\\l
                  United States
                  Environmental Protection
                  Agency	
Risk Reduction Engineering
Laboratory
Cincinnati OH 45268
                  Research and Development
EPA/600/S2-89/003  Feb. 1990
&ER&         Project  Summary
                   Capital  and O&M  Cost
                   Estimates  for Attached
                   Growth Biological  Wastewater
                   Treatment  Processes
                   Henry H. Benjes, Jr.
                   Data  for  projecting  process
                  capabilities  of  attached  growth
                  biological  wastewater treatment
                  systems and procedures for making
                  design calculations are presented in
                  this report. Carbonaceous oxidation
                  (secondary treatment) and single-
                  stage nitrification design examples
                  are given. Information for estimating
                  average  construction costs and
                  operation and maintenance (O&M)
                  requirements are  presented for
                  typical wastewater treatment plants
                  ranging in size from 1 to 100 mgd
                  capacity.
                   Estimated  average construction
                  costs and  O&M  requirements for
                  individual unit processes are related
                  graphically to appropriate  single
                  parameters  for each component.
                  Construction costs are broken down
                  into labor and materials components
                  to enable the costs to be  inflated
                  using readily available Bureau  of
                  Labor Statistics Wholesale Price
                  Indices. O&M requirements are given
                  for labor, energy, and  maintenance
                  materials and  supplies  so that
                  appropriate current, local unit costs
                  can be used to estimate annual
                  costs.
                   The data in this report provide  a
                  means for estimating anticipated
                  average performance and costs for
                  facilities, but they should  not be
                  substituted for detailed assessment
                  of local conditions or recognition of
                  changing design requirements.
  This Project  Summary was
developed by EPA's Risk Reduction
Engineering 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
  This report represents recommended
design procedures, reviews performance
design procedures, reviews performance
capabilities, and presents cost estimating
guidelines for municipal wastewater
treatment plants incorporating attached
growth biological  processes for
secondary treatment. Attached  growth
treatment processes are based  on the
development of biological growth on a
media  surface,  either by passing
wastewater over stationary media or by
moving media through a wastewater bath.
Attached growth  processes  are most
commonly exemplified by the trickling
filter. The rock media trickling filter has
been recognized and used since 1898.
There are nearly 4,000 municipal trickling
filter  wastewater treatment plans  in the
United States.

Objectives
  The objectives  of this report are to
develop suggested design procedures for
attached growth  biological  treatment
processes, assess the accuracy of those
procedures,  and present guidelines for

-------
estimating  capital  costs  and  O&M
requirements.
  Commonly used design procedures for
biological  treatment  processes  are
empirical  in  nature,  based  on
experienced  results.  Available design
procedures are reviewed  to assess their
utility  in  assisting  the  engineer in
predicting attached growth performance.
  Once a  design has  been developed
and  the  proposed  facilities  sized,
estimating  capital  costs  and  O&M
requirements  must  be considered.
Capital costs  include  the  cost of
construction; engineering,  legal,  and
administrative services; land; and interest
during construction.  This  report
emphasizes  the  development  of
construction costs. Other  costs,  except
land,  may be  related to  construction
costs. Land is a variable  that cannot be
typified. General information for plant
construction costs has been available for
some time;  however, this information is
often  presented  for an overall system,
rather than  in terms of unit processes.
The variability of combinations of several
unit processes  limits  the  use  of  these
data.  By  separating  plant costs  into
categories  of unit  processes, historical
cost data  from existing plants may be
applied to  similar processes in project
planning.

Attached Growth Processes
Considered
  Four attached growth  processes  are
examined  in the  report,  including rock
media trickling  filters,  plastic  media
trickling  filters,  rotating biological
contactors, and trickling filter/solids
contact, which is  an  attached  growth
process  enhanced  by a  coupled-
suspended  growth process.  The  six
process alternatives  analyzed  in  the
study  are listed  in  Table  1.  These
processes are generally incorporated into
liquid stream system designs that include
pretreatment  via  screening  and  grit
removal, primary and final sedimentation,
sludge pumping,  recirculation pumping,
and effluent disinfection. Sludge handling
selection  varies  depending on local
economic considerations.

