Comparison of Private Water Supply and Public Water Supply …dwi.defra.gov.uk/research/completed-research/reports/DWI70-2-306.… · purpose of discovering and delivering new and

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


© 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


© 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


© 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.


Part 2: Tests of design, function and disinfection effectiveness.


Part 3: Sensors for the photometric monitoring of UV

disinfection: Tests and calibration.

Defra


© 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


© 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


© 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


© Defra 2016

5

A licensing or approved contractor scheme should be implemented for installers of


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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© Defra 2016

14


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


© 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


© 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


© 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


© 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%


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


© Defra 2016

19

Supplier /

model(s) Lamp type

1

Flow rate

(m3/h)


2 Other information

DaRo UV

Systems

Saphir

systems

LP

0.84-7.23a

1.08-9.63b

Prefiltration


Lamp/electrical failure

alarm

Lamp life indicator

Hours run meter


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


© Defra 2016

20

Supplier /

model(s) Lamp type

1

Flow rate

(m3/h)


2 Other information

LIFF

AQA Pure

LP

0.84-7.23a

1.08-9.63b

Prefiltration

UVT: 98% @

40 mJ/cm2


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


© Defra 2016

21

Supplier /

model(s) Lamp type

1

Flow rate

(m3/h)


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


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)


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)


sleeve

Replace controller, UV


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


© 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


© 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


© 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


© 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


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


© 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.


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


© 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


© Defra 2016

28



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


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


© Defra 2016

29



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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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.

http://www.dwi.gov.uk/stakeholders/legislation/

Defra


© 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


© 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.

http://dwi.defra.gov.uk/private-water-supply/index.htm

Defra


© 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


© 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


© 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


© 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


© 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


© 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


© 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.


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


© 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?



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)

http://dwi.defra.gov.uk/private-water-supply/index.htm

http://www.privatewatersupplies.gov.uk/

http://www.dwi.gov.uk/private-water-supply/RHmenu/Updated%20Manual%20on%20Treatment%20for%20Small%20Supplies.pdf



Defra


© 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


© 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



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


© 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


© 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


© 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


© 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


© 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


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


Defra


© 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.


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


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


© 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


© 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



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


© 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

http://www.clarence.nsw.gov.au/file.asp?g=RES-WHM-86-71-52

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


© 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.

http://dwi.defra.gov.uk/stakeholders/guidance-and-codes-of-practice/uvirradiation.pdf

http://dwi.defra.gov.uk/research/completed-research/reports/DWI70-2-300.pdf

Defra


© 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.

http://nwri-usa.org/documents/UVGuidelines3rdEdition2012.pdf

http://www.biovir.com/Images/pdf061.pdf

http://viqua.com/bms/assets/389/Manual_Sterilight_Silver_S12Q-PA_EN_FR_520104_RevH.pdf

Defra


© 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


© 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.


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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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


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


No explicit quantification.

Defra


© 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


© 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


© 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


© 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


© 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


© 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


© 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


© 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

http://aquacure.co.uk/

Defra


© Defra 2016

101



ACUV554D 55 W


ACNUVS6S 8 W

ACNUV14S 15 W

ACNUV24S 21 W

ACNUV32S 39 W

ACNUV39S 65 W


ACUV62P 220V 50Hz 0.140 A

Lamp specification

Model Type Lamp life

(h)

Lamp power


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


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


ACUV62P ACUV6A 4,320 0.140 A low pressure mercury discharge

lamps

Defra


© 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


ACNUVS6S 5 µm

filtration

generally

required

98% 5 – 35oC max 8 bar


Iron < 0.2 mg/l


Total Hardness < 7gpg

Turbidity < 2 NTU

Colour / Tannins *

ACNUV14S

ACNUV24S

ACNUV32S

ACNUV39S


ACUV62P 5 µm

filtration

generally

required

98% 5 – 35oC max 8 bar


Iron < 0.2 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


© Defra 2016

103



ACNUVS6S Lamps

Quartz Sleeves

ACNUV14S

ACNUV24S

ACNUV32S

ACNUV39S


ACUV62P Lamps

D2 Aquafine Corp. (CA, USA)

http://www.aquafineuv.com/

Parent company: Trojan technologies

http://www.trojantechnologies.com/


Model

Flow rate

(>30 mJ/cm2)

