United States
Environmental Protection
Agency
Municipal Environmental Research ^» '"i
Laboratory ^Er*
Cincinnati OH 45268 '/1 i
Research and Development
EPA-600/S2-84-087 May 1984
&ERA Project Summary
Operation and Maintenance of
Selected Ozone and Ultraviolet
Disinfection Systems
Randy Junkins
A series of onsite evaluations were
made of wastewater and drinking water
treatment plants that use ozone or ultra-
violet (UV) light disinfection in place of
chlorine disinfection. The object was to
compile design and operational informa-
tion on such plants. The evaluations
were conducted at 10 municipal waste-
water treatment plants (7 that have
used or are using ozone for disinfection
and 3 that have used or are using UV
disinfection) and at 5 drinking water
treatment plants (all of which use ozone
for disinfection or odor control or both).
During these plant visits, operating data
were reviewed, operational practices
were observed, and operating personnel
were interviewed to establish factors
related to both poor and efficient pro-
cess performance. Typical operation
and maintenance problems are listed,
and the recommended remedial actions
for correcting those problems are pre-
sented.
This Project Summary was developed
by EPA's Municipal Environmental Re-
search Laboratory, Cincinnati. OH, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Alternatives to chlorine disinfection of
municipal wastewater have been given
increased attention in recent years. Two
alternative approaches that have gener-
ated particular interest are ozone and
ultraviolet (UV) light disinfection. This
project was initiated to identify operation
and maintenance (O&M) factors affecting
the performs nee of ozone a nd U V disi nf ec-
tion systems. The study was part of U.S.
Environmental Protection Agency (EPA)
efforts to compile and promulgate design
and operational information on ozone and
UV light disinfection of municipal waste-
water.
The object of the study was to deter-
mine, analyze, and set priorities for those
O&M factors that affect the operational
efficiencies of ozone and UV disinfection
systems. During the study, onsite evalua-
tions were conducted at 10 municipal
wastewater treatment plants and 5
municipal water treatment plants that
use ozonation or UV irradiation for disin-
fection or taste and odor control or both.
During these plant visits, operating per-
sonnel were interviewed, operational
practices were observed, and operating
data were reviewed to identify O&M
factors related to both poor and efficient
process performance. The information
compiled was documented in individual
plant evaluation reports, which are sum-
marized in the final project report. The
project report presents the O&M problems
encountered, conclusions drawn concern-
ing their cause, and recommendations for
their resolution.
Results
Fifteen sites were visited during the
study. Twelve of the facilities used ozone
systems for disinfection or taste and odor
control or both, and three of the plants
used UV disinfection systems. Data con-
cerning the ozone and UV systems evalu-
ated are presented in Tables 1 and 2,
respectively.
-------
Table 1.
Plant
No.
1
2*
3
4
5t
6f
7t
8
9
10
11
12
System Data
Number
of
Generators
17
2
3
13
2
3
3
2
4
6
4
4
for Ozone Plants
Generator
Manufacturer
Union Carbide
Welsbach
Emery
Union Carbide
Emery
Emery
Union Carbide
PCI
Welsbach
Trailigaz
Trailigaz
Degremont
Total
Capacity
kg/day
(Ib/day)
2,082
(4,590)
69
(152)
782
(1.725)
1,562
(3,445)
113
(250)
381
(840)
571
(1.260)
204
(450)
453
(1.000)
3,479
(7,680)
204
(450)
227
' '
Carrier
Gas
02
Air
Air
02
Air
02
°*
Air
Air
Air
Air
Air
Actual
03
Transfer
Efficiency*
50
50
60
95
67
84
—
85
80
--
94
"Transfer efficiency = OzfinJ-Osfout)
• Ozone system electronics are too
complicated for plant personnel to
perform routine maintenance and
repair work.
• Corrosion problems occur with Oa
analyzer valve components.
• Ozone contact tanks were constructed
belowthe system control room, and Oa
leaks cause instrumentation rubber
seals to corrode.
