UPGRADING EXISTING
WASTEWATER TREATMENT PLANTS
TECHNOLOGY TRANSFER DESIGN SEMINAR
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND MONITORING
NATIONAL ENVIRONMENTAL RESEARCH CENTER
CINCINNATI, OHIO
-------
UPGRADING EXISTING WASTEWATER TREATMENT PLANTS
Prepared for the
U. S. Environmental Protection Agency
Technology Transfer Design Seminar
Presented at
Denver, Co., October 31 - November 1 & 2, 1972
National Environmental Research Center
Advanced Waste Treatment Research Laboratory
Office of Research & Monitoring
Cincinnati, Ohio
-------
UPGRADING EXISTING WASTEWATER TREATHENT PLANTS
John M. Smith
Environmental Protection Agency
National En$ironinental Research Center
Advanced Waste Treatment Research Laboratory
Cincinnati, Ohio 45268
INTRODUCTI
It is estimated that an investment of 18.1 billion dollars will be required for
the construction of municipal wastewater treatment facilities in the United States
to meet the 1976 projected Federal State Water Quality Standards (1). Approxi-
mately 1/4 of this amount will be used for upgrading the performance of the
existing wastewater treatment facilities.
The distribution of these facilities according to population is presented in
Figure 1. This figure indicates that 32% of the population receiving treatment
is served by primary-intermediate treatment facilities, 317. by the activated
sludge process and 217. by the trickling filter process.
TOTAL POPULATNO 107 x
___________________ TUATED PIPULAT U 131 • 100
i 7 IP __
r Pr.i,,.Iuts..II4.t.
43-il pe.pI &.tI,sIH i Ii.
Treatment Classification by Population
EPA Survey 1968
,c .: ’ I7%J )f \\
Total Facilities 14,123
EPA Survey 1968
Another way of looking at the in-ground municipal treatment plant investment
is to examine the total number of existing plants and their distribution ac-
cording to type. This is shown in Figure 2. Comparison of Figures 1 and 2
indicates that there are more trickling filter plants than activated sludge
plants, but that more people are served by the activated sludge process. Simi-
FIG. 1
FIG. 2
-------
—2—
lar comparisons can be made for the remaining treatment types.
The application of todays technology in upgrading plant performance includes
(a) techniques that can be used to maintain the original treatment plant effi-
ciency under increasing organic and/or hydraulic loading, (b) the addition of
processes that can be used to increase overall plant removal efficiencies and
(c) process additions or modifications for specific contaminant removal. Physi-
cally, these upgrading procedures may be applied ahead of the plant; as modi-
fications of the treatment process itself; or as effluent polishing techniques.
Regardless of the techniques employed, cost effective treatment plant upgrading
requires efficient use of available tankage and equipment, along with implementa-
tion of the best possible operating and maintenance programs. Initial investiga-
tions of potential upgrading situations must include a thorough examination of
existing plant equipment and past performance history as well as a complete
understanding of existing plant deficiencies and future treatment requirements.
This information can then be used to formulate alternate courses of action, and
finally to select the most effective solution.
PRE-PLANT CONSIDERATIONS
Infiltration and Flow Reduction
Historically, the construction of municipal wastewater treatment plants has been
a “catch up” phenomenon. Under our current system of plant monitoring there has
been little if any motivation to improve treatment plant efficiency until the
plant is severely overloaded, many times to the point of routinely by-passing
untreated or poorly treated wastewater.
A first step then, in upgrading many facilities, is to examine techniques to
alleviate or reduce hydraulic overload. The first approach here is not at the
plant. Control of groundwater infiltration, the reduction of extraneous surface
sources into sanitary sewers, and the reduction of household water usage should
all be thoroughly examined before considering any in-plant changes.
Nationally, there are nearly 3 billion feet of public sewer, the majority of
which are constr,icted below the prevailing groundwater tables and therefore are
subject to infiltration. Defective sewer pipe, faulty pipe joints and poor
manhole construction are the principle causes of infiltration. In the United
States infiltration flows average 157. of the total sewage flow, and during
prolonged rainy periods amount to as much as 307. of the total flow. The entrance
of extraneous surface water from such sources as roof leaders, manhole covers,
cellar and foundation drains and other illegal connections can add another 207.
to the total sewage flow (2).
While the adoption of tighter (200 gal/day/inch diameter/mile) infiltration
specifications, better inspection during construction, improved construction
methods and materials, and legal control of house laterals insepction, can
virtually eliminate excessive infiltration for new systems, the reduction of
excess flows in older systems is a perplexing and costly propositon. The use
of closed circuit TV, dye testing and smoke testing have proven quite valuable
in locating faulty joints and illegal connections. Studies have shown that TV
inspection of 8” diameter sewer lines can be accomplished for about $0.20-0.30
per lineal foot, including labor, equipment, and supplies. Complete sealing of
all joints can be accomplished at a speed of about 300 ft/day using a three man
crew at a total cost including inspection of $1.70 per lineal foot. This is
equivalent to a joint cost of $5.79. In one study (2), sewer sealing by this
-------
—3—
method reduced the total sewage flow by 40%. Effective overall excess flow control
will require reduction of extraneous surface sources in addition to adequate in-
filtration control. Control of infiltration and extraneous surface sources can be
complimented by reduction of water usage in the home.
A recent EPA sponsored study (3) has shown that both water-saving devices and re-
cycling systems for non-potable use in the home can significantly reduce the per
capita sewage contribution from the average household. The installation of shallow-
trap and dual flush toilets resulted in a toilet water usage savings of 20%, while
flow restricting shower heads decreased the shower usage by as much as 35%. Wash-
water recycle systems have decreased total household water usage from 24 to 35%.