• Rock media trickling filters are  a
  simple, single-stage treatment process.
  Rock media varies in diameter from 1
  to 4 in.  and are designed with depths
  of  3 to   10 ft. Wastewater  is
  continuously  sprayed  over  the
  stationary  media,  which  supports
  biological growth. Treated wastewater
  is collected in  a underdrain  system
  where it is  recycled and/or directed to
  the  final settler. Biological  growth
  sloughs  from the media resulting in the
  need for final  sedimentation.  Rock
  media filters are usually employed  for
  secondary  treatment (carbonaceous
  removal) only.

• Plastic  media trickling filters were
  introduced  to overcome the limitations
  of rock  media.  Plastic  media  trickling
  filters may  be designed much deeper
  (commonly  21 ft deep) than rock filters
  since  the  media  is very light.
  Corrugated sheet  modules  are
  delivered in bundles  that are then cut
  to size and  placed in the media towers.
  Plastic rings, on  the other hand, are
  dumped, making installation simple.
    Recirculation  is   typically taken
  directly from  the  trickling  filter
  underflow,  although  some designs
  recycle  the final  sedimentation  tank
  underflow.  Plastic media  have been
  used for both carbonaceous  removal
  and nitrification.

• Rotating biological contactors (RBC's)
  were developed in  Europe  and
  introduced  in the  United States in the
  1970's.  The  media, which  supports
  biological growth, is generally 12 ft in
  diameter and rotated  through a bath of
  wastewater. The  media is alternately
  exposed to  the  liquid  and  to  the
  atmosphere. RBC effluent is  typically
  not recirculated.  Originally, the media
  was designed as a  series of closely-
  spaced,  parallel, flat discs  with a
  specific surface area  of 20 to 25 ft2/ft3.
  The newer lattice-structured media
Table f. Biological Treatment Process Alternatives
                      Treatment Process
offers about 50% more specific si
area than the disc-type  constru
The lattice-structured media, and
lesser  extent the disc structure
fragile  and should be  protected
direct  exposure  to  sun, wind,
weather. Therefore, the  medii
enclosed in either  superstructu
individual shaft covers. Media  re
can be provided by either mecri
drives  or air motivation. RBC's  m
used for either secondary treatm
secondary  treatment plus nitrifi
applications.

Trickling fitter/solids contact (TFI
a  development that enhance
reliability  of the trickling  filt
incorporating  suspended  g
treatment in the process.  There
been  other  "coupled"  atti
growth/suspended growth  pro<
that have been used in an atte
offset  the  disadvantages assc
with the two processes.  Wil
processes  there  can  be
variations  in  relative organic  I
rates,  locations  and quantit
recycle, and process arrangem
a consequence, there were an i
process alternatives under the
category  of  "coupled  tri
filter/suspended growth" proce;
TF/SC variation represents
processes in this report.
  The  TF/SC process uses the
filter (TF) as the primary me
remove organics and a  ven
hydraulic retention time aeratic
(SC)  to polish  that  tricklin
effluent. Where the  treat me
effluent quality needs only t
secondary  treatment stand;
conventional final sedimentatic
is used. Where higher quality
standards  are  required,
sedimentation basin with a flo<
center well is used. The TF/SC
is not  used for  nitrification.
nitrification  is required,  the
segment  must be  larger ;
process is  no longer categori;
TF/SC process.
                      Carbonaceous Removal Only
        Single-Stage Carb. Rem
                Nitrification
                  Rock Media Trickling Filter

                  Plastic Media Trickling Filter

                  RBC's

                  TricUing Filter/Solids Contact

-------
 ocedure
 i three-step approach was used to
 nduct the work of this study. The  first
 p  was to develop design criteria for
 atment plant liquid and solids handling
 it processes  applicable  to  attached
 iwth treatment systems. To thoroughly
 aluate a  complete treatment system
 ernative, it is necessary to consider the
 sign and costs of ancillary treatment
 its,  such as solids handling processes
 d functional parts of the  total project.
 tailed construction costs and O&M
 }uirements for typical additional  unit
 ocesses  and  functional units that
 mplete the  system alternatives  are
 :luded.
 The second  step  was  to collect,
 alyze, and formulate the construction
 st and O&M requirements for each  unit
 ocess. Comparative cost information is
 esented  for   certain  design
 odifications,  e.g.,  alternative  solids
 ocessing equipment.
 The third  step was to develop flow
 agrams  for  each  of  the systems.
 pical flow schemes meeting the state
 rformance  requirements  have been
 eluded.
 The attached   growth  processes
 timated have been sized to correspond
  design  flows ranging from  1  to  100
 gd.  Within this range, six nominal plant
 pacities have been evaluated: 1, 5, 10,
 , 50, and  100 mgd.
 The following unit process costs have
 en  included  in  the  complete final
sport:

