United States        Office of Water        September 1984
              Environmental Protection    Program Operations (WH-546)  430/9-8*4-008 '
              Agency           Washington DC 20460
vvEPA       Summary of
              Design Information
              on Rotating
              Biological Contactors

-------
                                           EPA-430/9-84-008
                 SUMMARY OF
           DESIGN  INFORMATION ON
       ROTATING BIOLOGICAL CONTACTORS
                     by

              James A. Heidman
             Richard C. Brenner

        Wastewater Research Division
Municipal Environmental Research Laboratory
           Cincinnati, Ohio 45268

                    and

             Walter G. Gilbert

      Municipal  Construction Division
     Office of Water Program Operations
           Washington, D.C. 20460
                                             Agency
     OFFICE  OF  WATER PROGRAM OPERATIONS
              OFFICE OF  WATER
    U.S.  ENVIRONMENTAL  PROTECTION AGENCY
          WASHINGTON,  D.C.  20460

-------
                              EPA REVIEW NOTICE
     This report has been reviewed by the Environmental Protection Agency
and approved for publication.  Approval does not signify that the contents
necessarily reflect the views and policies of the Environmental Protection
Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.

-------
                                    FOREWORD


      Rotating  biological  contactors  (RBCs)  are  relatively  new  to  secondary
 wastewater  treatment  in  the  United  States,  with over  500 municipal  facilities
 being installed  in  the  last  decade.   In  addition  to secondary  treatment  appli-
 cations,  RBCs  have  also  been used successfully  to upgrade  marginal  treatment
 facilities  and to provide nitrification.  Because of  a  variety of problems
 related to  the design, construction,  operation, and application of  RBC facili-
 ties  throughout  the country,  the U.S.  Environmental Protection Agency has
 undertaken  a number of research projects to investigate and identify the nature
 and extent  of  these problems  and to determine possible  solutions.   These
 efforts have indicated that  when properly designed, built, and operated, RBCs
 can provide an acceptable alternative  to conventional activated sludge systems.

      The  Wastewater Research  Division  of the Municipal Environmental Research
 Laboratory, located in Cincinnati, Ohio, has conducted a comprehensive array
 of research projects on various aspects of  the  RBC treatment process, including
 the theoretical  basis of  the  process,  process design and operational considera-
 tions, and  equipment reliability and design.  The results and findings of these
 projects  are detailed in  a Municipal Environmental Research Laboratory publica-
 tion  entitled  "Design Information on Rotating Biological Contactors" (EPA-600/2-
 84-106).  This publication is available from the National  Technical  Information
 Service (NTIS  PB84-199561).

      This Summary document is extracted from the above publication and presents
 in a  concise form essential information on  the design, construction, operation,
and application of the RBC process.   It is  intended to provide  a  basic under-
standing of the process  and its application  for  the treatment  of  municipal waste-
waters, and will  be of use to design engineers,  governmental  agency  review
personnel, municipal officials, operators,  and others  with  an  interest  in the
RBC process.

-------
                                    CONTENTS
 EPA Review Notice ............................ ii
 Foreword  .............................. ! *iii

      1.  Introduction ..........................  i
      2.  Process Description  .................... ] \  2
      3.  Equipment Description  .....................  4
              Introduction ........................  4
              Shafts ...........................  4
              Media  ...........................  4
              Drive systems  .......................  8
              Bearings ............ .  .............  3
              Load cells  .........................  8
              Covers ........................... 10
              RBC arrangement  .....................  [ iy
      4.  Equipment Reliability  ..................... n
              Introduction ........................ n
              Shafts  ......................... !  .* 11
              Media  ...........................  13
              Drive systems  .......................  13
              Bearings  .......................  ! !  !  13
              Summary   ........................ !  !  14
      5.   Organic Removal   ...................... !  !  15
              Wastewater  carbonaceous  characteristics   ........ !  !  15
              Mass  transfer of  oxygen  and  organics  ............  15
              First-stage  loading  limit   .................  19
              Design approaches   ................... !  !  21
              Secondary clarification   ................ !  !  25
      6.   Nitrification   ...... .  .........  ....!!!!!  27
              Introduction  .................  '.'.'.'.'.'.'.  27
              Influence of  organics  ...................  27
              Nitrification rates   ..................  !  !  28
      7.   Energy  Requirements   ....................  !  !  34
              Introduction  ......................  !  !  34
             Mechanical drive systems  ..................  34
             Air drive systems  ...................  !  !  36
     8.   Plant Design Considerations   ............. ! !  !  !  !  39
              Introduction  .....................  !  !  !  39
             Pretreatment  ......................  !  !  39
             Equipment specifications, maintenance, and reliability  .  !  .'  39
             Flow control  ........................  40
             Biomass monitoring and control ..............  *  41
             Miscellaneous considerations ...............  [42

