SrEPA
United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-81-222 Oct. 1981
Project Summary
Survey and Evaluation of
Fine Bubble Dome Diffuser
Aeration Equipment
D. H. Houck and A. G. Boon
This research project was initiated
with the overall objective of better
defining the oxygen transfer perfor-
mance, operation and maintenance
requirements, and proper design
approaches for fine bubble dome
diffuser aeration systems used in
activated sludge wastewater treat-
ment.
Working with the British Water
Research Centre of Stevenage, Eng-
land, the Association of Metropolitan
Sewerage Agencies surveyed 19
wastewater treatment plants with
dome diffuser aeration equipment and
reviewed the related literature. Thir-
teen of the plants were in the United
Kingdom, two in The Netherlands, and
four in the United States. The U.K.
plants were selected primarily on the
basis of long-term experience (5 yr or
longer) and were all municipal waste-
water treatment plants with varying
industrial flows. The Netherlands and
U.S. plants were chosen on the basis
of availability rather than longevity.
As nearly as possible, data on
influent and effluent wastewater
characteristics, power demand, air
supply, and process parameters were
compiled for a 5-yr period. Mainte-
nance personnel were interviewed to
develop a summary of long-term
operation and maintenance (O&M)
experience. Specific designs and plant
equipment for aeration, air cleaning,
and diffuser maintenance were studied.
Discussions were held with designers,
equipment manufacturers, and re-
search scientists to develop a better
understanding of design and per-
formance.
Although this survey clearly shows
the need for optimized design and
operating control strategies to realize
the full energy saving potential of this
type of equipment, dome diffuser fine
bubble aeration systems were provid-
ing relatively efficient, low-mainte-
nance service in the plants visited.
This Project Summary was developed
by EPA's Municipal Environmental
Research Laboratory, Cincinnati, OH,
to announce key findings of the
research project that are fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
As with other energy-intensive in-
dustries, energy-conserving design and
operation is receiving increased em-
phasis in the wastewater treatment
field. Aeration equipment employed in
activated sludge service is usually the
single largest energy consumer in a
wastewater treatment plant, normally
accounting for 60 to 80 percent of total
power demand. Because fine bubble
aeration equipment has the potential for
markedly higher oxygen transfer effi-
ciencies than the more traditional
coarse bubble spiral roll design, its use
is rapidly expanding in new or retrofitted
treatment plants.
Historically, fine bubble aeration
equipment was widely used in the
United States before 1950. It gradually
fell into disfavor because of its fairly
-------
intensive maintenance requirements
and was replaced by the very low
maintenance coarse bubble equipment
during the period of relatively cheap
power prior to 1972. Rapid escalation in
U.S. power costs since the 1974 Arab oil
embargo has renewed interest in fine
bubble aeration.
Because power costs have traditionally
been much higher in the United Kingdom
and Western Europe than in the United
States, fine bubble aeration equipment,
along with mechanical surface aerators,
continued to be widely used and
improved there. The ceramic dome
diffuser, which is the main subject of
this study, was first developed in 1954
and refined into its present form by
1961. In 1972, it became available in
the United States under a licensing
agreement. Although there are presently
only a handful of U.S. installations, the
dome diffuser is in use in several
hundred treatment plants around the
world and the last few years have seen
the evolution of competing devices in
either dome or disc form.
The purpose of this study was to
assess the long-term oxygen transfer
performance and O&M history of dome
diffuser aerators. A total of 19 treatment
plants were studied—13 in the United
Kingdom because of the large number
of major municipal treatment works
with 5 yr or greater operating experience
in that country. The British Water
Research Centre (WRC) cooperated in
the U.K. study and was able to add
substantially to the data base. Two
plants in The Netherlands were studied,
and the considerable Dutch research
effort on the various types of dome/disc
diffuser aerators was reviewed. Four
plants were visited in the United States;
three of these were running side-by-
side comparisons with other types of
aeration equipment. A literature review
was carried out in conjunction with the
WRC and EPA. Acorollaryactivity inthis
project was a review of the process
design of dome diffuser aeration systems
and the formulation of design recom-
mendations.
General Design Characteristics
of Surveyed Plants
Most of the visited plants had aeration
systems of the plug flow configuration,
using long, narrowchannels with one or
more passes. Several used step feeding
for better load distribution. All of the
surveyed plants were equipped with
dome diffusers manufactured by Norton/
Hawker-Siddeley.* Most of the plants in
the United Kingdom produced fully
nitrified effluents of high quality;
several practiced denitrification as well.
A list of the surveyed plants and
background data are provided in Table 1.