law wastewater pumping
'.hlorine feed & storage facilities
Derated grit removal & flow measurement
>O2 storage & feed equipment
'rimary treatment screens
 lotation sludge thickeners
 edimentation basins
 ludge handling tanks
Sludge pumping stations
Anaerobic digesters
Trickling filters
Filter presses
Rotating biological contactors
Centrifuges
Inplant & recycle pumping stations
Multiple hearth incinerators
Aeration basins
Sludge & ash lagoons
Mechanical aeration equipment
Land spreading of sludge
Blowers
Sand drying beds
Diffused air aeration equipment
Sludge composting
Effluent filtration
Pipeline transport
Chlorine contact basins
Truck transport

  The costs have been presented  in two
forms. The  construction cost components
have been  itemized for several sizes of
the unit process so they can be updated
according to the  inflation rates for the
individual components. The total updated
costs  (September  1987)  are  also
presented graphically  so they can  be
used for any size treatment system. Total
annual costs (September 1987) for the
different  size plants and  treatment
options are summarized.
  These  costs, presented in terms of
$/1,000 gal  wastewater treated, are the
sum of a plant's annual O&M costs and
its capital costs amortized for 20 yr at
10% divided  by  the  total  quantity of
wastewater treatment annually.
  The most  cost-effective  treatment
method is indicated in Table 2. RBC's are
estimated to be  the  most  economical
attached   growth  process  for
carbonaceous oxidation and  for  single-
stage nitrification. The estimated costs for
the TF/SC process are  essentially the
same as for the RBC process above 10
                       mgd. The relative ranking of these costs
                       for  estimating  purposes  should  be
                       tempered by site-specific conditions and
                       by engineering judgment.  The  use  of
                       these cost estimating procedures  results
                       in project estimates that are very close to
                       the  experienced costs to  as  much as
                       30% different than experienced  costs.
                       The  relative  accuracy,  comparing
                       competing alternatives should  be within
                       10%.
                       Design Approaches
                         Performance  data  from  operating
                       systems are used to evaluate the various
                       methods for designing attached growth
                       processes. This  is particularly important
                       since design must be based on achieving
                       specific effluent quality.  The  design
                       approaches for removal in rock  and
                       plastic  media  trickling filters  and  RBC's
                       are  evaluated first,  followed  by
                       performance evaluation  and  the design
                       approach for the TF/SC process
                         Attached  growth   processes  are
                       characterized by   a  decreasing
                       concentration of organics passing  over a
                       film of attached bacterial growth. Organic
                       and oxygen fluxes from the  carriage
                       water to the growth are proportional to
                       their concentrations. The surface area is
                       the major parameter in attached growth
                       process evaluation if the  organic loading
                       rate is not so high that either the rate of
                       organic  assimilation by bacteria or the
                       rate of oxygen  transfer would  limit the
                       removal rate.  Greater  surface area per
                       unit volume will support more  bacterial
                       growth and  provide  more  contact
                       opportunities  between  organics  and
                       bacteria. However,  there   are  many
                       complicating  factors  that  obviate the
                       effect  of media surface area. These
                       factors  have relegated attached growth
                       process design to empirical relationships
                       that are of limited usefulness.
ttile 2. Summary of Total Annual Costs for Plants Utilizing Attached Growth Treatment Processes
                                                                 Annual Cost Summary, $11000 gal
                                                                        Plant Size, mgd
                Process
                               10
                                            25
 'ingle-Stage Nitrification
 lastic Media
 IBC's
    3.28
    2.65
2.07
1.62
1.80
1.40
1.44
1.12
                                        50
                                       100
Carbonaceous Oxidation
toc/t Media
lastic Media
IBC's
f/SC