Flow Equalization
Equalization of the diurnal variation in incoming sewage flows to a treatment plant
can relieve hydraulic overload, and if properly designed, can significantly dampen
the variation in mass flow of contaminants Into the plant. It has long been recog-
nized that operation under these quasi-steady-state conditions is necessary and
desirable for optimum operation of both biological and chemical-physical plants.
One way of determining the required equalization volume for a given plant is to
plot an Inflow-mass diagram of the hourly fluctuations in sewage volume for a
typical day. This variation Is shown In Figure 3 along with the associated inflow
mass diagram. The ordinate of this diagram is obtained by accumulating the hourly
flows for a given plant and converting them into volumes. The slope of line “A”
in Figure 3 represents the average “outflow” pumping rate. The vertical distance
between lines “B” and “D”, which are drawn parallel to line “A” at the maximum
and minimum points of the inflow mass diagram, represent the minimum required
equalization volume for constant outflow. For most diurnal variations, this volume
amounts to 12-15% of the average daily flow. Either a flow-through or side-line
basin may be employed for flow equalization. The volume requirements are the same
in either case. In the flow-through tank, all the flow passes through the equaliza-
tion basin and much better mixing or mass-flow dampening is possible. In the side-
line tank only the amount of flow over the average is retained. This scheme
minimizes pumping requirements at the expense of less effective mixing. A flow
diagram for a treatment plant employing flow equalization using a side-line tank
is shown in Figure 4, and a summary of operating data for this plant is shown in
Table 1.
Flow equalization tanks should be designed as completely-mixed basins, using either
diffused air or mechanical surface aerators. For systems using surface aeration
equipment, the mixing requirements will vary from 0.02 to 0.04 HP/bOO gallons of
storage volume for a typical municipal wastewater having a suspended solids concen-
tration of 200 mg/i. Aeration to prevent septicity must also be provided. This
equipment should be sized to supply 10-15 mg 0 2 /l/hr.
-------
-4-
250
200
150
100
50
a
x
U I
UI
‘a ’
a
a
‘a’
1
=
a
I.,
FIG. 4
WILLED LAKE NDVI WASTEWATER TREATMENT PLANT 2 I : d
Walled-Lake Novi Wastewater
Treatment Plant 2.1 mgd
PIG. 3
EWIAUZATION REOWIEMEIIES
TINE OF DAY
Equalization Requirements
12 4 8PM 12 4 SAN
UI
a
a,
-a
a
a
1
a
U,
a
a
U I
a
UI
U.
-a
a
a
a
a
V&BLE 1
Performance Data - Walled Lake Novi
Plant
PERFORMANCE DATA
WALLED LAKE.NOYI PLANT
MONTH REMOVAL % EFF .
SOD 55 BOD SS
SEPT. 71 91.8 98.5 8 4.1
OCT. 71 98.8 18.2 2 3.4
NOV. 11 98.6 13.1 2.3 18.1
DEC. 71 98.8 82.3 • 2.3 18.0
JAN. 72 88.2 98.1 3.5 9.0
-------
-5-
IN-PLANT MODIFICATIONS AND ADDITIONS
Clarification
Improvement in primary and secondary clarification can be accomplished either by
structural modifications of the existing basin, by improved operation and sludge
management practices, or by the addition of chemicals to improve coagulation and
settling. Improper basin inlet design, high weir overflow rates, short circuiting
and lack of surface skimming devices are the principle causes of sub-standard
clarifier operation. Many of these deficiencies can be remedied by inexpensive
structural changes to the existing clarifier such as the addition of scum baffles,
strategically placed distribution plates, or additional weirs. West (4) has re-
ported that increasing the sludge recirculation capacity, measuring effluent turbidity,
and using a sludge blanket finder to control sludge wasting decreased the effluent
BOD from 20 to 10 mg/I and the effluent suspended solids from 35 to 13 mg/i at a
3.5 mgd activated sludge plant in Sioux Falle, South Dakota. West (4) reports
further that in St. Louis, Missouri the use of a sludge blanket finder, a turbidi-
meter, increased air supply and reduced sludge recirculation, decreased effluent
BOD from 40 to 9 mg/i and effluent suspended solids from 92 to 16 mg/I in a con-
ventional activated sludge plant.
The most recent innovation in clarifier design is the tube settler concept. A
discussion of this subject by Hansen and Cuip (5) and later papers by Cuip, Hansen
and Richardson (6), and Hansen, Culp and Stukenberg (7) have described the theoreti-
cal and practicable aspects of both the horizontal and steeply inclined tube settlers.
Tube settlers have applicability for upgrading both primary and secondary clarifiers.
Culp et.al. (8) has described the performance of several plant scale tube settler
installations. Results of these studies indicate that overflow rates of up to
5,000 gpd/ft 2 can be maintained in primary clarifiers without sacrificing removal
efficiency. Tube settlers in this application however, do not improve the basin
efficiency much beyond the 40-60% removal range. Conley (9) has recommended the
use of 1.0 gptn/sq.ft. as a maximum overflow rate and 35 lbs./sq.ft./day as a maxi-
mum solids loading for the design of secondary clarifiers with tube settlers. Since
the flocculating and settling characteristics of sludge vary from plant to plant,
each case should be evaluated separately for suitable design criteria.
