPA (UVDGM), ÖNORM, DVGW and NWRI/WRF standards apply to public water
supplies.
The BSI standard applies to LP UV devices intended for water conditioning in buildings; the
NSF/ANSI standard applies to point-of-entry and point-of-use LP UV equipment.
The BSI standard specifies a dose of 40 mJ/cm2 validated by biodosimetry; the NSF/ANSI
standard specifies a dose of 40 mJ/cm2 (disinfection) or 16 mJ/cm
2 (supplemental bactericidal
systems) validated by biodosimetry.
A reduction equivalence dose (RED) of 40 mJ/cm2 as required by the ÖNORM (and DVGW)
and BSI standards is the preferred validation criterion.
A UVI sensor is stipulated by all standards where UV is installed for disinfection applications.
Such a sensor is considered desirable, but not necessarily essential, for private supply
applications.
Design of a pilot study to evaluate a UV system for private water supplies
A pilot study based on either European (DVGW and ÖNORM) UV dose validation or US
(UVDGM) UV dose validation is proposed to evaluate a UV system spiked with surrogate
microorganisms under a range of flow, UV lamp intensities (doses) and water quality (UVT)
conditions.
v Recommendations
A number of key recommendations are suggested that would improve the reliability and
performance of UV disinfection for private supplies:
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
5
A licensing or approved contractor scheme should be implemented for installers of
equipment for private water supplies.
Copies of manufacturers’/suppliers’ operating and maintenance instructions should be
provided and retained by the supply owner. In addition, a maintenance log should be
maintained by the owner to record details of maintenance carried out and schedules for
future maintenance.
Audible and visual alarms should be more prominent, particularly where the UV system
is sited away from the user’s premises.
UV systems should include automatic shutdown of the water supply in the event of
power or lamp failure. An emergency valved by-pass line could be incorporated with
instructions to boil drinking water prior to consumption (whilst the UV system awaits
repair).
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
6
1. Introduction
1.1 Background
Drinking water supplied in England and Wales must be wholesome and safe to drink; this
applies whether the water is a public or private supply.
A minimum treatment requirement for public water supplies is that all supplies must be
disinfected. The requirement (or not) for private water supplies to be disinfected is informed
by risk assessments carried out by local authorities. Ultraviolet (UV) disinfection has been
used for many years on both public and private supplies. If properly designed and maintained,
UV disinfection will inactivate harmful microorganisms ensuring that water is safe to drink.
Regulatory sampling of drinking water shows greater than 99% compliance with
microbiological standards on public supplies but considerably lower compliance on private
supplies. Results of testing in England and Wales during 2014 showed private supplies to be
of unsafe microbiological quality, with 12.8% of samples containing E. coli and 13.4%
containing Enterococci (DWI, 2015a). One factor affecting private supplies is believed to be
inappropriate implementation of some UV disinfection systems.
One outcome of this study is an outline for pilot trials to establish the suitability and
performance of the most common UV system(s) used on private supplies. Such trials will
elucidate issues with UV disinfection and in the longer term help to improve drinking water
quality.
1.2 Objectives
The objective of this study was to understand and highlight the critical differences between
UV technologies used on public and private supplies. This study:
Established the range of UV technologies used on public and private supplies in
England and Wales, and established the critical differences in functionality and
application.
Reviewed international standards for UV treatment systems to compare validation
criteria and identified which criteria would demonstrate suitability for use in private
supplies.
Produced simple guidance for householders and local authorities to help in the
selection and assessment of UV systems used in private supplies.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
7
1.3 Report résumé
Following this brief introduction to the background and objectives of the study, this draft final
report includes the following sections:
Section 2 reviews UV technologies used in public water supplies and collates
information gathered from surveys of water utilities and UV equipment
manufacturers/suppliers.
Section 3 similarly reviews UV technologies used in private water supplies and collates
information gathered from 11 UV equipment manufacturers/suppliers, and presents the
findings of site visits made to 25 private supplies incorporating UV disinfection.
Section 4 compares the critical differences between the functionality and application of
UV technologies used in public and private supplies.
Section 5 reviews and compares current standards and guidelines applicable to UV
systems used in drinking water treatment, including validation criteria, and identifies
standards applicable to public supplies (ÖNORM, DVGW, UVDGM) and those
applicable to private installations (BSI, NSF/ANSI).
Section 6 reviews validation criteria for UV systems suitable for private supplies.
Section 7 presents a simple guide to help households select a suitable UV system.
Section 8 presents a simple guide to help local authorities assess the suitability of an
installed UV system.
Section 9 proposes a pilot study to evaluate the performance of a selected UV system
for a private supply based on the inactivation of Bacillus subtilis spores or MS2
Coliphages under a range of flow, UV lamp intensities (doses) and water quality (UVT)
conditions.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
8
2. UV Technologies in Public Water Supply
2.1 Objective
To establish the range of UV technologies employed by water companies in England and
Wales, and establish the functionality and application.
2.2 UV design and operating information provided by water companies for earlier DWI project
Details of UV technologies employed by water companies in England and Wales have been
collated as part of the present study, incorporating information collated for a previous
Defra/DWI study (DWI, 2015b). Additionally, information on the installed UV systems has
been gathered from the principal UV manufacturers and suppliers.
The questionnaire survey for the previous DWI project identified 139 UV plants (existing and
proposed) from returns from 16 water companies (including 6 nil returns) with a total UV
treatment capacity of 1,492 Ml/d. The UV treatment capacity represents approximately 23% of
the production capacity of the ten companies utilising UV (between 3-100% of capacity) and
approximately 17% of the production capacity of all 16 companies.
Table 2.1 Public supplies: Summary of information from previous DWI project
No. UV plants
Volume treated (Ml/d)
UV treatment by function1 (Ml/d)
(No. sites in brackets) UV treatment by source
2 (Ml/d)
(No. sites in brackets)
D C M GW LSW USW
139 1,492 1,084
(69)
996.4
(42)
39.8
(3)
613.6
(73)
896.3
(16)
28.7
(2)
Notes: 1. D = General Disinfection; C = Cryptosporidium risk; M = Micropollutants (e.g. pesticide, with UV included in an Advanced Oxidation Process
(AOP)). 2. GW = Ground Water; LSW = Lowland Surface Water; USW = Upland Surface Water.
Some of the larger companies using little or no UV are not included in the survey, so overall
UV use is likely to be less than 10% of UK output.
Of these, information was provided on 57 plants to allow the more detailed analysis given
below. This excludes the 3 AOP plants.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
9
Table 2.2 Public supplies: Summary of installations by works size and water type
Size (Ml/d) GW LSW GW+LSW USW Total
<10 23 0 0 0 23
10-19 12 0 1 2 15
20-39 6 2 1 0 9
40-59 0 3 0 0 3
60-79 0 4 0 0 4
80-99 1 1 0 0 2
100-149 0 0 0 0 0
150-200 0 1 0 0 1
All 42 11 2 2 57
Twenty-seven of these plants were installed prior to 2011, 9 are primarily for Cryptosporidium,
32 for general disinfection and the remainder for a combination of both.
Information on design dose, lamp type and supplier, summarised in Tables 2.3 to 2.5, was not
available for all of the 57 plants.
Table 2.3 Public supplies: Summary of installations by design dose
Design dose (mJ/cm2) Number of installations
25 1 (installed 2005)
40 16
45 1
48 10
60 1
>42 12
4 log removal of Cryptosporidium
(minimum 40 mJ/cm2)
1
Table 2.4 Public supplies: Summary of installations by lamp type
Lamp type Number of installations
LP 12
LPHO 1
MP 39
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
10
Table 2.5 Public supplies: Summary of installations by supplier
Supplier Number of installations
Wedeco 13
Trojan 15
Hanovia 4
ATG 2
Berson 11
Jabay 1
Xylem 3
The information on supplier and lamp type may not be fully up to date (Wedeco and Xylem
are now the same company), and may not be wholly representative of the water industry as a
whole. Further information is given directly from two UV suppliers later in this section.
In addition to this, further information was subsequently provided by one water company not
included in the original list. This company had 27 plants in the size ranges:
<10 Ml/d 23
10-19 Ml/d 2
20-39 Ml/d 1
40-49 Ml/d 1
All bar one were LPHO lamps, the other being MP.
Thirteen were treating upland water and the remainder treating borehole or spring water.
Twenty-one were Wedeco (or possibly Xylem) plants and 6 Trojan. All were installed primarily
for Cryptosporidium. Dose validation was by ÖNORM or DVGW for 10 plants, with a minimum
dose of 40 mJ/cm2, and the remainder were USEPA with a minimum dose of 17 mJ/cm
2.
2.3 Further design and operating information provided by water companies
A series of questions relating to UV design and operation was circulated to water company
contacts, and responses were received from 6 companies. The questions and responses are
summarised below; detailed responses are given in Appendix A.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
11
1. How is design dose established?
The European dose validation (ÖNORM or DVGW) is based on 40 mJ/cm2. Are higher doses
used to give a margin of safety, or because of higher microbial challenges from risk
assessments? Is target log removal taken into account (as for Crypto in USEPA dose
validation)?
A design dose of 40 mJ/cm2 is quoted by all water companies, with some reference to
ÖNORM and DVGW validation and use for “general disinfection. Two companies also quoted
lower doses (17 mJ/cm2 and 25 mJ/cm
2) for Cryptosporidium removal with USEPA validation.
One company described dose control based on UVT (USEPA validation) or UVI (ÖNORM/
DVGW validation).
2. Have situations arisen where numbers of units installed have limited flexibility and led to
higher doses than design at times of low flow?
Four water companies reported higher doses than design due to fluctuations in flow and/or
higher than design UVT values; this was not generally seen as a concern. Two companies
reported not experiencing higher than design doses, with one describing duty/assist/standby
arrangements if a large range in flow and/or UVT was expected.
3. Are flow rates and UVT controlled automatically to maintain the dose validation windows?
Three water companies reported automatic control of flow rate and UVT to maintain operation
within the dose validation window. Two water companies reported control of flow rate. Two
companies reported measurement of UVI to maintain dose. One company with UV operating
on stable good quality groundwaters based design dose on measured UVT which was then
monitored off-line as infrequently as monthly.
4. Is UVT commonly used as a feed-forward control parameter, or is control mainly based on
feedback from intensity monitors. Is this specific to UV plant suppliers?
Dose was controlled on both feed-forward control based on UVT and feedback from UVI
monitors; UVT was also monitored if not used for control. Different UV equipment suppliers
used different control systems.
5. Are intensity monitors recalibrated in accordance with manufacturer’s or dose validation
requirements? What is the typical frequency of recalibration?
In all cases UVI monitors were reported to be recalibrated according to manufacturer’s
recommendations/guidance, at intervals ranging from 1 to 12 months.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
12
6. Are lamps always replaced in accordance with manufacturer’s maximum hours-run
guidance? Is any allowance made for high frequency of stop/start which might shorten lamp-
life?
At five water companies lamps were replaced according to manufacturer’s
recommendations/guidance, with run times between 9,000 to 12,000 hours; one company
relied on UVI output to determine lamp replacement. Stop/start was not frequent and
generally there was no allowance for reduced lamp life.
7. What is the policy for routine cleaning of lamps? Is this based on time or can any
information from intensity/UVT be used to initiate cleaning? Are units taken off line for
cleaning, or can this be carried out while in operation?
At five water companies some degree of on-line automatic wiping of lamps was reported.
Automatic cleaning was backed-up by off-line manual cleaning in accordance with
manufacturer’s recommendations/guidance or UVI output.
8. Are there any other operational or maintenance issues?
Two water companies reported the formation of bromate where UV followed chlorination of
bromide-containing waters. One company reported difficulty in carrying out UVI monitor
calibration as replacement with the reference unit caused the system to shut down.
2.4 Information provided by UV companies
Additionally, a separate series of questions relating to UV design and operation was sent to
two of the main UV suppliers for the UK water industry. The questions and responses are
summarised in the tables below.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
13
1. Flow rates, unit/lamp type and configuration, and number of units.
UV Supplier 1 Flows from a few Ml/d to 120 Ml/d, average 20 Ml/d. Circa 45-50
municipal UV installations with United Utilities, Welsh Water, South
East Water, Cambridge and South Staffs, Bristol Water, Southern
Water and South West Water.
Most of our installations use LPHO lamps apart from a couple of
Medium Pressure lamp installations in South East Water (for
footprint reasons).
The lamps are generally installed parallel to the flow in U shape
and more recently L shape reactors.
The largest installation in UU uses the K reactor which has a 45°
angled lamp to the flow – low headloss.
UV Supplier 2 Delivered a number of UV units to water production facilities in the
UK, both private and municipal. These units treat flow rates ranging
from 5 m3/hr to over 2000 m
3/hr and include ten water plants using
medium pressure lamp-based units and over 50 using low
pressure, high output lamp-based units.
2. Which dose validation procedure was used, ÖNORM, DVGW or USEPA, and how was the
design dose decided? Was the UV primarily for general disinfection or Cryptosporidium?
UV Supplier 1 We have all three validation protocols installed depending on
specifications given by the consultant or end user. The larger sites
tend to be designed on USEPA, either for a minimum disinfection
dose of 40 mJ/cm2 MS2 RED or a required log crypto reduction.
The smaller units are often installed for ‘primary disinfection’ on
boreholes so would recommend ÖNORM or DVGW.
UV Supplier 2 Approximately 20% of the units were installed based on a USEPA
validation and the remaining units based on DVGW. All the units
with the exception of one or two were designed based on a UV
dose equal to or greater than 40 mJ/cm2. All DVGW units are
based on >40 mJ/cm2 dose.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
14
3. Are flow rates and UVT controlled automatically to maintain the dose validation windows?
UV Supplier 1 Yes, on USEPA systems. DVGW/ÖNORM don’t control based on
UVT but we recommend online UVT monitoring for DWI reporting.
All systems are flow paced to maintain the validation envelope.
UV Supplier 2 For systems utilizing USEPA “Calculated Dose” methodology,
online real-time UVT monitoring is recommended and to our
knowledge is being conducted.
4. Is UVT commonly used as a feed-forward control parameter, or is control mainly based on
feedback from intensity monitors?
UV Supplier 1 As per above.
UV Supplier 2 For systems utilising USEPA “Calculated Dose” methodology, UVT
and sensor intensity is utilized in real-time to determine the target
(calculated or theoretical) UV dose. Power level is adjusted to
meet the target UV dose. DVGW-based systems use intensity
monitoring as a primary control parameter (i.e. the intensity must
be above a given number at a given flow rate).
5. What cleaning systems are installed to prevent fouling problems?
UV supplier 1 Motorised chemical free wiping systems or CIP chemical if
required. Cleaning is not required on all water qualities. Our largest
installations in UU don’t have any wiping system.
UV Supplier 2 Medium pressure systems are equipped with a patented, chemical-
mechanical sleeve cleaning system. Low pressure, high output
lamp based systems utilize a mechanical sleeve cleaning system.
Both systems operate automatically at user-entered intervals. UV
Intensity sensor “windows” are also cleaned.
6. Was any form of mercury trap included?
UV Supplier 1 Not in our scope. A risk assessment is taken by the consultant to
assess the best mercury trap (if required) on the outlet pipework or
if a tank may be suitable to collect debris. There is a debate that
this may not be required for LP lamps and some papers were
published in the past considering the distance to the first user on
the distribution.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
15
UV Supplier 2 To our knowledge, mercury traps have not been typically installed.
Where specified by the end user, quite often this is provided for,
under the main contractor’s scope of supply, rather than the UV
Supplier.
2.5 Further comment from UV Supplier 2
There is on-going confusion regarding whether or not a target log reduction or a target UV
dose is required. For example, stating a requirement of 4-log crypto inactivation and a dose of
40 mJ/cm2 is not clear, as according to the USEPA, a validated dose of 22 mJ/cm
2 is required
for 4-log inactivation but the associated calculated/theoretical dose, will be in many cases
higher than 40 mJ/cm2 when accounting for the validation factor. Clarity on terminology is
required to improve communication.
To summarise the discussion points:
UV “costs” would be reduced, with better understanding of how the UV reactors
respond and perform, according to whatever is the validation protocol, to which they
have been certified.
Confusion is clearly evident from the interfaces of Regulatory body/End
customer/Consultant/Contractor/UV supplier, with each having their own (and often
slightly different take) on the interpretation of any regulatory requirements.
We have included the Regulatory Body here, as meaningful regulatory standards
(where they have been set) should not be open to wide interpretation1. Over recent
times, this most definitely has been improving, it is felt that this perhaps can now be
given more impetuous, as UV is seen more and more, as an effective primary
disinfection treatment process.
Significant additional UV operational savings can be made through better control of the
power requirements to maintain a UV reactor within its validated envelope.
All too often on operational UV sites, the UV reactors are simply “switched on” and
operated at full power, as an “added safety precaution”, in order to avoid possible
“insufficiently disinfected water” being sent into supply and so not incurring penalties
and fines from the regulator.
1 The regulations require that drinking water is adequately disinfected; DWI provide guidance only to
this end.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
16
As well as this additional power having little, or no impact, on increased disinfection, it
may have a more detrimental impact on nitrate/nitrites levels, depending on the UV
dose.
There are many other dynamics involved here, we have just tried to simplify the illustration, for
report purposes.
2.6 Summary
The majority of plants use MP or LPHO lamps.
Doses are mostly 40 mJ/cm2 or higher (apart from Water Company A where plants
aimed mainly at Cryptosporidium). There is not a clear consistency between the
USEPA crypto log inactivation approach and the ÖNORM/DVGW standard of
40 mJ/cm2.
Both USEPA and ÖNORM used roughly equally.
Type of control is generally consistent with dose validation requirements, with
calculated dose used for USEPA and intensity set point for ÖNORM/DVGW, although
there appear to be exceptions to this. UVT is sometimes used for shutdown in the event
of lower UVT than validated conditions, and often for regulatory reporting requirements.
Intensity monitor calibration carried out at monthly to 6-12 monthly intervals depending
on manufacturers’ recommendations. There is no indication that calibration is related to
dose validation protocol. Potential problem identified where only one monitor per unit
with taking off-line.
Lamp replacement mainly based on time and manufacturers’ recommendations.
Intensity used by one company (presumably with power monitoring).
Lamp cleaning frequency largely based on time, although intensity and power used by
one company.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
17
3. UV Technologies in Private Water Supply
3.1 Objective
To establish the range of UV technologies employed by private suppliers in England and
Wales, and establish the functionality and application.
Information on UV technologies for private supplies was collated from equipment
manufacturers/suppliers and from site visits made to private supplies.
3.2 UV equipment suppliers
A summary of UV suppliers and equipment used for small supplies is given in Table 3.1;
details of equipment are given in Appendix D.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
18
Table 3.1 Summary of UV systems available for small supplies
Supplier /
model(s) Lamp type
1
Flow rate
(m3/h)
Pretreatment Monitoring / control2 Maintenance
2 Other information
AquaCure
3 series
LP
0.48-3.063a
UVT: 98%
Lamp replacement
(4,320/8,760 hours)
Quartz sleeve
replacement
Economy SS
series
LP 0.12-2.763a
Plastic UV
steriliser
LP 0.183a
Aquafine
CSL series
LP4
9 / 115
UVT: 94/ 99%
Lamp status indicator
Lamp failure alarm
Lamp run time
UV intensity monitor
Water temp. monitor/
alarm
Lamp replacement
(8,000/9,000 hours)
Quartz sleeve
replacement (12 months)
Replace ballast as
required (not routine)
Optional auto-off at high
temperature (77°C)
eliminates on/off cycling
due to no flow
Optima series LPHO4
9 / 105 UVT: 94/ 99%
SP series LP4
0.226 UVT: 99%
SL series LP4 4.5 / 5.5
5 UVT: 94/ 99%
Bio-UV
UV home
series
2.2-3.23a
UVT: 98%
UV intensity monitor
Lamp replacement
(13,000 hours)
Replace UV intensity
monitor as required (not
routine)
UV home series includes
2 or 3 filters depending
on water quality: 60-µm
washable screen, 10-µm
cartridge, carbon filter IBP HO
series
4.6-543a
UVT: 99%
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
19
Supplier /
model(s) Lamp type
1
Flow rate
(m3/h)
Pretreatment Monitoring / control2 Maintenance
2 Other information
DaRo UV
Systems
Saphir
systems
LP
0.84-7.23a
1.08-9.63b
Prefiltration
Lamp status indicator
Lamp/electrical failure
alarm
Lamp life indicator
Hours run meter
UV intensity monitor
Remote monitoring of
lamp status (optional)
Lamp replacement
(8,000/8,760 hours)
Descale/clean quartz
sleeve
5-µm prefilter
recommended
Flow restrictor ensures
capacity cannot be
exceeded
Operating pressure: 10-
15 bar max
ECO series
LP
0.48-3.06
Hanovia
Pureline D
series
LP
7.3-893c
1.3-153d
UVT: >70%7
Power on LED
Unit tripped alarm
Lamp on/off
Lamp failure alarm
Total hours run
UV intensity monitor (%)
Low UV intensity alarm
Remote mode (flow
start/stop)
Lamp replacement
(12,000-16,000 hours)
Hydrotec
HydroPUR
series
LP 4.63e
2.93a
UV intensity monitor Lamp replacement
(8,000 hours)
Water temp: 5-50°C
Operating pressure:
10 bar max
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
20
Supplier /
model(s) Lamp type
1
Flow rate
(m3/h)
Pretreatment Monitoring / control2 Maintenance
2 Other information
LIFF
AQA Pure
LP
0.84-7.23a
1.08-9.63b
Prefiltration
UVT: 98% @
40 mJ/cm2
Lamp status indicator
Lamp life indicator
Lamp replacement
(8,760 hours/12 months)
5-µm cartridge filter
recommended
Water temp: 0-40°C
Operating pressure:
10 bar max
Flow restricted to
14-120 l/min
Prosep
SE series
1.32-2.70
Power indicator
Lamp on indicator
Lamp fail indicator/alarm
Hours run meter
Lamp replacement
(8,000-8,760 hours)
Operating pressure:
10 bar max
Solenoid valve (with
manual override) to shut
off water supply in event
of lamp or power failure
SS series
(single lamp)
LP 0.54-9.12
UVO3
Atlas series
0.7-4.83b
UVT: 95%
Lamp life indicator
Lamp failure alarm
Low UV alarm
Lamp change alarm
Lamp replacement
(9,000 hours)
Quartz replacement
Water temp: 2-40°C
Operating pressure:
10 bar max
Viqua
Sterilight
Sterilight
LP 2.5-8.93a
3.4-11.83b
UVT: >75%7
Fe <0.3 mg/l
Tot. hardness
Power on indicator
Lamp failure alarm
Lamp replacement
(9,000 hours/annually (or
biannually where use is
Water temp: 2-40°C
Available as ‘integrated
home system’ with 5-µm
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
21
Supplier /
model(s) Lamp type
1
Flow rate
(m3/h)
Pretreatment Monitoring / control2 Maintenance
2 Other information
Platinum
Range
<120 mg/l
Turb. <1 NTU
Mn <0.05
mg/l
Lamp life indicator
Lamp replacement alarm
Total run time indicator
UV intensity indicator (0-
99% with alarm at 50%)
seasonal))
Descale / clean quartz
sleeve
Replace controller, UV
monitor as required (not
routine)
cartridge filter
Low UV intensity can be
used to close inlet
solenoid valve
Cobalt series
LP
1.4-6.83a
1.8-9.13b
Sterilight
Silver series
LP 0.3-2.52a
0.4-3.42b
Viqua
UVMax
LP / LPHO
0.42-6.933b
UVT: >75%7
Fe <0.3 mg/l
Tot. hardness
<120 mg/l
Power supply indicator
Lamp operation indicator
Lamp age indicator
Lamp replacement
reminder
UV output sensor
Lamp replacement (9,000
hours)
Descale / clean quartz
sleeve
Water temp: 4-40°C
5-µm prefilter required
Solenoid valve flow shut-
off if UV dose insufficient
Wedeco
Aquada range
Altima range
Proxima
range
Maxima
range
LP
0.73-10.13a
0.98-13.43b
UVT: 80-98%
Lamp on/off indicator
Lamp life indicator
Audible & visual alarms
UV intensity indicator
Lamp replacement
(8,760 hours)
Descale / clean quartz
sleeve
Replace controller, UV
monitor as required (not
routine)
Water temp: 0-40°C
Operating pressure:
10 bar (max)
Optional automatic
solenoid safety shut-off
valve
Notes:
1. Lamp type: LP = low pressure; LPHO = low pressure high output.
2. Monitoring/control and maintenance will depend on model.
3. Flows quoted at the following UV doses: a) 40 mJ/cm2; b) 30 mJ/cm
2; c) 26 mJ/cm
2; d) 120 mJ/cm
2; e) 25 mJ/cm
2
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
22
4. Multiple lamps: CSL series – 4 lamps per unit; Optima series – 2 lamps per unit; SP series – 1 lamp per unit; SL series – 2 lamps per unit.
5. Flow at 30 mJ/cm2, 94 / 99% UVT.
6. Flow at 22 mJ/cm2, 99% UVT.
7. Flow rates quoted at 95% UVT.
8. Alarms may be audible or visual.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
23
3.3 Local Authority survey
DWI (Shaun Jones) sent a letter to all local authorities (LAs) in England and Wales at the
outset of this project to request permission for their contact details to be passes to WRc.
Contact details were provided to WRc for 28 LAs, listed below (including regions).
Babergh DC
(East of England)
Carlisle CiC
(North West)
North Norfolk DC
(East of England)
Swansea City & BC
(South Wales)
Bradford MDC
(Yorkshire &
Humberside)
Denbighshire CoC
(North Wales)
NW Leicestershire DC
(East Midlands)
Tameside MBC
(North West)
Braintree DC
(East of England)
Eden DC
(North West)
Northumberland CoC
(North East)
Taunton Deane BC
(South West)
Broadland DC
(East of England)
Gedling BC
(East Midlands)
Selby DC
(Yorkshire &
Humberside)
Vale of Glamorgan
Council
(South Wales)
Bromley
(Greater London)
Herefordshire
(West Midlands)
South
Buckinghamshire DC
(South East)
Waverley BC
(South East)
Calderdale MBC
(Yorkshire &
Humberside)
Lancaster CiC
(North West)
South Lakeland DC
(North West)
West Devon BC
(South West)
Cardiff Council
(South Wales)
Monmouthshire CoC
(South Wales)
Swale BC
(South East)
West Lancashire DC
(North West)
BC = Borough Council; CiC = City Council; CoC = County Council; DC = District Council;
MBC = Metropolitan Borough Council
It was proposed originally to visit 30 sites in three LAs (nominally 10 sites per LA), including
one LA in Wales. After discussion, it was agreed to visit 5-6 LAs (nominally 5-6 sites per LA),
including 1-2 in Wales. This would allow a greater geographical spread across England and
Wales, and may identify a greater range of installed plant and installation/operation/
maintenance procedures.
Initial contact was made with four LAs from this list. Responses varied with regard to
assistance offered, with concerns regarding the number of UV systems available for
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
24
inspection and staff resources required for assistance with visits. Subsequently the number of
LAs contacted was increased to include particularly LAs with the greatest number of UV
systems (as identified from previous work with DWI).
3.3.1 Site visits
Site visits were made to four LAs (identified in this report as LA ‘A’, LA ‘B’, LA ‘C’ and LA ‘D’).
Unfortunately, no LAs from Wales responded to requests to make visits.
During August and September 2015, visits were made to 25 premises, including single
domestic dwellings (SDDs), small domestic supplies and commercial supplies. The findings of
the visits are detailed in Appendix E and summarised below.
Installation and maintenance
Water treatment equipment, including UV, had been installed by local or regional specialist
companies or plumbers, and was generally maintained by the same or users. Standards of
mechanical and hydraulic installation were generally adequate, although some deficiencies at
boreholes were noted. Most users were aware of the basic maintenance requirements and
had service contracts in place, which included annual replacement of UV lamps and cleaning
of quartz sleeves.
Little, if any, manufacturers’ literature or operating/maintenance instructions had been
provided to users. Most users had limited knowledge of their treatment, including the function
of any units upstream of UV.
Maintenance logs were generally not kept, other than for the larger commercial supplies. Most
users kept copies of invoices that provided dates of maintenance and, to varying degrees, a
record of the work carried out. Spare filter cartridges and UV lamps were available at few
sites.
Most units, particularly serving the larger private supplies, were sited externally in purpose-
built enclosures, sheds or outbuildings. In one case, the UV enclosure was hidden behind
shrubbery that had to be pruned to allow access during the visit. Units for some SDDs were
located within the dwelling.
Pre-treatment
Pre-treatment depended on source water quality and size of the supply - larger commercial
private supplies tended to include more complex treatment.
Pre-treatment for SDDs and small supplies included particulate filters, nitrate filters and iron
(and possibly manganese) filters. Some of the larger commercial supplies also included ion
exchange softening, activated carbon, pH adjustment and chlorine dosing. The lack of
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
25
schematic diagrams, manufacturer’s literature and labelling often made it difficult to identify
the specific treatment.
Users were sometimes unaware of their treatment. Those users with service contracts in
place in particular were generally unaware of any maintenance requirements, such as the
replacement intervals for filter cartridges.
UV treatment
A range of UV equipment was installed, both branded and unbranded, but with little visible
information identifying design data such as maximum flow rate, operating pressure and
temperature.