3.86
2.93
2.53
3.09

2.59
1.74
1.48
1.69

2.32
1.41
1.22
1.38

1.88
1.20
0.99
1.02

1.67
0.97
0.82
0.85

1.60
0.94
0.77
0.79
1.18
0.94
1.17
0.94

-------
  Empirical models  based on  statistical
curve fitting  of data to variations  in
operating  conditions  and   physical
facilities are most commonly  used by
design  engineers.   The   actual
phenomenon  involved in  organics
removal may  or may not be understood
from  the  resulting  statistical model.
These empirical models yield varying
results  that  do  not reflect  the  true
removal phenomenon. It is  important to
realize this  limitation and  restrict  the
application of the empirical models to the
range of operating conditions  and
wastewater characteristics for which they
have been developed.
  Techniques  classified as  rational
approaches better describe the removal
mechanisms,  but  they also  present
difficulty  in application.  The Williamson
and  McCarty biofilm model  represents
the rational approach, although it is rather
complex and may be beyond general use
by design  engineers. This model
considers many  factors that  describe
substrate utilization by biofilms.  Basically,
it  predicts soluble substrate  removal
based on limitations of  oxygen  and
substrate diffusion through the liquid and
the biofilm  into  the  bacteria. It  also
considers  the simultaneous effects  of
biochemical  reactions. The biofilm
surface area is a key design parameter.

Empirical Models
  Empirical predictive design techniques
for trickling filters have  been presented
by several investigators. The  complete
project report  for this study  describes
several  empirical models including  the
National Research Council (NRC), Caller
and  Gotaas, modified  Velz,  and  the
rational  model  of Williamson  and
McCarty.
  A variation of the basic Velz equation is
presented in this summary report:
  , BOD_ out,
In  	5__  =K
  L BOD, in J
                        69SQ/A
                         —
                         6960j
                                   (1)
where:  K = coefficient related to  media
            volume gpm"/(ft3)"
        Q =flow  rate to  the  filter
            including recirculation, mgd
        A = filter surface area, ft2
        D = filter depth, ft
        n = hydraulic coefficient
                                           v = filter volume. 1,000 ft3

                                     The variation in  K with  varying media
                                   wetting rates (applied hydraulic loading to
                                   plan  surface  area of trickling  filter)  is
                                   predicted by the following equation for
                                   rock media trickling filters:
                                    K = 0.25 + (1nqw)/20
        (2)
                                    where:  qw = media wetting rate, gpm/ft2

                                    Model Evaluation
                                      The designer is faced with selecting a
                                    media  volume for which  the effluent
                                    criteria may be attained with a reasonable
                                    degree of confidence. In the following
                                    discussion,  data  are  presented  for
                                    existing  attached growth systems.  The
                                    Velz model generally is used to predict
                                    effluent soluble BOD5 from  the trickling
                                    filter. Sometimes  it is used  with influent
                                    soluble BOD5. Since influent BOD5 is
                                    hydrolyzed quickly, the author believes it
                                    is inappropriate  to use influent soluble
                                    BOD5. The model  has  been  applied in
                                    this  report to predict effluent total BOD5
                                    after the final clarifier. The model might
                                    be more precise  if used  to predict
                                    effluent  soluble  BOD5 and if effluent
                                    insoluble  BOD5 were estimated, but the
                                    precision of the model is not adequate to
                                    justify such refinements.
                                      Tables 3, 4, and 5 present field  data
                                    and  predicted results for  rock media,
                                    fabricated media,  and RBC systems,
                                    respectively. Variables  used  in   the
                                    equations  to  predict performance  are
                                    given in Table 6.
                                      It  is  noteworthy  that  the  K values in
                                    Equation 1 for rock media trickling filters
                                    approach those of plastic media at higher
                                    wetting rates:
                                     Wetting Rate (qj,
                                          gpm/ft2
K,
0.1
0.2
0.3
0.4
0.15
0.18
0.20
0.22
                                      An "n" value of 0.5 has been used in
                                    these comparisons. The performance of
                                    plastic  media  trickling filters  was
                                    predicted using a K of 0.21  gpmos/ft15
                                    for wetting rates varying from 0.5 to 2.27
                                    gpm/ft2  The  probable  reason that the
                                    specific  surface area  of plastic media is
                                    not  more  effectively utilized  at
                                    conventional  organic loading rates  in
                                    comparison  to  rock  media  is oxygen
                                    diffusion  limitations.  The  treatment
efficiency  achieved with  both typ
media  will  be  determined  b<
availability  of  oxygen  and
effectiveness of the media to aera
wastewater.
  Richards and  Reinhardt  invest
different configurations of plastic
using variable depths  with  the
media  volume  and  found
performance  improved with  depth
media specific surface areas used i
investigation did not vary. Their fi
indicated that the 45° and 60° cro.1
configurations  performed  better
either the vertical configuration or r.'
dump media. When they  evaluatec
plant data, they found an "n" of (
best mimic performance of soluble
removal. They  used an  "n" of
compare field data.