Fouling due to attachment and growth of biological slime on the sides of the tubes
is sometimes a problem. Some form of cleaning device (water jet or air) is re-
quired so that the solids build-up can be removed occasionally. The performance
of clarifiers using tube settlers at various installations is summarized in Table 2.
Little information is available at the present time to establish cost information
on tube settlers. However, an estimating cost figure of 12 to 20 dollars/sq.ft.
for tube settlers with an installation cost of 5 to 15 dollars/sq.ft. has been
recommended by the manufacturer (9).
Chemical Addition
The use of chemicals to improve treatment efficiency, or for specific contaminant
removal is now standard practice in many locations (l1)(l2). Chemicals commonly
used are the salts of iron and aluminum, lime, and synthetic organic polyelectrolytes.
-------
-6-
TABLE 2
Performance of Tube Settlers (10)
er.tional Data tiling
Tob. Rai.ting Facility Tithe Settler.
PJ.nt Location e Sin. Looption _ 9F 1 L. E l I SS Son ’ t If. SS
r€d gpr .Ioq ft 7i• gpo/sq ft
Philmooth. TvItkli ovg 0 1) Secondary 0.6 6070 3 3.4 6 60—70
Oregon Filter Clarifier
Philo m oth 0 Trickling 0 13 Pnioary 0.84 40.45 2 1—3 3 34—4l
Oregon Filte. Clarifier
Ilopevell Tonnahip. Activated 0 I ) Secondary 0 34 60-70 2-3 27
PeneayIvaoi Sledge ClarIfier
Kiami, Florid, Activated 1 0 Secondary 1 3 500 1.7 33
Sledge Clar llier
SCII - S,erface 0 .erflo ,. Rate
Ferceit removal rather than concentration
The metal salts (iron and aluminum) react with the alkalinity and soluble ortho-
phosphate in the vastewater to precipitate the metal hydroxides or phosphates.
Of equal importance in upgrading applications, is the ability of these hydroxides
to destabilize colloidal particles that would otherwise remain in suspension. The
destabilized colloids then flocculate and settle readily, thus improving the solids
removal efficiency during clarification.
Iron and aluminum salts have been widely used in many locations to precipitate
phosphorus from both activated sludge and trickling filter plants. The most common
locations within the plant for adding these minerals are (1) the primary clarifier,
(2) directly into the aerator of an activated sludge plant, and (3) before the final
clarifier. Although minimum metal dosage is dictated by incoming phosphorus levels
and desired removal efficiencies, this amount is normally sufficient for improved
flocculation of suspended solids. Results of metal salt addition on suspended solids
and BOD removal for some selected locations are shown in Table 3.
TABLE 3
Effect of Metal Salt Addition on
Suspended Solids and BOD Removal Efficiency
((.on ( roc.l E lF c.. rr (title> I I) or
or 168 rho. Color. or lA O) After
Loo.Liti. of Stat and Rotor. ‘keel Removal ‘1.1.1 Alter Ivoval S R,nov j
Lototlo. Type .1 Plant plinoral Addition Add ng/l Add n ell Add vt/I Add .Wi Inloron , .
Richard. ,., TOt.. Trickling Filter i.fore Final 30 Il 20 92 70 7 tO - - (II )
.td rat. Settler. *12( 504 ) 2
Chepol 11111. Trickling Filter 00 1cr. Fin.) AL/P 9.1. 1.0 70 47 Id 10 — . (I ))
.,,th CaeoIin. Itigh rat. 8.0.11.1 On..o. I lIt
Focal Cit7. Activated 8lAd 50 0010.0 Pri.ary 19 .811 F000 4 - 07 - - - (It)
T.va. Settling a. F.
‘3 (I)
v )n.in.. NiohiRan Triohlivl FIlt.r Odor. Pri..ry I) no/I F. - 80 — — 15 (I ))
Solo .. Fin .I 3-4 os/I F.’ 2
L k . Od.aaa. Hlchiian TrIthling FIle., Color. Pnln.rp R9 02 (2 ) oS 1 62(2) 1 1 4)
Ditiricool A.tiv.t.d Olod;. lv Lor010r AL/P 2/I 12 33 (3) (3)
Colonhia (Pilot Plant) Stop A.r.tion ot Ratio - - - 89 60 CI))
nata..a.. ‘a Activated Sludgo I, Aerator 50dm. Alto - hi - I I (14) 69 (ii)
high rate IA rg/l
Al
mm. Sledge In Ficat Stag. 50dm. Al,.. - 10 - - 29 (91) 00
Synt.. — 2 Stag. A.rator IA 5 3 vaJ 1
AOt ivalod Slvdgn Al
S ,. eitril Ivan mom
P.ot.ta. CAlifornia Coon... mntal In A.r.tnr Al,. Al/p (35 7) 74 9,0 (66 6) 29 7 (I I)
Activated Slodg. Molar Rail,
I 9/I
Cn .tonR intal In Aerator Iron Fe/P (35 7) 06 9 a fle A) I C 2
Actioated hod 8 . Molar Ratio 2 2 11
Fo/P I S/I (3) 1) 79 9 0 (37 2) II 0
F .m 3/i (23 7) 17 9 0 56 7 )A a
pen. Stale Coevontion.l Aerator Al/p 1 13 06 26 9 2 ? (I l)
A.tlvat.d SledS. EUlto.i Ratio 3/I
(I) removal acrou. privacy olarilIor
(2) ‘ reonval
it) r.ool .v.ndarp tr001nolit
-------
In general, the addition of minerals in the dosage range shown will produce
80-907. total phosphorus removal. If added to the plant ahead of the primary, the
addition of an anionic polymer is often recommended to minimize the carry-over of
insolubiiized phosphorus to the subsequent treatment stages. The addition of min-
erals to the aerator or aerator effluent of an activated sludge plant will generally
result in more efficient coagulant utilization than when added to the primary be-
cause, during biological treatment, the complex phosphorus forms are hydrolyzed to
the ortho form which is more easily precipitated. The addition of metal salts to
the primary tends to dilute the resulting primary sludge. If added to the aerator,
the metal has a weighting effect which results in a thickening of normal activated
sludge.