(94% UVT)

Other std flow

rate

(>30 mJ/cm2)

(99% UVT)

Inlet /

outlet

port

size


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


12R 30 m3/h 36 m

3/h No. of Lamps 12

8R 60 38 m3/h 46 m


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

http://www.aquafineuv.com/

http://www.trojantechnologies.com/

Defra


© Defra 2016

104

Model

Flow rate

(>30 mJ/cm2)

(94% UVT)

Other std flow

rate

(>30 mJ/cm2)

(99% UVT)

Inlet /

outlet

port

size


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


Model Voltage

System power

consumption (Watts)

120 V or (240) V AC


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


© Defra 2016

105

Model Voltage

System power

consumption (Watts)

120 V or (240) V AC


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


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



CSL Series

4R None stated Rated for

94% or 99%

Operating water temperature: 10-38oC

Max. operating pressure: 8 Bar

6R

8R

Defra


© Defra 2016

106


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


SP2

SL 10A Operating water temperature: 4-27oC


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


Defra


© Defra 2016

107


(optional)

SP and SL Series

SP1 Lamp status

indicator

Running

time

indicator

UV intensity

and

temperature

monitor /

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


© 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


© Defra 2016

109

D3 Bio-UV

http://www.bio-uv.co.uk


Model

Flow rate

(40 mJ/cm2)

(calculated at end of lamp

life with 98% UVT)

Inlet /

outlet

port

size


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” -


Model Voltage

System power

consumption

(Watts)



UV HOME 2 220-240V

50-60 Hz

36 1A fuse

UV HOME 3 61


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

http://www.bio-uv.co.uk/

Defra


© Defra 2016

110

Model Voltage

System power

consumption

(Watts)


IBP 4205 HO + 382 4 A fuse

Electrical cabinet ventilation IBP 5205 HO + 478

Lamp specification




UV HOME 2 HO 13,000 33 W Male threaded

UV HOME 3 55 W


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


Model Filtration

UV-T (T-10)

With Monitor

PRO3



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


© Defra 2016

111

Model Filtration

UV-T (T-10)

With Monitor

PRO3


Washable screen

filter 60 μm

Cartridge filter 10 μm

Carbon Filter


IBP 10 HO + - 85% -

IBP 30 HO +

IBP 40 HO +

IBP 2150 HO +

IBP 3150 HO +

IBP 4205 HO +

IBP 5205 HO +

System monitoring



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 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


© Defra 2016

112


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



UV HOME 2 UV lamp

Other replacement parts may include UV

monitor

-

UV HOME 3


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



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


© Defra 2016

113

Model Flow

control

UV

exposure

UV output


alarm report

or the cabling

of an

electronic

valve


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/


Model

USEPA std

flow rate

(40 mJ/cm2)

Other std flow

rate

(30 mJ/cm2)

Inlet /

outlet

port

size


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”

http://www.darouv.co.uk/

Defra


© Defra 2016

114

Model

USEPA std

flow rate

(40 mJ/cm2)

Other std flow

rate

(30 mJ/cm2)

Inlet /

outlet

port

size


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”




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


© Defra 2016

115

Lamp specification


(h)

Lamp power


Saphir Systems

Saphir 1 Low

pressure

(254 nm)

8,760 14 W GER15SE

Saphir 2 21 W GER25SE

Saphir 3 48 W GER25SEXO



ECO Series

ECO 1 Low

pressure

(254 nm)