Rotating Media Biological
Contractor (RBC's)
  The design approaches propos
RBC manufacturers are primarily
on "rational" models. One such ap
is  summarized  in  the  grai
relationship   between  effective
surface  area (expressed as  flow f
of surface area)  and effluent :
BOD5   shown  in  Figure  1.
relationship indicates benefits fron
media with  high  specific surface
The design approach shown in  F
is based on  soluble BOD5 in the
and effluent. Unfortunately, the
soluble  BOD5 portion  is  highly  v
For example,  the  following  havi
reported for  soluble BO05  in  |
effluents:
                                                                                   Plant
                                  Soluble BO
 Pewaukee, Wl
 Seattle, WA
 Tucson, AR
       66
31-50(41 av
50-71 (56 av
                The  use of influent  soluble
              assumes that insoluble BOD5 is r
              by some mechanism other than b
              stabilization. Some insoluble c
              may be incorporated  in biologi
              and removed by sedimentation, t
              will be hydrolyzed and metal
              Therefore,  a  design  approach
              based on only  soluble  organic lo
              a  liberal  one.  Since  hydr
              partjculate organics as well as
              organics are  available  substr;
              design  (substrate removal  ap
              empirical approach, or other) st
              based on total influent substrate.

-------
 To  provide a  design  approach  more
insistent with stationary media attached
•owth processes and to enable realistic
valuation of the available data, Equation
 has been applied to the  RBC process.
vailable data for mechanically  driven
sc  and lattice-type  RBC  systems have
een  evaluated using this  approach and
ere  summarized earlier  in  Table  4.
lese data represent both full-scale and
lot-plant installations.
 Because  lattice media  have  greater
jrface  area  per unit volume  than disc
ledia, an analysis of the  data was also
erformed  relating  BOD5 removal  to
»edia surface  area according  to the
illowing equation:
     BOD  out
In  - - - I = -K
                                 (3)
     BOD. in
         5
 The performance data in Table 4 have
een  evaluated  in  terms  of ks,  the
 (efficient related to media surface area,
pmn/(ft2)".  Figure  2  represents a
robability  distribution plot  of  the
alculated ks values that imply that media
pecific surface area is a more  significant
arameter for the design of RBC process
erformance than media volume. Using a
s value of  0.062  gpm°5/ft and  media
densities of 20 ft2/ft3 for the disc media
and  30 ft2/ft,3  for the lattice media,  the
standard error  of estimate would  be 5
mg/L.


Trickling  Filter/Solids Contact
  The inability  to  accurately  predict
trickling filter  process  performance and
the need  for  uniformly reliable  effluent
quality have led to the  development of a
variety of  combined  trickling  filter-
suspended growth systems. The TF/SC
process is one of the coupled processes
consisting of a trickling filter followed  by
an aeration basin. The trickling filter is
lightly  loaded, usually  20 to  50  Ib
BOD5/1,000 ft3/day.  The aeration  basin
detention time may be  10 min to as long
as 1  hr.
  The TF/SC  process  relies  on  the
trickling filter to stabilize the  majority of
the organics  while  the  aeration  basin
completes the  stabilization  of  the
organics and  conglomerates  the solids
into a settleable floe.
  The evaluation of coupled processes is
complicated by the difficulty in separating
the removal occurring  in the  individual
process units.  The data presented  by
most investigators  are  not  complete;
therefore,  a thorough  evaluation  is  not
possible.  The design  and  evaluation
procedures used in this report are based
on the following:

• Assume the trickling filter performs in
  the same manner  that it would  when
  operating alone.

• The trickling filer soluble BOD will exert
  a synthesis  oxygen demand of 0.5 Ib
  02/lb soluble BOD5 synthesized.

• The endogenous oxygen demand  will
  be  1.2  Ib 02/lb  insoluble BOD5  or
  synthesized  cellular material.

  An example  of the  application of these
concepts is presented in Table 7 for the
field data collected for the Corvallis, OR,
TF/SC plant. The Corvallis plant consists
of a trickling  filter followed by a  solids
contact  aeration basin of  0.02  mil  gal
volume. The final  report  describes
temperature consideration,  nitrification
design equations, and example design
illustrations.