• System equipment (i.e., generators and
compressors) is very noisy.
• The system includes numerous pieces
of equipment that must be maintained.
• System instrumentation must be con-
tinually calibrated.
• Ozone generators are a high mainte-
nance item and continually blowf uses.
• Corrosion problems with compressors
occur because of wet gases.
• Catalyst poisoning occurs in 03 de-
struction system.
• No dew point monitoring equipment is
provided with the ozone system.
• Excessive heat buildup in the O3
generator room causes generator
heaters to shut down units.
A MnnfnmnntiHIf* matariale ftret ncoH in
O3 fin)
Table 2. System Data for UV Plants
t/Vof in operation.
Plant
No.
13
14
15
UV
Manu-
facturer
Pure Water
Systems
Aquafine
Ultraviolet
Technology,
Inc.
No. of
Sections
6
4
3
Total
Number of
Lamps
392
128
48
Lamp
Length
m (in.)
1.5
(60)
1.5
(60)
0.8
(30)
Cleaning
Mechanisms
Mechanical
Wiper
Mechanical
Wiper
Chemical
Detergent
Flow/
Unit
m3/d
(mgd)
2.839
(0.75)
4,163
(1.25)
113
(0.03)
Arc Length
Unit of Flow.
m/m3/per
day
(in./gpm)
44,451
(4.6)
28,990
(330)
289.902
(30)
Costs
Annual costs for ozone plants with
compiled historical cost, data are pre-
sented in Table 3. The average ozone
production cost for the 12 plants surveyed
is $4.20/kg ($1.90/lb) of ozone gener-
ated. Very few operating cost data were
available for the UV plants surveyed
during the project. One plant reported a
UV disinfection operating cost of approxi-
mately $0.18/1,000 gallons treated.
Typical Problems Encountered
Typical O&M problems encountered
with ozone disinfection systems were
maintenance-related and are as follows:
• Multiple and frequent failures occur in
the ozone generator cell.
• Failures occur in the silicone control
rectifier (SCR).
• Severe foam problems occur with the
contact tank gas recovery system.
construction: Rubber sleeves are used
on compressors, and chlorinated,
rubber-based paint is used to seal the
ozone contact chamber.
Dew point meters are not reliable.
Improper (i.e., too high) dew point
setting causes dielectric failures.
Improper cleaning procedures for di-
electric tubes cause pitting and crack-
ing of tubes.
Typical O&M problems encountered
with UV disinfection systems included
the following:
• Ballasts on the UV lamps overheat and
shut down the system.
• Unit does not consistently disinfect
plant effluent to the degree required by
National Pollution Discharge Elimina-
tion System (NPDES) permit.
• Foam buildup interferes with operation
of the cleaning mechanism.
• Low flow rate causes the unit to
overheat.
• Algae accumulations in the unit inter-
fere with system operation.
-------
Conclusions
Most of the O&M problems experienced
with the ozone disinfection systems sur-
veyed were site-specific and not prevalent
problems that repeatedly occurred at all
the sites visited. The most common prob-
lems identified were the following:
• Inefficient ozone destruction in the
destruct unit.
• Ozone generator cell failure.
• Ozone leakage from the generator.
• Inadequate air-drying in the inlet air
desiccant dryer.
• Improper sealing of the ozone contact
tank.
• Malfunctioning of the ozone concen-
tration monitors.
Limited information was available con-
cerning the operation of UV disinfection
systems; but as with ozone disinfection,
most of the O&M problems identified
were site-specific and not widespread
occurrences.
Table 3. O&M Cost Summary for Ozone Systems
Plant
No.
