UV Disinfection for Drinking Water

David S. Briley, PE

Water JAM 2010

Virginia AWWA 2015 Operators Conference Virginia Beach, VA

Agenda  Principles of UV Disinfection  Applications of UV Disinfection for Drinking Water  Design Considerations  Validation Testing  Case Study – City of Raleigh, NC

AWWA NC 20102011 JAMWEA Water

 Obtaining Giardia/Crypto Credit for UV

Water JAM 2010

Principles of UV Disinfection

Principle of Ultraviolet (UV) Disinfection

AWWA NC 20102011 JAMWEA Water

 Physical process using electromagnetic energy to prevent DNA and RNA from further replication  Germicidal UV irradiation range - 200 to 300  UV produces no residual

Disinfection Mechanism  Dimerization of DNA (thymine bases)  Inability to Reproduce  Bug is Non-infective  Dark repair has been observed in some bacteria

Dimer

AWWA NC 20102011 JAMWEA Water

Dimer

5

AWWA NC 20102011 JAMWEA Water

Disinfection Effectiveness vs. Wavelength

6

Comparison of UV Lamps

Low-Pressure, High Output

Medium Pressure

Pulsed-UV

Nearly monochromatic

Polychromatic (185 – 1,387 nm)

Polychromatic (185 – 800 nm)

100-200

500 – 850

7,000 – 15,000

Lamp life (hr)

8,000 – 12,000

2,000 – 8,000

925 @ 30 Hz

Efficiency (200-300 nm)

35 – 40%

15 – 25%

85% UVT

Turbidity

Shields pathogens

< 1 NTU

Hardness

Cause scaling on quartz sleeves reducing UV intensity

< 200 mg/L CaCO3

pH

Affect solubility of metals, potentially affecting UVT and fouling

6.0 - 9.0

Iron

Fouling of lamp sleeves

Perform pilot test for lamp sleeve fouling

Suspended Solids

Absorption of UV light and shielding of pathogens

< 10 ppm

21

UV Transmittance (UVT)

AWWA NC 20102011 JAMWEA Water

 Most critical design parameter  Need large dataset to properly select design value (95th percentile)  Typical UVT Values for municipal WTPs: 85-95%

Site and Layout Considerations  Recommended location for UV system is post-filter  Reduces solids which can shield pathogens  Maximize WQ to achieve best UV performance and minimize fouling

 Optimal location is between filters and clearwell  Layout considerations  Sufficient straight pipe upstream of UV reactors  Flowmeter for each UV train  Ensure UV unit remains full under all conditions  Motorized isolation valves

AWWA NC 20102011 JAMWEA Water

 May require relocation of post-filter chemical feeds

Hydraulic Head Requirements  Existing WTPs have limited head b/w filters and clearwell  Need to identify available head in your WTP  Critical to understand headloss through UV to avoid impacting filter operations or clearwell volume  Headloss not directly related to UV technology  Ways to reduce headloss  Locate UV Facility close to filters and clearwell  Install larger reactors or more reactors ($$)

AWWA NC 20102011 JAMWEA Water

Typical Headloss UV Reactors

2 to 24 inches

UV Facility

36 to 48 inches

Hydraulic Considerations  Hydraulic design to allow for even distribution b/w units  Ensure UV unit does not exceed validated flowrate

 Ensure steady flow through UV units  Avoid rapid fluctuations in flow or pressure  Some UV units equipped with baffles to distribute flow through reactor  More complex for pumped systems

 Ensure UV units flooded at all times  Lamps can result in rapid heat buildup and damage components

AWWA NC 20102011 JAMWEA Water

 In large LPHO units, provided air release/vacuum valves

AWWA NC 20102011 JAMWEA Water

UV System Section

Electrical Design Considerations

 Power Supply  Varies significantly by UV technology  Recommend pre-selecting UV equip. or design bid alternates