References  ..............................    44

-------
                                  SECTION 1

                                 INTRODUCTION
     The relatively rapid introduction of rotating biological contactors
(RBCs) into the United States for municipal and industrial wastewater treat
ment has resulted in the widespread application of a technology with which
many design engineers are not intimately familiar.  Of necessity, many RBC
designs initially were based solely on empirical design procedures generated
by various manufacturers.  More recently, as interest in the process has
increased, alternative design approaches have begun to appear in the technical
literature.

     This document highlights design information and supplements commonly
accepted RBC design methodology, such as manufacturers' design manuals and
empirical and deterministic models found in the literature, by providing
additional information and summaries of operating and performance data not
readily available to the design community.  Most of the data used in evalua-
ting the RBC process were obtained from the technical literature, conference
proceedings, and the files of the manufacturers.  Important design parameters
and relationships (or lack of them) are discussed to promote a more rational
RBC design approach.

     The purpose of this document is to concisely summarize RBC design infor-
mation for municipal wastewater currently available in a number of EPA publi-
cations (1)(2)(3)(4).  When a more in-depth discussion of a particular topic
is desired, the reader may wish to consult the original publications for
further details.  Topics addressed include process and equipment descriptions,
equipment reliability, organic removal, nitrification, and energy requirements,
A major priority was given to emphasizing practical, usable design information
as well as important theoretical concepts.  This summary is not intended to
serve as a "cook book" design reference or to replace other available design
guides from either manufacturers or the technical  literature.

-------
                                  SECTION 2

                             PROCESS DESCRIPTION
     All RBC systems are cylindrical-type structures  consisting of plastic
media attached to and/or supported by horizontal  rotating shafts.   The first
commercial RBC system was installed in West Germany in 1960.   Units constructed
from this time to the early 1970s used flat 0.5-in. thick,  6.5- to 10-ft dia-
meter expanded polystyrene discs.  All present systems use  thin (0.04 to 0.06
in.), high density plastic media either formed as discs or  sections of discs
and aligned perpendicular to the shaft, or spirally wound onto and aligned
parallel to the shaft.  The media configuration,  method of  media attachment
to the shaft, available surface area  per unit shaft length, method of rotating
the shaft, etc., vary with each manufacturer.  An in-depth  discussion of avail-
able equipment is presented in Section 3.

     Approximately 70 percent of the  RBC systems  operating  in the United
States and Canada are designed for organic carbon removal only, 25 percent
for combined organics removal and nitrification,  and 5 percent for nitrifica-
tion of secondary effluent (5).  When used for these applications, the RBC
shafts are positioned over the wastewater surface such that about 40 percent
of the media is submerged at any time.  The shafts are rotated slowly (1 to 2
rpm) causing an alternating exposure  of the media to the atmosphere and the
wastewater.  Biological growth (biofilm) becomes  attached to the surface of
the media and forms a slime layer over the entire surface of the discs.  The
biological population that develops on each unit  reflects the environmental
and loading conditions unique to each shaft, and  simple visual observation
reveals a gradation in slime thickness and color  in staged systems.  The
first stage in a system operating within the proper organic loading range
exhibits  a characteristic brownish-grey color, while terminal stages that are
nitrifying normally have a characteristic reddish-bronze color (6).  The
rotation  of the discs alternately contacts the biofilm with organic material
in the wastewater and then with the air.  Shearing forces exerted on the bio-
film as it passes through the wastewater cause excess biological growth to be
sloughed  off the media into the stage liquor where the turbulence created by
disc rotation maintains the sloughed  biomass in suspension.