1
2
3
5
8
9
10
11
12
0.5
Plant Name
Rocky River WWTP
Upper Thompson WWTP*
Frankfort WWTP
Brookings WWTP*
Monroe WTP
Bay Metro WTP
Charles J. Des Baillets WTP
Pierrefonds WTP
Sherbrooke WTP
Current
Average
Flow
ma/day
(mgd)
45.420
(12.0)
2.271
(0.6)
17.033
(4.5)
8.327
(2.2)
26.469
(7.0)
37.851
(10.0)
798.656
(211)
56.800
(15)
58,700
(15.51
Ozone
Production
kg/day
(Ib/day)
1.010
(2.227)
22
(49)
85
(187)
27
(60)
36
(79)
97
(214)
1.740
(3.836)
135
(298)
113
(250)
Ozone
Production Cost
(/kg Ozone
Produced
(C/lb)
63
(28)
602
(272)
382
(173)
488
(221)
903
(408)
575
(261)
191
(86)
342
(154)
259
(116)
C/m3 of O2
Treated
(C/ 1.000 gal)
1.4
(5.2)
5.8
(22)
1.9
(7.1)
1.2
(5.9)
1.2
(4.8)
1.5
(5.8)
0.4
(1.4)
0.8
(2.9)
(2.0)
*Not in operation.
Remedial Actions
Recommended
Remedial actions recommended for
correcting O&M problems observed at
ozone and UV plants are presented in
Tables 4 and 5.
Table 4.
Recommended Remedial Actions for Correcting O&M Problems Observed at Ozone
Plants
Problem
Remedial Actions
Foam buildup in Oa contact tank
Os generator cell failure
Poor Oa transfer efficiency in the
Oa contact tank
Oa leakage from generator
Poor O3 destructor performance
Premature shutdown of feed-gas
compressor
Failure of silicone control rectifier
System air compressors discharge
more air than required
Refrigerant dryer, overloaded
Inefficient performance of inlet
air desiccant dryer
Provide foam sprayer in contact tank or separate foam
spray tanks between O3 contact tanks and O3 destruct unit.
Purge O3 generator before startup to ensure the unit is
clean and free of dust or rust particles.
Routinely inspect Oa diffusers and clean when required.
Replace damaged cells in generator; adequately ventilate
the Oa generator room.
Increase the quantity of catalyst in the destruct unit; install
an electric airpreheater before the destruct unit; check for
leakage from destruct unit; replace catalyst.
Replace faulty surge/vibration monitor.
Determine cause of problem and replace faulty silicone
control rectifier.
Change pulley size to reduce operating speed.
Check the inlet air compressor water seal system to
ensure that the compressor is not overheating and heating
the inlet air to the dryer.
Recharge the dryer with new desiccant material; check
that the desiccant generation heaters are not operating at
too high a temperature; install a dew point indicator/
controller to initiate the desiccant regeneration cycle
automatically when "wet air" is being discharged from
dryer; check the desiccant to ensure it is not being
contaminated with sulfate. chloride, iron (rust) or some
other constituent; clean the purge-air filter.
-------
Table 4. (continued)
Problem
Remedial Actions
Standby electrical generator will
only operate one Oa generator
at a time
Oa piping leakage
Ozone contact tank not covered
and/or properly sealed
High O3 levels in the WWTP
nonpotable water system
Insufficient number of Oa monitors
at WWTP
Dielectric tubes overheating and
breaking because of corrosion
problems
Malfunction of cooling H2O flow and
Oa generator gas pressure sensors/
alarms
Fluctuating ozone generation
Voltage surges that damage
electrical components and
instrumentation
Malfunction of Oa concentration
monitor
High noise levels around Oa
generation equipment
Drive belts on exhaust fans in Oa
generation room frequently fail
Oa generator side panel fasteners
frequently fail
Oa gas sampling equipment leaks
Oa generator shutdown because
of high temperatures in generator
room
Oa exhaust vent located too near
building air intake causing high
ambient 03 concentration inside
building
Ice buildup at outlet of Oa exhaust
vent caused blockage of vent
Switchover valves on inlet air
dryer towers ma/function
Dielectric fuses allow current
flow even when burned out
Residual ozone meter readings
O3 contact tank are unstable
decompression blowers at Oa
contact tank malfunction
Size standby electrical generator to operate sufficient
number of Oa generators.