Few installations included flow meters or monitoring and control; the larger commercial
supplies were generally better equipped. Some units included an indication of lamp life/days
operated, but at least one unit appeared to be reading correctly. Most units gave no indication
whether the UV lamp was functional; some users relied on observation of a ‘blue glow lamp’
to confirm operation of the lamp but did not necessarily understand that this was not an
indication of effective disinfection. With few exceptions, failure of the UV lamp would not
prevent flow (e.g. through activation of an automatic solenoid safety shut-off valve) and the
possibility of the consumption of non-disinfected water.
Post-treatment
There were few instances of any water treatment post UV. One supply incorporated activated
carbon and 2-µm particulate filtration post-UV, whilst another incorporated pH adjustment and
a second (older) UV system. UV-treated drinking water was supplied direct to taps and, at
some properties, to storage tanks. One user believed that cold water from storage was
supplied to the bathroom and used for brushing teeth and bathing.
General
Key findings arising from the visits:
There is a general lack of understanding amongst users regarding the treatment of their
private supplies. This is compounded by the lack of information provided by equipment
providers/installers.
There is no indication that UV equipment has been selected correctly for the flow (lack
of metering and control) or water quality (UVT, hardness, Fe). UVT measured >95% for
the majority of water samples (taken from before or after UV, including kitchen taps).
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
26
Some larger commercial private supplies incorporate more complex treatment systems.
Smaller private supplies and SDDs incorporate much simpler treatment, typically
particulate filtration and/or UV disinfection.
UV equipment is generally serviced by specialist companies, plumbers or the users,
with quartz sleeves cleaned at intervals between 2-12 months and lamps changed
around every 12 months; the frequency of maintenance of other equipment and
replacement of cartridge filters is less clear. Maintenance logs are not kept by users.
Monitoring and control of UV equipment is inadequate. Failure of a UV lamp may go
undetected for some time because a lack of a prominent alarm, and will generally not
prevent flow and the possibility of the consumption of non-disinfected water.
The potential for contamination of stored UV-treated water may not be well understood
by users.
There is currently no licensing or approved contractor scheme applicable to the
installation of equipment for private water supplies.
3.4 Survey of UV equipment installers
Subsequent to the site visits, contact was made with nine installers of UV equipment for
private supplies. An email asking for responses to 10 questions was sent to each of the
installers (see Appendix E).
Responses were received from two installers; these are summarised and included in Table
4.1.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
27
4. Critical differences between UV Technologies used in Public and Private Water Supplies
4.1 Objective
To establish critical differences in functionality and application between UV technologies used
in public and private water supplies in England and Wales.
4.2 Comparison of UV systems
A comparison of UV systems for public and private water supplies is given in Table 4.1, based
on UV systems for private supplies identified from the present study and in a study for the
Scottish Government (Scottish Government, 2015).
Table 4.1 Comparison of UV systems for public and private water supplies
Design and operating
factors Public supplies Private supplies
Size Smallest identified in survey
4.8 m3/d. Most are <10 Ml/d,
but many larger system in
operation, up to >100 Ml/d.
Mostly <10 m3/d, often much
smaller. Larger units usually
for commercial premises
rather than domestic.
Lamp types Mostly MP or LPHO Mostly LP
Design Usually based on detailed
feed water quality data.
Data on feed water quality
used to design system and
pretreatment, but insufficient
data will be available in many
situations. Heavy reliance on
pretreatment to deal with
poorer feed water quality.
One installer claimed that
units may be up-sized if poor
or variable water quality is
suspected.
Maintenance Lamp cleaning and
replacement carried out
routinely at supplier defined
intervals.
UV intensity monitors
Maintenance agreements
available with some
installers. Without these,
units are unlikely to be
maintained adequately in
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
28
Design and operating
factors Public supplies Private supplies
routinely calibrated. many situations, even where
suppliers provide guidance.
Some larger units may have
lamp hours run indicator and
a warning when replacement
is due.
Monitoring and control UV intensity monitors
standard for control. UVT
monitored and sometimes
used for control. Feed water
turbidity always monitored for
Regulation 26 requirements.
Control usually based on
maximum flow rate and
defined minimum UVT of the
water. Larger units may have
flow restrictors.
Unlikely to have UVI monitors
and definitely not UVT or
turbidity. May be UVI
monitors for larger units in
commercial premises, which
provide shut-down rather
than a control capability.
Dose 40 mJ/cm2 becoming widely
used where ÖNORM or
DVGW dose validation.
Others based on USEPA
crypto removal, sometimes
with lower dose. Dose
usually 40 mJ/cm2 or higher
for majority of units.
Variable between suppliers.
Some incorrectly define
30 mJ/cm2 as the “Industry
Standard”. Typically 30
mJ/cm2 for domestic and 40
mJ/cm2 for larger commercial
units.
Dose validation ÖNORM, DVGW or USEPA
standardised dose validation
becoming the norm. Some
older plants may not have
this, but likely to be replaced
over time.
Little dose validation by
biodosimetry to recognised
standard for most or all
systems. Some from larger
suppliers have microbial
challenge testing for generic
units, but this may not be
standardised dose validation.
Some have generic hydraulic
and UV intensity modelling
(Point Source Summation
modelling), again from larger
suppliers.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
29
Design and operating
factors Public supplies Private supplies
Feed water quality Turbidity always monitored
and controlled. UVT
monitored and sometimes
used in control (direct control
or shutdown if UVT falls
below dose validation limits).
Water quality standards will
help reduce fouling (e.g. from
Fe and Mn).
May be highly variable with
little control. Some units have
upstream filtration for
turbidity, but low UVT may
still occur for waters with
variable colour (not removed
by the usual filtration systems
used). Maintenance of
upstream treatment will often
be unreliable.
4.2.1 Comparison of UV mercury vapour lamps
Typical characteristics of the three types of UV mercury vapour lamp are summarised in Table
4.2.
LP lamps are currently the universal choice for small-scale UV systems intended for private
supply applications. Their principal advantage is that they have the highest energy efficiency
of the three types of mercury vapour lamp, but another important characteristic, given the
discontinuous usage pattern of the typical domestic water supply, is their relatively low
operating temperature; lamps which run hotter are more dependent on continuous water flow
to dissipate the heat generated.
LPHO lamps have heavy duty electrodes to allow operation at higher current and thus higher
output than LP lamps. They run hotter than LP lamps. Some contain a solid spot of mercury
amalgam on the lamp wall rather than free mercury2. The majority of public water supply UV
systems use LPHO lamps.
MP lamps have a much higher output than LP lamps but are less efficient in converting
electricity into germicidal UV. Fewer MP lamps are required for a given duty than LP or LPHO
lamps because of the higher output. They would normally only be considered for large-scale
public supply applications (Bolton and Cotton (2008) give an indicative minimum flow rate of
35 Ml/d), when the higher electricity cost can be offset against lower capital cost (smaller
plant, smaller building to house the plant) and lower maintenance cost (because of the lower
number of lamps). MP lamps are not used for private supplies, partly because of the low
electrical efficiency, but also because their much higher operating temperature and much
greater mercury content are characteristics which are not suitable for a domestic environment.
2 Bolton and Cotton (2008) consider amalgam lamps to be a distinct type, but note that some
manufacturers describe them as LPHO.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
30
Table 4.2 Typical characteristics of UV mercury vapour lamps (Bolton and Cotton,
2008)
Low pressure (LP) Low pressure, high
output (LPHO)
Medium pressure
(MP)
UV wavelength Monochromatic,
254 nm
Monochromatic,
254 nm
Polychromatic,
<200 nm to >400 nm
Mercury content (for
1.2 m lamp) ≈ 30 mg ≈ 30 – 75 mg ≈ 2 – 4 g
Operating lamp
temperature 30 – 50
oC 60 – 100
oC 600 – 900
oC
Input power 0.2 – 0.4 W/cm 0.6 – 1.2 W/cm 125 – 200 W/cm
Electrical to germicidal
UV conversion
efficiency
35 – 40 % 30 – 35 % 12 – 16 %
Lamp life 8,000 – 10,000 hr 8,000 – 12,000 hr 4,000 – 8,000 hr
4.2.2 UV LED technology
LED (light emitting diode) technology represents the most likely alternative to the mercury
vapour lamp in the future. LEDs are configurable, switchable and don’t require the warming
up period of mercury vapour lamps. They are safer to handle (no glass or mercury). However,
efficiencies are < 10%, lifetimes limited to c. 1,000 hr and production costs are high. Very
small-scale LED UV devices (e.g. for laboratory use) are available, but further development
will be necessary for larger-scale devices to be both technically and economically viable.
On the basis of development projections from a leading LED technology company, Chatterley
(2009) suggested cost equivalence between UV LP and UV LED for a point-of-use
disinfection application might be achieved by 2013, assuming LED output rising from 0.36 to
100 mW, lifetime from 1,000 to 10,000 hr and production cost falling from $664/mW to
$0.1/mW. While those projections proved optimistic – in 2015 the company produces UV
LEDs with an output of 10 mW, for example – the rate of progress does suggest viability at a
larger scale is achievable within the next decade.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
31
5. Comparison of Validation Criteria for different UV Systems
5.1 Objective
To compare current standards and guidelines applicable to UV systems used in drinking
water treatment, including validation criteria.
5.2 Current standards
Standards and guidelines applicable to potable water UV disinfection systems have been
published by:
US EPA
Austrian Standards Institute (ÖNORM)
DVGW Germany
National Water Research Institute/Water Research Foundation (NWRI/WRF)
British Standards Institute (BSI)
National Sanitation Foundation/American National Standards Institute (NSF/ANSI)
The standards are summarised in Table 5.1. The common objective is to provide independent
confirmation that a UV reactor achieves some specified level of performance within the range
of operating conditions defined by the supplier. All require dose validation by biodosimetry,
the principles of which are outlined in Appendix B.
Of the standards/guidelines listed above, four (US EPA, ÖNORM, DVGW and NWRI/WRF)
apply to public drinking water supply applications, either explicitly (US EPA, NWRI/WRF) or
by adoption by national regulators as required certification (ÖNORM, DVGW). US EPA have
adopted the concept of log removal credits, as applied in other US drinking water regulations,
and include tables of minimum dose necessary to ensure specified log removals of regulated
pathogens (primarily Cryptosporidium and Giardia but including a generic table for viruses
derived from Adenovirus sensitivity data); the UV system must then be validated against the
target log removal. The European standards, in contrast, stipulate that the UV reactor must be
validated for a dose of 40 mJ/cm2. NWRI/WRF provide design guidelines for both drinking
water and water reuse UV applications and describe a biodosimetry protocol suitable for
meeting the US EPA requirements.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
32
The BSI standard is the UK implementation of a European standard for LP UV devices
intended for water conditioning in buildings; the UV device being fitted either at the point of
entry of the mains supply into the building, or within the water distribution system inside the
building. It further defines devices intended for disinfection (‘killing or inactivating all types of
pathogenic bacteria to (…) at least 99.999% and all types of pathogenic viruses to (…) at
least 99.99%’) or bactericidal treatment (‘inactivating or killing bacteria present in water to an
unspecified degree’). Disinfection devices must be fitted with a UVI sensor linked to
alarm(s) and flow shut-off if measured intensity is too low, and the UVI sensor replaced
every year. A UVI sensor is not required for bactericidal treatment devices. The
biodosimetry protocol described in this standard is adapted from the Austrian ÖNORM
standard and requires validation of a 40 mJ/cm2 dose irrespective of whether the device
is intended for disinfection or bactericidal treatment.
The NSF/ANSI standard applies to point-of-entry and point-of-use UV equipment. The
standard defines two distinct classes of UV system: Class A, designed to inactivate ‘bacteria,
viruses, Cryptosporidium oocysts and Giardia cysts’ in water that is ‘not colored, cloudy, or
turbid’; and Class B, ‘designed for supplemental bactericidal treatment of disinfected public
drinking water or other drinking water that has been (…) deemed acceptable for human
consumption’. Class A systems are required to demonstrate a dose of 40 mJ/cm2; Class B
systems, 16 mJ/cm2. Class A systems must be fitted with a UVI sensor linked to alarm(s)
and/or water shut-off in the event of measured intensity falling below the minimum; testing the
alarm/shut-off functionality is part of the standard. Class B systems do not require a UVI
sensor, but if fitted must be tested as per Class A systems. This standard also extends to, and
specifies test procedures for, materials of construction and pressure integrity of the UV
system.
Acceptance of US EPA validation in European countries that have not developed their
own standard varies. French and Swiss regulations only recognise ÖNORM or DVGW
validation (Pilmis and Baig, 2009; Bucheli, 2009). Norwegian regulations accept US EPA,
ÖNORM or DVGW (Lund, 2009). Dutch regulations have no specific legal requirement for
validation, but require each installation to be approved by the national inspectorate;
biodosimetry will almost certainly be needed as part of the approval process.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
33
Table 5.1 Summary of standards and guidelines
Title Reference Dose validation test
Ultraviolet Disinfection Guidance Manual for the Final Long Term 2 Enhanced Surface Water
Treatment Rule (UVDGM)
EPA 815-R-06-007
November 2006
Biodosimetry
Plants for the disinfection of water using ultraviolet radiation – Requirements and testing – Low
pressure mercury lamp plants
M 5873-1
Austria ÖNORM
(March 2001)
Validated dose of 40 mJ/cm2 at
253.7 nm. Dose validation tests using
B subtilis spores.
Plants for the disinfection of water using ultraviolet radiation – Requirements and testing – Part
2: Medium pressure mercury lamp plants
M 5873-2
Austria ÖNORM
(August 2003)
As above
UV-Geräte zur Desinfektion in der Wasserversorgung – Teil 1: Anforderungen an die
Beschaffenheit, Funktion und Betrieb
[UV-devices for the disinfection of the water supply – Part 1: Requirements on the design,
function and action]
W 294-1 Germany
DVGW / DIN
(June 2006)
Not available in English.
Similar to Austrian standard in terms
of dose and use of B subtilis spores.
UV-Geräte zur Desinfektion in der Wasserversorgung; Teil 2: Prüfung von Beschaffenheit,
Funktion und Desinfektionswirksamkeit
[UV-devices for the disinfection of the water supply- Part 2: Tests of design, function and
disinfection effectiveness]
W 294-2 Germany
DVGW / DIN
(June 2006)
As above
UV-Geräte zur Desinfektion in der Wasserversorgung; Teil 3: Messfenster und Sensoren zur
radiometrischen. Überwachung von UV-Desinfektionsgeräten;
Anforderungen, Prüfung und Kalibrierung
[UV-devices for the disinfection of the water supply; Part 3: Sensors for the photometric
monitoring of UV-Disinfection; tests and calibration]
W 294-3 Germany
DVGW / DIN
(June 2006)
NA
UV Disinfection Guidelines for Drinking Water and Water Reuse, 3rd
Edition NWRI/WRF 2012 Challenge test using MS2
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
34
Title Reference Dose validation test
Water conditioning equipment inside buildings – Devices using mercury low-pressure ultraviolet
radiators – Requirements for performance, safety and testing
BS EN
14897:2006+A1:2007
European
(June 2007)
Similar to Austrian standard
Ultraviolet microbiological water treatment systems NSF/ANSI 55 – 2012
USA
Challenge test using MS2 or
Saccharomyces cerevisiae,
depending on type of device (T1
Coliphage was introduced as an
alternative to S. cerevisiae in 2012,
with the intention that S. cerevisiae
will be removed from the standard in
September 2017)
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
35
5.2.1 USEPA UV Disinfection Guidance Manual (UVDGM)
The UVDGM provides comprehensive guidance on the use of UV for water treatment. It
contains information applicable to users, equipment suppliers, and regulators. It is not a
statutory document, and US water utilities are not obligated to follow its recommendations for
good practice. Although written in the context of US water quality regulations, with particular
reference to the Long Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR), the
manual is essentially a good practice guide and, as such, its relevance is not restricted to the
US. The authority for the LT2ESWTR is derived from the Safe Drinking Water Act (SDWA) as
amended in 1996, which applies to public water systems defined as those serving at least 25
people.
The manual is arranged in six sections, the first of which is an introduction and summary of
the pertinent US water treatment regulations. The second section is an overview of UV
disinfection, including descriptions of microbial response to UV and of the components of UV
systems; and a discussion of other water quality effects and by-product formation. The
remaining sections consider the steps required to implement UV disinfection, from initial
planning and design through to operation and validation. Detailed supporting information,
case studies and a discussion of lamp break issues are appended.
The implementation sections are outlined below.
Section 3: Planning analyses for UV facilities
This section discusses what should be considered at the planning stage:
defining UV disinfection goals;
where to incorporate UV into a treatment train;
defining design parameters;
the characteristics of different types of UV lamp;
control strategies;
validation issues;
headloss constraints;
estimating footprint (in terms of what equipment to allow for);
estimating costs (in terms of what equipment to allow for).
Section 4: Design considerations for UV facilities
This section discusses the key factors that should be considered when undertaking detailed
design:
hydraulics;
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
36
operating approach;
instrumentation and control;
electric power supply;
layout;
specifications for equipment.
Section 5: Validation of UV reactors
This section, together with supporting appendices, describes in detail the UVDGM’s
recommended biodosimetry validation protocol:
minimum requirements for validation;
selection of challenge micro-organisms;
equipment requirements;
determining test conditions;
test methodology;
analysis of results;
reporting;
evaluating the need for re-validation.
The rationale behind the protocol is given. Quality assurance and quality control are
discussed.
Section 6: Start-up and operation of UV facilities
This section discusses commissioning and operation of UV plants:
commissioning;
operation;
maintenance;
monitoring and recording operating data;
staffing, training, safety.
5.2.2 Austrian standard ÖNORM 5873; Parts 1-2
ÖNORM 5873-1 ‘Plants for disinfection of water using ultraviolet radiation – Requirements
and testing: Low pressure mercury lamp plants (1/3/2001)’.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
37
ÖNORM 5873-2 ‘Plants for disinfection of water using ultraviolet radiation – Requirements
and testing – Part 2: Medium pressure mercury lamp plants (1/8/2003)’.
Scope
The ÖNORM standards set out the requirements for the design, testing, operation and
monitoring of UV systems for the treatment of drinking water. The standards include a
comprehensive definition of all of the technical terms used.
ÖNORM 5873-2 is derived from, and has much in common with, ÖNORM 5873-1, but does
include some important differences that reflect its application to medium pressure UV
systems.
Requirements
The standards require that a ‘Reduction Equivalent Fluence’ (REF) of 400 J/m2 (40 mJ/cm
2) is
delivered, relative to a wavelength of 253.7 nm, at a given flow rate and water quality (UV
transmittance). It is stated that this dose is sufficient to achieve a 6 log reduction of health
related water transmittable bacteria and a 4 log reduction of health related water transmittable
viruses ‘according to the state of the art’.
The water to be treated by UV must conform to the physical and chemical aspects of the EU
Drinking Water Directive, which has implications for the positioning of the UV system.
The standards set out requirements for:
the irradiation chamber;
monitoring;
control.
Type tests
The standards describe type tests to be used to independently verify that UV systems achieve
the performance claimed by the manufacturer (the operating conditions – UVT and flow rate -
which enable a Reduction Equivalent Fluence (REF) of 40 mJ/cm2). Tests can be performed
off-site or on-site. In the former case, the results are accepted for the particular system being
tested; results from an on-site test apply only to that installation.
Type tests have five parts:
compliance against manufacturers specification (REF);
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
38
general characteristics (e.g. electrical current);
radiation monitoring performance;
microbiological challenge test (Biodosimeter);
evaluation of the admissible operating conditions.
To allow for ageing, lamp output is adjusted to that expected at the end of guaranteed lamp
life. For off-site tests, the UV system inlet is fitted with a 90° bend to simulate a compromised
hydraulic installation.
The standards specify biodosimetry using Bacillus subtilis spores. A dose response curve
must be determined for each batch of spores, the UV sensitivity of which must lie within
stipulated limits. Protocols are given for determining the limiting operating conditions (flow
rate, UVT) at which the required REF of 40 mJ/cm2 is achieved, which can then be compared
against the manufacturer’s claims.
Operational Requirements
The standards require that operators of UV systems keep to servicing schedules set out by
the manufacturers, and keep appropriate records of operational and service actions.
Testing of a Production Series
ÖNORM 5873-1 sets out conditions under which a range of equipment of essentially the
same design but scaled for a different flow rates, referred to as a ‘Production Series’ can be
subjected to a reduced series of tests.
5.2.3 German standard DVGW W294 Parts 1-3
W294-1 UV-devices for the disinfection of the water supply - Part 1: Requirements on
the design, function and action (June 2006).
W294-2 UV-devices for the disinfection of the water supply - Part 2: Tests of design,
function and disinfection effectiveness (June 2006).
W294-3 UV-devices for the disinfection of the water supply - Part 3: Sensors for the
photometric monitoring of UV-Disinfection; tests and calibration (June 2006).
Scope
The 2006 German standards are not yet available as an English translation, however it is
understood that they are similar in concept to the Austrian standards, requiring:
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
39
validation of a dose of 400 J/m2 (40 mJ/cm
2);
validation by biodosimetry using Bacillus subtilis spores.
5.2.4 BS EN 14897:2006+A1:2007
BS EN 14987 is entitled:
Water conditioning equipment inside buildings – Devices using mercury low-pressure
ultraviolet radiators – Requirements for performance, safety and testing.
This British Standard is published by BSI as the UK implementation of EN 14987:2006, a
European Standard approved by CEN in 2006 and amended in 2007. The British Standard
reproduces the European Standard without alteration.
Scope
The scope of this standard is (emphasis added):
This document specifies definition, principles of construction, requirements and
methods for testing the performance of UV devices for drinking water installations
inside buildings which are permanently connected to the mains supply at the point of
entry into a building or within the water distribution system inside the building.
UV devices in the sense of this standard are UV bactericidal treatment devices or UV
disinfection devices.
The standard defines disinfection as ‘the killing of inactivating of all types of pathogenic
bacteria to a specified degree of at least 99.999% and all types of pathogenic viruses to a
degree of at least 99.99%’. Bactericidal treatment is defined as the ‘action of inactivating or
killing bacteria present in water to an unspecified degree’.
This standard, therefore, explicitly applies to privately installed UV devices which are treating
water supplied directly or indirectly from potable water mains, and as such excludes UV
devices treating water from private supplies.
Requirements
This standard draws heavily from ÖNORM M5973-1. Many of the definitions, parts of the text,
and some of the diagrams are essentially reproduced directly (which, prior to amendment in
2007, included an arithmetic error present in one of the ÖNORM tables).
The standard, in common with ÖNORM M5973-1, requires it to be demonstrated by
biodosimetry that the UV device applies a dose of 40 mJ/cm2 (400 J/m
2) at a wavelength of
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
40
254 nm over the defined operational range (flow rate, UVT) at the end of lamp service life.
This applies to both UV disinfection and UV bactericidal treatment devices.
The standard requires a UV disinfection device to display irradiance and sets out
requirements for the UVI sensor (Annex A) and guidance for the monitoring window (Annex
C), which largely reproduce equivalent requirements in ÖNORM M5973-1. The standard
defers to national regulations for sensor and window where such apply. The stability of the
sensor ‘shall be assured for at least one year’, after which it ‘shall be replaced by a new one’.
In addition to the provision of alarms, the standard requires that a ‘signal shall be provided’
which ‘allows the waterflow to be stopped’ when operation falls outside the validated limits,
when the device is shut down or in the event of failure of the power supply. Since the
validated limits include flow rate, the fitting of a flow meter is implied.
ÖNORM M5973-1 specifies Bacillus subtilis as the challenge micro-organism for the
biodosimetry, and also specifies the acceptable range of sensitivity (log inactivation v UV
dose) within which the calibration of the biodosimeter must fall. The only explicit reference to
B subtilis in the BSI standard is as an example challenge micro-organism in the list of
definitions, but it also specifies the acceptable range of sensitivity of the biodosimeter in terms
which are similar, but not identical, to those given in ÖNORM M5973-1. This effectively makes
the use of B subtilis an implicit requirement of the standard.
Testing
ÖNORM M5973-1 describes two test procedures, one for UV devices with a UVT sensor and
one for UV devices without a UVT sensor. Only the latter procedure is included in the BSI
standard, applicable for UV disinfection devices. The standard additionally includes a
simplified procedure for UV bactericidal treatment devices.
For UV disinfection devices the manufacturer supplies a table of flow rates and corresponding
UVT values (minimum flow rate/minimum UVT; maximum flow rate/maximum UVT; and
intermediate values) for which the device is expected to achieve the required performance.
The test procedure then derives the relationship between measured irradiance and UVT and
by biodosimetry determines whether the device achieves the required performance at flow
rates from minimum to maximum and corresponding irradiances. Ultimately a table of
permissible operating range is prepared listing the minimum irradiance for a given flow rate. In
operation the device must alarm if irradiance drops below the minimum for the pertaining flow
rate.
For UV bactericidal treatment devices the procedure omits the step deriving the relationship
between irradiance and UVT, since these devices are not required to be fitted with a UVI
sensor. The final table lists maximum flow rate for given UVT.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
41
One notable deviation from the ÖNORM M5973-1 procedures is that ÖNORM require a
correction factor to be applied to reported flow rates to allow for uncertainty in the UVI sensor
readings. This requirement is omitted from the BSI standard.
The standard stipulates the information to be provided by the manufacturer to the test
institute, and the performance information that must be provided to the user by the
manufacturer. Finally, it lists the installation, operation and maintenance information that must
be provided to the user by the supplier. The standard applies to water supplies drawn from
mains-treated water, but does not impose any explicit requirements on water quality. It does,
however, state that minimum UVT and maximum turbidity of the water to be treated have to
be taken into account when specifying a UV device, and it also identifies iron, manganese and
humic acids as examples of substances which impact UVT.
5.2.5 US NSF/ANSI Standard 55 – 2012
This standard is entitled:
Ultraviolet microbiological water treatment systems.
NSF International (NSF) is an independent body which supplies public health and safety-
based risk management solutions. The standards it publishes are intended to promote
sanitation and protection of public health. The American National Standards Institute (ANSI)
oversees the development and application of, and gives accreditation to, voluntary consensus
standards of products and services in the US; it does not itself develop standards.
Scope
This standard applies to point of entry and point of use UV equipment installed in single
private residences. Its purpose is to establish minimum requirements for the reduction of
micro-organisms using UV. It distinguishes between Class A systems, which are intended for
the inactivation of pathogenic micro-organisms, and Class B systems, which are intended only
for ‘supplemental bactericidal treatment of public or other drinking water that has been
deemed acceptable by a local health authority’.
Its scope also encompasses materials of construction, integrity (under pressure), product
literature, equipment labelling and service obligations of manufacturers.
Requirements
Class A systems must deliver a dose of 40 mJ/cm2 at a defined minimum UV transmittance
and must be fitted with a UV sensor that will trigger an alarm if an insufficient dose is being
applied. The alarm can comprise one or more of: visual warning, audible warning, automatic
shut-off of flow. Class B systems must deliver a dose of 16 mJ/cm2.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
42
The standard requires that a flow-limiting device be fitted that prevents the flow rate
exceeding the maximum specified for the system over the specified operating pressure range
of the unit. The system must be provided with a visual means to verify electrical operation of
each lamp.
Testing
Performance must be validated using biodosimetry in accordance with a proscribed protocol,
using either MS2 phage (Class A systems) or T1 Coliphage or Saccharomyces cerevisiae
(Class B systems)3. Collimated beam tests are required to determine the dose response
curve of each batch of challenge micro-organisms.
The protocol requires parallel testing of 2 UV units over 7 days. Flow rate must equal the
maximum allowed by the integral flow-limiting device. The quality of the test water is specified,
including a minimum UV transmittance of 96%. For Class A systems, the transmittance must
then be reduced using parahydroxybenzoic acid (PHBA) to 50% or until the alarm point is
reached, whichever results in the lower transmittance, and kept at this value for the duration
of the test4.
Samples must be taken during periods of steady-state operation and immediately on start-up
after overnight stagnation periods. The calculated log reduction is derived from the geometric
mean of all influent sample counts and the geometric mean of all effluent sample counts, and
must be equal to or greater than the log reduction at 40 mJ/cm2 read from the dose response
curve.
Class A systems validated in accordance with this standard can claim effective inactivation
specifically of Cryptosporidium oocysts and Giardia cysts. They cannot claim wider
effectiveness against cysts in general unless preceded by another treatment stage for
removal or inactivation of cysts that complies with the appropriate NSF/ANSI standard, nor
can they make claims of reduction of the challenge micro-organism. Class B systems can only
claim effectiveness for non-pathogenic, nuisance micro-organisms.
5.2.6 NWRI/WRF guidelines
The US National Water Research Institute (NWRI), in collaboration with the Water Research
Foundation (WRF) has produced UV disinfection guidelines for drinking water and water
reuse (NWRI, 2012). They are based largely on procedures adopted by the California
Department of Public Health for the review and approval of UV disinfection systems and
implicitly apply to large-scale installations.
3 The option of using T1 Coliphage for Class B system validation was introduced in 2012, with the
stated intention to eliminate the use of Saccharomyces cerevisiae after September 2017. 4 This requirement implies the expectation of a minimum design UVT ≤ 50% for all systems, since it
would preclude systems with a flow shut-off alarm having a minimum design UVT > 50%.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
43
Scope
These provide an overview of UV system design and operation, with outline guidance,
together with protocols for dose validation tests. They apply to LP and MP UV systems.