Aeration Systems
Aeration systems design data for the
surveyed plants are summarized in
Table 2. Average process performance
data for 1978-79 are presented in Table
3.
Plug flow aeration systems were in
use at all of the plants visited. Approxi-
mately one-half of the plants had two or
more passes per aeration tank. The
majority of the plants were operated in
the full plug flow mode with effective
length-to-width ratios up to 106 when
multiple passes were considered. The
U.K. plants exhibited very conservative
design approaches, owing principally to
very stringent discharge requirements.
Only three U.K. plants did not fully
nitrify. Most achieved treatment levels
exceeding 95 percent removal of BOD5,
suspended solids, and ammonia nitro-
gen. Food-to-microorganism (F/M)
loadings in U.K. plants typically ranged
from 0.1 to 0.2 kg BOD5/day/kg mixed
liquor suspended solids (MLSS), and
volumetric loadings ranged from 0.16 to
0.40 kg BODs/day/m3 (10 to 25 Ib/
day/1000 ft3) except in the higher rate
plants or those receiving strong in-
dustrial wastes. Similarly, the nitrifying
U.K. plants consumed two to three
times more air per unit of BOD5 removed
than did the conventional activated
sludge, non-nitrifying, U.S. plants.
Because volumetric loading rates were
lower, however, air flow rates per unit
volume of aeration tank were similar to
those in U.S. plants. Diffuser density
and air flow rates per diffuser were also
quite similar; this reflects the com-
monality of dome diffuser aeration design
in both countries. Tapered aeration, full
or partial, was used in all but four of the
19 plants surveyed.
Mixing power levels at minimum air
flow rates were relatively low in most of
the plants. Only one plant, Minworth,
reported any deposition of mixed liquor
solids; that occurred in the lightly mixed
anoxic zone. Significantly, all of the
lightly mixed plants had very effective
primary sedimentation. MLSS at all of
the plants except Oxford were less than
'Mention of trade names or commercial products
does not constitute endorsement or recommenda-
tion for use.
3500 mg/l. Oxford compensates for
higher-than-average volumetric loadings
by carrying 4500-5000 mg/l MLSS,
maintaining low F/M loadings to pro-
mote nitrification. The range of power
levels given reflects the practice of
tapered aeration, whereby air input (and
hence power input) is front loaded in the
plug flow plants. Often, mixing in the
lightly aerated section of plug flow
plants with tapered aeration was
enhanced by central placement of the
diffusers, along the tank length axis,
carrying a double spiral mixing pattern.
Prevention of Denitrification in
Final Clarifiers.
Single-stage nitrification (BOD re-
moval and nitrification in the same tank)
was being achieved in most of the U.K.
plants surveyed. To combat denitrifica-
tion in the final clarifiers, four plants
have been experimenting with partial
denitrification using anoxic zones in the
front ends of their respective aeration
tank batteries.
Experimental denitrification studies
have been conducted at Rye Meads by
the WRC and the Thames Water
Authority. It was determined that 50
percent removal of nitrate nitrogen was
the practical upper limit of the process |
as used at Rye Meads. Parallel labora-
tory studies suggested that the degree
of denitrification might be increased
another 10 to 20 percent by adding a
second anoxic zone at the beginning of
the third pass at Rye Meads. This has
not been fully supported by the experi-
mental results at Rye Meads.
Process modifications have been
undertaken at Coleshill to optimize
overall activated sludge performance
and reduce settling problems in the final
clarifiers caused by denitrification. In
the period June-December 1978, ni-
trate removal through the process
(including that occurring in final clari-
fiers) ranged from 42 to 57 percent.
Dramatic improvement in the problem
of rising sludge in the clarifiers was
reported. A change in the clarifier
desludging schedule, decreasing deten-
tion time during low flow periods, also
helped alleviate the problem.
Oxygen Transfer Performance
Method of Analyzing Oxygen
Transfer Efficiency
Currently, there are many methods
for measuring oxygen transfer efficiency,
including steady and nonsteady states
-------
Table 1. Surveyed Plant Characteristics
1978/1979 O&M
Average Flow Average Performance Expert-
Plant Location/Name Aeration System Description
United Kingdom
Basingstoke Nitrifying, 1 -pass plug flow.
symmetrical aeration
Beckton (New Plant) Nitrifying, 1 -pass plug flow, tapered aeration
Beddington Nitrifying, 2-pass plug flow, tapered aeration
Long Reach Non-nitrifying, 4-pass plug flow.