Conclusions
  This report  is a consolidated  volume
describing  the methodology  involved in
designing  attached  growth  biological
wastewater treatment  processes  to
achieve  carbonaceous  oxidation
able 3. Comparison of Predicated vs. Measured Effluent BOD5 Using Rock Media Trickling Filter Data
Plant Location
Aurora, IL
Dayton, OH
Oruham, NC
Madison, Wl
Richard, TX
"lainfield, NJ
Great Neck, NY
Oklahoma City, OK
-reemont, OH
Storm Lake, IA
lichland, WA
Misal, CA
Chapel Hill, NC
Dallas, TX
Bridgeport, Ml
Jass City, Ml
Charlotte, Ml
lillsdale, Ml
apeer, Ml
•tate Prison, Ml
assar, Ml
nglewood, CO
orvallis, OR
orvallis, OR
Depth, ft
6.0
7.5
7.0
10.0
6.5
6.0
4.0
6.0
3.3
8.0
4.5
32
4.25
7.5
6.0
6.0
6.0
6.0
5.8
8.0
5.6
4.4
8.0
8.0
R
—
—
—
—
-
06
1.0
1.0
1.5
2.1
2.8
3.1
2.0
0.5
1.2
1.3
—
—
0.3
0.1
1.7
1.0
2.4
05
0/A, mgd/ac
2.1
3.5
1.9
2.4
3.9
2.4
7.8
16.3
19.0
21.5
19.6
20.8
16.3
5.6
20.6
10.0
7.7
3.6
73.5
3.8
9.2
74.8
24.6
24.6
W/V, Ib
8005/7,000
fWday
4.4
12
13
6.4
73.3
25
20
78
41
62
44
53
19
21.4
29
23
29
10
22
13
6
60
16
19
BODj,
In
70
138
261
738
778
76
117
300
95
381
118
185
77
225
99
151
119
91
65
153
59
158
86
49
mg/L
Out
14
33
68
33
20
73
20
66
27
67
20
24
44
37
42
33
63
32
23
17
29
46
32
31
Predicted
Effluent BODj,
mg/L, from
Equation 1
20
34
66
27
32
15
32
78
32
63
25
49
79
45
26
30
39
26
27
34
77
49
32
18

-------
Table 4. Comparison of Predicted vs. Measured Effluent BOD5 Using Plastic Media Trickling Filter Data
Plant Location
Indianapolis, IN
Stockton, CA
Akron, OH
Buena Vista, Ml
Bay City, Ml
Essexville, Ml
Greenville, Ml
Rockwood, Ml
' Indio, CA
2 Linden Rochelle, NJ
3
3
Media
Plastic
Plastic
Plastic
dumped
Plastic
Plastic
Plastic
Plastic
Plastic
Plastic
Plastic
Plastic
Plastic
Depth, ft
21.5
21.5
25.5
20.0
21.5
21.5
21.5
22.0
33.0
21.5
20.0
10.0
q, gpm/ft2
2.0
0.28
0.36
0.46
0.90
0.75
0.46
0.32
0.27
1.10
1.4
0.6
Rate, gpm/ft2
2.0
0.71
0.75
1.20
1.1
1.50
0.50
0.97
--
2.77
1.4
0.06
' Drury, D. D., Carmota, III, J., and Degadillo, A., "Evaluation of High Density Cross Flow Media for
58(5):364, May 1986.
2Fillos, J., Nierstedt, R., and Donahur, A, "Full Scale Evaluation of Plastic Media Roughing Filters,
New Orleans, LA, October 1984.
3 Richards, T. and Reinhart, D., "Evaluation of Plastic Media in Trickling Filters," JWPCF, 58(7):774
8C
In
112
240
120
54
79
23
62
61
62
100
78
76
JDs, mg/L
Out
57
40
20
21
18
11
15
23
72
50
29
41
Predict
Effluent B
mrj/L fn
Equatio,
56
38
18
14
28
7
15
10
46
40
36
42
Rehabilitating an Existing Trickling Filter, " J
" Presented at 57th Annual WPCF Con fere
, July 1986
Table 5. Comparison of Predicted vs. Measured Effluent BOD5 Attached Growth Model Using RBC Data*
Plant Location
Pewaukee, Wl
Pewaukee, Wl
Edgewater, NJ
Gladstone, Ml
Gladstone, Ml
Woodland, WA
Kirksville, MO
Georgetown, KY
Brainerd, MN
Media
Disc
Disc
Lattice
Disc
Lattice
Lattice
Lattice
Lattice
Lattice
Volume, ft3
197
10,450
6,110
196
16,300
2,413
63,100
25,240
40.715
0, gpm
8.3
235.0
333.0
10.4
550.0
104.0
904.0
765.0
950.0
SOOg,
In
172
119
133
100
106
175
164
150
80
mg/L
Out
33
20
38
32
20
28
15
21
17
Predict
Effluent £
mg/L fr
Equatio
38
15
35
26
20
40
12
25
20
'Lehman, P. J., 'Start-up and Operating Characteristics of an RBC Facility in a Cold Climate," JWPCF, 55(10):1233, October 1983.
Table 6. Variables Used for Models Evaluation