Requirements
These guidelines do not quantify specific pathogen inactivation or UV dose requirements,
leaving it to appropriate regulatory agencies to determine these on a case-by-case basis.
Performance standards are specified for UVI sensors.
Testing
Validation by biodosimetry is required. MS2 phage is the recommended challenge micro-
organism where expected dose > 20 mJ/cm2, and bounds are defined within which the
calibration sensitivity curve for MS2 phage must fall. The guidelines do not, however, preclude
alternative challenge micro-organisms.
To simulate lamp ageing, the output must be reduced to 50% unless some other value can be
demonstrated by the manufacturer as representative of lamps at the end of their specified
service life.
5.3 Comparison of standards
A distinction can be made between those standards applicable to public water supplies
(ÖNORM, DVGW, UVDGM) and those applicable for private installations5 (BSI, NSF/ANSI).
The comparisons made in this section are therefore within these sub-groupings. The
NWRI/WRF guidelines are not considered further, as they do not themselves quantify
validation requirements. Some additional discussion of the implications of the differences
between standards is given in Appendix B.
5.3.1 Comparison of ÖNORM standard and UVDGM guidelines
A comparison of key elements of the UVDGM and ÖNORM validation methodologies is given
in Table 5.2 (the DVGW and ÖNORM standards being equivalent).
The UVDGM and European approaches are both designed to demonstrate that a UV reactor
will achieve a specified performance under given operating conditions. But in comparing the
two, it should be recognised that they have fundamental differences.
5 Note the distinction between private installation and private supply (also see p. 36). The BSI
standard applies to private installations treating water from the public supply, whereas the NSF/ANSI
standard applies to private installations treating water from a private supply.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
44
The UVDGM approach is concerned with validating UV for some specified log inactivation of a
given pathogen, which follows the established US EPA methodology of assigning log
inactivation credits to treatment processes. The lower the target inactivation, the smaller the
UV plant will be, since the required Validated Dose will be smaller.
The European approach is concerned with UV as the primary disinfection treatment stage. A
target REF (or RED) of 40 mJ/cm2 is stipulated, justified on the grounds that such a dose is
sufficiently high for adequate inactivation of health-related bacteria (6 log) and viruses (4 log)
according to current knowledge.
Both approaches require a UVI sensor, and allow for the optional inclusion of a UVT sensor.
Both approaches recognise that hydraulics impact performance. UVDGM provides three
options for ensuring that for off-site testing the inlet and outlet pipe arrangements are such
that hydraulic conditions are not better in the test rig than in the on-site installation. ÖNORM
requires that an off-site test stand includes a 90o elbow upstream of the UV device to induce
unfavourable hydraulic conditions.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
45
Table 5.2 Comparison between UVDGM and ÖNORM validation methodologies
UVDGM ÖNORM
Validation method Biodosimetry Biodosimetry
Target dose Depends upon target pathogen and log removal credit, for which values
of target dose are tabulated.
40 mJ/cm2
Challenge micro-
organism
Not specified. Bacillus subtilis ATCC 663 spores, with stipulated
bounds within which dose-response curve must lie.
UV intensity sensors Recommends that the reading of each plant sensor should differ by no
more than 10% from the mean reading of two or more recently calibrated
reference sensors, in the same sensor port with the same lamp, lamp
power and UVT. However, the methodology allows for a greater
uncertainty provided it is incorporated into the Validation Factor.
Stipulates that the uncertainty in plant sensor reading
shall be taken as 15% unless a higher value is
demonstrated. Specifications for measuring range and
resolution of plant sensors are given.
Lamp ageing Lamp output must be that expected at the end of the lamp utilisation
period. Simple turn-down is acceptable if either the manufacturer
confirms that this approach is adequate, or if tests demonstrate that
lamp ageing is uniform. If there is evidence of non-uniform lamp ageing,
then used lamps that have been operated under similar conditions
should be fitted for the validation tests.
Tests shall be conducted with new lamps that have in
service for ‘about 100 hours’. Lamp output must be
lowered to the value at the end of the lamp utilisation
period. The manufacturer must specify how output is to
be lowered (the fitting of mesh screens, or substitution of
an alternative ballast, are permitted), and by how much.
An illustrative figure of 30% is given only as an example.
Applicability of
validation
A recommended Validation Report structure and checklist are provided.
The Validated Dose, log removal credit achieved, validated operating
conditions, and validation test operating conditions (including flow rate,
UVT and lamp power) must be included.
The maximum flow, minimum UVT and minimum
reference irradiance as determined by the validation test
must be stated on identification plates attached to the
UV reactor. The operating range of the plant in terms of
these three parameters must be provided in graphical,
analytical and table form.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
46
5.3.2 Comparison of BSI and NSF/ANSI standards
A comparison of key elements of the BSI and NSF/ANSI standards is given in Table 5.3.
Significant differences between these standards to note include:
1. The BSI standard applies to UV devices treating water that has come from the public
water supply. The NSF/ANSI standard does not impose this restriction for Class A
devices, requiring instead that the water source be (qualitatively) clear and free of
obvious contamination.
2. Validation to the BSI standard produces a performance curve. Validation to the
NSF/ANSI standard produces a single point.
3. The BSI standard requires bactericidal treatment devices to achieve the same dose
(40 mJ/cm2) as disinfection devices. The NSF/ANSI standard requires a lower dose,
16 mJ/cm2, for Class B systems (supplemental bactericidal treatment) than for Class A
systems (40 mJ/cm2).
The NSF/ANSI standard also extends to materials of construction and structural performance
(integrity under pressure), aspects which are beyond the scope of the BSI standard.
Both standards require the UV device to be installed in the test rig in accordance with
supplier’s instructions. NSF/ANSI impose the additional stipulation that the inlet and outlet
pipe diameters are not smaller than the diameters of the respective connections.
5.4 Conclusions
Joint conclusions (with Section 6) are presented in Section 6.5.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
47
Table 5.3 Comparison between BSI and NSF/ANSI standards
BSI NSF/ANSI
Scope LP UV devices within buildings connected to mains supply,
located at point of entry to building or in the water distribution
system within the building.
LP UV devices for point-of-use and point-of-entry
applications.
Application Disinfection devices: those intended to kill or inactivate all types of
pathogenic bacteria by at least 5-log, and all types of pathogenic
viruses by at least 4-log.
Bactericidal treatment devices: those intended to kill or inactivate
bacteria to an unspecified degree.
Class A systems: those intended to kill or inactivate
micro-organisms, including bacteria, viruses,
Cryptosporidium oocysts and Giardia cysts.
Class B systems: those intended for supplemental
bactericidal treatment (reducing normally occurring non-
pathogenic nuisance micro-organisms) of disinfected
public drinking water or other drinking water deemed
acceptable for human consumption.
Water source Mains water (directly, for point-of-entry devices; indirectly for
devices located in the water distribution system within the
building).
Class A systems: Water must be ‘visually clear (not
coloured, cloudy or turbid)’ and free of ‘obvious
contamination’. Excludes devices intended to convert
wastewater to drinking water.
Class B systems: public water supply or other source
deemed acceptable for human consumption.
Target dose 40 mJ/cm2 Class A systems: 40 mJ/cm
2
Class B systems: 16 mJ/cm2
Validation method Biodosimetry Biodosimetry
Challenge micro-organism Bacillus subtilis spores, with stipulated bounds within which dose-
response curve must lie.
Class A systems: MS2 phage
Class B systems: T1 Coliphage or (until September
2017) Saccharomyces cerevisiae
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
48
BSI NSF/ANSI
UV intensity sensors Required for disinfection devices. Defers to national standards for
specification.
Class A systems: Required, must be linked to alarm
and/or automatic flow shut-off. Operation is tested as
part of the standard – the alarm must activate 100
consecutive times in response to reduction in UVT.
Class B systems: Not required, but if fitted must be
tested as per Class A systems.
Lamp ageing Tests to be performed with lamps that have been in service for
100 h. Lamp output must be adjusted to that expected at the end
of the lamp service life, an output of 70% is given only as an
example. Manufacturer must provide an appropriate method to
adjust the output.
Tests to be performed with lamps that have been in
service for 100 h. If UVI sensor isn’t fitted, lamp must be
turned down to 70% output.
Applicability of validation Tables of minimum irradiance v corresponding flow rate
(disinfection devices) or maximum flow rate v corresponding UVT
(bactericidal treatment devices) must be provided. These are
derived by testing the devices at the limiting points and at least
one intermediate point, and fitting a curve to the data points. For
disinfection devices, the maximum flow rate and corresponding
irradiance must be stated on the identification plate on the device.
For bactericidal treatment devices, the maximum flow rate and
corresponding UVT must be stated on the identification plate.
The test flow rate through the system is determined by
varying the inlet pressure at intervals up to the system’s
stated maximum working pressure; the maximum flow
rate so measured is the test flow rate.
Class A systems & Class B systems with UVI sensor:
The flow rate at which the applicable target dose is
achieved when UVT has been adjusted to the alarm
trigger point is the rated service flow.
Class B systems without UVI sensor: The flow rate at
which the applicable target dose is achieved with lamp
output at 70% is the rated service flow. Parameters must
be provided in graphical, analytical and table form.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
49
6. Review Standards for UV Systems and Identify Validation Criteria suitable for Private Supply
6.1 Objective
To review standards to identify validation criteria for UV systems suitable for private supplies.
6.2 Implications of water quality regulations
The statutory regulations applicable to private water supplies in England (The Private Water
Supplies Regulations 2009) and Wales (The Private Water Supplies (Wales) Regulations
2010, as amended) include in the definition of wholesomeness prescribed concentrations or
values (PCVs), some of which may be of relevance to UV disinfection (Table 6.1).
Table 6.1 PCVs for private water supplies of potential relevance to UV disinfection
Parameter PCV Unit
Turbidity 4 NTU
Colour 20 mg/l Pt/Co
Iron 200 g/l
Manganese 50 g/l
Turbidity
Turbidity is an indicator of the presence of particulate matter. Particulates can affect the
performance of UV reactors by sheltering pathogens from UV radiation and scattering UV
light. Some particulates might also absorb UV.
USEPA (2006) states that the effect of harbouring micro-organisms is not significant at
turbidity of up to 10 NTU. However, one reference given for this (Passantino et al., 2004) was
based a laboratory study using spiked MS2 phage, with turbidity increased by the addition of
clay. This would not necessarily simulate the nature of shielding that could occur in natural
waters, for example where micro-organisms have become enmeshed in, or coated in, some
inert material. Other studies have produced similar findings, but most have the same
limitations. One study (Amoah et al., 2005) used natural turbidity in lake water spiked with
Cryptosporidium and Giardia. A reduction in Cryptosporidium and Giardia inactivation (using
mouse infectivity) of up to 0.8 log and 0.4 log respectively was identified over the turbidity
range 0.3 to 20 NTU, when correction was made for the UVT of the water. However, the effect
was barely discernible below 10 NTU.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
50
The limited literature outlined above suggests that 4 NTU would not be expected to
compromise disinfection performance. Nevertheless, German regulations relating to the use
of UV disinfection give a guideline of ≤ 0.3 NTU (Eggers, 2009). French regulations require
≤ 0.5 NTU (Pilmis and Baig, 2009). Swiss regulations require ≤ 1.0 NTU where there is no
pre-treatment, and ≤ 0.3 NTU after filtration (Bucheli, 2009). VIQUA, manufacturer of Sterilight
UV units designed for residential and small-scale commercial use, recommend that turbidity
be < 1 NTU. It is normal practice amongst suppliers of small-scale UV units to recommend
that filtration to 5 m or better should precede the UV.
Colour
The significance of colour is that it is a regulated quality parameter that for a given raw water
source correlates with UVT. UVT is a critical parameter in determining performance of a UV
device, and the various validation protocols will test devices across the range of UVT
specified by the manufacturer. The validation protocols do not impose limits on acceptable
UVT, and in principle low UVT can be compensated for by increasing residence time
(reducing flow rate) and/or increasing irradiance. But outside of the UK some national water
quality regulations either stipulate or recommend the minimum acceptable UVT where UV
disinfection is used.
German regulations applying to UV disinfection give guideline values of UV254 absorbance
≤ 0.1 cm-1
, and UVT ≥ 70.8% (Eggers, 2009)6. Norwegian regulations require UVT ≥ 78.6%
(Lund, 2009). VIQUA, manufacturer of Sterilight UV units designed for residential and small-
scale commercial use, recommend that UVT should be >75%.
Examples of the correlation between colour and UVT in three UK upland raw water sources
are given in Figure 6.1. By comparison with the German, Norwegian and VIQUA UVT
guidelines, 20 oH approximates to the suggested limit of practical application for UV
disinfection. As an indication of the visual impact of colour, according to Australian water
quality guidelines (2013):
‘A true colour of 15 °H can be detected in a glass of water, and a true colour of 5 oH
can be seen in larger volumes of water, for instance in a white bath. Few people can
detect a true colour level of 3 °H, and a true colour of up to 25 °H would probably be
accepted by most people provided the turbidity was low.’
6 These values are inconsistent, since UVT (1 cm) = 70.8% corresponds to UV254 absorbance =
0.15 cm-1
.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
51
Figure 6.1 Example correlations between colour and UVT in UK upland raw waters
In upland waters colour also correlates with dissolved organic carbon (DOC) (in such waters it
is primarily the absorption by organic material of UV254 which reduces the UVT). A high
organic content is more likely to cause fouling of UV lamps (USEPA, 2006), resulting in the
reduction over time of the applied UV intensity and consequently disinfection efficiency.
Iron and manganese
Iron and manganese are two other substances with the potential to foul the external surfaces
of the lamp sleeves and other wetted components of UV reactors. UV devices exposed to
waters containing concentrations of iron >100 g/l are more susceptible to fouling (USEPA,
2006). German regulations give guideline values for iron (≤50 g/l) and manganese (≤20 g/l)
(Eggers, 2009). VIQUA, manufacturer of Sterilight UV units designed for residential and small-
scale commercial use, recommend maximum concentrations for iron (300 g/l) and
manganese (50 g/l).
Iron and manganese can also give colour to water and reduce UVT.
Hardness
Although hardness is not a regulated parameter for private supplies, it is of relevance to UV
applications because of its potential contribution to fouling by precipitation of calcium (or
magnesium) salts. According to USEPA (2006) UV devices exposed to waters of hardness
0
10
20
30
40
50
60
70
80
90
100
0 20 40 60 80 100 120 140 160 180 200
UV
T, %
Colour, Hazen
Source 1 Source 2 Source 3 Source 1 (coagulated/settled)
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
52
>140 mg/l CaCO3 are more susceptible to fouling. German regulations give a guideline value
for ‘calcite precipitation capacity’ of ≤ 50 mg/l CaCO3 (Eggers, 2009). VIQUA recommend that
if hardness >120 mg/l CaCO3 the water should be softened prior to UV.
6.3 Requirements for use of UV disinfection for public water supplies
The DWI guidance document on the use of UV for disinfection of public water supplies (DWI,
2010) requires that water suppliers installing UV for disinfection be able to provide supporting
evidence for the following:
the pathogen challenge
the minimum UV dose required
the capacity of the UV reactor(s) installed to achieve the minimum UV dose (validation)
verification that the minimum dose has been applied
The guidance document requires that validation be by biodosimetry but is not prescriptive as
to the protocol; it lists USEPA, ÖNORM and DVGW as examples. It reiterates the regulatory
requirement for disinfection that turbidity be less than 1 NTU (Regulation 26 in England,
Regulation 27 in Wales) but also refers to WHO guidance that median turbidity should ideally
not exceed 0.1 NTU.
6.4 Potential validation criteria for private supply
Validated dose
For disinfection applications, all European standards and the American NSF/ANSI standard
validate to a RED of 40 mJ/cm2. ÖNORM state that this assures ‘a 6-log-reduction of health-
related water transmittable bacteria, and a 4-log-reduction of health-related water-
transmittable viruses (…) according to the state of the art’. NSF/ANSI also adopted 4-log virus
reduction/6 log coliform bacteria criteria, in accordance with recommendations by Schaub
(1987). NSF/ANSI originally (2000) set 38 mJ/cm2 as the target RED, to achieve 4 log virus
reduction – this was derived from published data indicating 3-4 log reduction in both poliovirus
and rotavirus at 30 mJ/cm2 (6 log reduction in Escherichia coli at 30 mJ/cm
2 was also
projected in the same reference) and 5 log reduction in poliovirus at 40 mJ/cm2; but increased
this to 40 mJ/cm2 in 2002 ‘to be consistent with international standards’. The generic target
REDs for viruses given in the UVDGM allow only 0.5 log reduction at 39 mJ/cm2, because the
UVDGM targets were based on data for the UV-resistant Adenovirus.
As discussed in Appendix B, it should be noted that according to the UVDGM protocol, which
does not specify the challenge micro-organism to be used in the biodosimetry tests, a RED of
40 mJ/cm2 determined using Bacillus subtilis (as specified in the ÖNORM/DVGW and,
implicitly, BSI standards) will, all things being equal, result in a higher validated dose than a
RED of 40 mJ/cm2 determined using MS2 phage (as specified in the NSF/ANSI standard).
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
53
Few of the small UV units currently available for private supplies have been validated to one
of the recognised standards listed in Section 5. Sterilight and Wedeco both state that their
units have been validated using biodosimetry by third-parties, and provide performance
curves or tables which show operating conditions (flow rate, UVT) for a dose of 40 mJ/cm2.
LIFF units are rated by maximum flow rate for 40 mJ/cm2, at unspecified UVT, but it is not
known how this dose has been determined.
Across other suppliers (and also included in Sterilight and some Wedeco literature) a dose of
30 mJ/cm2 is often quoted. This is variously described as ‘(compliant) with international dose
standards’; ‘currently the ‘Industry Standard’’; and ‘UK protocol’. The justification for referring
to 30 mJ/cm2 as a ‘standard’ is unknown. At least one supplier (Aqua Cure) derives its
performance curves by mathematical modelling; this approach is not accepted by any of the
recognised validation protocols, all of which require biodosimetry. It is also of note that where
doses of 30 mJ/cm2 are quoted, the reference UVT tends to be high: Aqua Cure state 95%,
Filpumps 99%. It is evident from Figure 6.1 that this would require there to be no perceptible
colour, unlikely in upland areas of England and Wales. With respect to the adequacy of a
30 mJ/cm2 dose, according to the published inactivation data summarised in 0, it should be
sufficient for at least 4 log removal of protozoa and most bacteria; but for a number of viruses
is either too low for 4 log removal or provides minimal safety margin.
Instrumentation and control
The inclusion of a UVI sensor and/or UVT sensor will increase the cost of a UV device and
impose additional maintenance requirements and costs (sensor cleaning, calibration,
replacement) on the owner.
An indication of the cost implications of adding control functionality to a basic on/off UV device
is given in Figure 6.2. This chart is based on retail costs from one supplier of a range of UV
devices produced by a major manufacturer of UV equipment which are appropriately sized for
private supply applications. These devices are available with three levels of control
functionality:
1. Basic unit (on/off with visual indication that lamp is on).
2. PLC controller, display. As 1, with addition of control box with PLC, visual/audible
alarms, lamp run-time indicator.
3. UVI sensor, solenoid valve. As 2, with addition of UV intensity sensor linked to PLC, UV
intensity display, solenoid valve for shut-off of water flow.
The costs have been normalised to the cost of the smallest basic unit in the range.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
54
Figure 6.2 Relative cost of additional control functionality for small-scale UV
devices
It is evident from Figure 6.2 that the addition of control functionality as required by BS EN
14987 can more than double the cost of the smallest UV devices (suitable for single tap or
small dwelling).
Suppliers of small-scale UV devices generally advise lamp replacement after one year.
Annual lamp replacement has the benefit of simplicity. Frequent switching on/off is likely to
reduce lamp life. The cost of original replacement lamps for the UV devices represented in
Figure 6.1 is in the range 5-15% of basic unit cost (being proportionately higher for smaller
capacities) so any financial benefit from running a lamp for longer than the recommended one
year will not be great in absolute terms. If a UVI sensor is fitted, a lamp can be run until it
reaches the end of its useful life, but the cost of annual sensor replacement (as required by
the BSI standard) or recalibration (ÖNORM, DVGW) will likely exceed any saving made by
delaying lamp replacement.
Flow rate
For a given lamp output, flow rate is one of the two key factors which determine the applied
dose (the other being UVT). It is therefore essential that a UV device does not permit a
greater flow rate than the maximum specified by the manufacturer.
0
1
2
3
4
5
6
7
0.73 1.85 3.24 6.7 10.1
Rel
ativ
e co
st
Flow rate, m3/h, for 40 mJ/cm2 @ 98% UVT
Basic unit PLC controller, display UVI sensor, solenoid valve
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
55
Water quality issues
A UV device cannot be reliably sized for an application without knowledge of the UVT. This
may be problematic for a private supply, for which there is unlikely to be a record of historical
water quality. In this situation, there is a risk that a UV device will be selected without
adequate characterisation of the water quality – at best, based on short-term sampling (or
even single sample), at worst without any sampling. Water drawn from a borehole is more
likely to be of stable quality, and a single sample may adequately characterise it. But surface
water sources, or those under the influence of surface water, will likely be subject to variation
in quality in the short term (linked to rainfall) and seasonally.
6.4.2 Suitability of current standards for private supply applications
Of existing standards, only BSI and NSF/ANSI apply explicitly to small-scale (point of use,
point of entry) UV devices. The BSI standard is intended for UV devices treating water
obtained directly or indirectly from the public water supply, and thus by definition excludes
private water supplies. However, because its general requirement is identical to, and its test
protocol largely reproduces that of, ÖNORM 5873-1, validation to the BSI standard is
potentially of wider relevance. ÖNORM 5873-1 requires the water being treated by UV to
meet the physical and chemical regulatory quality standards, so there is little practical
difference in expected water quality.
Only the NSF/ANSI standard applies explicitly to private supplies. It is an American standard
which in addition to validation by biodosimetry incorporates materials of construction and
pressure integrity tests in accordance with American requirements. Its handling of
experimental uncertainty appears less extensive than other standards. While the UVDGM
guidelines account for experimental uncertainties most explicitly, by deriving correction factors
for each of three key aspects of the validation test process, the ÖNORM standard appears to
adequately account for these (by applying an equivalent factor in one case, and by defining
quality assurance checks for the other two). The BSI standard explicitly addresses two of
these three uncertainties, while the NSF/ANSI standard explicitly addresses one.
The UVDGM approach is unique amongst quantitative standards in that it validates for a
specified log inactivation of a specified pathogen. While this fits in with the American
regulatory framework for municipal water treatment, in which the concept of assigning log
removal/inactivation credits to treatment processes has been adopted, it is a distinction which
makes it less appropriate for UK private supply applications for which the UV device will be
relied upon to achieve primary disinfection.
Reduction in lamp output due to ageing is accounted for in the ÖNORM and BSI standards by
requiring the manufacturer to state the expected reduction in output over the life of a lamp and
provide a method of reducing output accordingly. The validation test is then performed at this
reduced output. The lamps used for the test must have been operated for 100 hours, i.e. are
new but run-in. UVDGM also requires the ageing to be accounted for, but allows either
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
56
‘artificial’ ageing (using new lamps but at reduced power input) or the use of old lamps.
UVDGM additionally require allowance for lamp fouling. NSF/ANSI makes no explicit
reference to ageing or fouling; instead, it tests that the alarm linked to the UVI sensor
functions and then relies on this alarm to respond to any situation that may cause the UVI to
drop below the (factory set) minimum.
All the standards considered require a UV device installed for primary disinfection to be fitted
with a UVI sensor with linked alarm(s). A UVI sensor is not a requirement for ‘bactericidal
treatment’ UV devices as defined in the BSI and NSF/ANSI Class B standards. Whereas in
the BSI standard the validated UV dose for bactericidal treatment UV devices is the same as
for disinfection devices (40 mJ/cm2), the NSF/ANSI Class B standard has a lower target
(16 mJ/cm2, compared with 40 mJ/cm
2 for disinfection devices); UV devices validated to
NSF/ANSI Class B must not be used for primary disinfection applications. The inclusion of a
UVI sensor does provide greater assurance that the validated dose is being delivered in
response to changes in lamp output (ageing and/or fouling) or water quality (UVT), but
imposes additional costs and maintenance requirements (on an on-going basis, cleaning;
annually, recalibration or replacement). Experience from the site visits (Appendix E) indicates
that currently few, if any, UV installations for single domestic dwellings include a UVI sensor;
that maintenance is often lacking; and that owners often have little understanding of how the
installed equipment works. In such circumstances, the inclusion of a UVI sensor might
encourage (or force, if it triggers the water supply to be shut off) owners to give more attention
to their UV equipment. The wider adoption of UVI sensors would require the availability of
replacement sensors and calibration services.
As noted in Appendix B, under the UVDGM validation protocol a dose of 40 mJ/cm2
determined using MS2 phage (the challenge micro-organism for biodosimetry specified by
NSF/ANSI) would result in a lower validated dose than using Bacillus subtilis (the challenge
micro-organism for biodosimetry required by ÖNORM and BSI), because UVDGM applies
correction factors (RED bias factors) to allow for differences in the UV sensitivity between the
challenge micro-organism and the target pathogen (Cryptosporidium or Giardia) for which the
UV device is being validated. NSF/ANSI adopted the value of 40 mJ/cm2 to be consistent with
‘International Standards’, having originally specified 38 mJ/cm2; the possible implication of
using a different challenge micro-organism to other standards is not discussed in the
NSF/ANSI standard.
ÖNORM provides the option of a ‘simplified procedure’ which validates at the maximum
design flow rate, rather than over a range of flow rates; this is similar to the NSF/ANSI
approach. UVDGM also has equivalent options of single and variable setpoints. The BSI
standard, which otherwise closely follows ÖNORM, does not include a single point option. In
practical terms, validation to a single point means the UV device will overdose if the flow rate
is below the maximum, but requires simpler control functionality. If validated over a range of
flow rates, a flow meter can be used to adjust UV intensity in response to changes in flow
rate, which is a more energy-efficient mode of operation but more complex to implement; but
for small UV devices, simple constant output operation is adequate.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
57
6.5 Conclusions
Joint conclusions (with Section 5) are:
The UVDGM approach is not considered appropriate.
A RED of 40 mJ/cm2 using Bacillus subtilis, as required by the ÖNORM (and DVGW)
and BSI standards, is the preferred validation criterion.
Although the scope of the BSI standard excludes private supply applications, its
requirements are similar to ÖNORM. One important difference is that ÖNORM includes
the option of a simplified test procedure based on operation at maximum flow rate. This
simplified procedure better reflects how privately installed UV devices are likely to be
operated in practice.
A UVI sensor is stipulated by all standards where UV is installed for disinfection
applications. Such a sensor is considered desirable, but not necessarily essential, for
private supply applications. Accordingly no specific standard is recommended per se
for private supplies.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
58
7. Self-help Leaflet for Households
7.1 Objective
To produce a simple self-help leaflet to help households select a suitable UV system.
The leaflet will explain the use of UV disinfection for private supplies in simple terms, including
the basics of UV disinfection, the importance of water quality and pre-treatment, system
design and required UV dose, operation and maintenance, and where to obtain further
information.
The leaflet will be produced as a pdf document to be published electronically.
7.2 Guide to the selection of UV disinfection systems for households
What is a private supply?
In general, a private water supply is any supply that is not provided by a water company. Most
private supplies are situated in remote rural locations, fed by a well, borehole, spring, stream,
lake, or similar. The supply may serve a single property, several properties, commercial or
public premises.
All private water supplies must meet regulations7 which include quality standards to ensure
that the water is safe to drink.
The regulations are implemented by local authorities who are responsible for monitoring
private supplies8 through inspections (‘risk assessments’) and sampling, and will advise of the
actions to be taken if a supply fails to meet the required standards.
What is UV disinfection?
UV disinfection inactivates harmful micro-organisms that could otherwise cause illness if
consumed in drinking water. The micro-organisms in the water are exposed to UV light
generated by a UV lamp enclosed in a stainless steel or (less commonly) plastic chamber.
The UV lamp operates optimally at a temperature of about 40°C, and a quartz sleeve normally
separates the lamp from the water to prevent the lamp from cooling.
7 For England, The Private Water Supplies Regulations 2009. For Wales, The Private Water Supplies
(Wales) Regulations 2010 (as amended). Available to download from:
http://www.dwi.gov.uk/stakeholders/legislation/. 8 A supply to a private single property is excluded from monitoring unless requested by the supply
owner.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
59
A UV system rated to provide a dose of 40 mJ/cm2 is recommended to achieve effective
disinfection; the actual dose delivered will depend on water quality and flow rate.
Why is water quality important?
For UV disinfection to be effective, the water must be clear and relatively free from certain
dissolved substances that may deposit on the quartz sleeve, reducing the amount of UV light
reaching the harmful micro-organisms.
The clarity of water is usually expressed in terms of the amount of UV light that can pass - its
‘Ultraviolet Transmittance’ (UVT). Minimum UVT values, typically greater than 90-95%, are
commonly specified by UV equipment manufacturers/suppliers.
Dissolved substances that may deposit on the UV sleeve include colour, iron and manganese.
What should be included in a treatment system?
This depends on the quality of the source water and the presence of any contaminants. In
general, a groundwater source (e.g. borehole) will be of better microbiological quality than a
surface water source (e.g. stream or lake). Treatment before UV should be sufficient to
ensure that the water being disinfected meets the required quality.