     Microorganisms respond to the environment surrounding them, and, in the
RBC, that environment is continually  changing.   The requirement for movement
of organics and nutrients from the liquid phase into the biofilm and oxygen
from the  atmosphere into the liquid film, biofilm, and bulk liquid makes it
necessary to consider physical mass transfer as well as microbial reaction
rates when rationally analyzing  RBC performance.    These considerations are
addressed in more detail in Section 5.

-------
      A typical  RBC  application  for  secondary  treatment  is  depicted  in  Figure  1.
 Here,  flow passes from  the  primary  clarifier  through  the series  arrangement
 of  RBC units  and then to  a  final  clarifier where  the  excess  biofilm sheared
 from the  RBC  unit is settled.   In some  installations, screens are used in
 place  of  primary clarifiers  with  a  resultant  increased  solids and biochemical
 oxygen demand  (BOD)  load  to  the RBC system.   If the RBC unit was receiving a
 secondary effluent  and  was  designed to  provide nitrification, a  clarifier
 following nitrification would not necessarily be  required.

     When the  dissolved oxygen  (DO)  level is  at or near zero, many  hetero-
 trophic microorganisms  are able to  reduce nitrate nitrogen to nitrogen  gas.
 This phenomenon is  used in wastewater treatment systems for nitrogen removal
 (denitrification).  Unoxidized nitrogen must  first be converted to  nitrate
 nitrogen  in an aerobic  environment.  Denitrification can then be accomplished
 in  separate-stage RBC systems where  carbonaceous oxidation and nitrification
 occur  in  the lead stage(s) of the RBC process train and a carbon source,
 commonly  methanol,  is added to provide the energy for microbial  denitrification
 in  the last (anoxic) stage.  RBC  systems can also be staged with the anoxic
 unit as the first stage of the RBC with the organic carbon naturally present
 in  the incoming wastewater used for  nitrate reduction.  Nitrate nitrogen must
 be  introduced to this stage by recirculation of nitrified wastewater from the
 downstream stages.

     As no DO is desired in the denitrification reaction,  RBC media  should be
 completely submerged in the wastewater for denitrification applications.  The
 submerged media are mechanically driven at a rotational  velocity of  about
 1.6 rpm according to the current design procedure  (7).  Only one full-scale,
municipal  RBC denitritrification facility was in  operation  in the United States
 (8) as of August 1984.
                               RBC UNITS
   PRIMARY TREATMENT
                                                  SECONDARY CLARIFIER
RAW
WASTE
WATER
SECONDARY
EFFLUENT
                            SOLIDS DISPOSAL
         Figure 1. Typical RBC plant schematic for secondary treatment application
               [from Autotrol design manual (7)].

-------
                                  SECTION 3

                            EQUIPMENT DESCRIPTION
INTRODUCTION
     The five major U.S. manufacturers marketing RBC equipment  at the present
time are Clow, Crane-Cochrane, Envirex (formerly Autotrol),  Lyco, and Walker
Process.  Each manufacturer designs the RBC with an individuality that is
unique to its firm.  Consequently, RBCs differ  from one another in practically
every component used in their assembly, including shafts,  plastic media con-
figurations, methods for separating the individual  discs,  methods of supporting
the plastic media, bearings, and drives.   Competition among  the manufacturers
encourages technical innovations that they expect to translate  into marketing
advantages.  Because of this competition,  most  of the manufacturers have pro-
gressed through several generations of design,  production, and  testing of  the
individual components that make up their  finished products.