     Modified Velz Parameters               Rock Media
                             Plastic Media
                                                               RBC's
 n
 K (Equation 1)
    0.5

(Equation 2)
0.5

0.21
0.5

0.308
(secondary treatment) and  nitrification of
domestic wastewater.  The  theoretical
considerations given  to  design  are
reviewed, and  detailed  examples using
the  most  accurate  approaches  are
presented. Cost analyses were facilitated
 by using  a computer;  however,  the
 procedures are  straightforward and  can
 easily be done manually.
    Several  mathematical  models have
 been used to  design attached  growth
 systems. None are particularly accurate
          in  predicting  process performai
          some  are  quite  complicatt
          carbonaceous removal, the Velz
          is  as accurate as  any  and
          conveniently  applied to  all
          growth processes. The Velz equ

-------
    30
    25
    20
Q
O
00
J>   15
to

0)
    70-
    5 -
/?flC Process Design Criteria

Domestic Wastewater Treatment

Wastewater Temperature = 5
4-Stage Operation
                                           Influent Soluhif «OOS mg/L

                                             ISO    120   100
         	1	1	1	1	1	1—

      0    0.5    1.0     1.5    2.0   2.5    3.0

                         Hydraulic Loading, gpd. ft2


Igun 1.    Manufacturer's design approach for RBC's.
                                          3.5
—i—

 4.0
—i—

 4.S
a  applied  to  rock media,  plastic or
spropriate modifications.

 Total plant capital and O&M  costs have
»en estimated for various size facilities
sing  attached  growth  biological
•ocesses for  secondary   treatment
 arbonaceous  oxidation)  and  for
                               nitrification in the final  report. RBC's are
                               estimated to be the most economical
                               attached   growth   process   for
                               carbonaceous oxidation and  for  single-
                               stage nitrification. The estimated costs for
                               the  TF/SC process  are  essentially  the
                               same as for the RCB process above 10
                               mgd. The relative ranking of these costs
                        for  estimating purposes  should be
                        tempered by site-specific conditions and
                        engineering judgment.

                          This report was submitted in fulfillment
                        of Contract No. 68-03-2556 by CWC/HDR
                        Engineers  under the sponsorship of the
                        U.S. Environmental Protection Agency.

-------
  0.088
  0.080
  0.072
   0.064
   0.056
  0.048
  0.040
               5     10       20    30   40  50  60    70    80     90
                   Percent of Time Ka is Equal to or Less Than Stated Value
Figure 2.    Probability of RBC performance based on media surface area

-------
'able 7. Evaluation of TF/SC Process
lorvallis TF/SC Plant (1983-1984)
Month
Q, mgd
Temp., °C
Influent
800$, mg/L
TSS, mg/L
TF Effluent
SODj, mg/L
SSOOg, mg/L
rSS, mg/L
Cn mg/L
Cg, mg/L
Or, mgd1
Solids
Aeration, Ib2
Reaeration, Ib3
Clarifier, Ib4
Total, Ib
SRT, days
BODs/TSS Ratio6
Oxygen Demand,
Ib/day
Synthesis7
Endogenous
Aeration^
Endogenous
Reaeration9
Oxygen Demand,
mg/Uhr
O2 Demand
Aeration10
O2 Demand
Reaeration"
Apr.
12.2
15