Common contaminants that might affect UV disinfection include:
Suspended solids/turbidity – removed by filtration in replaceable cartridges, typically
rated to remove particles larger than 5 µm, to around 1 NTU or lower.
Colour – removed by activated carbon cartridges or membrane filters to around 20°H or
lower.
Iron and/or manganese – removed by oxidation and filtration in proprietary units to
around 200 µg/l or 50 µg/l or lower, respectively
How should the treatment system be operated and maintained?
All treatment units must be operated and maintained according to manufacturers’/suppliers’
instructions. In particular, cartridges, filters and UV lamps must be replaced at recommended
intervals. Maintenance is often modest but essential and annual contracts with specialist
companies should be considered, particularly for more complex systems.
It is recommended that copies of manufacturers’/suppliers’ operating and maintenance
instructions be retained by the supply owner. In addition, a maintenance log should be
maintained by the owner to record details of maintenance carried out and schedules
for future maintenance.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
60
UV disinfection equipment is compact and simple to operate. Most household units have little
monitoring and control, often only a power on/off indicator and visual/audible alarms to
indicate power or lamp failure. The units should be left switched on at all times and the
operation of the lamp confirmed by regular and frequent observation.
Additional monitoring and control features are available, including lamp run time, UV intensity
monitor, and automatic water shut off in the event of lamp or power failure. Such features may
not be available on all systems and will inevitably increase system cost.
In the event of power or lamp failure, if the flow of water is not automatically
interrupted, the drinking water produced will not be disinfected.
Can UV disinfected water be stored?
UV disinfection does not provide a long-lasting disinfectant residual. UV disinfected water for
drinking or cooking should be supplied directly to an appropriate tap (usually the kitchen tap).
Any water storage facilities must be hygienically maintained to ensure good quality, but
should not supply drinking water.
Where can I obtain further information?
Further information can be obtained from:
Your local authority.
The Drinking Water Inspectorate (DWI)
(http://dwi.defra.gov.uk/private-water-supply/index.htm).
Checklists to Help Select a Suitable UV System
For UV disinfection to be effective, the water must be of suitable quality and the
applied UV dose must be sufficient. For these reasons, it is important that specialist
advice is sought prior to the purchase and installation of a UV system, including any
pre-treatment to remove contaminants such as suspended solids/turbidity, colour, iron
or manganese.
The following checklists will help the householder to select a suitable UV system.
Information required to Specify and Test a Water Treatment System
Information required by a competent installer to specify and test a suitable water treatment
system is listed below.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
61
Q: What information is required about raw water quality?
Normal and maximum levels of turbidity, colour and other chemicals (e.g. iron and
manganese) that might affect UV disinfection; microbiological quality (e.g. E. coli); seasonal
variation in water quality (if any); likely increase or decrease in future water quality.
Q: What information is required about water flow and demand?
Average and peak demand; future changes to demand such as due to additional properties or
change of use.
Q: What pre-treatment might be required?
This depends on the water quality: suspended solids/turbidity can be removed by filtration;
colour can be removed by activated carbon or membrane filtration; iron and/or manganese
can be removed by oxidation and filtration.
Q: How will treatment be proven to be effective?
Raw and treated water samples to be taken by the installer/local authority to verify effective
treatment1 for all contaminants of concern, e.g. turbidity, colour, iron and manganese.
Q: How will I know if there is a problem with treatment?
Loss of flow or pressure if filters or media become blocked, possibly associated with
discolouration and taste and odour problems; visual or audible alarms if UV lamp fails; UV
lamp replacement indicator (light or counter) if lamp not replaced. 1. Analysis should be carried out by a UKAS accredited laboratory.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
62
Pre-treatment Options
Contaminant Type of Pre-
treatment Comments
Particulate matter
(turbidity, colour)
Particulate filter
(cartridge)
Reduces turbidity and colour by filtration.
Over time the filter will block and must be
replaced when throughput is reduced or at
intervals recommended by the
manufacturer.
Colour Activated carbon
(cartridge)
Reduces colour by adsorption; will also
reduce turbidity if not preceded by a
particulate filter. Must be replaced at
intervals recommended by the
manufacturer. Bacterial growth may cause
taste and odour if not changed frequently
enough.
Iron / manganese Iron / manganese
filter
Dissolved iron and/or manganese is
oxidised and removed in a filter. Typically
supplied as a single proprietary unit
including automatic backwashing of the
filter.
Bacteria and other
microorganisms
UV disinfection Microorganisms are inactivated by UV light
as water passes through the UV unit. May
require upstream pre-treatment to remove
turbidity, colour, iron and/or manganese. UV
lamp must be replaced at intervals
recommended by the manufacturer and
quartz sleeve may require periodic cleaning.
Key Considerations when Purchasing a UV System
Water Quality
Water flowing to UV disinfection must be clear and relatively free from dissolved substances
that may deposit on the quartz sleeve. Typical guide values: UVT >90%, turbidity <1 NTU,
colour <20°H, iron <200 µg/l; manganese <50 µg/l. Values much higher than the guide values
(lower for UVT) will require appropriate pre-treatment.
Flow Rate
The UV system should be sized for a maximum flow to satisfy the peak demand allowing for
potential future increases. As a guide, typical water use per person is around 150 litres per
day and kitchen taps typically discharge at 6-10 l/min (depending on pressure). UV systems
are available for flow rates from 0.12 m3/h (2 l/min) upwards.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
63
UV Dose
A design UV dose of 40 mJ/cm2 is recommended to achieve effective disinfection; actual dose
is dependent on water quality and flow rate. Operating with poor quality water and/or at flow
rates above the design value will compromise disinfection.
Monitoring and Control
Simple UV systems have little monitoring and control, often only a power on/off indicator and
local visual/audible alarms to indicate power or lamp failure. More complex systems may
include lamp hours run, UV intensity monitors, and automatic water shut off in the event of
power or lamp failure or low UV intensity. Automatic water shut off prevents the flow of non-
disinfected water.
Maintenance
UV systems must be maintained according to manufacturers’/suppliers’ instructions. UV
lamps must be replaced and quartz sleeves cleaned at recommended intervals. UV lamps are
typically every 12 months, although frequent on-off operation will reduce the lamp life.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
64
8. Guide for Local Authorities
8.1 Objective
To produce a simple guide to help local authorities assess the suitability of an installed UV
system.
The guide will explain the use of UV disinfection for private supplies in simple terms, including
the basics of UV disinfection, the importance of water quality and pre-treatment, system
design and required UV dose, key design parameters and measurements (flow and UVT),
operation and maintenance, and where to obtain further information.
The leaflet will be produced as a pdf document to be published electronically.
8.2 Guide to the assessment of UV disinfection systems for local authorities
Summary of legislation, responsibilities and roles
The Drinking Water Directive (98/83/EC) requires water intended for human consumption to
be wholesome and clean and not a risk to public health. The Drinking Water Directive for
private supplies is implemented in England and Wales by the Private Water Supplies
Regulations 2009 and the Private Water Supplies (Wales) Regulations 2010 (as amended),
respectively.
The Drinking Water Inspectorate (DWI) is the competent authority for ensuring that the
Drinking Water Directive requirements are met in England and Wales. The DWI has a
statutory role to supervise local authorities in relation to the implementation of the Private
Water Supplies Regulations, including the provision of technical and scientific advice.
Local authorities are the regulators for private water supplies and have a number of statutory
duties under the Private Water Supplies Regulations. These duties include the requirement to
carry out risk assessments and monitor private water supplies to determine compliance with
drinking water standards. A supply to a private single dwelling as defined in Regulation 10 is
excluded from the risk assessment and monitoring requirement unless requested by the
supply owner or occupier.
The local authority has powers to require that a supply that is unwholesome or a potential
danger to human health is improved by the owners or people who control the supply.
Risk assessment
The risk assessment carried out by a local authority considers all aspects of the private water
supply:
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
65
the likelihood of contamination at the source of the supply and the surrounding area;
checks of any storage tanks, treatment systems and associated pipe work;
identification of actual and potential hazards that may affect the health of those using
the water for drinking purposes;
identification of where action is necessary to make sure the water supply is wholesome
and safe to drink.
Whilst the requirements of a local authority with regard to a private supply are wide ranging,
this guidance considers only those aspects related to UV disinfection.
What is UV disinfection?
UV disinfection inactivates harmful micro-organisms that could otherwise cause illness if
consumed in drinking water. The micro-organisms in the water are exposed to UV light
generated by a UV lamp enclosed in a stainless steel or (less commonly) plastic chamber.
The UV lamp operates optimally at a temperature of about 40°C, and a quartz sleeve normally
separates the lamp from the water to prevent the lamp from cooling.
To be consistent with standards that apply to UV disinfection for public water supplies, it is
recommended that UV equipment installed for a private supply should be rated to provide a
dose of 40 mJ/cm2.
The rated UV dose is related to the design flow rate and water quality (typically described by a
minimum UVT value). In practice, the dose delivered will depend on the actual water flow rate
and quality delivered to the system.
For UV disinfection to be effective, the water must be of good quality and the applied
UV dose must be sufficient. For these reasons, it is important that specialist advice
was sought prior to the purchase and installation of the UV system. Details of the
design specification, including water quality, flow rate, any pre-treatment and UV dose,
should be requested from the supply owner or occupier.
Why is water quality important?
UV disinfection should never be installed without determination of the source water
quality and its variation.
The majority of private water supplies are sourced from groundwaters, e.g. boreholes, wells,
springs, etc., and it is essential that the infrastructure associated with protection of the source
and abstraction is adequately maintained to avoid microbiological contamination from surface
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
66
water. The catchment of the source water should be identified; potential sources of
microbiological contamination include farming activities and discharges from septic tanks/soak
away systems.
For UV disinfection to be effective, the water must be clear and relatively free from certain
dissolved substances that may deposit on the quartz sleeve, reducing the amount of UV light
reaching the harmful micro-organisms.
The clarity of water is usually expressed in terms of the amount of UV light that can pass - its
‘Ultraviolet Transmittance’ (UVT). The higher the UVT, the lower the reduction of UV light as it
passes through the water and the greater the UV intensity to which micro-organisms are
exposed. As a guideline, UVT should be greater than 75% for UV disinfection to be
practicable. Minimum UVT values, typically greater than 90-95%, are commonly specified by
UV equipment manufacturers/suppliers.
Dissolved substances that may deposit on the UV sleeve include colour, iron, manganese and
hardness.
As an indication of the water quality required for UV disinfection, the water should at least
meet the statutory physical and chemical standards for “wholesomeness” in Schedule 1, Part
1 of the Private Water Supplies Regulations, including:
Colour – 20 mg/l Pt/Co (equivalent to °H)
Iron – 200 µg/l
Manganese – 50 µg/l
pH – 6.5-9.5
Turbidity must be reduced to 1 NTU prior to disinfection.
Hardness is not a regulated parameter; as a guideline hardness should not exceed 120 mg/l
CaCO3/l for UV disinfection to be practicable.
If source water quality is unsuitable for UV disinfection, it may be possible to pre-treat the
water to an acceptable quality. Water treatment equipment is available to remove turbidity,
colour, iron, manganese and hardness.
What should be included in the UV disinfection system?
This depends on the quality of the source water and the presence of any contaminants. In
general, a groundwater source (e.g. borehole) will be of better microbiological quality than a
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
67
surface water source (e.g. stream or lake). Treatment before UV should be sufficient to
ensure that the water being disinfected meets the required quality.
Figure 8.1 shows a typical treatment flowsheet. Pre-treatment and post-treatment are
optional, depending on the requirements of the supply.
Figure 8.1 Typical treatment flowsheet
Common contaminants that might affect UV disinfection include:
Suspended solids/turbidity – removed by filtration in replaceable cartridges, typically
rated to remove particles larger than 5 µm, to around 1 NTU or lower.
Colour – removed by activated carbon cartridges or membrane filters to around 20°H or
lower.
Iron and/or manganese – removed by oxidation and filtration in proprietary units to
around 200 µg/l or 50 µg/l or lower, respectively.
How should the system be operated and maintained?
All treatment units must be operated and maintained according to manufacturers’/suppliers’
instructions. In particular, cartridges, filters and UV lamps must be replaced at recommended
intervals. A simple system may be maintained by its owner, but specialist companies should
be used for more complex systems.
Copies of manufacturers’/suppliers’ operating and maintenance instructions should be
retained by the supply owner. In addition, a maintenance log should be maintained by
the owner to record details of maintenance carried out and schedules for future
maintenance.
Post-treatment
Source Drinking water
Drinking water
UV
disinfection
Chlorination
Pre-treatment UV disinfection
Iron/Manganese
removal
Solids/Turbidity
removal
Colour
removal
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
68
UV disinfection equipment is compact and simple to operate, and maintenance is modest but
essential. Most household units have little monitoring and control, often only a power on/off
indicator and visual/audible alarms to indicate power or lamp failure. The units should be left
switched on at all times and the operation of the lamp confirmed by regular and frequent
observation.
Some UV systems include automatic shutdown of the water supply in the event of power or
lamp failure, and this is recommended. Some systems also include manual override; it should
be made clear that if the UV lamp is not functioning correctly, the water provided is not
disinfected and should be boiled prior to consumption.
In the event of power or lamp failure, if the flow of water is not automatically
interrupted, the drinking water produced will not be disinfected.
Can UV disinfected water be stored?
UV disinfection does not provide a long-lasting disinfectant residual. UV disinfected water for
drinking or cooking should be supplied directly to an appropriate tap (usually the kitchen tap).
Any water storage facilities must be hygienically maintained to ensure good quality, but
should not supply drinking water.
Where can I obtain further information?
Further information can be obtained from:
The Drinking Water Inspectorate (DWI)
(http://dwi.defra.gov.uk/private-water-supply/index.htm)
Private Water Supplies (Technical and Sampling Manuals)
(www.privatewatersupplies.gov.uk)
Manual on Treatment for Small Water Supply Systems (updated report)
(http://www.dwi.gov.uk/private-water-
supply/RHmenu/Updated%20Manual%20on%20Treatment%20for%20Small%20Suppli
es.pdf)
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
69
Checklist to Help Assess an Installed UV System
The following checklist will help local authorities to assess the suitability of an installed UV
system.
Water Supply Specification and Design
The water supply should have been specified and designed to adequately treat the raw water
and provide a sufficient flow rate of treated water.
Request evidence from the supply owner or occupier of raw water analysis, particularly
parameters that might affect UV disinfection (turbidity, colour, iron, manganese, E. coli), any
seasonal variation in water quality, and average and peak water demand.
Pre-treatment
Water flowing to UV disinfection must be clear and relatively free from dissolved substances
that may deposit on the quartz sleeve, e.g. typically UVT >90%, turbidity <1 NTU, colour
<20°H, iron <200 µg/l, manganese <50 µg/l.
Where raw water analysis (see above) indicates that pre-treatment is required, request
evidence from the supply owner or occupier of the design specification (flow rate, contaminant
levels) and performance (analysis of treated water) of installed units.
UV System
The UV system should be sized to provide an adequate UV dose for a suitable water quality -
typically quoted as UVT but may include other contaminants (see above) - at a maximum flow
rate. A UV dose of 40 mJ/cm2 is recommended, but note that the actual dose delivered will be
lower if the water quality is poorer and/or the flow is higher than the design specification. The
actual UV dose will also be reduced if the lamp is not replaced at the recommended interval
and/or the quartz sleeve is not cleaned as required.
Request evidence from the supply owner or occupier of the design specification (UV dose,
flow rate, water quality) for the UV system. If manufacturers’/installers’ literature is not
available, some information may be available from invoices and on equipment components.
The UV equipment must be designed for drinking water use and, for equipment installed after
1 January 2010 in England, satisfy Regulation 5 of the Private Water Supplies Regulations
2009, or after 26 May 2010 in Wales, satisfy Regulation 4A of the Private Water Supplies
Regulations (Wales) 2010 as amended.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
70
Monitoring and Control
Simple UV systems have little monitoring and control, often only a power on/off indicator and
local visual/audible alarms to indicate power or lamp failure; more complex systems may
include facilities such as lamp hours run, UV intensity monitor and automatic water shutdown.
Automatic shutdown of the water supply in the event of power or lamp failure is
recommended; if manual override or bypass is provided, the water should be boiled prior to
consumption.
A UV intensity monitor measures the actual UV dose being delivered to the water. However, it
is recognised that this is a relatively expensive option available on few smaller systems.
Maintenance
UV systems and any pre-treatment must be maintained according to manufacturers’/suppliers’
instructions. UV lamps must be replaced typically every 12 months – although frequent on-off
operation will reduce the lamp life – and quartz sleeves cleaned at recommended intervals.
Request evidence from the supply owner or occupier of the maintenance history. Where the
supply is maintained under contract, ideally this will be an up-to-date maintenance log but
most likely will be a series of invoices indicating work carried out. Where the supply is
maintained by the owner, request records of maintenance and look for any obvious signs of
maintenance not being carried out, e.g. discoloured or odorous water.
Post-treatment and Storage
UV disinfection does not provide a long-lasting disinfectant residual. UV disinfected water for
drinking or cooking should be supplied directly to an appropriate tap (usually the kitchen tap).
For a longer distribution system, e.g. a commercial campsite, the UV disinfected water may
be dosed with chlorine. Any water storage facilities must be hygienically maintained to ensure
good water quality, but should not supply drinking water.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
71
9. Design of a Pilot Study to evaluate a UV System for Private Water Supplies
9.1 Objective
To design a pilot study to evaluate the performance of a selected UV system for private
supplies based on the inactivation of spiked surrogate microorganisms under a range of flow,
UV lamp intensities (doses) and water quality (UVT) conditions.
9.2 Test regime
The tests can be based either on the European (DVGW and ÖNORM) UV dose validation
procedure or the US one (UVDGM). The former, uses an appropriate strain of Bacillus subtilis
(ATCC 6633) spores, while the latter uses MS2 Coliphages. When choosing which method, it
would be beneficial to choose the surrogate microorganism, with inactivation characteristics
closely matching those of the target pathogen.
A dose response curve will be developed for the surrogate microorganism using a laboratory
UV collimated beam apparatus (CBA) at 254 nm wavelength in accordance with all validation
procedures.
The full scale UV unit will be evaluated using the surrogate microorganism of choice (used in
the CBA tests), spiked in test waters, and operated under a range of inlet surrogate
microorganism concentrations, flow rates, UVT values and UV lamp intensities. Flow rates
used should be designed to cover the range relevant to the particular application and include
at least 3 flow rates (low range, median and high range). Different UV lamp intensities are
required to be able to correlate to inactivation and subsequent RED calculations, and a
minimum of 5 are required in validation procedures to obtain a reliable dose interpretation
(e.g. 0%, 25%, 50%, 75%, 100%). The UVT values should cover the UVT ranges
encountered in the context of the real application. Inactivation of the surrogate microorganism
by the unit under each set of conditions will be used in conjunction with the dose response
curve to identify the UV Reduction Equivalent Dose (RED) delivered. Inactivation is assessed
by serial dilution and agar plate counts.
The UV unit selected for the tests should ideally be fitted with a UV intensity monitor and flow
meter, to enable continuous monitoring and control of the experimental conditions. UVI
sensors should be a built-in feature installed in the unit.
Suitable test waters will need to be prepared with a range of UVT. These could be prepared
from mixtures of raw and final water collected from a water treatment works, tap water and
groundwater. An alternative would be to artificially reduce the UVT of a good quality (high
UVT) tap water or groundwater using Lignin Sulfonate or instant coffee. Preparation of
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
72
suitable test waters will need to be investigated as a prelude to the test work. A mixture of
filter backwash water with high UVT water could be included to compare waters with different
turbidity but similar UVT, to identify the impact of turbidity on UV performance.
9.3 Procedure
1. Either acceptable test protocol for dose validation can be used (e.g. DVGW, ÖNORM or
USEPA UVDGM Appendix C). If any variations from the protocol are deemed to be needed,
the implications for the interpretation of the results should be clearly stated and taken into
account where possible. It may not be necessary to use the USEPA Bred factor as the test is
for general disinfection rather than for a specific pathogen.
The European dose validation protocol is more widely accepted in Europe, and the procedure
described below is based on that. The laboratory carrying out the work needs to fully
understand the requirements of using the protocol, at a level of detail which is beyond the
scope of this procedure to describe in full.
2. Prepare suspensions of B. subtilis spores or MS2 Coliphages (depending on choice of
validation method) in high UVT water for laboratory collimated beam tests, to be carried out in
accordance with published guidelines. The concentration of the spores should be high enough
to allow for reliable quantification of the level remaining after the highest UV dose used. The
highest dose is expected to achieve 5 log inactivation, the concentration should be one or two
orders of magnitude higher (6-7 logs), to account for errors in measurements and ensure a
standard deviation in log removal that is acceptable. The concentrations must be reproducible
with a standard deviation of no more than 0.2 log units.
3. Assess the inactivation at each UV dose with the collimated beam apparatus and produce
the dose response curve (UV dose vs log inactivation) for an appropriate range of UV doses,
at least 5 doses that would generate log inactivations of at least 0.5, 1, 2, 3, 4, 5. Each test
involves irradiating samples of the surrogate microorganism suspension in the test chamber
for a period of time to achieve the required UV dose, and evaluating the viability of the initial
and remaining microorganisms by plate counts. The dose distribution and resultant potential
variability of dose in the test chamber may need to be taken into account in the calculation,
based on published protocols or equipment suppliers’ information. Each test should be carried
out (minimum) in triplicate with a target reproducibility, e.g. ±10%.
4. Prepare a range of test waters for the full scale tests with UVT in the range 70-99%. It
would be recommended to include at least one water with increased turbidity with a UVT
value within this range.
5. Spike appropriate volume samples with the surrogate microorganisms, at a concentration
high enough to allow reliable detection after the maximum UV dosing, based on the maximum
expected/designed log inactivation. If tap water is used to prepare the test waters, checks
should be made that no chlorine remains before spiking.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
73
6. Carry out tests with the UV unit with (minimum) 3 flow rates, and 5 UV intensities (see
above) for each test water, and assess the inactivation for each test with (minimum) triplicate
samples of feed and treated water for each condition. Include a flow rate above the maximum
UV unit flow for each test water. Record the UVI reading if available. The UV unit should be
installed and operated based on the supplier’s instructions, including the inlet and outlet
hydraulic configurations.
7. Use the dose response curve to identify the RED delivered for each test condition and
compare these with the supplier’s stated doses for each flow rate and UVT.
9.4 UVT measurement
UVT is the % transmittance of UV through 1 cm of water, and for direct measurement a 1 cm
cell is used. If a cell longer than 1 cm is used, the indicated value must be scaled using the
formula
yUVTUVT
1
cmy
100100
where UVTy cm is the indicated UV transmittance in a cell of length y cm.
For example, if the indicated UV transmittance using a 4 cm cell is 80%, then
94.6% 100
80100
41
UVT
Pre-programmed spectrophotometers which use a cell length greater than 1 cm may make
this conversion automatically, so the user should check whether this is the case if using an
instrument with which they are unfamiliar.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
74
Figure 9.1 Outline procedure
Stock suspension of spores
Dilution for collimated beam
tests (if needed)
Spiking to test waters
Collimated beam tests
Dose response curve (UV
dose vs log inactivation)
UV unit trials at range of flow
rates and UVT
Inactivation at each set of
conditions
RED for each set of
conditions
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
75
10. Conclusions
UV technologies in public water supply
UV disinfection is widely used in public water supply, with most installations <10 Ml/d but also
larger installations >100 Ml/d.
Design is usually based on detailed feed water quality data, with a dose of 40 mJ/cm2 or
higher for the majority of units. Dose validation according to ÖNORM, DVGW or USEPA is
becoming the norm.
Monitoring and control is usually based on measurement of UV intensity; UVT is also
measured and sometimes used for control. Feed water turbidity is monitored according to
Regulation 26 requirements.
UV lamps are of the MP or LPHO type, with cleaning and replacement carried out routinely at
supplier defined intervals. UV intensity monitors are routinely calibrated.
UV technologies in private water supply
UV disinfection used in private water supply is mostly <10 m3/d (often much smaller); the
larger units are usually installed at commercial premises rather than domestic.
Design may be based on limited feed water quality data, with pre-treatment specified to deal
with poorer feed water quality. UV dose is typically 30 mJ/cm2 for domestic units and
40 mJ/cm2 for larger commercial units. Little, if any, biodosimetric dose validation; some larger
suppliers may carry out microbial challenge testing or hydraulic and UV intensity modelling.
Limited monitoring and control, particularly for domestic units, with control usually based on
maximum flow rate and specified UVT of the feed water. No measurement of turbidity or UVT;
some of larger commercial units may include UV intensity monitors which provide a shutdown
rather than a control capability.
UV lamps are of the LP type, with cleaning and lamp replacement carried out annually
(typically) by installers under service agreements in many cases; some simple systems may
be serviced by owners.
Key findings from site visits to private supplies incorporating UV disinfection
There was a general lack of understanding amongst users regarding the treatment of their
private supplies. This was compounded by the lack of information provided by equipment
providers/installers.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
76
There was no indication that UV equipment had been selected correctly for the flow or water
quality.
Smaller private supplies and SDDs incorporated simple treatment, typically particulate
filtration and/or UV disinfection. Some larger commercial private supplies incorporated more
complex treatment systems.
UV equipment was generally serviced by specialist companies, plumbers or the users, with
quartz sleeves cleaned at intervals between 2-12 months and lamps changed around every
12 months; the frequency of maintenance of other equipment and replacement of cartridge
filters was less clear. Maintenance logs are generally not kept by users.
Monitoring and control of UV equipment is inadequate. Failure of a UV lamp may go
undetected for some time because a lack of a prominent alarm, and will generally not prevent
flow and the possibility of the consumption of non-disinfected water.
The potential for contamination of stored UV-treated water may not be well understood by
users.
There is currently no licensing or approved contractor scheme applicable to the installation of
equipment for private water supplies.
Review and comparison of standards and validation criteria for UV systems
UK (BSI) and international standards (USEPA (UVDGM), ÖNORM, DVGW, NWRI/WRF and
NSF/ANSI) have been reviewed and compared.
The USEPA (UVDGM), ÖNORM, DVGW and NWRI/WRF standards apply to public water
supplies.
The BSI standard applies to LP UV devices intended for water conditioning in buildings; the
NSF/ANSI standard applies to point-of-entry and point-of-use LP UV equipment.
The BSI standard specifies a dose of 40 mJ/cm2 validated by biodosimetry; the NSF/ANSI
standard specifies a dose of 40 mJ/cm2 (disinfection) or 16 mJ/cm
2 (supplemental bactericidal
systems) validated by biodosimetry.
A reduction equivalence dose (RED) of 40 mJ/cm2 as required by the ÖNORM (and DVGW)
and BSI standards is the preferred validation criterion.
A UVI sensor is stipulated by all standards where UV is installed for disinfection applications.
Such a sensor is considered desirable, but not necessarily essential, for private supply
applications.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
77
Design of a pilot study to evaluate a UV system for private water supplies
A pilot study based on either European (DVGW and ÖNORM) UV dose validation or US
(UVDGM) UV dose validation is proposed to evaluate a UV system spiked with surrogate
microorganisms under a range of flow, UV lamp intensities (doses) and water quality (UVT)
conditions.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
78
11. Recommendations
A number of key recommendations are suggested that would improve the reliability and
performance of UV disinfection for private supplies:
A licensing or approved contractor scheme should be implemented for installers of
equipment for private water supplies.
Copies of manufacturers’/suppliers’ operating and maintenance instructions should be
provided and retained by the supply owner. In addition, a maintenance log should be
maintained by the owner to record details of maintenance carried out and schedules for
future maintenance.
Audible and visual alarms should be more prominent, particularly where the UV system
is sited away from the user’s premises.
UV systems should include automatic shutdown of the water supply in the event of
power or lamp failure. An emergency valved by-pass line could be incorporated with
instructions to boil drinking water prior to consumption (whilst the UV system awaits
repair).
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
79
References
Amoah, K., Craik, S., Smith, D.W. and Belosevic, M. (2005). Inactivation of Cryptosporidium and
Giardia cysts by UV light in the presence of natural particulate matter. Aqua- Journal of Water Supply:
Research and Technology, 54, (3), pp 165-178.
Australian Drinking Water Guidelines Version 2.0 (2013). Physical and Chemical Characteristics – Fact
Sheets: Colour (True).
http://www.clarence.nsw.gov.au/file.asp?g=RES-WHM-86-71-52
Blatchley, III E.R. et al. (2008). Validation of large-scale, monochromatic UV disinfection systems for
drinking water using dyed microspheres. Water Research, 42, 977-688.
Boeker, M. (2009). How the UVDGM 2006 Impacts UV Disinfection System Performance). 5th IUVA
World Congress, European Regulatory Workshop, Amsterdam, 23 September 2009.
Bolton, J.R. (2000). Calculation of ultraviolet fluence rate distributions in an annular reactor:
significance of refraction and reflection. Water Research, 34, (13), 3315-3324.
Bolton, J.R. and Cotton, C.A. (2008). The Ultraviolet Disinfection Handbook. American Water Works
Association. ISBN 1-58321-584-0
Bolton, J.R., Beck, S.E. and Linden, K.G. (2015). Protocol for the determination of fluence (UV dose)
using a low-pressure or low-pressure high-output UV lamp in bench-scale collimated beam ultraviolet
experiments. IUVA draft protocol published for comment.