The addition of metal salts to the aerator of an activated sludge plant produces
moderate increases in suspended solids removal efficiency in some instances, but
on many occasions results in significant increases in suspended solids carry-over.
Mulbarger (16) states that this is due more to an incorrect ratio of volatile
solids/aluminum added than to a strict overdose of chemicals. Results at Pomona,
California indicate that increased turbidity is due to lowering the pH beyond the
optimum limits for alum flocculation. The addition of 2 mg/i of polyelectrolyte
prior to final clarification reduced the effluent turbidity to its previous value.
The addition of polyelectrolytes alone, in the 0.25-1.0 mg/i range has been shown
to improve suspended solids removal of primary clarifiers from 37.77. to 64.77., and
BOD removal from 31.07. to 46.77. (10). The effectiveness of lime addition in in-
creasing the efficiency of primary clarifiers at several locations is shown in
Table 4.
ThBLE 4
Effect of Lime Addition on
Primary Clarifier Performance
Percent Percent
Removal Sciore Removal After
Location Lime Added Lime Addition Lime Addition Remarks
mg/I CaD QJ ss
Duluth. 75 50 70 60 75
Minnesota 125 55 70 75 90
Rochester, 140 — — SO 80.90 Jar tests
New York
Lebanon, 145 — — 66 74 Pilot plant
Ohio
The addition of lime to primary clarifiers can nearly double the mass of primary
sludge to be handled, depending on the alkalinity of the incoming wastewater, and
will raise the primary effluent pH to the 9-Il range, depending on the lime dosage.
The addition of lime at this location will decrease the organic loading to the sub-
sequent biologically stage in many instances to where complete nitrification will
occur. If this happens, the nitric acid and carbon dioxide produced by the nitrif i-
cation and carbon oxidation reactions are adequate to maintain the pH within the
aerator at a near optimum level for nitrification (12). A principal consideration
in selecting lime addition to the primary clarifier as an upgrading technique is
whether or not the existing sludge handling facilities are adequate and if not,
whether alternate sludge handling schemes are possible.
-------
-8-
Process Modifications
Examination of the various modifications of the activated sludge and trickling filter
processes have identified situations where overall plant performance can be improved
by process modifications within the plant, or by adding another biological stage
in series with the existing plant.
Many such examples of in-plant modifications are contained in EPA process design
manual for upgrading wastewater treatment plants (10) Three of these examples
are shown below to illustrate the principles involved and the anticipated improve-
ment in effluent quality.
(1) Upgrading Using Chemical Addition to Primary Clarifiers and Conversion of
Conventional Activated Sludge to Contact Stabilization
This plant was originally a parallel activated sludge (1.2 mgd) and trickling
filter plant. Flow into this plant at the time of upgrading had increased to 6.0
mgd and treatment efficiency had declined to 75% removal of BOD and SS.
The activated sludge portion of the plant was upgraded from 1.2 to 3.0 mgd
by conversion of the primary settlers to final settlers, and by changing the exist-
ing aeration basin from conventional activated sludge to contact stabilization.
This was accomplished by increasing the capacity of the mechanical aerators from
40 to 110 HP (60 HP in the contact basin and 50 HP in the stabilization basin) and
by appropriately modifying the basins piping. This type of modification is possible
since experience has shown that the contact stabilization process is well suited to
a feed containing a high portion of BOD in the suspended or colloidal form. These
alterations were designed to increase BOD removal and suspended solids from 75 to
907.. With this modification, the two stage filters were changed from parallel opera-
tion to series flow, and were again able to remove 70-807. of the primary effluent
BOD. Overall trickling filter plant BOD removal was increased from 82 to 90%.
(2) Use of a Roughing Filter to Upgrade an Existing Low Rate Trickling Filter
Plant
The existing single stage trickling filter plant was designed for an average
flow of 0.7 mgd with 85% BOD and suspended solids removal. At the time of upgrading,
the flow had increased to 1.15 mgd with some deterioration in suspended solids and
BOD removal efficiency.
The plant was upgraded by replacing the existing comminuters and primary
clarifiers with three 1.0 mgd Hydrosieve units. A plastic media roughing filter
designed for an application rate of 2.5 gpm/ft 2 and an organic loading of 520
#BOD/l000 cu.ft. was installed following the Hydrosieve units and in series with
the existing trickling filters to bring the plant capacity up to 2.3 mgd. The
abandoned primary clarifiers were converted into sludge thickeners and the secondary
clarifiers and chlorine contact chambers were enlarged to handle the increased flow.
The anaerobic digester was upgraded to a high rate unit by the addition of a gas
recirculation system for more efficient mixing. The upgraded facility now provides
85% removal of suspended solids and BOD at twice the original hydraulic design
capacity.