tapered aeration
Mogden (Battery B) Nitrifying, 4-pass plug flow, some aeration taper
Oxford (1 969 Plant) Nitri/denit, 1 -pass plug flow, tapered aeration
Rye Meads Nitri/denit, 4-pass plug flow, tapered aeration
(Stage III)
Coalport Nitrifying, 2-pass step feed.
symmetrical aeration
Coleshill (Stage III) Nitri/denit, 1 -pass plug flow, tapered aeration
Finham (South) Non-nitrifying, 1 -pass plug flow.
symmetrical aeration
Hartshill Non-nitrifying, 1 -pass plug flow.
tapered aeration
Minworth Nitri/denit, 1 -pass plug flow, tapered aeration
Strongford Nitrifying, 1 -pass plug flow, some aeration
(New Plant) taper
The Netherlands
Holten-Markelo Nitri/denit, 2-pass plug flow, tapered aeration
Steenwijk Nitrifying, 2-pass plug flow, tapered aeration
United States
Glendale, Calif. Non-nitrifying, 1 -pass plug flow.
tapered aeration
Madison, Wise. Non-nitrifying, 3-pass step feed,
tapered aeration
Fort Worth, Tex. Non-nitrifying, 1 -pass plug flow.
tapered aeration
Tallman Island, N.Y. Non-nitrifying, 2-pass plug flow, step feed
mgd
4.9
174
25.5
52.8
45.2
5.3
10.4
3.2
13.5
7.5
5.7
72.4
10.6
4.7
11.8
3.0t
14.5
99$
68
m3/sec
0.22
7.6
1.12
2.31
1.98
0.23
0.46
0.14
0.59
0.33
0.25
3.17
0.46
0.21
0.52
0.1 3}
0.64
4.3$
2.98
%fBODs)n
97
95
96
94
97
98
98
95
96
90
94
96
95
93
96
90
88
--
86
Q£7"CO_ on/*o*
fOI WOR trl flsC
97 A
94 +
97
91 +
97 +
96 +
98 +
95 +
96 +
92 +
94 +
96 +
+
92 +
95 +
90 +
92 +
..
+
*A - average: B - better than average; - = worse than average.
^10-mo data.
%3-mo data.
procedures. For this study, oxygen
transfer efficiency was estimated using
a mass balance technique based on
empirically derived oxygen consump-
tion values for BOD5 removed and
ammonia nitrogen oxidized and on a
similarly derived oxygen credit for
nitrate nitrogen denitrified. The method
was developed by Boon and Hoyland of
the WRC based on the work of Ecken-
felder and has an estimated ± 20
percent accuracy.
The oxygen mass balance technique
used in this project is represented by the
following equation:
G,
where: Gt
= 10'3f[R(B,-Be)+ 1.64
(N8-Ne) + 2.83Ne»]
= overall rate of oxygen
consumption by
microorganisms in an
aeration tank, kg/sec
f = average wastewater
flow rate, mVsec
Bs = primary effluent
BOD5, mg/l
Be = final effluent BOD5
mg/l
N8 = primary effluent
NH4+-N, mg/l
Ne = final effluent NH4+-N,
mg/l
Ne* = final effluent NOa'-N,
mg/l
R =0.75+0.05/(F/M),for
0.1
-------
Table 2. Aeration System Design Data
Aeration Basin
Dimensions
Plant Location/Name
Lgth
Wdth
(m)*
Dpth
fm)*
L/W
Diffuser
Density Aeration Taper
Idomes/rrfft 1%)
Minimum
Mixing
Power Level Avg. Air Flow/
Min. Air Flow
United Kingdom
Basingstoke 79.2 6.7 2.5 12 3.9 none 20.8 1.5
Beckton (New Plant) 223 41.2 3.1 5.4 2.8-1.9 46/31/23 13.6-6.8 1.5
Beddington (New Tanks) 67 7.3 2.4 18.4 2.7-1.1 34/28/23/15 16.1-6.4 1.8
Long Reach 80 6.0 3.8 53 7.8-3.5 35/27/23/15 58.7-25.7 2.0
Mogden /Battery B) 122 4.6 3.7 106 5.0-3.1 34/22/22/22 29-18.5 1.0
Oxford (1969 Plant/ 37.8 6.9 2.4 5.5 3.8 43/28 18.8 1.5
Rye Meads (Stage III) 70 4.3 3.0 65 4.6-2.3 21/33/28/18 29-13.8 2.4
Cos/port 65 4.6 4.3 27.8 2.8 none 25-16.7 1.2
Coleshill(Stage III) 64 18.3 2.9 3.5 3.9-2.0 § 36.9-16.7 2.0
Finham (South) 61 3.0 3.6 20.3 4.3 none 36.0 1.5
Hartshill 27.4 9.2 3.2 3.0 5.9-4.1 59/41 89.0-62.0
Minworth 178 18.3 3.0 9.7 0.4/1.9-0.9 § 26.8-13.4 1.25
Strongford (New Plant) 108 9.3 3.0 46.4 2.3-1.9 §
The Netherlands
Holten-Markelo 30 6.6 4.0 9.1 1.9-0.9 34/25/25/16
Steenwijk 100 6.75 4.0 29.6 2.8-1.5 34/25/25/16 20-10
United States
Glendale, Calif.