Bucheli, M. (2009). UV disinfection of drinking water in Switzerland: situation, regulations and practice.
5th IUVA World Congress, European Regulatory Workshop, Amsterdam, 23 September 2009.
Bucheli-Witschel, M., Bassin, C. and Egli, T. (2010). UV-C inactivation in Escherichia coli is affected by
growth conditions preceding irradiation, in particular by the specific growth rate. Journal of Applied
Microbiology, 109, pp. 1733-1744.
CDPHE (Colorado Department of Public Health and Environment), 2013. Basis for Acceptance of
NSF/ANSI Standard Class “A” Ultraviolet Disinfection Equipment for Use in Small Public Water
Systems in Colorado.
http://www.colorado.gov/cs/Satellite?blobcol=urldata&blobheadername1=Content-
Disposition&blobheadername2=Content-
Type&blobheadervalue1=inline%3B+filename%3D%22UV+Position+Paper.pdf%22&blobheadervalue2
=application%2Fpdf&blobkey=id&blobtable=MungoBlobs&blobwhere=1251917854611&ssbinary=true
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
80
Chatterley, C. (2009). UV-LED irradiation technology for point-of-use water disinfection in developing
communities. MSc Thesis, Department of Civil, Environmental and Architectural Engineering, University
of Colorado at Boulder.
Cooper, J.R. and Cavender, G. (2011). Disinfection performance testing of high-efficiency ultraviolet
water treatment chamber. Disinfection White Paper v8. NeoTech Aqua Proprietary Information.
Deguchi, K. et al. (2005). UV-disinfection reactor validation by computational fluid dynamics and
relation to biodosimetry and actinometry. Journal of Water and Environment Technology, 3, (1), pp 77-
84.
Dicks, C. and Sief, R. (2009). CFD as a tool to predict certification results (DVGW, ÖVGW, SVGW). 5th
IUVA World Congress, European Regulatory Workshop, Amsterdam, 23 September 2009.
Dykstra, T.S. et al. (2002). Hydraulic calibration and fluence determination of model ultraviolet
disinfection system. Journal of Environmental Engineering, November, pp 1046-1055.
DWI (2010). Guidance on the use of Ultraviolet (UV) Irradiation for the Disinfection of Public Water
Supplies.
http://dwi.defra.gov.uk/stakeholders/guidance-and-codes-of-practice/uvirradiation.pdf
DWI (2015a). Drinking Water 2014. Private Water Supplies in England_July 2015. A report by the Chief
Inspector of Drinking Water.
http://dwi.defra.gov.uk/about/annual-report/2014/pws-eng.pdf
DWI (2015b). Effect of UV on the Chemical Composition of Water including DBP Formation (DWI
70/2/30).
http://dwi.defra.gov.uk/research/completed-research/reports/DWI70-2-300.pdf
Eggers, J. (2009). Drinking Water Disinfection by UV Light in Germany. 5th IUVA World Congress,
European Regulatory Workshop, Amsterdam, 23 September 2009.
Hijnen W.A.M., Beerendonk E.F. and Medema G.J. (2006). Inactivation credit of UV radiation for
viruses, bacteria and protozoan (oo)cysts in water: A review. Water Research, 40, (1), pp 3-22.
Ho, C.K. (2009). Evaluation of reflection and refraction in simulations of ultraviolet disinfection using the
discrete ordinates radiation model. Water Science and Technology, 59, (12), pp 2421-2428.
Jin, S. et al. (2006). Polychromatic UV fluence measurement using chemical actinometry, biodosimetry,
and mathematical techniques. Journal of Environmental Engineering, August, pp 831-841.
Johnson, J.D. and Qualls, R.G. (1984). Ultraviolet disinfection of secondary effluent: Measurements of
dose and effects of filtration, EPA 600/2-84-160.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
81
Lund, V. (2009). Ultraviolet Disinfection Regulations in Norway. 5th IUVA World Congress, European
Regulatory Workshop, Amsterdam, 23 September 2009.
Morita S. et al. (2002). Efficacy of UV irradiation in inactivating Cryptosporidium parvum oocysts.
Applied and Environmental Microbiology, 68, (11), pp 5387-5393.
NWRI (2012). Ultraviolet Disinfection Guidelines for Drinking Water and Water Reuse, 3rd
Edition.
NWRI, Fountain Valley, California.
http://nwri-usa.org/documents/UVGuidelines3rdEdition2012.pdf
Park, G.W., Kinden, K.G. and Sobsey, M.D. (2011). Inactivation of murine norovirus, feline calicivirus
and echovirus 12 as surrogates for human norovirus (NoV) and coliphage (F+) MS2 by ultraviolet light
(254 nm) and the effect of cell association on UV inactivation. Letters in Applied Microbiology, 52, pp.
162-167.
Passantino, L., Malley, J., Knudson, M., Ward, R. and Kim, J. (2004). Effect of low turbidity and algae
on UV disinfection performance, Journal of the American Water Works Association, 96, (6), pp 128-137.
Pilmis, V. and Baig, S. (2009). UV Regulations in France. 5th IUVA World Congress, European
Regulatory Workshop, Amsterdam, 23 September 2009.
Schaub, S.A. (1987). Guide Standard and Protocol for Testing Microbiological Water Purifiers, Report
of Task Force submitted to USEPA. http://www.biovir.com/Images/pdf061.pdf
Scottish Government (2015). Ultraviolet disinfection for private water supplies. WRc Report
UC10061.02 (Awaiting publication).
Shen, C. et al (2009). Validation of medium-pressure UV disinfection reactors by Lagrangian
actinometry using dyed microspheres. Water Research, 43, pp 1370-1380.
Shepherd, D., Gee, R., Hall, T., Rumsby, P. and Dillon, G. (2014). Effect of UV on the Chemical
Composition of Water including DBP Formation, WRc Report No. Defra10459.04.
USEPA (2006). Ultraviolet Disinfection Guidance Manual for the Final Long Term 2 Enhanced Surface
Water Treatment Rule, EPA 815-R-06-007.
VIQUA. Sterilight Silver Owner’s Manual, 520104_RevH.
http://viqua.com/bms/assets/389/Manual_Sterilight_Silver_S12Q-PA_EN_FR_520104_RevH.pdf
Wright, H. et al. (2009). Comparing UV Validation Using USEPA Drinking Water and NWRI/AwwaRF
Guidance. 5th IUVA World Congress, European Regulatory Workshop, Amsterdam, 23 September
2009.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
82
Appendix A UV Technologies in Public Water Supply: Further Design and Operating Information
A series of questions relating to UV design and operation was circulated to water company contacts,
and responses were received from six companies. The questions and detailed responses are
summarised in the tables below.
1. How is design dose established? The European dose validation (ÖNORM or DVGW) is based on
40 mJ/cm2. Are higher doses used to give a margin of safety, or because of higher microbial challenges
from risk assessments? Is target log removal taken into account (as for Crypto in USEPA dose
validation)?
Water Company A All of our units are now installed on the basis that they may need to
treat Crypto. This means that since DWI guidance came out they
have all been validated based on 25 mJ/cm2 by the USEPA
guidelines. Prior to that, they had been designed for a dose of 40
mJ/cm2 based on the DVGW recommendations. Crypto has been
seen as the worst case scenario for us. The potential presence of
viruses is covered by having two-stage disinfection on a risk-based
approach.
Water Company B 27 plants, 17 with USEPA dose validation (17 mJ/cm2), dose control
by calculated dose from UVT.
10 with ÖNORM or DVGW (40 mJ/cm2) using intensity set point
dose control.
Water Company C European dose of 40 mJ/cm2 for general disinfection.
Water Company D Design around 40 mJ/cm2, Log removal not taken into account, we
do overdose as a safety margin.
Water Company E Target design dose of 40 mJ/cm2.
Water Company F Our older UV sites were designed on a minimum of 48 mJ/cm2,
later systems use 40 mJ/cm2 as a minimum. Validation is by
USEPA/UVDGM on Wedeco/Xylem systems, USEPA for Trojan.
Systems typically installed to replace contact time/tank, so
disinfection rather than Crypto. Where installed for Crypto
specifically we have opted for 40mJ/cm2 for added security.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
83
2. Have situations arisen where numbers of units installed have limited flexibility and led to higher
doses than design at times of low flow?
Water Company A Probably – not too concerned about higher doses. Certainly some
of the earlier units installed run at a much higher dose than would
be required.
Water Company B Yes. Units are designed for worst case scenarios for UVT and flow,
many of which have high seasonal fluctuation in flow. Also, some
works have since had an additional organics removal process,
substantially improving UVT. Not overly concerned with overdosing
as chlorine is dosed post-UV at majority of sites.
Water Company C No, typically installations have been designed with due regard for
the validated envelope vs flow and UVT range. Where a large
range of flow and UVT is expected then duty/assist/standby
arrangements are considered.
Water Company D No
Water Company E Yes, at the majority of our sites the actual dose is higher than the
target dose. This higher actual is due to a combination of UVT
being maintained above 90% and ensuring the duty UV reactor
remains within validated conditions for flow and power.
Water Company F Yes, AMP4 standard design was for 100% standby, for smaller
sites this required multiple units running concurrently to avoid warm
up time issues on duty/standby change-over. The minimum UV
dose is often well in excess of requirements. Some sites have been
modified to provide duty/standby operation with a plant shut-down
as required.
3. Are flow rates and UVT controlled automatically to maintain the dose validation windows?
Water Company A Dose is controlled to the flow rate. We don’t measure UVT on-line.
All of our UV sites are stable good quality groundwater, so we
measure the UVT and design to that, assuming little movement
over time. Lab sampling confirms this, but this can be as infrequent
as monthly.
Water Company B All are monitoring continuously for dose (and UVI) and UVT, and
linked to SCADA. UVT not always alarmed. Emergency shut down
on dose for 26 plants.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
84
Water Company C Single reactors are sized to cope with the nominal works flow at
nominal low end UVT. Duty/Assist is operated between nominal
and maximum works flow or exceptionally low UVT. Automatic shut
down is initiated if the UVT falls below the validation window at the
minimum works flow.
Water Company D Yes the PLC control “ensures” unit runs inside its validation window
although I have seen excessively high doses at times.
Water Company E Flowrates are controlled to automatically maintain the dose
validation window. The UVT is monitored continuously at the inlet
to the UV process and is used to determine the target dose.
Water Company F Generally older systems operate on fixed flow and UVT is
monitored to ensure >90%. Newer systems will increase intensity to
maintain dose requirement automatically.
4. Is UVT commonly used as a feed-forward control parameter, or is control mainly based on feedback
from intensity monitors. Is this specific to UV plant suppliers?
Water Company A N/A – see question 3.
Water Company B 10 plants (USEPA validated) with dose control based on calculated
dose from UVT (feed-forward), with remainder (ÖNORM/DVGW)
using intensity set point.
Water Company C UVT is not used as a feed-forward control parameter, just for
validation purposes. Control is based on feedback from intensity
monitors and flow.
Water Company D Only on surface water sites, we find on groundwater the UVT is
constant.
Water Company E UVT is used as a feed-forward control parameter on our Trojan
plants. At one site we have a Wedeco plant. I am not so familiar
with this. Although there are UVT monitors I believe the target dose
is based on feedback from the UV intensity monitors.
Water Company F As above, generally UVT monitored rather than used for control.
Newer systems have dual validated instruments and linked to plant
shut-down when falling outside setpoints. Intensity monitors used to
maintain actual UV dose (covers lamp output and water quality).
Yes, different suppliers use different control systems.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
85
5. Are intensity monitors recalibrated in accordance with manufacturer’s or dose validation
requirements? What is the typical frequency of recalibration?
Water Company A As per the manufacturers requirements. I think 6 monthly.
Water Company B As per manufacturer’s recommendation, typically recalibrated by
external servicing contract on 12 monthly basis.
Water Company C Intensity monitors are recalibrated in accordance with
manufacturer’s instructions. Typically every 6 to 12 months.
Water Company D Yes by use of reference radiometer monthly as per manufacturer.
Water Company E Yes intensity monitors are recalibrated in accordance with
manufacturer’s guidance. At the majority of our sites this is every
35 days but at one site it is six monthly.
Water Company F Intensity instruments checked/validated frequently and calibrated in
line with manufacturers requirements. Instrument issues detected
by plant control systems.
6. Are lamps always replaced in accordance with manufacturer’s maximum hours-run guidance? Is any
allowance made for high frequency of stop/start which might shorten lamp-life?
Water Company A The manufacturer’s guidance is a starting point. However, we rely
on the intensity of the output to determine replacement.
Water Company B Yes, replaced in accordance with manufacturer’s recommendation
of 12,000 hours. Normally done as part of full service. Duty change
normally at weekly frequency.
Water Company C Yes – it is very clear that if this is not undertaken then the
disinfection efficacy is not assured. No sites have frequent
stop/start so no allowance for that is given. Where start/stop has
occurred frequently (during commissioning) we have seen
increased lamp failure.
Water Company D Lamps changed when time expired.
Water Company E Yes, the lamps are replaced every 9,000 hours on our Trojan sites
and 12,000 hours for the Wedeco system at one site. There is no
allowance made for a high frequency of stop/start but to alleviate
this to some extent a duty change is carried out every 4 days.
Water Company F Lamps replaced in line with manufacturer’s recommendations, plant
control system ensures start/stops are kept to a minimum and
frequency logged.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
86
7. What is the policy for routine cleaning of lamps? Is this based on time or can any information from
intensity/UVT be used to initiate cleaning? Are units taken off line for cleaning, or can this be carried out
while in operation?
Water Company A This varies. Some of our “dirtier” sites (particularly those with iron)
will have lamp wipers in place. There may be an issue with these
giving an instantaneous fall in UV dose, though. Otherwise
maintenance and cleaning will take place off-line. Intensity (or in
particular increase in power to maintain intensity) is a key
parameter to pick this up. We don’t have on-line UVT.
Water Company B All units except single-lamp units for spring sources have automatic
mechanical wiping which is triggered on time approximately twice a
day. A full clean is also done as part of the annual service. Clean
can also be triggered based on deterioration in performance based
on log of unit power or intensity which is reviewed weekly. Units are
duty / standby so clean is always offline. One site which has a
particular fouling issue has had a dedicated contract set up with
service supplier for monthly clean.
Water Company C Lamp sleeves are cleaned depending on risk, based on feed water
quality. Automatic cleaning with wipers is specified in certain
situations. The trigger is time based but they would also be cleaned
if there were difficulties in achieving the required intensity/UVT.
Automatic wiping of sleeves is undertaken on-line. Manual cleaning
of sleeves is undertaken off-line. UV disinfection is not generally
specified at sites where heavy fouling of lamps is deemed likely.
Water Company D All units have mechanical wiping, manual cleaning on time basis
(not sure of the frequency).
Water Company E An automatic cleaning regime is used while the units are on line. A
manual clean may be carried out in response to UVT.
Water Company F Routine cleaning undertaken as per manufacturer’s
recommendations and site specific based on time and incorporated
into UV maintenance schedule. Intensity trends/alarms are also
used to indicate cleaning required before this time, but typically this
does not occur on our sites. Units are taken out of service for
cleaning.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
87
8. Are there any other operational or maintenance issues?
Water Company A Not sure if this is what you’ve got in mind, but we’ve taken out
mercury/glass traps when there is a contact tank or reservoir
downstream of the UV. We now only have them where the water
goes directly to supply. This has reduced the head-loss across the
unit (marginally but critically in a couple of places), while reducing a
maintenance burden.
Water Company B No major problems. Had issues with stocking critical spares initially
but this has since been resolved. Only one works with no
redundancy i.e. run in duty/duty so will have impact on works
throughput and operability if a unit is offline. One instance of Cl
dosing followed by UV causing bromate formation.
Water Company C None as yet.
Water Company D Yes we had difficulty in doing the UV intensity monitor checks as
the systems were designed with only one sensor and removing that
to replace with reference unit caused system to shut down, we
have had to install a limited time maintenance switch to enable
calibration.
Water Company E I didn’t get too much feedback from operational colleagues
regarding this question. One mentioned the cost of routine
maintenance and cleaning. Another comment mentioned that if
hours run or sensor days runs are exceeded it invalidates the unit. I
suspect this latter comment means it requires a more rigid
maintenance regime compared to some other processes and
monitors.
Water Company F Issues with raw water bromide levels, Cl dosing followed by UV
causing bromate formation.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
88
Appendix B Biodosimetry
Where chemical disinfection is employed it is possible to measure the residual concentration
and use that, in conjunction with contact time, to judge the sufficiency of disinfection. With UV
there is no measurable residual, so there is no equivalent means of monitoring the efficacy of
disinfection. UV intensity varies within a reactor, and micro-organisms passing through do not
follow the same flow path; consequently, they do not all receive the same UV dose. To
provide the necessary confidence that UV reactors are providing effective disinfection, all
current standards require equipment suppliers to validate performance of their equipment by
biodosimetry, and provide evidence of this validation to end users.
B1 Principles of biodosimetry
Biodosimetry is a validation procedure in which the UV reactor is challenged with a non-
pathogenic surrogate test micro-organism under a range of operating conditions (e.g. flow
rate, lamp output, UVT). There are differences between the test protocols specified in the
various standards, but the principles, outlined below and illustrated in Figure B1, are the
same:
(1) Experimental tests
(a) The UV dose-response curve (log inactivation as a function of dose) is determined
for the surrogate micro-organism using a laboratory collimated beam UV source.
(b) The reactor is challenged with the surrogate micro-organism under a defined matrix
of operating conditions, and the log inactivation determined for each set of
conditions.
(2) The Reduction Equivalent Dose (RED)9 for each set of challenge test operating
conditions is determined by comparing the log inactivation against the dose-response
curve. The RED is the dose from the dose-response curve which corresponds to the
log inactivation observed in the challenge test.
Under the UVDGM protocol, correction factors are applied to the RED to determine the
validated dose; these factors account for the difference in UV sensitivity between the
surrogate micro-organism and the target pathogen and for experimental uncertainties. Under
the ÖNORM/DVGW, NSF/ANSI and BSI protocols, experimental uncertainties are handled (to
different extents) in the derivation of the RED, and the RED so determined is the validated
dose; under these protocols the target validated dose of 40 mJ/cm2 implicitly makes due
allowance for the UV sensitivity of the specified surrogate micro-organism.
9 In European terminology ‘fluence’ is often used rather than ‘dose’, and hence REF instead of RED.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
89
Figure B.1 Biodosimetry validation procedure
…
1a Dose-response curve tests 1b Reactor challenge tests
Challenge microorganism
Reactor
Influent sample
Effluent sample
Collimated beam
Challenge microorganism
sample
Dose
Log inactivation
Log inactivation Condition
1 2 …
… …
…
2 Reduction Equivalent Dose (RED)
Log inactivation
Dose
RED
n
Interpretation of dose as determined by biodosimetry is not straightforward. Strictly speaking,
a RED determined by biodosimetry is meaningful only with reference to the challenge micro-
organism with which it was determined. The reasons for this are as follows. Except in
hydraulic conditions of perfect plug flow, the exposure to UV of each individual micro-
organism passing through a reactor is different, because UV intensity within the reactor is not
uniform, each micro-organism takes a different path through the reactor, and the retention
time of each micro-organism is different. Consequently, there will in practice be a probability
distribution of UV doses, and the observed log inactivation will represent the overall effect of
this distribution. The inactivation resulting from a given dose is determined from the dose-
response curve, which is different for each type of micro-organism. Hence for a given reactor
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
90
under identical operating conditions, the RED determined using one type of challenge micro-
organism will not necessarily be the same as that determined using a different type. Only
under the theoretical condition of perfect plug flow, in which every individual micro-organism
has equal exposure to UV, will the value of RED be the same for different types of challenge
micro-organism. One of the correction factors10
applied to the RED in the UVDGM protocol is
intended to make allowance for the difference in sensitivity between the challenge micro-
organism and the target pathogen; there is no equivalent factor in the other protocols.
Attempting to quantitatively compare test protocols is further complicated by differences in
methodology, not least how experimental uncertainties are accounted for. UVDGM quantify
experimental uncertainty by applying up to three uncertainty factors, relating to:
UVI sensor reading
Goodness of fit of the dose-response curve
Reproducibility of replicate biodosimetry sample results
ÖNORM account for these uncertainties in different ways. UVI sensor uncertainty is applied
as a correction factor to the maximum flow rates permitted for the UV device. An acceptable
envelope is specified within which the dose-response curve must lie. And each biodosimetry
test point must be repeated five times, with a defined maximum acceptable standard deviation
for the log colony counts. NSF/ANSI only accounts for goodness of fit of the dose-response
curve, by specifying the envelope within which the curve must lie, but does stipulate that UVI
sensors should have a measurement uncertainty of not more than 9%. The BSI standard
draws extensively on the ÖNORM standard, but omits the correction factor based on UVI
sensor uncertainty. A summary of how the different standards address the experimental
uncertainties quantified in the UVDGM is given in Table B.1.
One US state (CDPHE, 2013) investigated whether UV reactors validated in accordance with
the NSF/ANSI Class A standard (40 mJ/cm2 using MS2 phage
11) should be permissible for
small public water supplies, which would normally require equipment validated in accordance
with the UVDGM protocol. The conclusion was that UV reactors with 40 mJ/cm2 NSF/ANSI
10 RED bias factor. This factor is influenced by UVT, because UVT affects the distribution of UV
intensity within a UV reactor. 11
Note: According to the UVDGM protocol, using the factors provided to make allowance for
differences in UV sensitivity between challenge micro-organisms and target pathogens, and all
experimental uncertainties being equal, for inactivation of Cryptosporidium or Giardia a RED of
40 mJ/cm2 determined using MS2 will result in a lower validated dose than a RED of 40 mJ/cm
2
determined using Bacillus subtilis (the challenge micro-organism used for ÖNORM/DVGW and BSI
standards). For 4-log inactivation the difference is < 10% if UVT ≥ 98%, increasing to c. 30% if UVT
is in the range 80 – 85%.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
91
Class A validation would only be awarded treatment credits equivalent to a UVDGM validated
dose of 1.5 mJ/cm2. In arriving at this conclusion some conservative (worst case)
assumptions were made in relation to experimental uncertainties requiring quantification by
UVDGM but not by NSF/ANSI. The UVDGM guidelines claim similar reasons for only allowing
ÖNORM/DVGW-validated units a 3 log credit for Cryptosporidium. In their comparison of
NSF/ANSI and UVDGM, CDPHE (2013) assumed that for NSF/ANSI uncertainties could
occur equally in each of the three areas, but as indicated in Table B.1, NSF/ANSI does place
constraints on two out of the three; arguably, therefore, CDPHE’s conclusions are over
cautious.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
92
Table B.1 How standards address experimental uncertainties
Standard UVI sensor Sensitivity (dose-response) curve Biodosimety test results
UVDGM Factor applied if measurement uncertainty
>10%.
Factor applied if uncertainty (based on
95% confidence interval) >30% (using
approximate method) OR >15% (using
standard statistical method).
Factor applied based on 95% confidence
interval.
ÖNORM Factor applied. Minimum uncertainty of
15% must be applied.
Boundaries are specified within which the
sensitivity curve must lie.
A maximum permissible standard
deviation is specified for parallel (5 off,
before and after) log(cell count)
enumerations.
BSI No explicit quantification. Boundaries are specified within which the
sensitivity curve must lie.
A maximum permissible standard
deviation is specified for parallel (5 off,
before and after) log(cell count)
enumerations.
NSF/ANSI Specifies maximum total uncertainty of
± 9% for sensor, but not used in validation
procedure.
Boundaries are specified within which the
sensitivity curve must lie.
No explicit quantification.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
93
B2 Alternative approaches to biodosimetry
Alternatives to biodosimetry are summarised below, to indicate their historical context and
current potential.
B2.1 Point Source Summation method for estimating average UV intensity
The point source summation (PSS) method is a mathematical method for estimating average
UV intensity (Johnson and Qualls, 1984) from a model of the irradiance field. The method
consists of treating a UV lamp as a series of point sources, and the reactor volume as a
series of point receivers. The UV light from each point source that reaches each point receiver
is summed to give the total intensity at that point in the reactor; and then the average of all
point receiver total intensities is calculated to give the average intensity of the reactor. The
method is extended to multiple lamp reactors by including for each point receiver the point
sources along each lamp.
Bolton (2000) modified the PSS method to allow for refraction and reflection and concluded
that in potable water treatment applications where the transmittance is high (> 90%)
neglecting these effects can result in an over-prediction of average intensity by as much as
25%.
B2.2 Direct measurement of intensity (radiometry)
UV intensity can be measured at points in a reactor using sensors. UV sensors are limited in
that they can only accurately measure light that is normal to the sensor surface (Jin et al,
2006), and also require periodic recalibration. Whilst UV sensors are normally used for control
purposes to maintain dose validated conditions, their role for practical dose validation itself
would be limited.
B2.3 Actinometry
Using photochemical reactions to measure UV dose is well established. Chemicals that have
been used for this purpose include potassium ferrioxalate (Dykstra et al., 2002) and uridine
(von Sonntag and Schuchmann, 1992). An attraction of uridine in relation to water disinfection
is that its response to UV light is similar to that of DNA, which is particularly beneficial in the
context of MP and pulsed lamps (von Sonntag and Schuchmann, 1992; Jin et al., 2006). Jin
et al. (2006) suggested that a mixture of potassium iodide and potassium iodate has
advantages over uridine for use with LP reactors.
Deguchi et al. (2005) demonstrated the use of free chlorine at a concentration of about 1 mg/l
as an actinometer (for their tests they used tap water to which no chlorine was added). They
described the UV-chlorine reaction as first order. The kinetics were relatively slow and less
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
94
than 10% of the chlorine reacted (C/Co > 0.9 where C = outlet concentration and Co = inlet
concentration). In accordance with the dispersion model for non-ideal hydraulics (Levenspiel,
1972) they expected in these circumstances that the measured performance would be similar
to that of an ideal plug-flow reactor, and therefore it would not be necessary to model the
hydraulics of the reactor to use chlorine for dose validation. They found that the measured
average intensity derived from the chlorine actinometry was within 1% of the average intensity
calculated by the point source summation method, which was consistent with expectations,
and concluded that the combination of actinometry with free chlorine and tracer testing might
be useful for on-site validation. Chlorine would be convenient for on-site testing given the
absence of any concerns regarding discharge of the actinometer chemical into supply, but
one potential difficulty is achieving a reliable result when dependent upon differences in
chlorine concentration of < 0.1 mg/l. It is also not known whether the condition of less than
10% chlorine reaction is generally to be expected in commercial UV reactors.
A limitation of actinometry is that when used directly it may overestimate the actual dose that
microorganisms would receive, because of non-ideal hydraulics within the reactor. Dykstra
et al. (2002) developed an axial-dispersion model, calibrated using tracer tests, to improve the
actual dose estimated from ferrioxalate actinometry. However, they described the ferrioxalate
photoreaction as zero order. In developing the general form of the dispersion model for non-
ideal hydraulics, Levenspiel (1972) noted that backmixing does not affect performance for
zero-order reactions. On that basis, non-ideal hydraulics should impact reactor performance
for zero-order reactions only if dead zones are created which reduce effective residence time.
B2.4 Numerical modelling
Numerical modelling of UV dose distributions generally combines a CFD (computational fluid
dynamics) representation of the hydraulics with a numerical representation of the irradiance
field (Blatchley et al., 2008). The UVDGM recognises the potential of such models, but is
cautious (explicitly in the context of the state-of-the-art at the time the manual was published,
2006) about accepting them for validation purposes, three areas of concern being:
the absence of a standard approach to assessing model credibility;
the absence of consensus for which modelling approaches are most appropriate for the
specific case of UV reactors;
the limited pool of multi-disciplinary expertise required to model UV reactors.
It states that modelling “…should not be used in lieu of validation for prediction of the actual
RED magnitude as a means of granting pathogen inactivation credit”. Numerical modelling (a
combination of CFD and Bolton’s extended PSS method) was used, however, to derive the
RED bias factors for the UVDGM.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
95
Modelling has developed further since 2006 (e.g. Ho, 2009). Deguchi et al. (2005) used CFD
modelling to estimate the dose distribution of a pilot-scale reactor and derived an estimate of
actual dose that was 94.9% of the value derived from biodosimetry using Bacillus subtilis
spores. Dicks and Sief (2009) have described how a manufacturer (Aquafides GmbH) is using
numerical modelling during product development to predict performance in validation testing.
They presented results showing agreement between predicted performance and ÖNORM
validated performance of better than ±2.5% in waters of high transmittance (> 90%) but a
greater deviation of -15.5% at a lower transmittance (79%).
B2.5 Dyed microspheres
Blatchely et al. (2008) noted that numerical models can yield accurate predictions, but gave
similar reasons to those listed in the UVDGM for why validation based on measured
behaviour is still preferred. However, they have been developing a practical method which
can characterise UV dose distribution and be used to validate numerical models, using dyed
microspheres.
The method is based on the attachment of a UV-sensitive compound to microspheres. The
compound reacts when exposed to UV light to yield a product that is fluorescent, which
makes the extent of reaction readily measureable. The microspheres are similar in size and
density to the pathogens of concern, so follow similar trajectories in the prevailing hydraulic
conditions. Dose delivery to the microspheres should therefore mimic that to pathogens, such
that the method yields a representative measurement of the dose distribution. Blatchely et al.