-------
-9-
(3) Upgrading a Contact Stabilization Package Plant to a “Modified” Completely
Mixed Flow Pattern
The original plant was a 0.87 mgd contact stabilization plant designed for
2.6 and 6.5 hour detention times respectively, in the contact and stabilization
zones. Investigation indicated the plant was performing poorly because of excess
detention time in the contact zone. This increased detention time in the contact
zone allowed partial stabilization of the adsorbed organics which resulted in a
poorly settling mixed liquor. Before upgrading, the effluent BOD and suspended
solids were 26 and 24 mg/i respectively.
To improve the plants performance, the influent piping was modified so
that the raw wastewater was evenly distributed into what originally was the
stabilization zone. No wastewater was introduced into the former contact zone.
Mixing liquor in the upgraded system proceeded from the former stabilization zone
through the former contact zone to the secondary clarifier.
The return sludge was introduced into the former stabilization zone at one point
resulting in a “modified” completely-mixed flow pattern with an overall detention
time of 9.1 hours at an average flow of 867,000 gpd. The upgrading procedure
lowered the effluent BOD and suspended solids concentrations to 13 and 6 mg/i
respectively for a total BOD reduction of 907 and suspended solids reduction of
967..
The above examples illustrate only a few of the many process modifications that
can be employed to maintain or moderately increase process efficiency under in-
creased loadings. Recently, the use of oxygen aeration systems have gained wide
recognition as another useful and efficient means of upgrading existing overloaded
treatment plants. Because of the increased oxygenation capacity of these systems,
a high level of biologically active solids can be maintained in the aeration tanks.
This allows the biological capacity of a hydraulically or organically over-loaded
plant to be doubled or tripled in size. Total plant capacity is increased by ap-
propriately enlarging the final clarifiers as required by plant loading and sludge
management considerations. A bonus advantage of the oxygen aeration system is
that under proper loading conditions, significantly lower amounts of waste sludge
is produced.
The largest application of oxygen aeration for upgrading the performance of an
existing vastewater treatment plant is the 320 mgd Nevtown Creek plant in New Yoik
City. One of 16-20 mgd aeration bays of this “modified air” plant has been con-
verted to oxygen aeration to improve the BOD and suspended solids removal from 65
to 90%. This oxygen aeration system has been on stream at this location for one
month and preliminary data indicate that target removal efficiencies can be met.
The aeration basin contact time at average design flow of 20 mgd is 1.5 hours in-
cluding 25% sludge recycle.
Another popular and successful treatment plant modification is the conversion of
single stage biological systems to two stage systems. This is necessary in most
instances to insure consistent nitrification which is required by many states now,
either directly as a nitrogen control standard, or indirectly because of more
stringent disinfection or very low total oxygen demand requirements.
-------
-10-
Upgrading to two stage treatment systems has been successful using (I) two stage
trickling filters, (2) trickling filters preceded by activated sludge, (3) activated
sludge preceded by trickling filters, or by (4) two stage activated sludge plants.
In each case, 60-70% of the BOD is removed during the first stage treatment. This
lowers the organic loading sufficiently so that nitrification can proceed in the
subsequent stage.
EFFLUENT POLISHING
The use of effluent polishing techniques is recognized as one of the most cost
effective methods of upgrading existing plants to obtain increased organic and
suspended solids removal. Overall plant performance can be improved from the
70-857. efficiency range to the 95-997. range depending on the process employed.
Four unit processes are considered here for effluent polishing. They are: 1) fil-
tration,(2) microscreening, (3) granular carbon adsorption, and (4) reverse osmosis.
Filtration
Simple filtration of secondary effluent provides a positive method of suspended
solids control, and as such, is the most widely used and the most efficient single
unit process for upgrading treatment plant performance today.
Contemporary filtration systems can be broadly classified as either deep-bed or
surface filters. The most popular trend recently in deep-bed filter design is
the use of the dual or tn-media filter. Here, the use of two or more layers of
different media having increasing specific gravity with increasing bed depth allows
gradation of the filter bed from coarse to fine in the direction of flow. This
allows more efficient utilization of the total bed depth for solids storage than
conventionally graded, single medium filters.
The approach taken in the development of surface filters on the other hand, is to
allow filtration to take place on or near the top of relatively shallow single
medium filters, and to optimize removal of the accumulated solids. In addition
to the standard pressure and gravity surface filters, several innovative techniques
for providing a continuous clean filter surface have been developed. The moving
bed filter developed by Johns-Manville Products, the radial flow filter developed
by the Dravo Company, the radial flow-external wash filter developed by the Hydro-
mation Corporation, and the Hardinge traveling bed filters are typical examples.
The performance of each of the above filters in polishing secondary effluent will
depend on such factors as surface loading rate, temperature, floc size and strength,
degree of biological or chemical flocculation, media depth, grain size, solids load-
ing, run length, and method of filter operation. Because of the numerous variables
involved, and the eaèe of obtaining reliable design data from small pilot filters,
it is recommended that final filter selection be based on pilot plant results where
possible. Operating results from typical filter installations are shown in Table 5.