SHAFTS

     RBC shafts are used to support and rotate  the plastic media.  Maximum
shaft length is presently limited to approximately 27 ft,  with  25 ft occupied
by media.  Shorter shaft lengths are also available.  The  shafts are fabricated
from steel and are covered with a protective coating suitable for water and
high humidity service.  Proper protective coating procedure  requires sand
blasting of the steel prior to the coating application.  A coal tar epoxy  is
normally used as the protective coating with a  minimum film  thickness of 14
mils (0.014 in.).

     Each manufacturer designs its own shape, size, and thickness of shaft.
The wall thickness of the shaft is governed by  structural  requirements, and
the shape is highly dependent on the method the manufacturer employs in sup-
porting the plastic media from the shaft.   The  five manufacturers each utilize
a shaft that differs from the others in either  thickness,  size, or shape,  or
in some cases all three.  Structurally, these differences  are readily apparent
as shown in Figure 2 and identified in Table 1.  Lyco currently manufactures
Series 300 circular shafts.  The previous Lyco/Hormel Series 200 octagonal
shaft is also included in Table 1 because of the large number of installations
still using it.

MEDIA

     The heart of the RBC process is the plastic media.   In  1972, the high
density polyethylene  (HOPE) disc was introduced as a cost reduction alterna-
tive to the previously used 0.5-in. thick polystyrene disc.   The major

-------
in
                  CLOW
WALKER PROCESS
                                                 16-m.  square
                                                      (outside
                                                     dimension)
                                                  Weld'
                                                  Locations
                                                                    in
                  LYCO
                                              ENVIREX/AUTOTROL
                                          l-igure 2. Cross-sections of RBC shafts.
                                                                                       CRANE-COCHRANE
                                    30 in.
                                                                                                 outer face
                                                                                                    to
                                                                                                 outer face
                                         ENVIREX

-------
                       TABLE 1.   SHAFT CHARACTERISTICS
Manufacturer
Clow
Crane-Cochrane
Envirex
Envirex/Autotrol
Lyco
Lyco/Hormel
Walker Process
Shape
Round
Round
Octagonal
Square
Round
Octagonal
Round
Size
(in.)
30
30
30
16 x 16
28
24
30
Thickness
(in.)
0.625
0.75
0.625
1.00
0.75
0.75
0.75
Section Modulus
(in.3)
415
492
434
282
426
344
492
Ref.
9
10
11
7
12
12
13
advantage of polyethylene is its ability to be formed into various configura-
tions that require a thickness of only 40 to 60 mils  (0.04 to 0.06 in.).
This innovation enabled 100,000 to 180,000 sq ft of surface area to be provided
on a 27-ft shaft with 12-ft diameter media.  Today, all  U.S.  manufacturers of
RBCs utilize polyethylene as their plastic media.

     Various media configurations or corrugation patterns  have been selected
by the manufacturers, each with its own claimed advantages.  Several  examples
are shown in Figure 3.  There are several reasons  for using corrugations.
Corrugations add stiffness to the sheets and enable these  sheets to be formed
with diameters as large as 12 ft.  Corrugations increase the available surface
area by 15 to 20 percent.  Corrugations cause the  wastewater to follow a
tortuous path through the media, thus increasing wastewater exposure time  to
the air for greater oxygen transfer in the atmospheric sector of the rotational
cycle.  Finally, corrugations are used as spacers  to keep  the sheets separated.

     Standard density media are normally used in the lead  stages of an RBC
train.  Standard density media are defined as media with a surface area of
100,000 sq ft supported on or from a 27-ft shaft in which  the media diameter
is approximately 12 ft.  By reducing the space required for the repeating
plastic corrugations by 33 percent, the available  surface  area can be effec-
tively increased by 50 percent; this results in shafts with 150,000 sq ft  of
media, commonly called high density media.  Some manufacturers are now also
offering media with densities of 120,000 and 180,000 sq ft per 27-ft long
shaft for increased design flexibility.