(2008) have demonstrated this method on a full-scale LP UV reactor, and extended it for use
with MP reactors (Shen et al., 2009). The developers are confident that the method, as well
as being useful for validating numerical models, also represents an alternative to
biodosimetry. In principle, validating with dyed microspheres should result in smaller safety
factors (there will still be experimental uncertainty, but the method should allow for a lower
RED bias factor) which means smaller reactors, lower capital costs and lower operating costs.
The Water Research Foundation (formerly AwwaRF), in collaboration with the AWWA and
USEPA, has funded two projects (#4112, #4217) with the objective of developing and
demonstrating this method in the context of the UVDGM, project completion being due in
2015.
B2.6 Comparison with biodosimetry
Microorganisms passing through a UV reactor are discrete particles which follow trajectories
that are dependent on the hydraulics and the physical properties of the microorganisms. A
dissolved chemical, assuming it is well mixed, exists everywhere throughout the water. Thus a
chemical actinometer does not represent exactly the same physical system as a
biodosimeter. Under plug flow conditions, or when the combination of reaction rate, residence
time and extent of backmixing is such that deviations from plug flow are mitigated, the mean
dose determined from actinometry should equate to that derived from the average intensity
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
96
given by the PSS method in combination with the hydraulic residence time, but this will not be
the same as the dose determined using biodosimetry (as demonstrated by Deguchi et al.,
2005). Under other conditions, some form of dispersion model is needed to interpret
actinometry results (as indicated by Dykstra et al., 2002).
There might be some benefit in applying the PSS method to older non-validated systems, in
that the mean dose derived by this method will be greater than the actual dose. Actinometry
using chlorine might also be useful for the same purpose.
Numerical modelling requires a representation of the irradiance field, for which the PSS
method remains useful. It is difficult to envisage numerical modelling being accepted for
validation without further demonstration of reliability and agreement of protocols. However,
there might be scope for using this approach for older non-validated systems.
Dyed microspheres, in principle, should provide an equivalent of biodosimetry but with the
advantage of relative ease of quantification. Whether the method is cost effective remains to
be seen.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
97
Appendix C UV sensitivity of micro-organisms
Examples of inactivation by UV, for a range of micro-organisms, are given in Table C.1 and
Table C.2.
Table C.1 UV dose (mJ/cm2) for inactivation of protozoa and viruses
Target Log10 Inactivation
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Protozoa
Giardia cysts1
1.5 2.1 3.0 5.2 7.7 11 15 22
Cryptosproridium
cysts1
1.6 2.5 3.9 5.8 8.5 12 15 22
Viruses
‘Viruses’1
39 58 79 100 121 143 163 186
Adenovirus type 402
56 111 167
Poliovirus2
7 15 22 30
Adenovirus type 413
112
Hepatitis A3
21
Coxsackie virus B53
36
Poliovirus type 13
27
Rotavirus SA113
36
Murine norovirus4 7.3 14.6 21.9 29.2
Feline calicivirus4 6.3 12.5 18.8 25
Echovirus 124 7.4 14.8 22.2 29.6
1 USEPA (2006)
2 Hijnen WAM, Beerendonk EF and Medema GJ. (2006)
3 Bolton JR and Cotton CA. (2008)
4 Park GW, Linden KG and Sobsey MD. (2011)
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
98
Table C.2 UV dose (mJ/cm2) for inactivation of spores and bacteria
Target Log10 inactivation
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Spores
B. subtilus spores1
28 39 50 62
B. subtilus spores2 56 111 167 222
Bacteria
Campylobacter jejuni3
4.6
Campylobacter jejuni2 3 7 10 14
Clostridium perfringens3
23.5
Clostridium perfringens2
45 95 145
Enterobacter cloacae3
10 (33)
Enterocolitica faecium3
17 (20)
E. coli1
3 4.8 6.7 8.4
E. coli O157:H73
6 (25)
E. coli O1572 5 9 14 19
E. coli wild type3
8.1
E. coli wild type4 6 - 8.5
E. coli wild type2 5 9 14 19
Klebsiella pneumoniae3
20 (31)
Legionella pneumophila3
9.4
Legionella pneumophila2
3 - 8 6 - 15 8 - 23 11 - 30
Mycobacterium
smegmatis3 20 (27)
Pseudomonas
aeruginosa3 11 (19)
Salmonella typhi3
8.2
Salmonella typhi2 6 12 17 51
Shigella dysenteriae
ATTC290273 3
Shigella dysenteriae2 3 5 8 11
Shigella sonnei2 6 13 19 26
Streptococcus faecalis3
11.2
Streptococcus faecalis2
9 16 23 30
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
99
Target Log10 inactivation
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Vibrio cholerae3
2.9 (21)
Vibrio cholerae2 2 4 7 9
1 USEPA (2010)
2 Hijnen WAM, Beerendonk EF and Medema GJ. (2006)
3 Bolton JR and Cotton CA. (2008) - values in brackets include photoreactivation data
4 Bucheli-Witschel, Bassin C and Egli T. (2010)
The inactivation values for bacteria proposed by Hijnen et al. (2006) are higher than those
reported by other sources. Hijnen et al., reviewed the relative UV sensitivity of seeded and
environmental (wild) micro-organisms, and inflated doses required for a given log removal
derived using the former by a factor, unspecified for individual bacteria but typically 3, to
account for the lower sensitivity of the latter.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
100
Appendix D UV Suppliers
D1 Aquacure
http://aquacure.co.uk/
System Rating and connection
Model
USEPA std
flow rate
(40 mJ/cm2)
Inlet /
outlet
port size
Other related information
3 Series
ACUV153D 8 l/min ¾” BSP
male
thread
ACUV303D 19 l/min
ACUV553D 36 l/min
ACUV554D 51 l/min 1” BSP
male
thread
Economy Stainless Steel Series
ACNUVS6S 2 l/min ¼”nptf
ACNUV14S 6 l/min ¼”nptf
ACNUV24S 16 l/min ½”nptf
ACNUV32S 23 l/min ½”nptf
ACNUV39S 46 l/min ¾”nptf
6 Watt Plastic Ultra Violet Steriliser (ACUV62P)
ACUV62P 3.01 l/min ¼ “ push
fit
Electrical Requirements
Model Voltage System power
consumption Other related information
3 Series
ACUV153D 15 W
ACUV303D 30 W
ACUV553D 55 W
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
101
Model Voltage System power
consumption Other related information
ACUV554D 55 W
Economy Stainless Steel Series
ACNUVS6S 8 W
ACNUV14S 15 W
ACNUV24S 21 W
ACNUV32S 39 W
ACNUV39S 65 W
6 Watt Plastic Ultra Violet Steriliser (ACUV62P)
ACUV62P 220V 50Hz 0.140 A
Lamp specification
Model Type Lamp life
(h)
Lamp power
consumption Other related information
3 Series
ACUV153D ACUV15 8,760 15 W low pressure mercury discharge
lamps ACUV303D ACUV30 30 W
ACUV553D ACUV55 55 W
ACUV554D ACUV55 55 W
Economy Stainless Steel Series
ACNUVS6S ACUVLHR60 8,760 8 W low pressure mercury discharge
lamps ACNUV14S ACUVLHE120 15 W
ACNUV24S ACUVLHC360 21 W
ACNUV32S ACUVLHC720 39 W
ACNUV39S ACUVLFC15 65 W
6 Watt Plastic Ultra Violet Steriliser (ACUV62P)
ACUV62P ACUV6A 4,320 0.140 A low pressure mercury discharge
lamps
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
102
Pre-treatment requirements
Model Filtration UV-T Other related information
3 Series
ACUV153D 5 µm
filtration
generally
required
98% 5 – 35oC max 10 bar
Recommended WQ limits:
Iron < 0.2 mg/l
Manganese < 0.05 mg/l
Total Hardness < 7 gpg
Turbidity < 2 NTU
Colour / Tannins *
ACUV303D
ACUV553D
ACUV554D
Economy Stainless Steel Series
ACNUVS6S 5 µm
filtration
generally
required
98% 5 – 35oC max 8 bar
Recommended WQ limits:
Iron < 0.2 mg/l
Manganese < 0.05 mg/l
Total Hardness < 7gpg
Turbidity < 2 NTU
Colour / Tannins *
ACNUV14S
ACNUV24S
ACNUV32S
ACNUV39S
6 Watt Plastic Ultra Violet Steriliser (ACUV62P)
ACUV62P 5 µm
filtration
generally
required
98% 5 – 35oC max 8 bar
Recommended WQ limits:
Iron < 0.2 mg/l
Manganese < 0.05 mg/l
Total Hardness < 7 gpg
Turbidity < 2 NTU
Colour / Tannins *
Maintenance
Model Items for regular replacement Comment
3 Series
ACUV153D Lamps
Quartz Sleeves
ACUV303D
ACUV553D
ACUV554D
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
103
Model Items for regular replacement Comment
Economy Stainless Steel Series
ACNUVS6S Lamps
Quartz Sleeves
ACNUV14S
ACNUV24S
ACNUV32S
ACNUV39S
6 Watt Plastic Ultra Violet Steriliser (ACUV62P)
ACUV62P Lamps
D2 Aquafine Corp. (CA, USA)
http://www.aquafineuv.com/
Parent company: Trojan technologies
http://www.trojantechnologies.com/
System Rating and connection
Model
Flow rate
(>30 mJ/cm2)
(94% UVT)
Other std flow
rate
(>30 mJ/cm2)
(99% UVT)
Inlet /
outlet
port
size
Other related information
CSL Series
4R 9 m3/h 11 m
3/h 2” No. of Lamps 4
6R 14 m3/h 18 m
3/h No. of Lamps 6
8R 18 m3/h 23 m
3/h 3” No. of Lamps 8
12R 30 m3/h 36 m
3/h No. of Lamps 12
8R 60 38 m3/h 46 m
3/h 4” No. of Lamps 8
10R 60 49 m3/h 59 m
3/h No. of Lamps 10
12R 60 59 m3/h 72 m
3/h No. of Lamps 12
Optima HX
02 ADS 9 m3/h 10 m
3/h
(99% UVT)
2” No. of Lamps 2
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
104
Model
Flow rate
(>30 mJ/cm2)
(94% UVT)
Other std flow
rate
(>30 mJ/cm2)
(99% UVT)
Inlet /
outlet
port
size
Other related information
SP and SL Series
SP1 0.22 m3/h
(22 mJ/cm2)
0.22 m3/h
(22 mJ/cm2)
3/8” No. of Lamps 1
SP2 0.44 m3/h
(22 mJ/cm2)
0.44 m3/h
(22 mJ/cm2)
3/8”
SL 10A 0.68 m3/h
(30 mJ/cm2)
0.9 m3/h
(30 mJ/cm2)
½”
SL 1 2.3 m3/h
(30 mJ/cm2)
2.7 m3/h
(30 mJ/cm2)
1
MP2 SL 4.5 m3/h
(30 mJ/cm2)
5.5 m3/h
(30 mJ/cm2)
1½” No. of Lamps 2
Electrical Requirements
Model Voltage
System power
consumption (Watts)
120 V or (240) V AC
Other related information
CSL Series
4R 240V 50-60Hz
or
120V 50-60Hz
190
6R 265
8R 370
12R 540
8R 60 590
10R 60 730
12R 60 865
Optima HX
02 ADS 240V 50-60Hz
or
120V 50-60Hz
265
SP and SL Series
SP1 240V 50-60Hz
or
17 or (38)
SP2 46 or (84)
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
105
Model Voltage
System power
consumption (Watts)
120 V or (240) V AC
Other related information
SL 10A 120V 50-60Hz 48
SL 1 61 or (65)
MP2 SL 96 or (151)
Lamp specification
Model Type Lamp life (h) Lamp power
consumption Other related information
CSL Series
4R LP 8,000
(for rated
output)
47.5 V (190 V/ 4)
6R 44 V (265 V/ 6)
8R 46.25 V (370 V/ 8)
12R 45 V (540 V/ 12)
8R 60 73 V (590 V/ 8)
10R 60 73 V (730 V/ 10)
12R 60 72 V (865 V/ 12)
Optima HX
02 ADS LPHO 9,000 132.5 V (265 V/ 2)
SP and SL Series
SP1 LP
8,000 (for
rated output)
17 1 lamp / unit
SP2 46
SL 10A 48
SL 1 61
MP2 SL 48 V (96 V/ 2) 2 lamps / unit
Pre-treatment requirements
Model Filtration UV-T Other related information
CSL Series
4R None stated Rated for
94% or 99%
Operating water temperature: 10-38oC
Max. operating pressure: 8 Bar
6R
8R
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
106
Model Filtration UV-T Other related information
12R
8R 60
10R 60
12R 60
Optima HX
02 ADS None stated Rated for
94% or 99%
SP and SL Series
SP1 None stated Rated for
94% or 99%
Operating water temperature: 4-27oC
Max. operating pressure: 8 Bar
SP2
SL 10A Operating water temperature: 4-27oC
Max. operating pressure: 10 Bar
SL 1
MP2 SL Operating water temperature: 10-38oC
Max. Operating Pressure: 10 Bar
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
CSL Series
4R Lamp status
indicator
Running
time
indicator
UV intensity
and
temperature
monitor /
alarm
(optional)
Lamp failure alarm (optional)
6R
8R
12R
8R 60
10R 60
12R 60
Optima HX
02 ADS Lamp status
indicator
Running
time
indicator
UV intensity
and
temperature
monitor /
alarm
Lamp failure alarm (optional)
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
107
Model Feature 1 Feature 2 Feature 3 Other related information
(optional)
SP and SL Series
SP1 Lamp status
indicator
Running
time
indicator
UV intensity
and
temperature
monitor /
alarm
(optional)
Lamp failure alarm (optional)
SP2
SL 10A
SL 1
MP2 SL
Maintenance
Model Items for regular replacement Other related information
CSL Series
4R UV lamp (8,000 h), quartz sleeve (12 months) Ballast (not routine)
6R
8R
12R
8R 60
10R 60
12R 60
Optima HX
02 ADS UV lamp (9,000 h), quartz sleeve (2 years) Ballast (not routine)
SP and SL Series
SP1 UV lamp (8,000 h), quartz sleeve (12 months) Ballast (not routine)
SP2
SL 10A
SL 1
MP2 SL
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
108
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
CSL Series
4R - - - Optional auto-off at high
temperature (77°C), serves to
reduce on/off cycling due to no
flow.
6R
8R
12R
8R 60
10R 60
12R 60
Optima HX
02 ADS - - - Optional auto-off at high
temperature (77°C), serves to
reduce on/off cycling due to no
flow.
SP and SL Series
SP1 - - - Hot water sanitizer available.
SP2
SL 10A
SL 1
MP2 SL
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
109
D3 Bio-UV
http://www.bio-uv.co.uk
System Rating and connection
Model
Flow rate
(40 mJ/cm2)
(calculated at end of lamp
life with 98% UVT)
Inlet /
outlet
port
size
Other related information
UV Home Series Reactors
UV HOME 2 2.2 m3/h ¾” -
UV HOME 3 3.2 m3/h ¾” -
IBP HO + Series Reactors
IBP 10 HO + 4.6 m3/h 1” -
IBP 30 HO + 6.6 m3/h 1½“ -
IBP 40 HO + 9.3 m3/h 1½” -
IBP 2150 HO + 13 m3/h 2” -
IBP 3150 HO + 22 m3/h 2” -
IBP 4205 HO + 39 m3/h 2½” -
IBP 5205 HO + 54 m3/h 2½” -
Electrical Requirements
Model Voltage
System power
consumption
(Watts)
Other related information
UV Home Series Reactors
UV HOME 2 220-240V
50-60 Hz
36 1A fuse
UV HOME 3 61
IBP HO + Series Reactors
IBP 10 HO + 220-240V
50-60 Hz
96 2 A fuse
IBP 30 HO +
IBP 40 HO + 191
IBP 2150 HO +
IBP 3150 HO + 287
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
110
Model Voltage
System power
consumption
(Watts)
Other related information
IBP 4205 HO + 382 4 A fuse
Electrical cabinet ventilation IBP 5205 HO + 478
Lamp specification
Model Type Lamp life (h) Lamp power
consumption Other related information
UV Home Series Reactors
UV HOME 2 HO 13,000 33 W Male threaded
UV HOME 3 55 W
IBP HO + Series Reactors
IBP 10 HO + HO 13,000 87 W Male threaded
IBP 30 HO + 87 W
IBP 40 HO + 105 W
IBP 2150 HO + 2 x 87 W
IBP 3150 HO + 3 x 87 W
IBP 4205 HO + 4 x 87 W
IBP 5205 HO + 5 x 87 W
Pre-treatment requirements
Model Filtration
UV-T (T-10)
With Monitor
PRO3
Other related information
UV Home Series Reactors
UV HOME 2 2 Filters Kit
UV HOME 2 Sanitizer
Washable screen
filter 60 μm
Cartridge filter 10 μm
or
3 Filters Kit
UV HOME 2 Sanitizer
85% 2 or 3 filters depending on the
water quality UV HOME 3
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
111
Model Filtration
UV-T (T-10)
With Monitor
PRO3
Other related information
Washable screen
filter 60 μm
Cartridge filter 10 μm
Carbon Filter
IBP HO + Series Reactors
IBP 10 HO + - 85% -
IBP 30 HO +
IBP 40 HO +
IBP 2150 HO +
IBP 3150 HO +
IBP 4205 HO +
IBP 5205 HO +
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
UV Home Series Reactors
UV HOME 2 Optional
PTFE UV
sensor and
PRO3
monitor
offering data
reporting by
a diode and
contact type
alarm
- - PTFE sensor which limits the
fouling of the sensor and as a
consequence the maintenance
operations.
independent electrical cabinet sold
with a 1.5 m cable:
easy installation
no overheating risk
easy maintenance: yearly
calibration with a setting screw on
the panel
UV HOME 3
IBP HO + Series Reactors
IBP 10 HO + Optional
PTFE UV
sensor and
PRO3
monitor
offering data
- - PTFE sensor which limits the
fouling of the sensor and as a
consequence the maintenance
operations.
independent electrical cabinet sold
with a 1.5 m cable:
IBP 30 HO +
IBP 40 HO +
IBP 2150 HO +
IBP 3150 HO +
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
112
Model Feature 1 Feature 2 Feature 3 Other related information
IBP 4205 HO + reporting by
a diode and
contact type
alarm
easy installation
no overheating risk
easy maintenance: yearly
calibration with a setting screw on
the panel
IBP 5205 HO +
Maintenance
Model Items for regular replacement Other related information
UV Home Series Reactors
UV HOME 2 UV lamp
Other replacement parts may include UV
monitor
-
UV HOME 3
IBP HO + Series Reactors
IBP 10 HO + UV lamp
Other replacement parts may include UV
monitor
Use of single-base lamps,
patented sealing system and
vertical design for an easy
maintenance
IBP 30 HO +
IBP 40 HO +
IBP 2150 HO +
IBP 3150 HO +
IBP 4205 HO +
IBP 5205 HO +
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
UV Home Series Reactors
UV HOME 2 None None Volt-free
contacts
available on
the card
within the
electrical
cabinet
allowing an
Alarm:
- green diode: working ok
- orange diode: pre-alarm
(threshold <75%)
- red diode: main-alarm
(threshold <50%)
UV HOME 3
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
113
Model Flow
control
UV
exposure
UV output
flow control Other related information
alarm report
or the cabling
of an
electronic
valve
IBP HO + Series Reactors
IBP 10 HO + None
None
volt-free
contacts
available on
the card
within the
electrical
cabinet
allowing an
alarm report
or the cabling
of an
electronic
valve
Alarm:
- green diode: working ok
- orange diode: pre-alarm
(threshold <75%)
- red diode: main-alarm
(threshold <50%)
IBP 30 HO +
IBP 40 HO +
IBP 2150 HO +
IBP 3150 HO +
IBP 4205 HO +
IBP 5205 HO +
D4 DaRo UV Systems
http://www.darouv.co.uk/
System Rating and connection
Model
USEPA std
flow rate
(40 mJ/cm2)
Other std flow
rate
(30 mJ/cm2)
Inlet /
outlet
port
size
Other related information
Saphir Systems
Saphir 1 13.5 l/min 18 l/min ¾” Single ended UV Lamps.
5 micron pre filter units recommended
supplied by DaRo systems but not
included.
Tailor made UV systems available.
Saphir 2 19 l/min 25.5 l/min ¾”
Saphir 3 33 l/min 45 l/min 1”
Saphir 7 66.5 l/min 90 l/min 1”
Saphir 10 120 l/min 160 l/min 1½”
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
114
Model
USEPA std
flow rate
(40 mJ/cm2)
Other std flow
rate
(30 mJ/cm2)
Inlet /
outlet
port
size
Other related information
Saphir + system offers more telemetry
and control options.
ECO Series
ECO 1 8 l/min
10.5 l/min ¾” Designed primarily for longevity. Should
last up to 20 years with only minor
components possibly having to
be replaced.
ECO 2 19 l/min 27.5 l/min ¾”
ECO 3 36 l/min
46.5 l/min ¾”
ECO 5 51 l/min 70 l/min 1”
Electrical Requirements
Model Voltage System power
consumption Other related information
Saphir Systems
Saphir 1 240V 50Hz
single phase
15 W ¾ inch bsp male thread
Saphir 2 25 W ¾ inch bsp male thread
Saphir 3 45 W ¾ inch bsp male thread
Saphir 7 75 W 1 inch bsp male thread
Saphir 10 75 W 1.5 inch bsp male thread
ECO Series
ECO 1 240V 50Hz
single phase
28 W ¾ inch bsp male thread
ECO 2 38 W ¾ inch bsp male thread
ECO 3 75 W ¾ inch bsp male thread
ECO 5 95 W 1 inch bsp male thread
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
115
Lamp specification
Model Type Lamp life
(h)
Lamp power
consumption Other related information
Saphir Systems
Saphir 1 Low
pressure
(254 nm)
8,760 14 W GER15SE
Saphir 2 21 W GER25SE
Saphir 3 48 W GER25SEXO
Saphir 7 75 W GER36SEXO
Saphir 10 75 W GER36SEXO
ECO Series
ECO 1 Low
pressure
(254 nm)
8,000 15 W
ECO 2 30 W
ECO 3 55 W
ECO 5 55 W
Pre-treatment requirements
Model Filtration UV-T Other related information
Saphir Systems
Saphir 1 5 µm pre-
filter
98% Max. 10 bar
Recommended WQ units:
5 µm pre-filter
(recommended but not included)
Saphir 2
Saphir 3
Saphir 7
Saphir 10
ECO Series
ECO 1 5 µm pre-
filter
98% Max. 15 bar
Recommended WQ units:
5 µm pre-filter
(recommended but not included)
ECO 2
ECO 3
ECO 5
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
116
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related
information
Saphir Systems
Saphir
1-10
Lamp on indicator Lamp/electrical
failure
Saphir + UV Lamp status
Indicator:
Lamp status
indicator, Remote
lamp on indicator
via internal volt
free contacts.
The display gives
a hidden
indication of the
age of the lamp.
The "Lamp On"
and "Lamp
Status" indicators
will blink off for a
half second
period once every
minute. The
number of blinks
show how many
year quarters
(three months)
have passed,
before the "Lamp
Status" indicator
alternates from
green to red in
the twelfth month.
UV Lamp Running
Indicator:
The control box has
a blue "lamp
running" indicator
on the front panel
display, which is on
if and only if lamp is
running correctly.
System status
indicators:
The system status
indicators will tell you
how long the system
has been running with
the current UV lamp
i.e. how long before a
new replacement UV
lamp is required. It will
also tell you whether
the power is on to the
unit, and whether the
UV lamp is working or
not.
Volt free Contact
output for remote
display of lamp
indicator:
This is for remote
monitoring of lamp
status, via a plug
and socket (which is
sealed and capped
when not in use.
Both capped and
plugged
configurations are
rated to >IP65).
ECO Series
ECO 1 UV monitors
available
ECO 2
ECO 3
ECO 5
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
117
Maintenance
Model Items for Regular Replacement Other related information
Saphir Systems
Saphir 1 Clean/ Descale quartz sleeve.
Replace lamp and o rings
Saphir 2
Saphir 3
Saphir 7
Saphir 10
ECO Series
ECO 1 Clean / Descale High purity
quartz sleeve. Replace lamp and
o rings
Typically the system will only cost around 14p per
day to run
ECO 2
ECO 3
ECO 5
Operational control
Model Flow
Control
UV
Exposure
UV output
flow
control
Other related information
Saphir Systems
Saphir
1-10
Flow
regulators
available.
Alarms
available.
Saphir +
ECO Series
ECO 1 Flow
regulators
available.
Alarms
available.
ECO 2
ECO 3
ECO 5
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
118
D5 Filpumps
http://shop.filpumps.co.uk/
System Rating and connection
Model
USEPA std
flow rate
(40 mJ/cm2)
Other std flow
rate
Inlet / outlet
port size Other related information
Azzurri systems range
1 10 l/min ½” BSP Male Thread
2 20 l/min ¾” BSP
3 45 l/min 1” BSP
4 60 l/min 1” BSP
5 85 l/min 1½” BSP
6 100 l/min 1½” BSP
7 200 l/min 1½” BSP
Electrical Requirements
Model Voltage System power
consumption Other related information
1 220/240V
50/60Hz
- single phase
2 - single phase
3 - single phase
4 - single phase
5 - single phase
6 - single phase
7 - single phase
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
119
Lamp specification
Model Type Lamp life (h) Lamp power
consumption Other related information
Azzurri systems range
1 FPSUV403 9,000 hours
(~1 year)
-
2 FPSUV405 9,000 hours
(~1 year)
-
3 FPSUV412
9,000 hours
(~1 year)
-
4 FPSUV440 9,000 hours
(~1 year)
-
5 FPSUV480 9,000 hours
(~1 year)
-
6 FPSUV550 9,000 hours
(~1 year)
-
7 FPSUV80/2 9,000 hours
(~1 year)
-
Pre-treatment requirements
Model Filtration UV-T Other related information
Azzurri systems range
1 99% per
1 cm
2
3
4
5
6
7
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
120
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
Azzurri systems range
1 Lamp on /
Fail indicator
2
3
4
5
6
7
Maintenance
Model Items for regular replacement Comment
Azzurri systems range
1 Replace lamp / and o rings
2
3
4
5
6
7
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
Azzurri systems range
1 - - - -
2
3
4
5
6
7
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
121
D6 Hanovia
http://www.hanovia.com/
Parent company: Halma plc
http://www.halma.com/
System Rating and connection
Model
Flow rate
(26 J/cm2)
(95% UVT)
Other std flow
rate
(120 J/cm2)
(95% UVT)
Inlet /
outlet
port
size
Other related information
Pureline D Range
D 0007 7.3 m3/h 1.3 m
3/h 40 mm Pureline D 0083-D 0850 models for flow
rates of 84 – 1000 m3/h
Cabinet dimensions allow for door
isolator, cable entry, bracket space and
fan space on D 0089
D 0013 12.7 m3/h 2.4 m
3/h 50 mm
D 0023 23 m3/h 4.6 m
3/h 50 mm
D 0047 46 m3/h 9 m
3/h 80 mm
D 0089 89 m3/h 15 m
3/h 150 mm
Electrical Requirements
Model Voltage
System power
consumption
(Watts)
Other related information
Pureline D Range
D 0007 230V or
115V
(except D
0089)
50/60Hz
80 -
D 0013 140
D 0023 270
D 0047 270
D 0089 550
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
122
Lamp specification
Model Type Lamp life (h) Lamp power
consumption Other related information
Pureline D Range
D 0007 Low
pressure
amalgam
/ high
purity
quartz
12,000-
16,000
- Process (mating) connections:
Flange DN series PN16 rated D 0013
D 0023
D 0047
D 0089
Pre-treatment requirements
Model Filtration UV-T Other related information
D 0007 - >70% -
D 0013
D 0023
D 0047
D 0089
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
Pureline D Range
D 0007 Remote
mode
Warning and
trip
messages:
Lamp fail
Low UV %
intensity
Unit Tripped
UV intensity
%
Total hours
run
2 line x 16 character backlit LCD
with indication of System Status D 0013
D 0023
D 0047
D 0089
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
123
Maintenance
Model Items for regular replacement Other related information
Pureline D Range
D 0007 UV Lamps
D 0013
D 0023
D 0047
D 0089
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
Pureline D Range
D 0007 Remote
start/stop
Lamp
on/off
Low UV
warning
Safety Features:
Door interlocked cabinet isolator
Separate door lock
Resettable circuit breaker
Power on LED
D 0013
D 0023
D 0047
D 0089
D7 Hydrotec
http://www.hydrotec.co.uk/
System Rating and connection
Model Flow rate
(250 J/m2)
Other std flow
rate
(400 J/m2)
Inlet /
outlet
port
size
Other related information
HydroPUR
2E 4.59 m³/h 2.87 m³/h 1” BSP
External
Thread
Operating Pressure 0 – 10bar
Pressure Loss (at design
flowrate) <100mbar
5E 7.21 m³/h 4.51 m³/h 1½”
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
124
Model Flow rate
(250 J/m2)
Other std flow
rate
(400 J/m2)
Inlet /
outlet
port
size
Other related information
BSP
External
Thread
10E 14.9 m³/h 9.25 m³/h 2” BSP
External
Thread
Electrical Requirements
Model Voltage
System power
consumption
(Watts)
Other related information
HydroPUR
2E 230V
50Hz
60 W A 230v/1Ph/50Hz fused power supply is to be
provided for the unit. 5E 100 W
10E 130 W
Lamp specification
Model Type Lamp life
(h)
Lamp power
consumption Other related information
HydroPUR
2E Ecolux
20N
Mercury
LP with
indium-
amalgam
8,000 17 -
5E Ecolux
30N
27
10E Ecolux
40N
38
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
125
Pre-treatment requirements
Model Filtration UV-T Other related information
HydroPUR
2E - - Water Temperature 5 – 50°C
Ambient Temperature (max.) 40°C
Operating Pressure 0 – 10bar
Pressure Loss (at design flowrate)
<100mbar
The flow path of the water should
guarantee shadow-free
radiation of the water.