Madison, Wise.
Fort Worth, Tex.
Tallman Island. N.Y.
73.2
41.2
83.8
110
9.75
9.1
36.6
28
4.9
4.7
4.3
4.9
7.5
13.6
23
7.9
3.0-0.9
9.1-3.6
5.4-3.0
1.3
57/43
48/29/23
34/27/21/18
none
24-7.5
33.7-13.4
-.
--
1.4
1.5
1.4
1.2
*; m = 3.28 ft.
t/ dome/m* - 9.29 domes/100 ft".
t/ W/m3 = 0.038 wire hp/1000 ft3.
§See Appendix B of main report.
Table 3. Aeration Process Performance Data
Average Flow Design
& Data Year DWF
Plant Name/Location (m3/sec)* (of/sec)"
United Kingdom
Basingstoke
Beckton (New Plant)
Beddington
(New Tanks)
Long Reach
Mogden (Battery B)
Oxford (1969 Plant)
0.22/78-79
7.6/78-79
1. 12-78-79
2.31/78-79
1.98/78-79
0.23/78-79
Rye Meads (Stage III) 0.45/78-79
Coalport
Coleshill (Stage III)
Finham (South)
Hartshill
Minworth
Strongford (New
Plant)
The Netherlands
Holten-Markelo
Steenwijk
United States
Glendale. Calif.
Madison. Wise.
Fort Worth. Tex.
Taltman Island. NY
0 14/78-79
0.42/78-79
0.32/1979
0.25/1979
3.17/1978
0.47/1979
0.21/1978
0.52/1978
0.13/78-79
0.63/1979
4.3/pt. 1979
3.0/78-79
0.26
8.8
0.96
1 97
1.53
0.17
0.42
0.20
062
0.26
0.28
2.11
0.77
0.15
0.62
-
4.2
3.5
BODs (mg/l)
' Raw
281
169
320
334
238
367
310
—
321
500-700
-
250
400
312
220
213
—
91
Primary
157
96
149
18O
99
165
144
157
158
162
400-500
142
50-100
182
102
158
156
64
Volumetric
Loading
(Ib BODi/day/
Effluent
4
8
12
20
8
7
5
9
12
32
20-40
6
10
21
12
11
19
--
13
1000 ft3n
22.4
20.6
11.7
44.0
9.8
41.0
24.3
36.0
22.9
70.0
112
22.2
-
30.4
24.2
31.9
27.0
29.6
F/M
Loading
A verage Air Flow
(kg BODs/day/kg2) ft3/* SODst cfm/1OOOft3§ Remarks
0.08
0.13
0.20
0.30
0 18
0.10
0.08
0.14
0.10
0.45
0.30
0.09
0.05
0.18
0.11
0.35
0.30
0.24
1910
1110
1785
612
1392
1046
1416
1402
1000
693
747
689
--
--
--
748
732
--
--
28.9
16.8
13.4
16.6
208
26.2
23.0
11.0
15.8
34.4
43.8
10.1
—
—
15.4
80.0
—
-
Non-nitrifying
Non-nitrifying
Initial anoxic
zone
Initial anoxic
zone
Initial anoxic
zone
High rate, non-
nitrifying
1 mo data
Initial anoxic
zone
Figures
approx.
Non-nitrifying
Partial
nitrification
Non-nitrifying
Non-nitrifying
Non-nitrifying
Non-nitrifying
*1m3/sec = 22.8 mgd.
t/ lb/day/1000 ft3 = 0.016 m3/day/m2.
tl ff/lb = 0.062 rr?/kg.
§1 cfm/1000 ft3 = 0.017 1/m3/sec.
-------
Table 4. Oxygen Transfer Performance Data Summary
Percent Saturation of
Plant
Beckton
Basingstoke
Mogden
Oxford
Rye Meads
Coalport
Coleshill
Minworth
Strongford
Beddington
Harts/?///
Long Reach
Finham
Steenwijk
Glendale
Madison
Aeration
Tank L/W
5.4
12
106
5.5
65
27.8
3.5
9.7
46.4
18.4
3.0
53
20.3
29.6
7.5
13.6
MLSS
(mg/l)
2900
4900
2300
5500
4700
2500
3000
3200
5000
2300
3000
1700
2000
3300
2000
2000
Mixed Liquor D. O.