8,000 15 W

ECO 2 30 W

ECO 3 55 W

ECO 5 55 W



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


© 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


© 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


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


© Defra 2016

118

D5 Filpumps

http://shop.filpumps.co.uk/


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




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

http://shop.filpumps.co.uk/

Defra


© Defra 2016

119

Lamp specification




1 FPSUV403 9,000 hours

(~1 year)

-


(~1 year)

-

3 FPSUV412

9,000 hours

(~1 year)

-


(~1 year)

-


(~1 year)

-


(~1 year)

-

7 FPSUV80/2 9,000 hours

(~1 year)

-




1 99% per

1 cm

2

3

4

5

6

7

Defra


© Defra 2016

120

System monitoring



1 Lamp on /

Fail indicator

2

3

4

5

6

7

Maintenance



1 Replace lamp / and o rings

2

3

4

5

6

7

Operational control

Model Flow

control

UV

exposure

UV output



1 - - - -

2

3

4

5

6

7

Defra


© Defra 2016

121

D6 Hanovia

http://www.hanovia.com/

Parent company: Halma plc

http://www.halma.com/


Model

Flow rate

(26 J/cm2)

(95% UVT)

Other std flow

rate

(120 J/cm2)

(95% UVT)

Inlet /

outlet

port

size


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


Model Voltage

System power

consumption

(Watts)


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

http://www.hanovia.com/

http://www.halma.com/

Defra


© Defra 2016

122

Lamp specification



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



D 0007 - >70% -

D 0013

D 0023

D 0047

D 0089

System monitoring


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


© Defra 2016

123

Maintenance


Pureline D Range

D 0007 UV Lamps

D 0013

D 0023

D 0047

D 0089

Operational control

Model Flow

control

UV

exposure

UV output


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/


Model Flow rate

(250 J/m2)

Other std flow

rate

(400 J/m2)

Inlet /

outlet

port

size


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”

http://www.hydrotec.co.uk/

Defra


© Defra 2016

124

Model Flow rate

(250 J/m2)

Other std flow

rate

(400 J/m2)

Inlet /

outlet

port

size


BSP

External

Thread

10E 14.9 m³/h 9.25 m³/h 2” BSP

External

Thread


Model Voltage

System power

consumption

(Watts)


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


(h)

Lamp power


HydroPUR

2E Ecolux

20N

Mercury

LP with

indium-

amalgam

8,000 17 -

5E Ecolux

30N

27

10E Ecolux

40N

38

Defra


© Defra 2016

125



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


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


HydroPUR

2E UV lamps

Quartz sleeves

-

5E

10E

Defra


© Defra 2016

126

Operational control

Model Flow

control

UV

exposure

UV output


HydroPUR

2E - - - -

5E

10E

D8 LIFF

http://www.lifffilters.co.uk/lifffilters/index.asp

Parent company: BWT Ltd.

http://www.bwt-uk.co.uk/


Model

USEPA std

flow rate

m3/h

(40 mJ/cm2)

Other std flow

rate

m3/h

(30 mJ/cm2)

Inlet /

outlet

port

size


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.

http://www.lifffilters.co.uk/lifffilters/index.asp

http://www.bwt-uk.co.uk/

Defra


© 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


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


Model Voltage

System power

consumption

(Kw) approx.


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 1 230V 50Hz 0.015

Defra


© Defra 2016

128

Model Voltage

System power

consumption

(Kw) approx.


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




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 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


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


© Defra 2016

129

Model Filtration UV-T Other related

information

P55N Ultra Violet

Disinfection


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


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



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


© Defra 2016

130


Disinfection


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



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 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


© Defra 2016

131

Model Flow

control

UV

exposure

UV output


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/


Model Flow rate Other std flow

rate

Inlet /

outlet

port

size


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

http://www.prosep.co.uk/

http://www.parker.com/

Defra


© Defra 2016

132

Model Flow rate Other std flow

rate

Inlet /

outlet

port

size


SS4875 18.6 m3/h 2”

SS1055 21.3 m3/h 3”

SS61075 34.8 m3/h 3”


Model Voltage

System power

consumption

(Watts)


SE Series

SE1 230/240 V

50 Hz

21 W Supply fuses protection 3A

SE2 28 W

SE3 40 W


SS15 230/240 V

50 Hz

15 W

SS30 30 W

SS55 55 W

SS75 75 W

SS1475 75 W

SS1575 75 W


SS4855 230/240 V

50 Hz

220 W

SS4875 300 W

SS1055 330 W

SS61075 450 W

Defra


© 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


SS15 LP Mercury Vapour Approx.