-------
-11-
ThBLE 5
Filtration Performance
Paid Medic Filter Hydraulic 88 8 Pffl..ene Effluent
Filter Tone j &5L 2ss h._ Lq.dtnx .L M s.L S I l2 Esferenc .
ft gp.I .q.ft. % % .gIi n g11
Deep—led
Gravity Dounf low TV. UI. 1.0—2.0 — 6 70 53 3-7 - (18)
Gravity Dounflow TV. 1ff. 0.9—1.7 2—3 3 67 38 — 2.3 (19)
Pr.e.ur. tlpficw T F 1ff. 0.9—1.7 5 3 83 74 3 0 2.5 (19)
Preleure Upf low A $ 1ff. 0.9—1.7 5 3 77 — • — (19)
Praeoure Opt Ion AS. Elf 1.0—2.0 5 2 2 3 0 62 7.0 6.4 (20)
Preuour. UpIlow AS Elf. 1.0—2.0 3 4.0 67 73 4 9 6.4 (20)
Pr.eonr. UpfIou A S Elf 1.0—2.0 3 4.9 36 65 37 7.1 (20)
Mixed Media S A Elf. 0.25-2 0 5 0 74 80 4 6 2 5 (215
Mixed Media A S Elf 2.0 73 74 3.8 6.0 (20)
Mixed Media AS 5ff 4.0 73 85 4.3 3.9 (20)
Mined Media A S. Elf 0.25.2 0 2 5 2 3 (22)
Surtace Filters
Moving Sad TV Elf 6-08 4.2 2 47 71 - (2 )
Moving led I F 0ff.t2)0.6 0.8 4 2 2 67 80 - - (23)
Gravity Donnflcn A S 511 — — 2 2 55 64 7 2 7.3 (20)
Cr.vIty Dounflo.. A S Off - - 40 69 70 4.8 74 (20)
Gravity Donnilcu. AS 511 - - 8.0 48 64 61 67 (20)
Gravity Dounflo.. AS Off 0.9-17 2.0 16-40 72—93 52.70 — — (19)
Grevity Dounflot, AS 8ff 0.93 10 2.0 46 57 - (24)
Granity Donnflou AS. 8ff 0.58 - 2 0-6 0 70 80 - - (75)
Grevity Do..nflx .. CS 1ff 043 1 0 53 62 78 4 (26)
I F . Trickling Filter 8 A - Extended Aeration
A.S. - Activated Sledge C.S — Contact Stabilization
(1) 100 .g/l alun & 0 2—0.75 ng/i anionic polyx.er added
(2) 200 .ng/l slow & 0 2—0.75 ng/l anionic polyowr added
As the table indicates, all of the fllter8 mentioned produce an effluent with sus-
pended solids and BOD 5 generally less than 7 mg/l. As a rule, the deep-bed filters
are better suited to treating strong biological floc, will yield longer run lengths,
and are less sensitive to solids loading than are the surface filters. The surface
filters are better adapted to removing the more fragile chemical flocs, yield shorter
run lengths, and require less backwash water than the deep-bed filters. A key ele-
ment in efficient effluent polishing is to match correctly the type of filter used
with the flocculant nature of the solids to be removed, and also to design the filter
for solids-loading and run lengths that are compatible with normal operating schedules.
Microscreening
The microstrainer is another surface filtration device that has found increasing
utility for polishing secondary effluents. The system consists of a specially
woven stainless steel fabric mounted on the periphery of a partially submerged
horizontal revolving drum. Influent enters through the upstream end of the drum
and flows radially outwards through the fabric leaving the intercepted suspended
solids behind. Microstrainers are available in sizes ranging from 5’O” diameter x
P0” wide having a 0.06 to 0.6 mgd capacity to 10’O” diameter x lO’O” wide with a
capacity of 4.0 to 12.0 mgd. Filtration efficiency depends primarily on fabric
size and the character of the solids being removed. For wastevater polishing ap-
plications, these microstrainers are available with automatic controls to increase
drum speed and backwash pressure to accommodate variations in flow, and to a lesser
extent, variations in solids loading. One of the chief advantages of using a micro-
strainer for polishing secondary effluent is its low head requirement of 1 to 1-1/2
feet. Pumping secondary effluent prior to microstraining tends to shear the biologi-
cal floc and decrease solids removal efficiency.
Microstrainers are washed continuously at 20-50 PSIC and require 4-6% of the filter
throughput. Continuous backwash is advantageous for upgrading small plants since
it eliminates sur-charging of the upstream units which must be considered when using
conventional intermittently backwaahed filters. Operational data from various
microstrainer installations are presented in Table 6 (10).
-------
—12—
TABLE 6
Microatrainer Performance
Plant Fled Fabric 58 Effleesti I ) Effluent
l .ties . !. s±. !op.! 55 Somonal D leckesat .
.d microns percent .zJI percent . m/1 of f lee.
Ir ton, Dotarin 0.! A 8.1 23 37 34
Effluent
Labenon. ,io Pilot A 5. 23 59 1.9 SI - 5.3
Effluent
Pilot A S 35 73 7.3 61 — S 0
Effluent
ir.1n. Illinois 3.0 A $ 23 71 3.0 74 3 0 3.0
if I lum.t
Lotion. Do 1 Isnd 3.6 Effluent 35 35 7 3 30 — 3 0
from A.8
end T F
Drorkooll. En lamd 7 2 5 F 35 66 5 7 32 8.4
if fiumor
8. — Activated Sludie
27 F — TrLckIin Flit..
Carbon Adsorption
The necessity to upgrade secondary effluent quality beyond the levels that can
be obtained by implementation of the process modifications previously discussed,
or by the application of tertiary filtration, will require a substantial invest-
ment in additional plant equipment as well as a 3O-5O increase in operating cost.
It is doubtful that this level of expenditure can be justified except as part of
a major plant expansion where the expected lifetime of the new facility is suf-
ficiently long to permit reasonable amortization of the high capital investment
required. The situation is especially difficult for small plants.in the 0.5 mgd
range that serve rapidly growing urban areas.