AWWA NC 20102011 JAMWEA Water

 LPHO Systems: 0.5 to 1.4 kW/mgd  MPHO systems: 2.0 to 2.4 kW/mgd  Uninterruptible Power Supply  Power conditioning (electronic ballasts sensitive to power spikes)  Ensures UV operation (and continued disinfection) until standby power starts up  In event of generator failure, allows for controlled UV shutdown to prevent flow entering clearwell

On-Site vs Offsite Validation  Significant challenges with on-site testing      

Operational limitations with online WTP Limited available water for test matrix May require partial or full WTP shutdowns Capacity of backwash handling facilities to handle test water Consistency of water quality and UVT during testing Issues with chlorine in UV influent

AWWA NC 20102011 JAMWEA Water

 Offsite Validation Testing allows for better control of test parameters  UV manufacturers have validated reactors for most applications

Validation Testing • Cannot directly measure UV residual or inactivation • Validation testing required to confirm performance

AWWA NC 20102011 JAMWEA Water

• Validation testing should cover design conditions (max flow, min UVT, etc)

UV Dose Response Curves MS-2 Reduction [Log (No/N)]

2.5 MS-2 - low pressure

2 MS-2 - 255 nm UV-LEDS

1.5

1

0.5

0 0

10

20

30

40

50

60

AWWA NC 20102011 JAMWEA Water

UV Dose (mJ/cm2)

30

UV Definitions - UV Dose  Reduction Equivalent Dose (RED) – Inactivation measured during full-scale reactor testing correlated to UV dose-response curve from collimated beam testing.  Required Dose (Dreq) – UV dose specified in LT2ESWTR to achieve target log inactivation for target pathogen.  Validated Dose (Dval) – UV dose delivered by UV reactor as determined through validation testing. Compared to the required dose to determine log inactivation credit.

AWWA NC 20102011 JAMWEA Water

 Calculated Dose – The RED calculated using the dosemonitoring equation that was developed through validation testing.

Dval

RED   Dreq VF

31

UVDGM Requirements for UV Design  Relate UV facility design to validation system setup  UVDGM Option 1  Straight pipe upstream of UV unit during validation testing (X)  +5 pipe diameters  Requires coordination with UV suppliers during concept design

AWWA NC 20102011 JAMWEA Water

X

Relating Validation Setup to Installation  Option 2 – Site Specific Validation Test  Identical piping layout for 10D upstream and 5 D downstream  Costly!!

 Option 3 –Velocity Profiles

AWWA NC 20102011 JAMWEA Water

 CFD modeling to demonstrate velocity profile is similar or better than during validation testing

Validated Reactor

Installed Reactor

Regulatory Approval for UV Disinfection  UV systems have been granted inactivation credit for Crypto. and Giardia in:      

Arizona California Tennessee Utah Washington Wisconsin

AWWA NC 20102011 JAMWEA Water

 NC: Granted credit to UV system at D.E. Benton WTP (City of Raleigh) in Oct. 2013.  Other States are currently developing guidance or protocols such as New York.

Case Study: Raleigh, NC  Dempsey E. Benton WTP  Permitted Capacity: 16 mgd  Design Capacity: 20 mgd WTP  WTP online May 2010  Constructed to provide reliability in City’s water system

AWWA NC 20102011 JAMWEA Water

 EM Johnson WTP  Capacity = 86 mgd

Raleigh’s Goals for UV Disinfection  To provide multiple barrier disinfection  Urbanized watershed  Source water quality variability  Some Crypto. hits but still Bin 1

 To lower DBPs – simultaneous compliance with LT2ESWTR and Stage 2 DBPR  Receive inactivation credit for Crypto. and Giardia  DBP Compliance Strategy

AWWA NC 20102011 JAMWEA Water

 Reduce free chlorine contact in WTP  Can be effective if in combination with chloramines

UV Disinfection System Design  Three 10-mgd UV reactors  N+1 redundancy  Design Flowrate = 20 mgd  Design dose = 40 mJ/cm2  Design UV