Install Oa resistant piping (stainless steel); use welded
joints, not threaded joints.
Completely cover and tightly seal the Oa contact tank; do
not use the Oa contact tank for a filter backwash supply
tank; install additional seal/gasket material.
Do not use ozonated water as a nonpotable water supply
source.
Provide Oa monitors for both the lab and other plant
working areas; provide standby Oa monitor.
Check for leaky air release parts in Oa generator cooling
water system; add corrosion prevention chemicals to the
cooling water; thoroughly clean dielectrics and final rinse
with alcohol.
Check for rust particle fouling; add rust prevention
chemical to cooling HxO; check for ozone destruction of
diaphragms in pressure sensors.
Check operation of Oa meters; increase frequency of
cleaning and calibrating meters; check generator opera-
tion and power source.
Check for flow surges to plant resulting from on-off
operation of pumping stations that cause Oa dosage
control to fluctuate; dampen flows to WWTP; install
voltage-control transformer.
Increase frequency of cleaning and calibration check to
once per day if required.
Provide noise abatement provisions around compressors
and generators; install acoustical material around
equipment and in equipment rooms.
Install fans so drive belts have minimum exposure to
exhausted air; use more ozone-resistant belts.
Install heavy-duty, screw-type fasteners on side panels.
Ensure pumps, valves, connections, and piping are
constructed of Oa-resistant material.
Provide ventilation in O3 generator room.
Relocate exhaust vent; extend height of exhaust vent.
Heat trace and insulate the exhaust vent.
Pneumatic cylinders must be rebuilt.
Replace fuses with different type.
Relocate probe closer to section of contact tank where Oa
is being injected.
Check seals and bearings for deterioration as a result of
Oa attack.
-------
able 5.
Recommended Remedial Actions for Correcting O&M Problems Observed at UV
Plants
Problem
Remedial Actions
Slowly open inlet valve into UV chamber; position
mechanical wiper mechanism in the middle of the
chamber for added support.
Replace powerstat with unit that allows finer adjustment-
check minimum power requirements for UV lamps;
replace powerstat with shut-off switch that shuts off
lamps for close control.
Check for faulty LED, burned out LED, or malfunctioning
ballast.
Before startup of unit, be sure that the wiper has not dried
out and become stuck to UV lamps; routinely check wiper
for proper alignment.
Provide adequate difference in elevation between high-
and low-level settings; use time-delayed relays for water
level sensors.
Routinely check flow meter calibration; be sure that it
accurately monitors entire flow range.
Provide foam control measures such as water sprays
before UV chambe/; thoroughly clean tubes before startup
to prevent scum buildup and jamming of lamp wiper.
Slightly slope the teflon tubes to ensure that they remain
full of water during no-flow periods.
udden flow surge into UV
iamber causes damage to
V lamps
tulty powerstat does not allow
jeration of UV system at low
iwer levels resulting in electrical
lergy wastage
ght-emitting diode (LED) does
it give true indication of whether
V lamp is actually operational
V lamp wiper mechanism
equentlyjams
'ater level monitoring system in
V chamber malfunctions.
V chamber influent flow meter
faulty
lam builds up in UV chamber
^logical growth on teflon tubes
'.curs during intermittent flow
inditions
Randy Junkins is with Roy F. Weston. Inc., West Chester, PA 19380.
Francis L. Evans, III is the EPA Project Officer (see below).
The complete report, entitled "Operation and Maintenance of Selected Ozone and
Ultraviolet Disinfection Systems," (Order No. PB 84-180 124; Cost: $22.00,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
P O 0 0 •) J 3 d ^
'.* b civVlK HrciHECTiUN
W t C* I '-I I'< 3 L I ^ K A K Y
<^iO 5 -J^rtKauWN bTKht
C h 1 C ^ o u I L o 0 o 0 4
U.S. GOVERNMENT PRINTING OFFICE: 1984-759-102/9*
------- |