8,760

15 W 1

SS30 LP Mercury Vapour 30 W 1






SS4855 LP Mercury Vapour Approx.

8,760

220 W 4x55

SS4875 LP Mercury Vapour 300 W 4x75





SE Series

SE1 - - Max 0 – 10bar

SE2

SE3


SS15 - - Max 0 – 10bar

SS30

SS55

SS75

SS1475

Defra


© Defra 2016

134


SS1575


SS4855 - - Max 0 – 10bar

SS4875

SS1055

SS61075

System monitoring


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


SS15 Power indicator Lamp

run/fail

indicator

Hours run

metre

Lamp operating; the green




illuminates

SS30

SS55

SS75

SS1475

SS1575


SS4855 Power indicator Lamp

run/fail

indicator

Hours run

metre

Lamp operating; the green




illuminates

SS4875

SS1055

SS61075

Defra


© Defra 2016

135

Maintenance


SE Series

SE1 UV lamps

O rings

Quartz sleeves

-

SE2

SE3


SS15 UV lamps

O rings

Quartz sleeves

-

SS30

SS55

SS75

SS1475

SS1575


SS4855 UV lamps

O rings

Quartz sleeves

-

SS4875

SS1055

SS61075

Operational control

Model Flow

control

UV

exposure

UV output


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


© Defra 2016

136

Model Flow

control

UV

exposure

UV output



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


AC alarm contacts. SS30

SS55

SS75

SS1475

SS1575


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


AC alarm contacts. SS4875

SS1055

SS61075

Defra


© Defra 2016

137

D10 Silverline UK Limited

http://silverlineuk.co.uk/


Model flow rate

30mJ / cm2 (UVT 98%)

Inlet /

outlet

port

size


UV-DS Steriliser Range

UV-DS08 4 ½” BSP

UV-DS15 8 ¾” BSP

UV-DS30 21 ¾” BSP

UV-DS55 36 ¾” BSP





UV-DS08 220/240 V 8 W

UV-DS15 15 W

UV-DS30 30 W

UV-DS55 55 W

Lamp specification




UV-DS08 Low

pressure

8,000 8 W

UV-DS15 15 W

UV-DS30 30 W

UV-DS55 55 W

http://silverlineuk.co.uk/

Defra


© Defra 2016

138




UV-DS08 5 µm 98% 100 psi

UV-DS15

UV-DS30

UV-DS55

System monitoring



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



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


© Defra 2016

139

Operational control

Model Flow

control

UV

exposure

UV output



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


Model

USEPA std

flow rate

(40 mJ/cm2)

Other std flow

rate

Inlet /

outlet

port size


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

http://viqua.com/sterilight/

http://trojanuv.com/uvmax-sterilight

Defra


© Defra 2016

140

Model

USEPA std

flow rate

(40 mJ/cm2)

Other std flow

rate

Inlet /

outlet

port size


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





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


© Defra 2016

141




SCM-200 100-240V

50-60Hz

35 W

SCM-320 42 W

SCM-600 70 W

SCM-740 82 W


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




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


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


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


© Defra 2016

142




SP320-HO 5 µm pre-

filter

> 75% 2-40°C, max 8.62 bar.


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


SCM-200 5 µm pre-

filter

> 75% 2-40°C, max 8.62 bar.


Iron: < 0.3 mg/L


Turbidity: < 1 NTU


Tannins: < 0.1 mg/L

SCM-320

SCM-600

SCM-740


S1Q-PA 5 µm pre-

filter

> 75% 2-40°C, max 8.62 bar.