AWWA NC 20102011 JAMWEA Water

Transmittance = 90%

UV Disinfection System Design

 LPHO Reactors (Wedeco K Series)  Space for 2 future reactors  UV Reactor  5 lamp rows  12 lamps per row  60 lamps total

 Splitter weirs upstream

AWWA NC 20102011 JAMWEA Water

of each UV unit  36” mag flowmeter upstream of each UV unit

UV Disinfection System Design  UPS  Operate 2 UV reactors for 15 mins  Generator w/ ATS  Effluent valves on each UV reactor powered by UPS  Complete shutdown if standby generators do not

AWWA NC 20102011 JAMWEA Water

start to prevent plant flow without UV disinfection  Clean-in-place system for periodic cleaning of UV lamp sleeves

Key Issues for Approval for Inactivation Credit

AWWA NC 20102011 JAMWEA Water

 Most States are following UVDGM  Off-Spec operations  UV Lamp Breakage Risks and Response Plan  Combined lamp aging/fouling factor (CAF)  Monitoring and controls to ensure disinfection  Backup plan in the event of UV system failure  Monitoring and reporting forms  Validation Testing  Low wavelength action spectra (applies to MPHO)

Off-Spec Operations  EPA UVDGM: at least 95% of water delivered through UV reactors operating within validated conditions  UV System controls can limit off-spec operations  Some regulators don’t like 5% if UV is for primary disinfection

AWWA NC 20102011 JAMWEA Water

 Off-spec for no more than 15 mins at a time (NC, UT, WA)  Off-spec for no more than 0.1% (WI)

UV Off-Spec Operations

AWWA NC 20102011 JAMWEA Water

0.35

0.14%

0.30

0.12%

0.25

0.10%

0.20

0.08%

0.15

0.06%

0.10

0.04%

0.05

0.02%

0.00

0.00% J

F M A M J

J

A S O N D J

Total Off-Spec Flow (MG)

F M A M J

J

Total Off-Spec Flow (%)

A S

Percent of Treated Flow

58 off-spec events in 20 months Each event < 5 mins Most were due to faulty UVT analyzer at the time UVT analyzer issues have been since corrected

Off-Spec Volume (MG)

• • • •

UV Transmittance Analyzer 100 99

UV Transmittance (%)

98 97 96 95 94 93 92 91

AWWA NC 20102011 JAMWEA Water

90 N

F

J

S

UV Transmit (%) Minimun

D

A

J

UV Transmit (%) Average

O

J

Mercury Release Findings - LPHO

AWWA NC 20102011 JAMWEA Water

 Following a lamp break, the concentrations of mercury in the water passing through the reactor will be an order of magnitude or more less than the regulatory MCL of 2 μg/L.  With amalgam lamps, the majority of the mercury will be within the solid amalgam.  Liquid or amalgam mercury will settle to the bottom of the reactor.  Mercury and quartz shards can be captured using isolation valves and low velocity zones downstream of reactor.

Lamp Breakage Risks  Regulators concerned about lamp breakage and potential mercury release  Identified lamp breakage risks and demonstrate how design and operation will mitigate these risks

 Debris:  Downstream of filters, and splitter weirs  Baffle at UV reactor inlet

 Water Hammer:  Free water surface upstream and downstream

 Partially full reactor:

AWWA NC 20102011 JAMWEA Water

 Downstream weir set to keep reactor flooded  Level switch at top of reactor shuts down reactor when activated

Lamp Breakage  Hg has high density and low solubility  LPHO lamps have order of magnitude less Hg than MP lamps  WRF Research showed much of Hg will be trapped in reactor and Hg release MAX VALIDATED UVT, UV dose calc will clamp at MAX VALIDATED UVT If UVT < 90% [DESIGN UVT], all rows on at 100%, UV dose calc will clamp at 90% If low intensity is detected, new lamp row will start Flowmeter failure, design flow (10 MGD) used in UV dose calc.