5E
10E
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
HydroPUR
2E An LED
display
indicates UV
intensity for
visual
indication
on the
control box.
- - Volt free connections to a BMS
system
are provided.
5E
10E
Maintenance
Model Items for regular replacement Other related information
HydroPUR
2E UV lamps
Quartz sleeves
-
5E
10E
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
126
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
HydroPUR
2E - - - -
5E
10E
D8 LIFF
http://www.lifffilters.co.uk/lifffilters/index.asp
Parent company: BWT Ltd.
http://www.bwt-uk.co.uk/
System Rating and connection
Model
USEPA std
flow rate
m3/h
(40 mJ/cm2)
Other std flow
rate
m3/h
(30 mJ/cm2)
Inlet /
outlet
port
size
Other related information
Liff UV filtration units
P15N Ultra
Violet
Disinfection
Unit
8 1” LIFF recommend a 5 µm filter.
Completely eliminates e-coli, cysts and
bacteria as well as coliforms,
campylobacter, legionella and
pseudomonas.
P30N Ultra
Violet
Disinfection
19 1”
P55N Ultra
Violet
Disinfection
54 1” LIFF recommend a 5 µm filter.
Completely eliminates e-coli, cysts and
bacteria.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
127
Model
USEPA std
flow rate
m3/h
(40 mJ/cm2)
Other std flow
rate
m3/h
(30 mJ/cm2)
Inlet /
outlet
port
size
Other related information
AQA Pure and AQA Pure + Range
AQA Pure 1 13.5 l/min 18 l/min ¾” BSP
male
thread
AQA Pure 2 22.5 l/min 30 l/min ¾” BSP
male
thread
AQA Pure 3 40 l/min 53 l/min ¾” BSP
male
thread
AQA Pure 4 45 l/min 60 l/min 1” BSP
male
thread
AQA Pure 7 90 l/min 118 l/min 1” BSP
male
thread
AQA Pure 10 120 l/min 160 l/min 1½”
BSP
male
thread
Electrical Requirements
Model Voltage
System power
consumption
(Kw) approx.
Other related information
P15N Ultra Violet
Disinfection Unit
230V 50Hz
0.015 UV electrical standard CE
P30N Ultra Violet
Disinfection
0.030
P55N Ultra Violet
Disinfection
0.055
AQA Pure and AQA Pure + Range
AQA Pure 1 230V 50Hz 0.015
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
128
Model Voltage
System power
consumption
(Kw) approx.
Other related information
AQA Pure 2 0.025
AQA Pure 3 0.025
AQA Pure 4 0.036
AQA Pure 7 0.036
AQA Pure 10 0.036
Lamp specification
Model Type Lamp life (h) Lamp power
consumption Other related information
Liff UV filtration units
P15N Ultra Violet
Disinfection Unit
LP 6,500-8,000 15 Lamp should be changed yearly to
maintain optimum performance.
P30N Ultra Violet
Disinfection
30
P55N Ultra Violet
Disinfection
55
AQA Pure and AQA Pure + Range
AQA Pure 1 LP 8,760 15
AQA Pure 2 25
AQA Pure 3 25
AQA Pure 4 36
AQA Pure 7 36
AQA Pure 10 36
Pre-treatment requirements
Model Filtration UV-T Other related
information
P15N Ultra Violet
Disinfection Unit
NP1 10" Filter Housing c/w NSW5 5 µm
filter cartridge (to be purchased
separately)
90% @
40 mJ/cm2
0-40°C, max 5 bar
P30N Ultra Violet
Disinfection
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
129
Model Filtration UV-T Other related
information
P55N Ultra Violet
Disinfection
AQA Pure and AQA Pure + Range
AQA Pure 1 FSS114 30" Filter Housing c/w SB30-5 5
µm filter cartridge (to be purchased
separately)
98% @
40 mJ/cm2
0-40°C, max
10 bar
AQA Pure 2
AQA Pure 3
AQA Pure 4
AQA Pure 7
AQA Pure 10
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
P15N Ultra Violet
Disinfection Unit
n/a
P30N Ultra Violet
Disinfection
P55N Ultra Violet
Disinfection
AQA Pure AQA Pure +
Standard - Lamp running indicator only AQA PURE + RANGE offers Lamp Life Clock,
Lamp Status Indicator & Volt Free Contact
Maintenance
Model Items for regular replacement Comment
Liff UV filtration units
P15N Ultra Violet
Disinfection Unit
Low pressure lamp every 9 to 12 months
FP20N Ultra Violet
Disinfection Unit
P30N Ultra Violet
Disinfection
P55N Ultra Violet
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
130
Model Items for regular replacement Comment
Disinfection
AQA Pure and AQA Pure + Range
AQA Pure 1 Low pressure lamp every 12 months
AQA Pure 2
AQA Pure 3
AQA Pure 4
AQA Pure 7
AQA Pure 10
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
Liff UV filtration units
P15N Ultra Violet
Disinfection Unit
Flow
restricted to
0.13 l/s
P30N Ultra Violet
Disinfection
Flow
restricted to
0.32 l/s
P55N Ultra Violet
Disinfection
Flow
restricted to
0.9 l/s
AQA Pure and AQA Pure + Range
AQA Pure 1 Flow
restricted to
13.5 l/min
AQA Pure 2 Flow
restricted to
22.5 l/min
AQA Pure 3 Flow
restricted to
40 l/min
AQA Pure 4 Flow
restricted to
45 l/min
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
131
Model Flow
control
UV
exposure
UV output
flow control Other related information
AQA Pure 7 Flow
restricted to
90 l/min
AQA Pure 10 Flow
restricted to
120 l/min
D9 Prosep
http://www.prosep.co.uk/
Parent company: Parker Process Filtration Division
http://www.parker.com/
System Rating and connection
Model Flow rate Other std flow
rate
Inlet /
outlet
port
size
Other related information
SE Series
SE1 22 l/min ¾” BSP Male connection
SE2 30 l/min
SE3 45 l/min
Single Lamp Chambers
SS15 9 l/min ¾” BSP Male connection
SS30 23 l/min ¾”
SS55 37 l/min ¾”
SS75 56 l/min 1”
SS1475 117 l/min 1½”
SS1575 152 l/min 1½”
Multiple Lamp Chambers
SS4855 12.3 m3/h 2” BSP Male connection
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
132
Model Flow rate Other std flow
rate
Inlet /
outlet
port
size
Other related information
SS4875 18.6 m3/h 2”
SS1055 21.3 m3/h 3”
SS61075 34.8 m3/h 3”
Electrical Requirements
Model Voltage
System power
consumption
(Watts)
Other related information
SE Series
SE1 230/240 V
50 Hz
21 W Supply fuses protection 3A
SE2 28 W
SE3 40 W
Single Lamp Chambers
SS15 230/240 V
50 Hz
15 W
SS30 30 W
SS55 55 W
SS75 75 W
SS1475 75 W
SS1575 75 W
Multiple Lamp Chambers
SS4855 230/240 V
50 Hz
220 W
SS4875 300 W
SS1055 330 W
SS61075 450 W
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
133
Lamp specification
Model Type Lamp
life (h)
Lamp power
consumption Number of Lamps
SE Series
SE1 CHACSE1LAMP 8,000 17 W 1
SE2 CHACSE2LAMP 27 W 1
SE3 CHACSE3LAMP 38 W 1
Single Lamp Chambers
SS15 LP Mercury Vapour Approx.
8,760
15 W 1
SS30 LP Mercury Vapour 30 W 1
SS55 LP Mercury Vapour 55 W 1
SS75 LP Mercury Vapour 75 W 1
SS1475 LP Mercury Vapour 75 W 1
SS1575 LP Mercury Vapour 75 W 1
Multiple Lamp Chambers
SS4855 LP Mercury Vapour Approx.
8,760
220 W 4x55
SS4875 LP Mercury Vapour 300 W 4x75
SS1055 LP Mercury Vapour 330 W 6x55
SS61075 LP Mercury Vapour 450 W 6x75
Pre-treatment requirements
Model Filtration UV-T Other related information
SE Series
SE1 - - Max 0 – 10bar
SE2
SE3
Single Lamp Chambers
SS15 - - Max 0 – 10bar
SS30
SS55
SS75
SS1475
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
134
Model Filtration UV-T Other related information
SS1575
Multiple Lamp Chambers
SS4855 - - Max 0 – 10bar
SS4875
SS1055
SS61075
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
SE Series
SE1 Lamp
operating/failure
warning lights
and audible
alarm
- - Lamp operating; the green
indicator lamp illuminates
Lamp failure: the buzzer sounds
and the red indicator lamp
illuminates
SE2
SE3
Single Lamp Chambers
SS15 Power indicator Lamp
run/fail
indicator
Hours run
metre
Lamp operating; the green
indicator lamp illuminates
Lamp failure: the buzzer sounds
and the red indicator lamp
illuminates
SS30
SS55
SS75
SS1475
SS1575
Multiple Lamp Chambers
SS4855 Power indicator Lamp
run/fail
indicator
Hours run
metre
Lamp operating; the green
indicator lamp illuminates
Lamp failure: the buzzer sounds
and the red indicator lamp
illuminates
SS4875
SS1055
SS61075
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
135
Maintenance
Model Items for regular replacement Other related information
SE Series
SE1 UV lamps
O rings
Quartz sleeves
-
SE2
SE3
Single Lamp Chambers
SS15 UV lamps
O rings
Quartz sleeves
-
SS30
SS55
SS75
SS1475
SS1575
Multiple Lamp Chambers
SS4855 UV lamps
O rings
Quartz sleeves
-
SS4875
SS1055
SS61075
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
SE Series
SE1 Solenoid
valve (with
manual
override) to
shut off
water
supply in
the event of
a lamp or
power
failure.
- Hours run
meter to
monitor lamp
life
Volt free contacts, auxiliary 230V
AC alarm contacts. SE2
SE3
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
136
Model Flow
control
UV
exposure
UV output
flow control Other related information
Single Lamp Chambers
SS15 Solenoid
valve (with
manual
override) to
shut off
water
supply in
the event of
a lamp or
power
failure.
- Hours run
meter to
monitor lamp
life
Volt free contacts, auxiliary 230V
AC alarm contacts. SS30
SS55
SS75
SS1475
SS1575
Multiple Lamp Chambers
SS4855 Solenoid
valve (with
manual
override) to
shut off
water
supply in
the event of
a lamp or
power
failure.
- Hours run
meter to
monitor lamp
life
Volt free contacts, auxiliary 230V
AC alarm contacts. SS4875
SS1055
SS61075
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
137
D10 Silverline UK Limited
http://silverlineuk.co.uk/
System Rating and connection
Model flow rate
30mJ / cm2 (UVT 98%)
Inlet /
outlet
port
size
Other related information
UV-DS Steriliser Range
UV-DS08 4 ½” BSP
UV-DS15 8 ¾” BSP
UV-DS30 21 ¾” BSP
UV-DS55 36 ¾” BSP
Electrical Requirements
Model Voltage System power
consumption Other related information
UV-DS Steriliser Range
UV-DS08 220/240 V 8 W
UV-DS15 15 W
UV-DS30 30 W
UV-DS55 55 W
Lamp specification
Model Type Lamp life (h) Lamp power
consumption Other related information
UV-DS Steriliser Range
UV-DS08 Low
pressure
8,000 8 W
UV-DS15 15 W
UV-DS30 30 W
UV-DS55 55 W
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
138
Pre-treatment requirements
Model Filtration UV-T Other related information
UV-DS Steriliser Range
UV-DS08 5 µm 98% 100 psi
UV-DS15
UV-DS30
UV-DS55
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
UV-DS Steriliser Range
UV-DS08 A contact
that goes
live if the
lamp fails
this could be
used to run
a flashing
beacon or
sound an
alarm.
UV-DS15
UV-DS30
UV-DS55
Maintenance
Model Items for regular replacement Comment
UV-DS Steriliser Range
UV-DS08 Quartz sleeve
Lamps
O Rings
The quartz sleeve within the unit
should be periodically cleaned
or replaced to ensure it does not
impair the UV light.
Recommended 2-3 yearly but
sooner if water clarity is
questionable.
UV-DS15
UV-DS30
UV-DS55
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
139
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
UV-DS Steriliser Range
UV-DS08 A solenoid
valve to
close off the
water
supply if the
lamp fails.
A solenoid
valve to
close off the
water if the
power fails.
A UV control
box with lamp
run and fail
lights and an
inbuilt alarm.
UV-DS15
UV-DS30
UV-DS55
D11 Viqua Sterilight
http://viqua.com/sterilight/
Parent company: Trojan Technologies
http://trojanuv.com/uvmax-sterilight
System Rating and connection
Model
USEPA std
flow rate
(40 mJ/cm2)
Other std flow
rate
Inlet /
outlet
port size
Other related information
Sterilight Platinum range
SP320-HO 2.5 m3/h 3.4 m
3/h
(30 mJ/cm2)
¾” NPT Supplied with integrated pre-treatment
system (pre-filtration and GAC)
Also SPV range for validated systems
SP410-HO 3.2 m3/h 4.5 m
3/h
(30 mJ/cm2)
1” NPT
SP600-HO 5.9 m3/h 7.9 m
3/h
(30 mJ/cm2)
1” NPT
SP740-HO 7.0 m3/h 9.5 m
3/h
(30 mJ/cm2)
1½” NPT
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
140
Model
USEPA std
flow rate
(40 mJ/cm2)
Other std flow
rate
Inlet /
outlet
port size
Other related information
SP950-HO 8.9 m3/h 11.8 m
3/h
(30 mJ/cm2)
1½” NPT
Sterilight Cobalt range
SCM-200
1.4 m3/h 1.8 m
3/h
(30 mJ/cm2)
1“ NPT Also available as integrated ‘home
systems’ with cartridge filters.
SCM-320 2.3 m3/h 3.0 m
3/h
(30 mJ/cm2)
SCM-600 5.5 m3/h 7.3 m
3/h
(30 mJ/cm2)
SCM-740 6.8 m3/h 9.1 m
3/h
(30 mJ/cm2)
1½” NPT
Sterilight Silver range
S1Q-PA 0.3 m3/h 0.4 m
3/h
(30 mJ/cm2)
¼” NPT Also available as integrated ‘home
systems’ with cartridge filters.
S2Q-
PA/SSM-17
0.4 m3/h 0.7 m
3/h
(30 mJ/cm2)
½” NPT
S5Q-
PA/SSM-24
1.0 m3/h 1.4 m
3/h
(30 mJ/cm2)
¾” NPT
S8Q-
PA/SSM-37
1.8 m3/h 2.3 m
3/h
(30 mJ/cm2)
¾” NPT
S12Q-
PA/SSM-39
2.5 m3/h 3.4 m3/h
(30 mJ/cm2)
1” NPT
Electrical Requirements
Model Voltage System power
consumption Other related information
Sterilight Platinum range
SP320-HO 90-265V
50-60Hz
48 W
SP410-HO 60 W
SP600-HO 78 W
SP740-HO 90 W
SP950-HO 110 W
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
141
Model Voltage System power
consumption Other related information
Sterilight Cobalt range
SCM-200 100-240V
50-60Hz
35 W
SCM-320 42 W
SCM-600 70 W
SCM-740 82 W
Sterilight Silver range
S1Q-PA 100-240V
50-60Hz
19 W
S2Q-PA/SSM-17 22 W
S5Q-PA/SSM-24 30 W
S8Q-PA/SSM-37 46 W
S12Q-PA/SSM-39 48 W
Lamp specification
Model Type Lamp life (h) Lamp power
consumption Other related information
Sterilight Platinum range
SP320-HO LP (254
nm)
9,000 37 W Sterilume™ - HO (high-output)
SP410-HO 46 W
SP600-HO 65 W
SP740-HO 75 W
SP950-HO 90 W
Sterilight Cobalt range
SCM-200 LP (254
nm)
9,000 25 W Sterilume™ - HO (high-output)
SCM-320 34 W
SCM-600 58 W
SCM-740 70 W
Sterilight Silver range
S1Q-PA LP (254
nm)
9,000
(Annually,
Bi-annual is
seasonal use
only)
14 W Sterilume™ - EX (standard output)
S2Q-PA/SSM-17 17 W
S5Q-PA/SSM-24 25 W
S8Q-PA/SSM-37 37 W
S12Q-PA/SSM-39 39 W
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
142
Pre-treatment requirements
Model Filtration UV-T Other related information
Sterilight Platinum range
SP320-HO 5 µm pre-
filter
> 75% 2-40°C, max 8.62 bar.
Recommended WQ limits:
Iron: < 0.3 mg/L
Total hardness*: < 120 mg/L
Turbidity: < 1 NTU
Manganese: < 0.05 mg/L
Tannins: < 0.1 mg/L
SP410-HO
SP600-HO
SP740-HO
SP950-HO
Sterilight Cobalt range
SCM-200 5 µm pre-
filter
> 75% 2-40°C, max 8.62 bar.
Recommended WQ limits:
Iron: < 0.3 mg/L
Total hardness*: < 120 mg/L
Turbidity: < 1 NTU
Manganese: < 0.05 mg/L
Tannins: < 0.1 mg/L
SCM-320
SCM-600
SCM-740
Sterilight Silver range
S1Q-PA 5 µm pre-
filter
> 75% 2-40°C, max 8.62 bar.
Recommended WQ limits:
Iron: < 0.3 mg/L
Total hardness*: < 120 mg/L
Turbidity: < 1 NTU
Manganese: < 0.05 mg/L
Tannins: < 0.1 mg/L
S2Q-PA/SSM-17
S5Q-PA/SSM-24
S8Q-PA/SSM-37
S12Q-PA/SSM-39
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
Sterilight Platinum range
SP320-HO UV intensity
display
(0-99%, with
audible and
visual alarm
at 50%)
‘Lamp life
monitor
(display as
remaining
days)’.
Audible
lamp failure
alarm.
System diagnostic self-test at start-
up.
UV sensor failure alarm.
Pre-warning that lamp needs
changing when last 30 days
reached.
SP410-HO
SP600-HO
SP740-HO
SP950-HO
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
143
Model Feature 1 Feature 2 Feature 3 Other related information
Audible ‘lamp replacement
reminder’
‘Total controller running time’
display.
‘Power on’ display
Low UV intensity can be used to
close inlet solenoid valve.
Sterilight Cobalt range
SCM-200 UV intensity
display
(0-99%, with
audible and
visual alarm
at 50%)
‘Lamp life
monitor
(display as
remaining
days)’.
Audible
lamp failure
alarm
Audible ‘lamp replacement
reminder’
‘Total controller running time’
display.
‘Power on’ display
Low UV intensity can be used to
close inlet solenoid valve.
SCM-320
SCM-600
SCM-740
Sterilight Silver range
S1Q-PA UV intensity
on SSM
systems
only. Display
(0-99%, with
audible and
visual alarm
at 50%)
‘Lamp life
monitor
(display as
remaining
days)’.
Audible
lamp failure
alarm
Audible ‘lamp replacement
reminder’
‘Total controller running time’
display.
‘Power on’ display
Low UV intensity can be used to
close inlet solenoid valve.
S2Q-PA/SSM-17
S5Q-PA/SSM-24
S8Q-PA/SSM-37
S12Q-PA/SSM-39
Maintenance
Model Items for regular replacement Comment
Sterilight Platinum range
SP320-HO Descale / clean quartz thimble. Replace lamp
/ and o rings.
(Other replacement parts include controller,
UV monitor)
Safety-Loc™ connector with
interlock that ensures power is
disconnected before lamp can
be removed.
SP410-HO
SP600-HO
SP740-HO
SP950-HO
Sterilight Cobalt range
SCM-200 Descale / clean quartz thimble. Replace lamp
/ and o rings.
Safety-Loc™ connector with
interlock that ensures power is SCM-320
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
144
Model Items for regular replacement Comment
SCM-600
(Other replacement parts include controller,
UV monitor)
disconnected before lamp can
be removed. SCM-740
Sterilight Silver range
S1Q-PA Descale / clean quartz thimble. Replace lamp
/ and o rings.
(Other replacement parts include controller,
UV monitor)
S2Q-PA/SSM-17
S5Q-PA/SSM-24
S8Q-PA/SSM-37
S12Q-PA/SSM-39
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
Sterilight Platinum range
SP320-HO ? Output for
closing
inlet
solenoid
valve (if
UV% low)
Flow pacing
sensor,
reduces lamp
power when
no flow
SP410-HO
SP600-HO
SP740-HO
SP950-HO
Sterilight Cobalt range
SCM-200 Flow
restrictor on
outlet to
limit
throughput
Output for
closing
inlet
solenoid
valve (if
UV% low)
?
SCM-320
SCM-600
SCM-740
Sterilight Silver range
S1Q-PA None
None
S2Q-PA/SSM-17 Output for
closing
inlet
solenoid
valve (if
UV% low)
?
S5Q-PA/SSM-24 Flow
restrictor on
outlet to
limit
throughput
S8Q-PA/SSM-37
S12Q-PA/SSM-39
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
145
D12 VIQUA
http://viqua.com/uvmax/
System Rating and connection
Model
Flow rate
(>30 mJ/cm2)
(95% UVT)
Other std flow
rate
(40 mJ/cm2)
Inlet /
outlet
port
size
Other related information
UV Max
UVMax A 0.42 m3/h -
3/8”
NPT
UVMax
F4plus
6.93 m3/h
1” NPT
Electrical Requirements
Model Voltage
System power
consumption
(Watts)
Other related information
UV Max
UVMax A 120-240V 22
UVMax F4plus 130
Lamp specification
Model Type Lamp life (h)
Lamp power
consumption
(Watts)
Other related information
UV Max
UVMax A LP 9,000 14
UVMax F4plus LPHO 9,000 110 Sterilume
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
146
Pre-treatment requirements
Model Filtration UV-T
(%)
Water
temperature
(°C)
Other related information
UV Max
UVMax A 5 µm
prefilter
required
75 (minimum) 4 - 40 Max hardness 120 mgCaCO3/l
Iron 0.3 mgFe/l maximum UVMax F4plus
System monitoring
Model Feature 1 Feature 2 Feature 3 Other related information
UV Max
UVMax A - - Lamp
operation
indicator
Power supply operation indicator.
UVMax F4plus Lamp age
indicator /
replacement
reminder
UV output
sensor
UV sensor with diagnostic test.
Power supply operation indicator.
Audible alarm.
Maintenance
Model Items for regular replacement Other related information
UV Max
UVMax A
UVMax F4plus Quartz sleeve, lamp, O rings
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
UV Max
UVMax A - - - -
UVMax F4plus - - Solenoid
valve flow
shut-off if UV
dose
insufficient
COMMcenter unit for Pro series
systems displays UV dose, alarm
history, lamp hours, and other
performance parameters for up to
nine systems.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
147
D13 Wedeco
http://www.xylem.com/treatment/ca/brands/wedeco
Parent company: Xylem
http://www.xyleminc.com/en-us/Pages/default.aspx
System Rating and connection
Model
USEPA std
flow rate
m3/h
(40 mJ/cm2)
Other std flow
rate
m3/h
(30 mJ/cm2)
Inlet /
outlet
port
size
Other related information
Aquada UV Range ( Same for Altima, Proxima and Maxima)
Aquada 1 0.73 0.98 ½”
Aquada 2 1.85 2.47 ¾”
Aquada 4 3.24 4.32 ¾”
Aquada 7 6.70 9.00 1”
Aquada 10 10.10 13.4 1½”
Electrical Requirements
Model Voltage
System power
consumption
(kW)
Other related information
Aquada UV Range ( Same for Altima, Proxima and Maxima)
Aquada 1 230V 50-
60Hz (TN-S-
net, TN-C-
net)
35 UV electrical standard CE
Aquada 2 55
Aquada 4 55
Aquada 7 85
Aquada 10 85
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
148
Lamp specification
Model Type Lamp life (h) Lamp power
consumption Other related information
Aquada UV Range (Same for Altima, Proxima and Maxima)
Aquada 1 LP (254
nm)
8,760 20
Aquada 2 40
Aquada 4 40
Aquada 7 80
Aquada 10 80
Pre-treatment requirements
Model Filtration UV-T Other related information
Aquada UV Range ( Same for Altima, Proxima and Maxima)
Aquada 1 80-98% 0-40°C, max 10 bar.
Aquada 2
Aquada 4
Aquada 7
Aquada 10
System monitoring
Model Altima Proxima Maxima Other related information
Aquada UV Range ( Same for Altima, Proxima and Maxima)
Aquada 1 Moulded
control unit.
Glow-cap
lamp
operation
indicator.
Moulded control
unit.
Glow-cap lamp
operation
indicator.
Audible Alarm
buzzer.
Visual alarm
Moulded control
unit.
Glow-cap lamp
operation
indicator.
Audible Alarm
buzzer.
Visual alarm
Aquada 2
Aquada 4
Aquada 7
Aquada 10
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
149
Model Altima Proxima Maxima Other related information
display.
Digital lamp life
display.
Push button
alarm/computer
reset.
display.
Digital lamp life
display.
Push button
alarm/computer
reset.
UV intensity
monitor.
Digital UV
intensity display.
Maintenance
Model Items for regular replacement Comment
Aquada UV Range ( Same for Altima, Proxima and Maxima)
Aquada 1 Descale / clean quartz thimble. Replace lamp
/ and o rings.
(Other replacement parts include controller,
UV monitor)
Aquada 2
Aquada 4
Aquada 7
Aquada 10
Operational control
Model Flow
control
UV
exposure
UV output
flow control Other related information
Aquada UV Range ( Same for Altima, Proxima and Maxima)
Altima Safe-T-Cap
lamp
connector
system.
Proxima Safe-T-Cap
lamp
connector
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
150
Model Flow
control
UV
exposure
UV output
flow control Other related information
system.
Micro-
computer
controller.
Maxima Power
connection
for optional
automatic
solenoid
safety shut
off valve.
Safe-T-Cap
lamp
connector
system.
Micro-
computer
controller.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
151
Appendix E Local authority site visits
E1 Introduction
During August and September 2015, visits were made to four local authorities to inspect
private water supplies incorporating UV treatment. Twenty-five premises were visited,
including single domestic dwellings, and small domestic and commercial supplies. The
findings of the visits are reported in Sections E2-E5.
Subsequent to the site visits, contact was made with nine installers of UV equipment for
private supplies (eight identified from the site visits and one identified by a water utility as
installing and servicing UV equipment in its region). An email was sent to each installer asking
for responses to a number of questions. Details are provided in Section E6.
E2 Local Authority ‘A’
A summary of the site visits is shown in Table E1.
Site visits were made to 11 private supplies incorporating UV on 18-19 August 2015. The
supplies included single domestic dwellings (SDD), small domestic and commercial sites. In
all cases the water source was a hard groundwater drawn from wells or boreholes. As well as
water hardness, other potential problem parameters included microorganisms (wells and
shallow boreholes), nitrate, iron and manganese.
E2.1 Installation and maintenance
Water treatment equipment, including UV, had been installed by local specialist companies or
plumbers, and was generally maintained by the same. Most users had service contracts in
place, which included annual replacement of UV lamps and cleaning of quartz sleeves.
Little, if any, manufacturers’ literature or operating/maintenance instructions had been
provided to users. Most users had little knowledge of their treatment, including the function of
any units upstream of UV.
Maintenance logs were generally not kept. Most users kept copies of invoices that provided
dates of maintenance and, to varying degrees, a record of the work carried out.
Most units were sited externally in purpose-built enclosures, sheds or outbuildings. In one
case, the UV enclosure was hidden behind shrubbery that had to be pruned to allow access.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
152
E2.2 Pre-treatment
Pre-treatment included particulate filters, nitrate filters and iron (and possibly manganese)
filters. The lack of manufacturer’s literature and labelling often made it difficult to identify the
specific treatment.
Users were sometimes unaware of their treatment, and generally unaware of any
maintenance requirements, such as the replacement intervals for filter cartridges.
E2.3 UV treatment
A range of UV equipment was installed, both branded and unbranded, but with little visible
information identifying design data such as maximum flow rate, operating pressure and
temperature.
Few installations included flow meters or monitoring and control; two units included an
indication of lamp life/days operated. Most units gave no indication whether the UV lamp was
functional; some users relied on observation of a ‘blue glow lamp’ to confirm operation of the
lamp. With one reported exception, failure of the UV lamp would not prevent flow and the
possibility of the consumption of non-disinfected water.
E2.4 Post-treatment
There was no evidence of any water treatment post UV. UV-treated drinking water was
supplied direct to taps and, at some properties, to storage tanks. One user believed that cold
water from storage was supplied to the bathroom and used for brushing teeth and bathing.
E2.5 General
Key findings arising from the visits:
There is a general lack of understanding amongst users regarding the treatment of their
private supplies. This is compounded by the lack of information provided by equipment
providers/installers.