Range
10-80
10-60
10-100
10-40
20-100
-
20-50
—
20-100
--
reported low
10-40
--
--
10-30
10-30
Average
40
30
50
20
60
--
35
--
80
15
--
20
—
-
20
20
High
(kg/kWh)
1.95
1.20
1.62
2.34
1.14
--
—
--
--
1.25
--
--
--
—
--
1.99
Aeration Efficiency*
Low
(kg/kWh)
1.54
1.08
1.12
1.93
1.04
--
--
--
--
1.05
--
-
—
—
--
1.56
Average:
Average
(kg/kWh)
1.75
1.16
1.37
2.13
1.09
1.08
2.12
1.71
1.49
1.11
1.11
2.07
1.76
0.78
1.14
1.77
1.48
Average
(Ib/hp-hr)
2.88
1.91
2.25
3.50
1.79
1.78
3.49
2.81
2.45
1.83
1.83
3.40
2.89
1.28
1.87
2.91
2.43
Years
of
Data
3
5
5
5
3
1
1
1
1 wk
10
1 mo
1
1
1
10 mo
2
*Defined as mass of Oi transferred per unit of power input as measured by the line draw.
data, such as Beddington, exhibit fairly
constant performance data over the
period of record. The two U.S. plants for
which performance could be estimated
seem to be similar in both process
design and performance. The Dutch
plants are more closely related to U.S.
plants in design; however, the estimated
performance at Steenwijk is somewhat
less for unknown reasons.
Operation and Maintenance
General Maintenance
Experience
Maintenance observations at the 19
survey plants are summarized (Table 5).
Generally, the plants have had good,
and often exceptional, reliability from
dome diffuser equipment. After initial
shakedown, the plastic pipe mounted
systems have performed well. Earlier
plants used dome diff users mounted on
a cast iron air distribution grid. Rusting
of the interior surfaces of the air lines
led to rust and scale deposits on the
interiors of the domes and caused
plugging after 5 to 6 yr. Most of the
plants with iron pipe are retrofitting to
plastic pipe with generally good results.
Several of the retrofitted plants have
experienced minor problems with some
of the anchors that hold the plastic pipe
saddles to the tank floor coming loose
and pulling out. The cause of this seems
to be spalling of concrete around the
mounting holes in the floors. This has
i not been reported as a problem in
systems where tank concrete is new
and apparently less vulnerable to
spalling.
Several plants have also reported
scattered failures of other plastic parts,
notably the pipe coupling straps and
orifice bolts. Beckton had major prob-
lems on startup with the coupling
straps. Mogden has had considerable
problems with failure of the orifice bolts,
probably related to over tightening
during installation. Most of the plants,
however, reported few or no startup
problems of this nature. Careful super-
vision of installation to avoid over-
tightening of plastic parts was cited as
the key to trouble free startup by most of
the plant personnel. It was also noted
that the plastic parts were much less
costly to replace than the previously
used brass bolts.
Formation of Biological Slimes
on Diff users
The major operational problem asso-
ciated with the dome diff users was the
formation of biological slimes on
diffusers operating in zones of high
volumetric loading and/or low dissolved
oxygen (D.O.). Beddington continues to
have major problems with slime forma-
tion, which manifests itself as coarse
bubbling at the surface of the aeration
tank. The slime growth does not cause
an increase in air pressure; rather, it
induces an apparently wholly external
surface fouling that causes the air
bubbles to coalesce after exiting the
surface of the diffuser domes. The
resulting coarse bubbling lowers oxygen
transfer efficiency, thereby lowering
mixed liquor D.O. and further encourag-
ing slime growth.
When first confronted with the
problem, Beddington removed and
refired their fouled domes. On startup of
a cleaned tank, the problem quickly
recurred, however, and it was soon
obvious that other, less costly solutions
were needed. I n further tests, a vigorous
brushing of the dome surface accom-
panied by high air flow rates was found
to return the dome to nearly new perfor-
mance levels. Periodic tank cleaning
and dome brushing have allowed
Beddington to control (not eliminate) the
problem at moderate cost.
Although Beddington's sliming prob-
lem was intensified by the presence of
strong industrial wastes, which
depressed oxygen transfer efficiency
and caused low D.O. in the first passes
of the multi-pass plug flow tanks, it was
not the only plant that exhibited sliming.