 If off-spec condition is detected  Auto start standby reactor  Approx 5 mins to start new reactor and achieve 100% power

AWWA NC 20102011 JAMWEA Water

 Power Loss  Standby generators called to start  UPS can power UV system for up to 15 mins  If generators fail to start, PLC will close effluent valves and shutdown UV system after set delay (~10 mins)

Contingency Planning  NC PWS very focused on contingency planning  Recommend focus on redundancy and reliability during design and development of control logic  Minimize under-disinfected water (off-spec) from entering distribution system  City of Raleigh developed Plan for Total UV System Failure

AWWA NC 20102011 JAMWEA Water

 Convert back to meeting Giardia CT via free chlorine disinfection  Requires moving point of ammonia feed and converting clearwell from chloramines back to free chlorine

Monitoring and Reporting UV Reactor 1 Date

AWWA NC 20102011 JAMWEA Water

1/1/11 1/2/11 1/3/11 1/4/11 1/5/11 1/6/11 1/7/11 1/8/11 1/9/11

UVT

UV Reactor 2

OffMin. Runtim Flow Spec RED Runtim Flow e (Hrs) (MG) Flow (mJ/cm2 e (Hrs) (MG) (MG) ) 24.11 4.38 0.00 82.04 24.04 4.34 24.04 4.41 0.00 82.06 24.04 4.34 24.12 4.41 0.00 82.04 24.10 4.34 24.06 4.44 0.00 82.05 24.06 4.38 24.05 3.84 0.00 82.06 24.04 3.80 24.02 3.94 0.00 82.06 24.05 3.87 24.00 4.44 0.00 82.26 24.05 4.34 23.90 4.34 0.00 82.04 23.90 4.28 24.02 4.28 0.00 82.06 24.02 4.28

UV Reactor 3

OffSpec Min. RED Runtim Flow (mJ/cm2) e (Hrs) (MG) 0.00 82.00 0.00 0.00 82.00 0.00 0.00 82.00 0.00 0.00 82.00 0.00 0.00 82.00 0.00 0.00 82.00 0.00 0.00 82.00 0.00 0.00 82.00 0.00 0.00 82.00 0.00

OffMin. Flow Spec RED Runtim (MG) Flow (mJ/cm2 e (Hrs) (MG) ) 0.00 0.00 0.00 N/A 0.00 0.00 0.00 N/A 0.00 0.00 0.00 N/A 0.00 0.00 0.00 N/A 0.00 0.00 0.00 N/A 0.00 0.00 0.00 N/A 0.00 0.00 0.00 N/A 0.00 0.00 0.00 N/A 0.00 0.00 0.00 N/A

1/10/11

24.03 4.69

0.00

82.05

17.03 2.91

0.00

82.00

0.00

0.00

0.00

0.00

N/A

1/11/11

24.08 4.28

0.00

82.05

24.03 4.19

0.00

82.00

0.00

0.00

0.00

0.00

N/A

1/12/11

21.89 3.19

0.00

82.07

24.01 3.72

0.01

82.00

0.00

0.00

0.00

0.00

N/A

1/13/11

24.09 3.05

0.00

82.06

24.01 2.89

0.00

82.00

0.00

0.00

0.00

0.00

N/A

1/14/11

24.03 3.41

0.00

82.07

24.01 3.28

0.00

82.00

0.00

0.00

0.00

0.00

N/A

Summary  UV Disinfection can be effective for LT2 and Stage 2 DBP Rule Compliance  UV has been approved for log inactivation credit in several states. With more installations, regulatory agencies are developing a comfort level with reliability of UV  Talk to regulatory agency early in the process and about your unique design/circumstances  If you want to reduce chemical CT,

AWWA NC 20102011 JAMWEA Water

 Don’t expect it to be a quick process  Prepare to dive into the details with regulatory agenday  Prepare to develop more in-depth SOPs and more elaborate controls to ensure CONTINUED DISINFECTION

Questions

AWWA NC 20102011 JAMWEA Water

David S. Briley, P.E. (919) 833-7152 [email protected]