There is no indication that UV equipment has been selected correctly for the flow (lack
of metering and control) or water quality (UVT, hardness, Fe). UVT measured >95% for
8 out of 10 water samples (taken from before or after UV, including kitchen taps); two
UVT values <90% were sampled from kitchen taps.
UV equipment is generally serviced by specialist companies or plumbers, with quartz
sleeves cleaned and lamps changed around every 12 months; the frequency of
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
153
maintenance of other equipment and replacement of cartridge filters is less clear.
Maintenance logs are not kept by users.
Monitoring and control of UV equipment is inadequate. Failure of a UV lamp may go
undetected for some time because a lack of a prominent alarm, and will generally not
prevent flow and the possibility of the consumption of non-disinfected water.
The potential for contamination of stored UV-treated water may not be well understood
by users.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
154
Table E.1 Summary of site visits to Local Authority ‘A’
Property ref /
type Water source Pre-treatment UV treatment Post-treatment Comments
1
Single domestic
dwelling
Well In-line nitrate
filter
Unbranded UV sited externally
UVT = 97.5%
Flow meter
No monitoring or control
12-month service contract (with local
installer)
None – direct
supply to taps
and to storage
tank
Hard groundwater, colour 0.5°H, turb. 0.17 NTU; historic
concerns re hydrocarbons (due to local heating oil
contamination), nitrate and micro-organisms (E. coli).
No user instructions/manufacturers’ information or maintenance
log.
Risk assessment underway.
2
Commercial
(bed &
breakfast)
Well In-line nitrate
filter
Wedeco UV-C / Aquada sited
externally
UVT = 87.3%
No flow meter, monitoring or control
12-month service contract (with local
installer)
None – direct
supply to taps
Hard groundwater, colour 1.0°H, turb. 0.14 NTU; historic
concerns re hydrocarbons (due to local heating oil
contamination), nitrate and micro-organisms (E. coli).
No user instructions/manufacturers’ information or maintenance
log.
Risk assessment underway.
3
Commercial
(various
business units)
Borehole None Wedeco Aquada AG (Type 2) sited
internally
UVT = 97.6%
No flow meter, monitoring or control
12-month service contract (with local
plumber)
None – direct
supply to taps
Hard groundwater, colour 0.3°H, turb. 0.12 NTU.
No user instructions/manufacturers’ information or maintenance
log.
Risk assessment completed.
4
Commercial
(Holiday
cottages and
function centre)
Borehole (deep) Fe filter SS55 UV Steriliser (Type A) sited
externally
No flow meter, monitoring or control
12-month service contract
None – direct
supply to taps
No water sample.
Parallel treatment streams (Fe filtration/UV).
Some user instructions/manufacturers’ information, no
maintenance log.
Risk assessment underway.
5
Small Domestic
(three
properties)
Borehole Particulate filter Unbranded UV sited externally
UVT = 97.0%
No flow meter
Viqua countdown monitor / total days
operated
None known –
direct supply to
taps
Hard groundwater, colour 0.5°H, turb. 0.16 NTU; historic
concerns re nitrate and micro-organisms.
No user instructions, manufacturers’ information or
maintenance log.
Difficult access to UV unit.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
155
Property ref /
type Water source Pre-treatment UV treatment Post-treatment Comments
User (new property owner) uncertain re service contract.
Risk assessment programmed.
6
Single domestic
dwelling
Borehole Shakesby Fe
filter
UV Steriliser (55W) sited externally
UVT = 96.1%
No flow meter, monitoring or control
None known –
direct supply to
taps
Hard groundwater, colour 1.8°H, turb. 0.30 NTU; historic
concerns re iron and micro-organisms.
“Do not adjust” tag on flow valve.
No user instructions, manufacturers’ information or
maintenance log.
Uncertain re service contract (some service dates recorded
locally on shed wall).
7
Commercial
(Single property
but has paying
guests)
Well None Wedeco UV-C / Aquada sited
internally
UVT = 84.0%
No flow meter, monitoring or control
Service contract (with local plumber)
None known –
direct supply to
taps and storage
tanks
Colour 3.6°H, turb. 0.10 NTU;
No user instructions, manufacturers’ information or
maintenance log.
Risk assessment programmed.
8
Commercial
(stables and
tenanted
properties)
Shallow
borehole
None Wedeco UV-C (5.5 m3/h, 80 W) /
Aquada sited externally
UVT = 96.8%
No flow meter, monitoring or control
Service contract (with local plumber)
None known –
direct supply to
taps and storage
tanks
Hard groundwater, colour 0.5°H, turb. 0.11 NTU;
No user instructions, manufacturers’ information or
maintenance log.
Risk assessment programmed.
9
Commercial
(farm and
tenanted
properties)
Deep borehole Fe filter –
particulate filter
Wedeco UV-C (5.5 m3/h) sited
externally
UVT = 97.1%
No flow meter
Days remaining / lamp out indicators
6/12-month service contract (with local
installer) for filters/UV
None known –
direct supply to
taps and storage
tanks
Hard groundwater, colour 0.5°H, turb. 0.12 NTU; no known
quality issues.
No user instructions, manufacturers’ information or
maintenance log.
Risk assessment programmed.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
156
Property ref /
type Water source Pre-treatment UV treatment Post-treatment Comments
10
Commercial
(farm and
tenanted
properties)
Borehole Particulate filter -
Mn filter - Fe
filter –
particulate filter
– scale inhibitor
Wedeco Aquada UV sited externally
UVT = 97.3%
No flow meter
Days remaining / lamp out indicators
12-month service contract (with local
installer)
None known –
direct supply to
taps and storage
tanks
Hard groundwater, colour 0.5°H, turb. 0.12 NTU.
No user instructions, manufacturers’ information or
maintenance log.
Risk assessment programmed.
11
Commercial
(two properties,
one is a holiday
let)
Shallow
borehole
Fe filter Unbranded UV sited externally
UVT = 96.4%
No flow meter
No monitoring or control
12-month service contract (with local
installer)
None – direct
supply to taps
Hard groundwater, colour 1.0°H, turb. 0.20 NTU; previous
failures for Fe, Mn, turbidity and coliforms.
No user instructions, manufacturers’ information or
maintenance log.
Risk assessment programmed.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
157
E3 Local Authority ‘B’
A summary of the site visits is shown in Table E2.
Site visits were made to 6 private supplies incorporating UV on 20-21 August 2015. The
supplies included single domestic dwellings (SDD), small and large domestic sites, and a
commercial site. Water sources consisted mainly of boreholes with some spring water
sources. Problem parameters included low pH, microorganisms (wells, springs and shallow
boreholes), iron and manganese.
E3.1 Installation and maintenance
Water treatment equipment, including UV, had been installed by local specialist companies or
plumbers and was either maintained by the specialist companies or users. Most users had
service contracts in place, which included annual replacement of UV lamps and cleaning of
quartz sleeves.
The majority of users had not been provided with operating/maintenance instructions and
instead relied on specialist companies with whom they had service contracts to maintain
equipment. Most users had little knowledge of their treatment, including the function of any
pre-treatment stages.
Maintenance logs were generally not kept. Most users kept copies of invoices that provided
dates of maintenance and, to varying degrees, a record of the work carried out.
For the large domestic sites, most units were sited externally in purpose-built enclosures. For
the single domestic dwellings, units were located where most convenient within the dwelling.
E3.2 Pre-treatment
Pre-treatment included particulate filters and one instance of activated carbon for chlorine
removal. There was also pH correction, chlorine dosing, and iron and manganese filters.
Users were sometimes unaware of their treatment. Those users that had service contracts in
place were generally unaware of any maintenance requirements, such as the replacement
intervals for filter cartridges.
E3.3 UV treatment
A range of UV equipment was installed, both branded and unbranded, but most had no visible
information identifying design data such as maximum flow rate, operating pressure and
temperature.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
158
Few installations included flow meters or monitoring and control. Only one unit included an
indication of lamp life/days operated which appeared to be reading incorrectly. This same unit
was known to have an audible alarm and automatic solenoid safety shut-off valve but the
users were unaware of this. Most users relied on observation of a ‘blue glow lamp’ to assess if
the unit was working. Most are unaware that this is not an indication that the lamp is actually
performing the required disinfection, but rather an indication only that the lamp is on.
E3.4 Post-treatment
There was only one instance of water treatment post UV. This seemed to be a retrospective
installation as the treatment incorporated activated carbon and particulate (2 µm) filter
cartridges, both of which should have been located before UV to improve water quality (rather
than after due to the potential for bacterial growth in both units). UV-treated drinking water
was supplied direct to taps and, at some properties, to storage tanks.
E3.5 General
Key findings arising from the visits:
There is a general lack of understanding amongst users regarding the treatment of their
private supplies.
There is no indication that UV equipment has been selected correctly for the flow (lack
of metering and control) or water quality (UVT, Fe or Mn). UVT measured >95% for all
6 water samples (taken from before or after UV, including kitchen taps).
UV equipment is generally serviced by specialist companies or the users themselves,
with quartz sleeves cleaned (as per manufacturers’ recommendations – some annually,
some every 2–3 months) and lamps changed around every 12 months. The frequency
of maintenance of other equipment and replacement of cartridge filters is less clear.
Maintenance logs are not kept by users.
Monitoring and control of UV equipment is inadequate. Failure of a UV lamp may go
undetected for some time because of a lack of a prominent alarm, and will generally not
prevent flow and the possibility of the consumption of non-disinfected water.
The potential for contamination of stored UV-treated water may not be well understood
by users.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
159
Table E.2 Summary of site visits to Local Authority ‘B’
Property ref /
type Water source Pre-treatment UV treatment Post-treatment Comments
1
Commercial
(farm and
tenanted
property)
Spring 5 µm activated
carbon and
spun propylene
filter
Unbranded UV (GE branded lamp)
sited internally
UVT=98.1%
No water meter
No monitoring or control
None – direct
supply to taps
Colour 0.77°H, turb. 0.13 NTU
User of unit was informed by previous owner how to maintain
the UV treatment unit.
Maintenance log kept.
2
Large Domestic
(Caravan Park)
Borehole Mn/Fe removal,
pH correction,
Cl dosing,
5µm filters x2,
Activated
carbon filter
Unbranded UV sited externally
UVT = 97.3%
Water meter
No flow audible and visible alarm
located on exterior of purpose built
enclosure
Service contract (with local installer)
Activated carbon
5µm filters x2
Colour 1.536°H, turb. 0.27 NTU
No user instructions, manufacturers’ information.
Maintenance log kept.
3
Large Domestic
(Caravan Park)
Borehole Mn removal
(Perhaps with
activated
carbon in Mn
removal
column)
Unbranded UV sited externally
UVT= 98.8%
Water meter
No monitoring or control
Service contract for Mn removal (with
local installer)
None – direct
supply to taps
Colour 0.26°H, turb. 0.12 NTU
No user instructions, manufacturers’ information or
maintenance log.
4
Small Domestic
Borehole Mn/Fe removal Wedeco UV unit sited externally
UVT= 99.6%
Water meter
Wedeco Aquada UV digital lamp life
display
Service contract (with local installer)
None – direct
supply to taps
Colour 0°H, turb. 0.17 NTU
No user instructions, manufacturers’ information or
maintenance log.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
160
Property ref /
type Water source Pre-treatment UV treatment Post-treatment Comments
5
Single domestic
dwelling
Spring pH modifier,
filter
Wedeco UV unit sited internally
UVT= 95.6%
Water meter
Wedeco Aquada UV digital lamp life
display
Service contract (with local installer)
None – direct
supply to taps
Colour 3°H, turb. 0.26 NTU
User instructions and manufacturers information supplied to
user with UV unit
6
Single domestic
dwelling
Spring +
Borehole
None Shann Chi UV unit sited internally
UVT=99.3%
None – direct
supply to taps
Colour 2.6°H, turb. 0.1 NTU
User instructions and manufacturers information supplied to
user with UV unit
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
161
E4 Local Authority ‘C’
A summary of the site visits is shown in Table E3.
Site visits were made on 3rd
September 2015 to 4 private water supplies incorporating UV
treatment. The supplies included small domestic and commercial sites. In all cases the water
source was from boreholes. Potential problem parameters included micro-organisms, arsenic,
nitrate, ammonia, metals, turbidity and odour.
E4.1 Installation and maintenance
Water treatment equipment, including UV, had been installed by local or regional specialist
companies, and was generally maintained by the same. Standards of installation were
generally adequate. Users were aware of the basic maintenance requirements, such as the
replacement of the lamps and filter cartridges. Most users had arranged for 12-monthly
service of the UV systems, which included replacement of UV lamps. Fouling of the quartz
sleeves was not noted as an issue generally.
Limited manufacturers’ literature or operating/maintenance instructions were available locally.
In general, users had some limited knowledge of the treatment system. Maintenance logs
were available for some systems, with degrees of detail varying from a label showing when
the next service was due, to a record of previous service dates and actions. Spare filter
cartridges were available at site for the largest commercial supply only.
E4.2 Pre-treatment
A wide range of pre-treatment was installed, always including single- or two-stage particulate
filters. In addition, pre-treatment included dosing with sodium hypochlorite, ion exchange
softening, GAC adsorption and pH correction. The lack of adequate schematic diagrams or
labelling of equipment often made it difficult to identify the pre-treatment process stages.
E4.3 UV treatment
A range of UV equipment was installed, including a 2-stream system (it was not clear whether
this operated as duty/standby or duty/duty). Information on the UV systems installed was in all
cases very limited, at best a label showing the system model and lamp power requirement.
No information on rated flow rate was available for any of the systems inspected.
The largest commercial system included a volumetric water meter. At best, system monitoring
was limited to ‘System on’, ‘Hours run’ and/or ‘Lamp OK’, whilst other systems had no
functionality indication. In all cases, failure of the UV lamp would not prevent flow and the
possibility of the consumption of non-disinfected water.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
162
E4.4 Post-treatment
One site had post-UV treatment in the form of pH correction and a second (older) UV system,
although this could not be inspected due to restricted access (in loft) and time constraints.
Recent works to the supply (new reservoir and UV) were in response to insufficiency of
supply, but the reason for the newer UV system was not clear. UV-treated drinking water was
supplied direct to taps after being stored in 3 header tanks situated in loft.
E4.5 General
Key findings arising from the visits:
Relatively complex treatment systems have been installed for some larger commercial
private supplies. These systems did not have detailed schematic diagrams available at
site, and did not have labelled components such as valves and pumps, and appropriate
pipe labels (direction of flow, stream ID). Information on the simpler systems was not
usually available at site.
There is currently no licensing or approved contractor scheme employed where works
are carried out on these private supplies to public / commercial premises. Absence of
competency approved schemes may increase risks.
The need for additional disinfection with chlorine had historically been identified at 1
site, where the UV treated water was blended with water from the public supply,
presumably to maintain an effective free chlorine residual in the blended water storage
tank. The order of treatment at this site (chlorine followed by UV) will tend to destroy
some of the chlorine and could lead to a small increase in formation of disinfection by-
products.
There was no local indication of the specification for the UV equipment (e.g. UV dose,
flow rate, water UVT). UVT measured >95% for all water samples (taken from before or
after UV, including kitchen taps).
UV equipment was generally serviced by specialist companies, with lamps and
particulate filters changed every 12 months. The frequency of replacement of cartridge
filters appears to be linked to the need to change the lamp, rather than filter condition.
Some single-stage filters were visually very much in need of replacement, other sites
with multiple-stage filtration appeared to be in better condition. It was difficult to
ascertain the porosity of filters and whether raw water quality was taken into account
when specifying the design.
Continuous monitoring of UV equipment functionality was generally inadequate. Failure
of a UV lamp may go undetected for some time because a lack of a prominent alarm or
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
163
automatic shut-off valve, and will generally not prevent flow and the possibility of the
consumption of non-disinfected water.
The potential for contamination of stored UV-treated water may not be well understood
by users.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
164
Table E.3 Summary of site visits to Local Authority ‘C’
Property ref /
type Water source Pre-treatment UV treatment Post-treatment Comments
1
Commercial
(caravan holiday
park, ~73 m3/d)
2 boreholes
(3rd b/h taken out
of use)
GAC, ion
exchange
softener,
2-stage
particulate
filters,
sodium
hypochlorite.
UV sited in dedicated treatment
building adjacent to untreated water
storage tank. Installed 2012 (?)
UVT = 96.5%
Water volumetric meter.
Power on indicator (not lit, not
functional (?).
Run hours indicator (indicates ~ 3 yrs,
not reset?).
12-month service visits (with local
installer).
Electromagnetic
scale prevention
device (non-
contact type).
Blended with
public water
supply at storage
reservoir.
Colour 0.5 °H, turb 0.12 NTU; historic concerns re ‘H2S’ odour /
ammonia from de-commissioned borehole.
No P&I D drawing at site, no valve identification. Simple
schematic provided to council. Some manuals available (GAC
and ion exchange systems).
Label indicates when next service due.
Risk assessment carried out previously.
GAC likely to be a vestige from previous odour issues.
2
Commercial
(public venue
with 27 bedroom
accommodation
in 2 buildings,
~13 m3/d)
2 boreholes Ion exchange
softener,
1-stage
particulate filter.
Daro Saphir, (Installed Dec 2012, 2
parallel streams) sited within main
property.
UVT = 97.2%
No volumetric meter. System ‘On’ and
‘Lamp OK’ indication.
12-month service visits (with local
installer).
None. Supplies a
number of
storage tanks.
Colour < 0.5°H, turb 0.13 NTU. Historic concerns re coliforms /
enterococci (2013)
No P&I D drawing at site, no valve identification. No user
instructions /manufacturers’ information.
Label indicates service history.
Risk assessment carried out previously.
3
Domestic
(farmhouse) and
Commercial
(business units)
1 Borehole Ad-hoc
chlorination of
feed reservoir
(hypochlorite
solution or
tablets).
1-stage
particulate filter.
Wedeco Aquada Altima sited
externally in small plastic cabinet.
Installed 2001.
UVT = 95.2%
No volumetric meter, control or status
indication.
12-month service visits (with local
installer).
None – direct
supply to taps
Colour 0.8°H, turb 0.13 NTU.
No P&I D drawing at site, no valve identification. No user
instructions /manufacturers’ information.
Limited service history record provided.
Risk assessment carried out previously (2012)
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
165
Property ref /
type Water source Pre-treatment UV treatment Post-treatment Comments
4
Commercial
(public house /
restaurant)
1 Borehole 10 m3 untreated
storage tank.
1 stage
particulate filter.
SS75 UV Steriliser, sited in wooden
shed adjacent to untreated storage
tank. (Installed Dec 2012). No flow
meter, control or status indication.
UVT = 97.5%
12-month service visits (with local
installer).
2nd
stage trt in
main building loft
(not seen). pH
correction (acidic
-> neutral), UV
(Pre 2009, not
seen).
Storage tanks
drained down and
disinfected 6
monthly.
Colour 2.3°H, turb 0.09 NTU.
No P&I D drawing at site (council have made a sketch). No
valve identification. No user instructions /manufacturers’
information.
Limited service history record provided.
Risk assessment carried out previously.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
166
E5 Local Authority ‘D’
A summary of the site visits is shown in Table E4.
Site visits were made on 9th September 2015 to 4 private water supplies incorporating UV
treatment. The supplies included small domestic and commercial sites. In all cases the water
source was groundwater drawn from boreholes in a chalk aquifer. Potential problem
parameters included micro-organisms and nitrate.
E5.1 Installation and maintenance
Water treatment equipment, including UV, had been installed by local or regional specialist
companies, and was maintained by the same for 3 of the 4 sites. Standards of installation
were generally good. Users were generally aware of the basic maintenance requirements,
such as the replacement of UV lamps and filter cartridges, although 1 user who had taken on
responsibility for maintenance of the UV system themselves, had misunderstood the effective
lifespan of UV lamps and the potential impact of fouling of the lamp thimble. The 3 other sites
had arranged for 12-monthly service of the UV systems, which included annual replacement
of UV lamps. Scale formation on the quartz sleeves was noted as an issue for this hard
borehole water.
The UV manufacturers’ system literature was available at only 1 site, the other 3 relied upon
the contractors knowledge for service and maintenance. In general, owners had limited
knowledge of the treatment system. A comprehensive maintenance log was kept by 1 of the
sites, and a very uninformative log was kept at a second site; the other 2 sites relied upon the
contractors to record details of maintenance. With the exception of 1 site, spare filter
cartridges were not normally available at site. Old lamps were stored at 1 site, possibly due to
uncertainty about the correct disposal route.
E5.2 Pre-treatment
One site had an uncapped borehole within the UV system building, open to the atmosphere
and therefore at significant risk of contamination. Furthermore, this site had no pre-filtration
stage upstream of the UV system. A single-stage pre-filtration was installed at the other 3
sites. No other forms of pre-treatment were used.
E5.3 UV treatment
A range of UV equipment was installed, including a 2 lamp / single stream system and a small
industrial system. Information on the UV systems, as installed, was very limited at 2 of the 4
sites. The small industrial system included detailed UV reactor design specifications. Another
smaller system had a copy of the UV system manual available.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
167
One or more volumetric meters were installed at each site. System monitoring ranged from
the small industrial system with an integrated UV sensor and UVT measurement as well as a
range of other system monitoring and alarms, to a simpler ‘System on’, ‘Hours run’ and/or
‘Lamp OK’ indication. In all cases, failure of the UV lamp would not prevent flow and the
possibility of the consumption of non-disinfected water.
E5.4 Post-treatment
None of the sites visited had any post-UV treatment. 3 of the 4 sites supplied treated drinking
water directly to taps via a pressure accumulator vessel. Storage reservoir tanks for treated
water had been decommissioned at these sites, as a result of advice from Winchester
Council.
E5.5 General
Key findings arising from the visits:
One of the commercial supplies had recently been equipped with a small industrial UV
system with relatively sophisticated monitoring. Whilst the system was not complex, the
pipe layout was confusing and would have benefited from some clear labelling of pipes
and valves, together with a corresponding P&I diagram. An older UV installation had an
unusual and uncertain pipe configuration, which appeared to include a manual isolating
valve which could allow bypass of the UV reactor if wrongly positioned and a non-return
valve which appeared to be incorrectly orientated. Clear labelling of pipes and valves,
together with a corresponding P&I diagram, was necessary.
Decommissioning of old storage tanks for treated water has minimised the risk of
contamination of treated water downstream of UV.
2 of the 4 UV systems showed the design specification for the system. UVT measured
>95% for all water samples (taken after UV, including kitchen taps).
UV equipment was generally serviced by a single specialist company, with lamps and
particulate filters changed every 12 months. Hours run indication on the UV equipment
was not always reset at lamp change. The owner of 1 system, which was maintained by
themselves, had decided that 3 yearly replacement of the lamp was acceptable,
because the lamp was still ‘glowing blue’ after this time period. The frequency of
replacement of cartridge filters appears to be linked to the need to change the lamp,
rather than filter condition. All of the filter modules were single stage and had opaque
housings, preventing in-situ assessment of fouling.
Continuous monitoring of functionality of UV equipment was minimal for 3 of the 4
systems. Failure of a UV lamp may go undetected for some time because a lack of a
prominent alarm or automatic shut-off valve, and will generally not prevent flow and the
possibility of the consumption of non-disinfected water.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
168
Table E.4 Summary of site visits to Local Authority ‘D’
Property ref /
type Water source Pre-treatment UV treatment Post-treatment Comments
1
Small domestic
(Farm,
farmhouse and
additional
cottages)
1 boreholes 1 stage
particulate filter.
UV Water Sterilzer (Dual lamp system)
sited in shed, installed 2011(?)
UVT = 96.8%
Power on indicator
Lamp operating / failed indicators x 2
Run hours indicators x 2 (both indicate
0, owner has reported problem).
12-month service visits (with local
installer).
None Colour 0.5 °H, turb 0.14 NTU; historic concerns re
bacteriological failures (boil water notice served before UV
installed). Poor condition of ‘large’ storage tank for treated
water was previously identified – now relined, fenced off from
livestock.
No P&I D drawing at site, no valve identification. No manual or
records available at site.
Risk assessment carried out previously.
2
Commercial
(Farm with other
commercial
activities such
as fishing,
camping, ~100
m3/d)
1 borehole 1 stage
particulate filter.
Wedeco Spektron 15, installed Dec
2015, rated 20m3/h, 400 J/m
2, >90%
UVT. Sited inside isolated farm
building.
UVT = 97.2%
Measured intensity =
79.1 W/m2
Volumetric meter fitted. System
monitored (UV intensity, UVT,
temperature, pressure, flowrate, lamp
run time, alarmed locally but not
external to building.
12-month service visits (with local
installer).
None. Direct feed
to supply via
small pressure
accumulator
vessel.
Colour 2.8°H, turb 0.46 NTU. Livestock in field adjacent to
borehole, now better protected by raised blockwork and locked
cover. Had bacteriological failures before storage reservoir for
treated water was decommissioned.
No P&I D drawing at site, no valve identification. No user
instructions /manufacturers’ information at site.
New risk assessment to be carried out this year.
3
Commercial
(Farm,
Farmhouse,
nursery school,
other
properties)
1 Borehole 1 stage
particulate filter.
Wedeco Aquada Altima (rated 1.77
m3/h, 400 J/m
2, UVT >94%). Sited in
field, over top of borehole, inside small
GRP cabinet. Installed April 2012.
UVT = 98.5%
Volumetric meter fitted. No control or
None. Direct feed
to supply via
small pressure
accumulator
vessel.
Colour 1.3°H, turb 0.54 NTU.
No P&I D drawing at site, no valve identification. Manual for UV
available at site.
Detailed service history record provided.
Risk assessment carried out previously.
Owner believes that 3 yearly interval for lamp replacement is
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
169
Property ref /
type Water source Pre-treatment UV treatment Post-treatment Comments
status indication except ‘lamp on’.
Self-maintained, detailed records kept
and maintenance plan.
adequate as lamp checked operational every month. 18
monthly replacement of pre-filter (spare held in stock).
4
Small domestic
(Several rented
properties)
1 Borehole None Hanovia UV (unknown model, >10 yrs
service). Sited in brick built shed over
top of borehole.
Volumetric meter, lamp status
indication (on / failed), hours run
indication (85029 hrs, presume not re-
zeroed at lamp change).
UVT = 97.4%
Some doubt about servicing
responsibilities.
None. Direct feed
to supply via
small pressure
accumulator
vessel.
Colour 0.8°H, turb 0.10 NTU.
Borehole not capped within building, so open to atmosphere
and potential ingress of contamination.
No P&I D drawing at site. Confusing pipe layout, possibly N/R
valve wrongly orientated. Manual isolating valve could allow
bypass of UV, not labelled or locked closed. No user
instructions /manufacturers’ information.
Very limited service history record – inspection date and
signature only.
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
170
E6 Survey of UV equipment installers
The following email was sent to nine installers of UV equipment for private supplies, mostly
identified from the LA site visits:
Dear Sir/Madam,
WRc is carrying out work for the Drinking Water Inspectorate (DWI) to identify the suitability,
design and operation of UV disinfection systems for private water supplies. We have carried
out a number of visits to premises with UV systems on private supplies, which has helped us
to understand how systems have been installed, and to get a user’s perspective on operation
and maintenance. What we also need to do is to get installer’s views on system selection,
design and operation, and assess how operating and maintenance requirements are
conveyed to users.
In summary, what we would like to understand is:
how the required UV dose is identified, and how suitable units are selected to meet the
dose requirements,
how this dose is maintained for defined feed water quality conditions,
how variations in water quality are dealt with to avoid the production of inadequately
disinfected water.
To address these issues, we would be very grateful if, as a recognised installer of UV systems
for private supplies, you could provide your views and any supporting information on as many
of the following questions as possible. Could you email your responses to me
([email protected]) at your earliest convenience please, and copy to Tom Hall
([email protected]), who is copied in on this email. We will be producing a
summary of the responses obtained for inclusion in the report to be produced for DWI, which
may be available on their website. If you would like us to maintain any confidentiality or
anonymity with regard to your replies, please let me know.
If you need any further information or clarification on any of this, please contact me.
Thank you in anticipation.
Glenn Dillon
Technical Consultant
Direct line: 01793 865045
Defra
Report Reference: DEFRA10932.05/16375-0 June 2016
© Defra 2016
171
Installer Questions:
1. Are there any types or models of UV unit which you normally or most regularly select,
and is there a reason for favouring any units?
2 For the units you install, how is the UV dose defined, and has the dose been verified or
validated by the supplier in any way?
3. In selecting a unit or dose level, is level of microbial risk considered for specific
installations?
4. What water quality parameters are considered when designing an installation, and how
are these taken into account (e.g. UVT, turbidity, colour, hardness, iron, manganese)?
5. How is varying water quality and/or flow addressed in maintaining the target dose?
6. Are the UV systems installed subject to any QC or certification?
7. Are installed systems "failsafe", i.e. in case of failure (e.g. power failure or lamp failure) is
water flow stopped?
8. Is a maintenance/service log provided to customers, e.g. containing manufacturers'
product information, operating and maintenance instructions, service records, etc?
9. Who is responsible for initiating servicing, e.g. replacement of filters or UV lamps, users or
installers (as part of a service contract)?
10. Are installers required to hold specific qualifications or be members of specific trade
organisations or registers, e.g. similar to gas engineers and the Gas Safe Register (previously
CORGI)?