Indeed, every plant visited showed
some signs of coarse bubbling, which
was probably attributable to slime
growth on domes. Without exception,
the phenomenon occurred at the primary
effluent feed points or at the transition
from anoxic to aerobic treatment. It was
particularly severe in the first 20 to 25
percent of the first pass of two-to-four-
pass plug flow systems. Tapering the
aeration helped somewhat but did not
fully solve the problem.
-------
Table 5. Maintenance Data Summary
Plant Name/Location
United Kingdom
Basingstoke
Beckton
New Plant
Old Plant
Beddington
(New Tanks)
Long Reach
Mogden (Battery B)
Oxford
Rye Meads
Coalport
Coleshill (Stage III}
Finham /South)
Warts/7///
Minworth
Strongford (New Plant)
The Netherlands
Holten-Markelo
Steenwijk
United States
Glendale, Calif.
Madison. Wise.
Fort Worth, Tex.
Tallman Island. N.Y.
Started Up
1964-71
1970
1959
1969
1978
1961
1969
1956-70
1970
1968
1974
1973
1971
1972
1978
1977
1978
1977
1978
1979
Startup Experience
Some problems with plastic
tank bottom mounts
Problems with plastic holddowns
No significant problems
No significant problems
No significant problems
No significant problems
Some problems with plastic
tank bottom mounts
Some problems with retrofitted
plastic piping
No significant problems
No significant problems
No significant problems
No significant problems
No significant problems
No significant problems
No significant problems
No significant problems
Several blowoff lines failed
No significant problems
Some problems with blowoffs
No significant problems
Cleaned
Every 5 yr
Every 8 yr
Twice in 15 yr
Every 4 yr*
Not yet
Every 6 yr
Not yet
Every 6 yr
Not yet
Not yet
Not yet
Not yet
Not yet
Not yet
Not yet
Not yet
Not yet
Not yet
Not yet
Not yet
Operating Experience
Fair, scale problems
Good after initial problems
Gradual plugging due to rust in cast iron
pipes
Poor but improving major slime problem
Good, new plant
Plastic retrofit in Battery B 11968) has
not yet required cleaning
Good, no apparent loss of effluent quality
after 10 yr
Fair, plugging due to rust in older lines.
Plastic system good
Good
Good, tanks cleaned once/year and domes
brushed
Good, only have had to repair several small
line leaks
Fair, some slime growth
Good, tanks cleaned once/ year and domes
brushed
Good
Good
Good
Good, small evidence of slime
Substantial sliming problem in mid-1980
after 3 yr of operation
Some line breaks and problems evident, but
overall performance stable
Good
*Initially. Cleaning has not been required for the last 6 yr.
To summarize, slime growths
appeared to occur in zones of heavy
organic loading, or low D.O., or both.
The occurrence of these growths was
exacerbated by extreme plug flow
aeration tank design and the presence
of strong industrial wastes.
Conclusions
In general, dome diffuser fine bubble
aeration systems were providing relative-
ly efficient, low-maintenance service in
the surveyed plants. However, the plant
visits and related study clearly indicated
a need for optimized design and oper-
ating control strategies if the full energy
saving potential of the equipment is to
be realized. Listed below are the
principal conclusions resulting from
this study.
1. Assessment of data from the
surveyed plants resulted in widely
varying estimates of field oxygen
transfer performance for the
dome diffuser. Generally, field
performance was lower than
might be expected from clean
water oxygen transfer data. With
the use of a mass balance tech-
6
nique (based on empirically de-
rived oxygen consumption values
for BOD5 removed and ammonia
nitrogen oxidized and a similarly
derived oxygen credit for nitrate
nitrogen denitrified), the process
(i.e., dirty water or mixed liquor)
aeration efficiency for the 16 of
19 plants with adequate data to
make predictive estimates aver-
aged 1.48 kg 02 transf erred/kWh
consumed (2.43 Ib 02/wire hp-
hr). The highest and lowest
observed aeration efficiencies
were 2.13 kg 02/kWh (3.50 Ib
02/wire hp-hr) and 0.78 kg 02/
kWh (1.28 Ib 02/wire hp-hr) at
Oxford and Steenwijk, respec-
tively. For the three plants (Fin-
ham, Madison, and Glendale)
with a reasonably sufficient
comparative data base, fine
bubble dome diffuser process
aeration efficiency was approxi-
mately 1.65 times higher than for
side-by-side coarse bubble dif-
fuser systems: 1.56 kg 02/kWh
(2.56 Ib 02/wire hp-hr) vs. 0.95
kg 02/kWh (1.56 Ib 02/wire hp-
hr).