Iron: < 0.3 mg/L


Turbidity: < 1 NTU


Tannins: < 0.1 mg/L

S2Q-PA/SSM-17

S5Q-PA/SSM-24

S8Q-PA/SSM-37

S12Q-PA/SSM-39

System monitoring



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


© Defra 2016

143


Audible ‘lamp replacement

reminder’

‘Total controller running time’

display.

‘Power on’ display

Low UV intensity can be used to

close inlet solenoid valve.


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


reminder’


display.




SCM-320

SCM-600

SCM-740


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


reminder’


display.




S2Q-PA/SSM-17

S5Q-PA/SSM-24

S8Q-PA/SSM-37

S12Q-PA/SSM-39

Maintenance



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


SCM-200 Descale / clean quartz thimble. Replace lamp

/ and o rings.

Safety-Loc™ connector with

interlock that ensures power is SCM-320

Defra


© Defra 2016

144


SCM-600


UV monitor)

disconnected before lamp can

be removed. SCM-740


S1Q-PA Descale / clean quartz thimble. Replace lamp

/ and o rings.


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



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


SCM-200 Flow

restrictor on

outlet to

limit

throughput

Output for

closing

inlet

solenoid

valve (if

UV% low)

?

SCM-320

SCM-600

SCM-740


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


© Defra 2016

145

D12 VIQUA

http://viqua.com/uvmax/


Model

Flow rate

(>30 mJ/cm2)

(95% UVT)

Other std flow

rate

(40 mJ/cm2)

Inlet /

outlet

port

size


UV Max

UVMax A 0.42 m3/h -

3/8”

NPT

UVMax

F4plus

6.93 m3/h

1” NPT


Model Voltage

System power

consumption

(Watts)


UV Max

UVMax A 120-240V 22

UVMax F4plus 130

Lamp specification

Model Type Lamp life (h)

Lamp power

consumption

(Watts)


UV Max

UVMax A LP 9,000 14

UVMax F4plus LPHO 9,000 110 Sterilume

http://viqua.com/uvmax/

Defra


© Defra 2016

146


Model Filtration UV-T

(%)

Water

temperature

(°C)


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


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


UV Max

UVMax A

UVMax F4plus Quartz sleeve, lamp, O rings

Operational control

Model Flow

control

UV

exposure

UV output


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


© 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


Model

USEPA std

flow rate

m3/h

(40 mJ/cm2)

Other std flow

rate

m3/h

(30 mJ/cm2)

Inlet /

outlet

port

size


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”


Model Voltage

System power

consumption

(kW)



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

http://www.xylem.com/treatment/ca/brands/wedeco

http://www.xyleminc.com/en-us/Pages/default.aspx

Defra


© Defra 2016

148

Lamp specification



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




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 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


© 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



Aquada 1 Descale / clean quartz thimble. Replace lamp

/ and o rings.


UV monitor)

Aquada 2

Aquada 4

Aquada 7

Aquada 10

Operational control

Model Flow

control

UV

exposure

UV output



Altima Safe-T-Cap

lamp

connector

system.

Proxima Safe-T-Cap

lamp

connector

Defra


© Defra 2016

150

Model Flow

control

UV

exposure

UV output


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


© 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


© 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


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


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



private supplies. This is compounded by the lack of information provided by equipment



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


© Defra 2016

153

maintenance of other equipment and replacement of cartridge filters is less clear.

Maintenance logs are not kept by users.


undetected for some time because a lack of a prominent alarm, and will generally not



by users.

Defra


© 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


installer)

None – direct

supply to taps


concerns re hydrocarbons (due to local heating oil

contamination), nitrate and micro-organisms (E. coli).


log.


3

Commercial

(various

business units)

Borehole None Wedeco Aquada AG (Type 2) sited

internally

UVT = 97.6%



plumber)

None – direct

supply to taps

Hard groundwater, colour 0.3°H, turb. 0.12 NTU.


log.