66
75

25
6
63
13,075
3,110
3.8

520
2,180
17,300
20,000
3.1
0.3


90

63

266


38


67
May
74
18

90
82

34
5
72
8,091
2,150
2.7

360
1,350
7,600
9,310
2.1
0.4


71

72

270


36


68
June
7.3
20

87
74

32
6
61
8,180
1,940
2.3

324
1,364
6,450
8,138
2.2
0.4


91

75

154


42


39
July
6.2
20

78
68

28
5
60
6,345
1,768
2.4

295
1,060
5,280
6.635
1.9
0.4


69

68

244


34


61
Aug.
6.2
22

72
63

29
5
57
5,437
1,557
2.5

260
907
4,700
5,867
1.8
0.4


74

69

240


36


60
Sept.
5.7
22

94
68

39
8
59
5,415
1.675
2.6

280
903
4,800
5,983
1.9
0.53


112

98

170


53


43
Oct,
5.6
21

114
66

38
8
56
70,293
2,948
2.2

490
1,720
8,040
10,250
3.3
0.54


108

164

572


68


143
Nov.
15.2
17

56
56

33
6
55
13,703
3,571
5.4

595
2,285
25,560
28,340
4.0
0.5


106

139

534


61


134
Dec.
17.9
14

35
58

26
4
59
76,739
4,278
6.3

703
2,690
35,520
38,973
4.5
0.36


63

707

368


43


92
Jan.
13.4
14

49
56

26
4
54
17,170
4,777
5.2

797
2,870
30,800
34,464
5.5
0.42


58

127

457


46


774
Feb.
16.6
13

56
64

22
3
59
76,523
4,832
6.9

806
2,760
39,390
42,956
4.8
0.32


44

91

312


34


78
Mar.
12.7
13

48
64

22
3
58
75,353
4,982
6.1

830
2,560
32,550
35,940
5.3
0.33


41

97

299


35


75
 '  Q, = <
 2  Aeration Ib solids  = CaxVax 8.34  = Cax 0.02 x 8.34
 3  Reaeration Ib solids = CrVrx8.34 = Crx 0.02 x 5.34
 4  Clarifier Ib solids = (0, * Q,) Ca x 8.34/24, assuming 1-hr time to achieve Cr
 5  Total Ib solids/(Q in x TF TSS out x 8.34)
 «  BODs/TSS ratio  =  (TF 8O05 out - TF SBOD5 out)/TF TSS out
7 Synthesis Oxygen, 0.5 xTF SBOD5 out x 24 x Vax 8.34
                                                           KSK,
                                                           KSK,  ta+1
 » Endogenous Oxygen, Aeration, Ibid = 1.2 Cax 8.34 x (BODs/TSS ratio)
 ' Endogenous Oxygen, Reaeration, Ib/d =  1.2 Crx 8.34 x (BODJTSS ra
 '° Oxygen Demand Aeration, mg/Uhr = (Synthesis + Endogenous Aeration)/(V, x 8.34 x 24)
 '' Oxygen Demand Reaeration, mg/Uhr = (Endogenous Aeratin)/(Vr x 8.34 x 24)
 Vhere: Q, = RAS flow mgd
        0, = In flow mgd
        Ca = MLSS, mg/L
        Cr = RAS concentration, mg/L
        V, = Aeration basin volume, mil gal
        Vr = Reaeration basin volume, mil gal
        TF SS out    - trickling filter effluent suspended solids, mg/L
        TT BOD out   = trickling filter effluent 800$, mg/L
        TF SBOD out = trickling filter effluent soluble 800$, mg/L
        Ks= lShri@20'C
        Ke = 0.02 hr> @ 20"C
        Kt = 1.072 (T-20)
        t  = aeration detention time, hr

-------
Henry H. Benjes, Jr., is with CWC/HDR Engineers, Dallas, TX 75230.
John J. Convery is the EPA Project Officer (see below).
The complete report, entitled "Capital and O&M Cost Estimates for Attached
Growth Biological Wastewater Treatment Processes," (Order No. PB 89-148
3241 AS; Cost: $36.95, 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:
         Risk Reduction Engineering Laboratory
         U.S. Environmental Protection Agency
         Cincinnati, OH 45268
                                                       10

-------
                                             11
                                             Is
                                             «
                                             C
                                             in
     	  o
     HOOWO
     OHO
     O
     00
2
o


f.
3




I
  HI  H
H»  H

     O
CM4
O
ON

s
W

«B

O
x

                                                   r'l'


                                                  O

-------