2. Methods of plant operation fre-
quently contributed to less-than-
optimum oxygen transfer per-
formance.
• In the U.K. plants particu-
larly, volumetric and F/M
loading rates were often
lower than required for
nitrification, or high levels
of BOD removal, or both.
The least energy efficient
plants, with two exceptions,
were underloaded volumet-
rically.
• A number of the plants were
also overaerating the mixed
liquor and had taken no
steps to monitor D.O. con-
centrations and reduce air
flows to more efficient op-
erating levels. The two most
energy efficient plants,
Oxford and Beckton, closely
monitored mixed liquor D.O.
and adjusted air flows ac-
cordingly.
3. Lowered oxygen transfer effici-
ency could also be traced to
design practices that make it very M
-------
diff icu It for operators to ru n treat-
ment plants effectively.
• When multiple-pass plug
flow systems are used, the
air supply capability is poorly
matched with the oxygen
demand, particularly in the
second and subsequent
aeration channels. This leads
to overaeration in the latter
passes and localized organic
overloading and diffuser
sliming in the first pass.
Step feeding only partially
alleviated the overaeration
problem. Tapering the aera-
tion dome configuration
was also of limited value in
suppressing overaeration in
the second and subsequent
passes of multiple-pass
systems; however the ta-
pering significantly helped
suppress diffuser sliming.
In terms of overall oxygen
transfer performance, tapered
aeration apparently had no
apparent advantage over
the nontapered systems.
• The full practical operating
range attainable with the
equipment, in terms of air
flow per dome, is not properly
used in selecting diffuser
density. Providing too many
domes creates a situation
where the minimum total
aeration system air flow is
controlled by the minimum
allowable air flow rate per
dome (0.014 mVmin or 0.5
cfm, defined by control
orifice headless character-
istics) for large portions of
the day; this produces ex-
tended periods of overaera-
tion. The recommended
maximum unit dome air
flow rate of 0.057 mVmin
(2.0 cfm) is consequently
rarely approached in opera-
tion.
• Many of the plants had
shallow aeration tanks, 3.7
m (12 ft) or less, which
reduces attainable oxygen
transfer efficiency.
• Most of the plants lacked air
flow monitoring capability
for individual aeration grids,
and air control valves, where
provided, were usually too
coarse in their adjustability
to be of use in controlling air
flows. Plant operators were
often prevented from cor-
recting overaeration condi-
tions because of equipment
limitations.
4. Significant industrial waste frac-
tions in municipal wastewater
may substantially lower dome
diffuser oxygenation efficiency
via a reduction in the alpha
factor. Alpha is especially affected
in the first segment of long, plug
flow aeration tanks (to values
reportedly as low as 0.3 to 0.4)
where detergents and other
surfactants haven't had suffi-
cient contact time to be biode-
graded. As these surfactants are
oxidized in passing through the
aeration process, alpha reportedly
increases to values of 0.8 or
higher at the effluent end of the
tank. Beddington and Hartshill
are two examples of plants that
are adversely affected by indus-
trial waste discharges.
5. The authors believe that, with
enhanced design and operating
techniques, aeration efficiencies
of dome diffuser plants with no
unusual alpha depressing wastes
present could be increased 25 to
75 percent over the average
value of 1.48 kg Oz transferred/
kWh (2.43 Ib 02/wire hp-hr)
estimated from the survey.
6. The limited data evaluated in this
study indicate some parity of
performance among the ceramic
dome and disc diffusers presently
being marketed in the United
States. There appears to be a
definite correlation between
dome or disc diameter (of the
horizontal surface) and specific
oxygen transfer per diffuser.
Data from clean water tests
suggest that fewer of the larger
diameter units may be required
to transfer equivalent amounts of
oxygen at the same oxygen
transfer efficiency.
7. Generally, maintenance experi-
ence with dome diffusers ranged
from good to excellent. Both of
the plants reporting significant
maintenance problems. Bedding-
ton and Basingstoke, had devel-
oped operating strategies that
were effectively controlling the
problems without excessive costs
or downtime. It is concluded that
the generally quite good mainte-
nance experience is directly attri-
butable to two principal factors:
• Conscientious (though not
labor intensive) attention to
aeration system operation,
particularly that relating to
air cleaning and repair of
infrequent equipment fail-
ures.
• Steady improvement and
refinement of the dome
diffuser equipment and its
application over the course
of its history, particularly in
piping and air cleaning.