Risk assessment completed.

4

Commercial

(Holiday

cottages and

function centre)

Borehole (deep) Fe filter SS55 UV Steriliser (Type A) sited

externally


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.


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


concerns re nitrate and micro-organisms.

No user instructions, manufacturers’ information or

maintenance log.

Difficult access to UV unit.

Defra


© Defra 2016

155

Property ref /


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%


None known –

direct supply to

taps


concerns re iron and micro-organisms.

“Do not adjust” tag on flow valve.


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%


Service contract (with local plumber)

None known –

direct supply to

taps and storage

tanks

Colour 3.6°H, turb. 0.10 NTU;


maintenance log.


8

Commercial

(stables and

tenanted

properties)

Shallow

borehole

None Wedeco UV-C (5.5 m3/h, 80 W) /

Aquada sited externally

UVT = 96.8%


Service contract (with local plumber)

None known –

direct supply to

taps and storage

tanks

Hard groundwater, colour 0.5°H, turb. 0.11 NTU;


maintenance log.


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.


maintenance log.


Defra


© Defra 2016

156

Property ref /


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


installer)

None known –

direct supply to

taps and storage

tanks

Hard groundwater, colour 0.5°H, turb. 0.12 NTU.


maintenance log.


11

Commercial

(two properties,

one is a holiday

let)

Shallow

borehole

Fe filter Unbranded UV sited externally

UVT = 96.4%

No flow meter



installer)

None – direct

supply to taps

Hard groundwater, colour 1.0°H, turb. 0.20 NTU; previous

failures for Fe, Mn, turbidity and coliforms.


maintenance log.


Defra


© Defra 2016

157

E3 Local Authority ‘B’


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.


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


temperature.

Defra


© 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



private supplies.


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.


undetected for some time because of a lack of a prominent alarm, and will generally not



by users.

Defra


© Defra 2016

159

Table E.2 Summary of site visits to Local Authority ‘B’

Property ref /


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


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


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


No user instructions, manufacturers’ information.

Maintenance log kept.

3

Large Domestic

(Caravan Park)

Borehole Mn removal

(Perhaps with

activated

carbon in Mn

removal

column)


UVT= 98.8%

Water meter


Service contract for Mn removal (with

local installer)

None – direct

supply to taps



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


None – direct

supply to taps

Colour 0°H, turb. 0.17 NTU


maintenance log.

Defra


© Defra 2016

160

Property ref /


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


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


User instructions and manufacturers information supplied to

user with UV unit

Defra


© Defra 2016

161

E4 Local Authority ‘C’


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.



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


Defra


© 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


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


© Defra 2016

163

automatic shut-off valve, and will generally not prevent flow and the possibility of the

consumption of non-disinfected water.


by users.

Defra


© Defra 2016

164

Table E.3 Summary of site visits to Local Authority ‘C’

Property ref /


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.


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.


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.


installer).

None – direct

supply to taps

Colour 0.8°H, turb 0.13 NTU.



Limited service history record provided.

Risk assessment carried out previously (2012)

Defra


© Defra 2016

165

Property ref /


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%


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.


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.


Defra


© Defra 2016

166

E5 Local Authority ‘D’


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.



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


© 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


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


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


Defra


© Defra 2016

168

Table E.4 Summary of site visits to Local Authority ‘D’

Property ref /


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).


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.


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.


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.


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.


No P&I D drawing at site, no valve identification. Manual for UV

available at site.

Detailed service history record provided.


Owner believes that 3 yearly interval for lamp replacement is

Defra


© Defra 2016

169

Property ref /


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.


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


Very limited service history record – inspection date and

signature only.

Defra


© 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

mailto:[email protected]

mailto:[email protected]

Defra


© 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)?

Documents

Comparison of Private Water Supply and Public Water Supply …dwi.defra.gov.uk/research/completed-research/reports/DWI70-2-306.… · purpose of discovering and delivering new and