The effectiveness of granular activated carbon for upgrading the treatment ef-
ficiency of larger plants is well established. Operating experiences at Pomona
and South Lake Tahoe, California; Nassau County, New York and Colorado Springs,
Colorado have left little doubt regarding process efficiency, operating cost and
reliability of these systems. Operating results and principle design parameters
for three of these locations are shown in Table 7 (10).
TABLE 7
Tertiary Granular Carbon Adsorption
Design Parameters and Operating Results
l.aks T.lto. Nassau Coonty
Operating Data
Capacity 200 pm 1.500 pm 400 pm
Sourc, of Vast. Dcmsstit D stic Domestic
Iscoedary Tvs.tme.t !t.nd.rd Activatad Pludg. St.nd.rd Activated Sl. .d 1 e Nt h-rste Aottu.t.d $ludg.
Pre —t r eat eot Chlorination Co. ulatlon A Filtration Coagulation A Filtration
CarbonType 16 .4Oms.h D .3 0 e.h Sa3 Oma.h
Colton Ccnflm , .r.tion 4 — S i . p. DonAtIon S-OptIon in Parallel 4—Stage Daunilo..
Cotton Dimensions 6 di i. a 9’ deep Ii ’ di .. 14’ deep B’ die a 6’ d.sp
Nonioal Contact Time 36 minutes 13 miflutss 24 minute.
leading Pat. 7 epnleq.ft B p.laq.ft. 7.3 pm/sq. ft
Carbon Colon. P ,rfocmgnt, 5 j4 3 i U I lsa .t m l issue ifflusot t.nlluant lIiiesnt
C . /l 47 10 20—30 2-10 • S
I X. is/I — - 3-20 2— 5 —
Colon, Pt-Co Mite 30 3 20-50 3
Csrbon Doug. 330 lbs/million subs 230 lbs/million pun 100 lbs/million pllon
-------
-13—
The capital cost of tertiary granular carbon systems will vary widely depending
on the particular system design and pretreatment provided. Direct application
of secondary effluent to downflow carbon adsorption columns as practiced at Pomona,
California will result in a smaller capital investment than the tertiary system
used at Lake Tahoe, but may increase operating costs due to more frequent column
backwashing.
Operating cost will depend primarily on the organic loading and associated carbon
dosages. Table 8 compares the actual capital and operating costs for the 7.5 mgd
conventional activated sludge plant at South Lake Tahoe with the corresponding
tertiary carbon adsorption costs at that location (27). Other investigators have
estimated capital and operating costs for tertiary carbon adsorption systems that
are considerably higher than those shown. The above data was used to estimate
operating costs for carbon dosages of 350 #/mg and 500 #1mg. These higher dosages
would be anticipated from most activated sludge effluents applied directly to
adsorption columns.
The following tabulation of costs illustrates that at the 7.5 mgd scale the use
of granular carbon adsorption with regeneration can increase total plant operating
cost by 60% and require a capital expenditure as great as 30% of the original plant
investment. CuIp and Cuip (27) estimate that the total costs for tertiary carbon
adsorption for a 2.5 mgd plant would be 50% greater than for the 7.5 mgd Tahoe
system. Little operating information is available on carbon adsorption operating
costs below this level.
TABLE 8
Tertiary Granular Carbon Adsorption Costs
PrL. .nry md Activated S1 . .d 8 e
0v .ntc $l,d . Hindu . 8
and ChlorInation 7.5 .gd
Op.ritt .g Co.t
$,
103
CapItal Co.t
Sloe
67.50
Tot.I Co.t
$l.e
170 50
Grandly Carbon Adaorpti.n
and R.Ben.rati.n 7 5 . d
2 SOdioe
30
21.3
51 5
35O#lwg
42*
21 5
63 5
5OO#/
60
21 5
I I S
* eatinated ba.ed on Sovth
Like Tahoe Data
Recently some investigators have advocated adding powdered activated carbon directly
into the aerator of an activated sludge plant to upgrade the organic removal ef-
ficiency. The DuPont Company has developed a PACT (powdered activated carbon treat-
ment) process for this purpose. Preliminary data obtained in a parallel 0.45 mgd
study showed that the application of 308 mg/i of powdered activated carbon into a
completely mixed aerator of a conventional activated sludge plant reduced the soluble
effluent BOD from 20 to 11 mg/i, for an overall BOD removal of 96% (28). In another
test, the flow to one side of a parallel activated sludge plant receiving 295 mg/l
of powdered carbon was more than doubled without sacrificing soluble ROD removal
efficiency. Total BOD removal decreased from 98 to 96% with suspended solids in-
creasing from 58 to 89 mg/i (28). These experiences indicate that powdered acti-
vated carbon addition to activated sludge plants will produce excellent soluble BOD
-------
-14-
removal, but these plants must be followed by filtration, to produce high suspended
solids removals.
The operating cost for the addition of 300 mg/i of powdered activated carbon on
a once through basis would be about $220/mg which is prohibitively high for con-
tinuous use. Regeneration of this carbon could bring the cost down to about $70/mg
which is competitive with granular carbon adsorption systems. Although a great
deal of work has been completed on powdered carbon regeneration schemes (29,30,31,
32), they have not been successfully.used for this application. In any case, it
is doubtful that powdered carbon adsorption should be used to compete with biologi-
cal oxidation in the aerator of an activated sludge plant except for the removal
of “hard” or refractory BOD substances. The most practical application would be
to use the powdered carbon as an operational tool to improve treatment efficiency
during times of “biological upset” due to toxic materials, or during short periods
of extremely heavy organic loads.