8. Diffuser sliming, causing exter-
nal fouling, is apparently produced
by conditions of high F/M loading,
or low D.O., or both—conditions
that can occur when strong
industrial wastes are introduced
into a plant. Three plants, Beckton
(temporary reduction of loading),
Beddington (brushing), and
Madison (steam cleaning) have
developed somewhat effective
responses to sliming.
9. In designing new plants, close
attention should be given to
required air flow at minimum
loading. Use of a wider range of
air flows in the design of dome
diffuser systems, as now recom-
mended by the manufacturer,
will improve operational flexibil-
ity and thereby improve overall
system efficiency. Aeration effi-
ciency is only one parameter of
diffuser performance; high reli-
ability and flexibility of operation
should also be considered in
conjunction with operational and
capital costs.
10. Careful attention should be given
to air cleaning to avoid internal
fouling of dome diffusers. Manu-
facturer's recommendations in
this area should be followed
carefully. When dome diffuser
systems are retrofitted into exist-
ing plants, existing air piping
should be carefully checked for
rusting or scaling and should be
cleaned or coated as needed to
avoid particle shedding from the
pipe walls into the air stream
where it can cause internal
diffuser fouling.
Recommendations
This study has identified a number of
significant research needs that should
be addressed as soon as practicable:
-------
1. The question of alpha sensitivity
as it relates to the relative perfor-
mance under field operating con-
ditions of dome/disc diffusers
versus other aeration devices
should be a high priority research
need.
2. The opportunity to develop useful
side-by-side comparison data for
dome diffusers, coarse bubble
aerators, and fine bubble tube
diffusers (in wide band spiral flow)
exists at three U.S. treatment
plants: Madison, Wisconsin; Tall-
man Island (New York City); and
Fort Worth, Texas. In conjunction
with ongoing process (dirty water)'
testing at the Los Angeles County
Sanitation Districts, data should
be developed from these plants.
3. Oxygenation performance studies
of plants that have been modified
to optimize application of dome or
disc diffusers should be conducted
as soon as possible. Such studies
could possibly be rapidly imple-
mented in cooperation with the
WRC. In addition, one or more
major tests in U.S. plants should
be initiated in the near future,
possibly under EPA's Innovative
Technology Program.
4. The Nokia and Degremont dif-
fusers, which have experienced
significant overseas application,
are now being marketed in the
United States. Afollow-upeffortto
evaluate the O&M performance of
this equipment is recommended.
The Nokia dome, in particular,
represents a radical departure
from conventional ceramic dome
technology and should be of prime
interest in further studies.
Data evaluated during this project
appear to predict substantial
performance equivalence between
the Norton/Hawker-Siddely dome,
the Sanitaire disc, the Degremont
disc, and the Nokia disc. The larger
diameter Sanitaire and Degremont
units may transfer more oxygen
per diffuser, allowing the use of
fewer diffusers, when compared
with the smaller Norton/Hawker-
Siddely dome. Available data are
too limited for final judgment,
however, and further evaluation is
strongly recommended.
Diffuser cleaning is a labor inten-
sive and costly process that can
usually be forestalled by careful
O&M. Providing for diffuser clean-
ing was the usual practice in the
United Kingdom, however, and
appears prudent in light of British
experience. Alternatives to refiring,
notably ultrasonic cleaning, need
further development. Further study
of ultrasonic cleaning might be
carried out in cooperation with the
Fort Worth, Texas, plant to docu-
ment labor requirements, cleaning
effectiveness, and equipment
reliability.
These recommendations have been
stated in terms of urgency. Inviewofthe
increasing number of dome and disc
diffuser systems being designed and bid
in the United States, it is believed that
expedited research is necessary to avoid
repeating the deficiencies observed at
the surveyed plants.
The full report was submitted in
fulfillment of Grant No. R806990 by the
Association of Metropolitan Sewerage
Agencies under the partial sponsorship
of the U.S. Environmental Protection
Agency.
D. H. Houck was formerly with the Association of Metropolitan Sewerage
Agencies, Washington, DC 20036; A. G. Boon is with the Water Research
Centre, Stevenage, England SGI 1TH.
Richard C. Brenner is the EPA Project Officer (see below).
The complete report, entitled "Survey and Evaluation of Fine Bubble Dome
Diffuser Aeration Equipment," (Order No. PB 82-105 578; Cost: $15.50,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
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
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
RETURN POSTAGE GUARANTEED
MERL0063240
LOU W TILLEY
REGION V EPA
LIBRARIAN
230 S DEARBORN ST
CHICAGO IL 60604
Third-Class
Bulk Rate
GflVFDNMFNT PPTNTTNC fl FF T TF • 1 Qft 1 - - ^ 50 -00? / '
------- |