-------
REFERENCES
1. “Water in the News,” Soap and Detergent Association, New York, N.Y.
(June 1972).
2. “Control of Infiltration and Inflow into Sewer Systems,” EPA Publication,
11022-EFF (December 1970).
3. S. Cohen, “Demonstration of Waste Flow Reduction from Households,” Report
for EPA, 68-01-0041, (February 11, 1972).
4. A. F. West, “Case Histories of Plant Improvement by Operations Control,”
Nutrient Removal.and Advanced Waste Treatment Symposium, FWPCA, OBR,
Cincinnati, Ohio (1969).
5. S.P. Hansen and C. L. Cu].p, Jour. Amer. Water Wks. Assoc, Vol. 59 , p. 1134
(1967).
6. C. L. Culp, S. P. Hansen, and C. H. Richardson, Jour. Amer. Water Wks.
Assoc. Vol. 60 , p. 681 (1968).
7. S. P. Hansen, C. L. Cuip and J. R. St ikenberg, Jour. Water Poll. Cont.
Fed. Vol. 4 , p 1421 (1969).
8. C. L. Cuip, I C. Y. Hsiung, and W. R. Conley, Jour. Sanit. Engr. Div.,
ASCE, 95, Vol. 5 , p. 829.
9. V. R. Conley and A. F Slechta, Presented at the 43rd Annual Wat. Poll.
Cont. Fed. Conference, Boston, Mass. (October 1970).
10. “Process Design Manual for Upgrading Existing Waste Water Treatment
Plants,” EPA, Contract 14-12-933 (October 1971).
11. “Process Design Manual for Phosphorus Removal,” EPA, Contract 14-12—936
(October 1971).
12. D. S. Parker, “Process Development for Nitrogen Removal at the Central
Contra Costa Water Reclamation Plant, Contra Costa County, California,”
Paper presented at the EPA, AWT Design Seminar held at the University of
C lifornia, Riverside, California (March 24, 1972).
13. R. C. Brenner, “Advances in Treatment of Domestic Wastes,” Paper presented
at the National EPA Training Course held in Athens, Georgia, (October 18-
22, 1971).
14. J. J. Convery, “The Use of Physical-Chemical Treatment Techniques for
the Removal of Phosphorus from Municipal Waste Water,” Presented at
FWQA, AWT Seminar in San Francisco, (October 28-29, 1970).
15. A. B. Hais, et.al; “Alum Addition to Activated Sludge with Tertiary
Solids Removal,” EPA, Advanced. Waste Treatment Research Laboratory,
National Environmental Research Center, Cincinnati, Internal Report
(March 1971).
16. M. C. Mulbarger and D. C. Shifflett, Chem. Engr. Prog., Vol. 67 , No. 107
(1970).
-------
17. L. S. Directo, R. P. Miele, A. N. Masse, “Phosphate Rexnoval by Mineral
Addition to Secondary and Tertiary Treatment Systems,” 27th Purdue
Industrial Waste Conference, (May 2-4, 1972).
18. Private Commurdcation with Peter Kaye, Municipal Sales Manager, Dravo
Corporation, Pittsburgh, Pennsylvania, June 2, 1971.
19. J. J. Convery, Solids Removal Processes, Nutrient Removal and Advanced
Waste Treatment Symposium, FWPCA, Cincinnati, Ohio (April 29-30, 1969).
20. D. R. Zeny, “Hanover Tertiary Plant Studies,” April 9-June 30 Quarterly
Report WPRD Grant 92-01-68 (Unpublished).
21. C. L. Cuip and S. Hansen, Water & Sewage Wkà., 114, No. 2, p. 46 (1967).
22. Private Communication with Mr. Michael Strachow, Johnson & Anderson,
Consulting Engineers, Pontiac, Michigan, (March 1972).
23. G. R. Bell, D. V. Libby and D. T. Lordi, “Phosphorus Removal Using
Chemical Coagulation and a Continuous Countercurrent Filtration Process,”
FWQA, No. 17010-EDO (June 1970).
24. F. B. Laverty, R. Stone and L. A. Meyerson, Jour. Sanit. Engr. Div.,
ASCE, 87, 6, 1 (November 1961).
25. B. Lynam, G. Ettelt and T. J. McAloon, Jour. Water l’oll. Cont. Fed.,
Vol. 41 p. 247, (February 1969).
26. Performance Data Contained in Hydroclear Corporation Catalogue, Avon Lake
Ohio as tested by the Clark County Utilities Department, Springfield,
Ohio, May 1969.
27. R. L. Cuip and G. L. Cuip, “Advanced Wastewater Treatment,” Van Nostrand-
Reinhold Company, New York (1971).
28. DuPont PACT Process “Advertised Pilot Plant Results,” Jackson Laboratory,
P. 0. Box 525, Wilmington, Delaware, 19899.
29. R. Bloom, Jr., et aJ.; Env. Sci. Technology, Vol. 3 , p. 214 (March 1969).
30. E. L. Berg, R. Villiers, A. N. Masse and L. Winslow, Chem. Engr. Prog.
Symp. Series, Vol. 67 , No. 107, p. 154, (1970).
31. D. S. Davies and R. A. Kaplan, Chem. Engr. Prog. Syinp. Series, 60, 12,
p. 46, (1964).
32, FWPCA Contract 14-12-400 (GATX), Infilco Products Co.
US C CPJIM(NTPRINIINGOff ,CE. 1C72— ?59.548/1019
------- |