     RBC manufacturers employ various methods for  supporting plastic media
from their shafts.  Clow, Crane-Cochrane, and Lyco rely on a coated steel  or
stainless steel radial arm system to support the plastic media.  Envirex

-------
FROM WALKER PROCESS BROCHURE (13)
 COURTESY OF CRANE-COCHRANE
 FROM AUTOTROL
DESIGN MANUAL (7)
                                           COURTESY OF LYCO
                      COURTESY OF CLOW

                 Figure 3. RBC media configurations.

                               7

-------
employs plastic media hubs that fit onto its square and octagonal  shafts with
the plastic media sheets thermally welded to the hubs.   Walker Process attaches
the plastic media to the shaft with an epoxy bonding agent and stainless steel
strips with the media then spirally wound onto the shaft in 35-in.  wide strips.

DRIVE SYSTEMS

     Historically, RBCs have been driven mechanically and the drive assembly
has proven to be a very reliable equipment component.  RBC manufacturers
specify factory-assembled drive packages for all mechanical drive  equipment,
consisting of motors, speed reducers, and drive systems.  Reduction of motor
output speed down to approximately 1.6 rpm can be accomplished through the
use of various combinations of multi-V-belts, gear boxes, and chain-and-sprocket
units.  The electric motors presently used for mechanical drive RBCs are
normally high efficiency, 3-phase, 60-hertz units.  The motors, designed with
protective coatings for high humidity environments, are capable of providing
long-term, reliable service.  Energy requirements for mechanical  drive systems
are discussed in Section 7.

     If the designer desires, the manufacturers can modify their standard
drive packages to include variable speed capability.  This provides the opera-
tor with additional flexibility for DU and biofilm thickness control.  Methods
for achieving variable speed capability include positive infinitely variable
(P.I.V.) speed changers and variable frequency controllers, among others.
P.I.V. speed changers are hand-adjusted units and work on the principle of
changing mechanical gear ratios to obtain desired rotational velocity.
Variable frequency controllers enable a.c. motor speed to be changed directly
by varying the frequency of input current.

     Envirex offers an air driven RBC unit.  The air drive assembly consists
of 4- or 6-in. deep plastic cups welded around the outer perimeter of the
media and an air header placed below the media  (Figure 4).  Air is released
at a pressure of 3 to 4 psig into the attached cups, creating a buoyant force
that causes the shaft to turn.  Approximately 20 to  30 percent of the air is
not captured by the cups and escapes into the radial passages where it flows
upward through the corrugated media.  Air flow requirements are discussed in
Section 7.

BEARINGS

      Some  early RBC designs experienced deflections  of  longer  shafts
causing unequal wearing of the shaft ends and bearings.   The  use of self-
aligning  bearing  units  appears to  have  eliminated  this  condition.   Pro-
tection from corrosion  by the  use  of moisture resistant  bearings and  cover
plates on  the  idle end  of the  shaft  has minimized  another  potential problem.
To permit  easy access  for  lubrication and maintenance,  the  bearings should
be located outside the  media  covers.

LOAD  CELLS

      Hydraulic  load  cells  are  available and in  use for  periodically measuring
total  shaft weight.   The  shaft weighing device  consists  of  a  load  cell  bearing

                                       8

-------
                                           RADIAL PASSAGES-
                                                                   AIR CUPS
            Figure 4. Air drive RBC schematic [from Autotrol design manual (7)].
 installed  on  the  idle end  of  a mechanically driven  shaft.  A hand-operated
 hydraulic  pump  is attached to the  load  cell and used to  lift the bearing off
 its  base while  the  shaft continues to rotate or is  momentarily stopped.  The
 resulting  hydraulic pressure  is read from a gauge and the reading converted
 to shaft weight, which  in  turn can be used to estimate biofilm thickness.
 Such measurements are useful  in determining conditions that may cause exces-
 sive fatigue  stress on  the shaft and increased energy consumption.

     The electronic strain gauge load cell is a recent development that
 enables shaft load to be measured continuously without lifting the idle end
 bearing off its base.   A companion converter unit is available that when
 plugged into this type  of  load cell provides a direct readout of total shaft
weight.  Electronic strain gauge load cells are primarily applicable to new
 installations as it is  difficult and costly to modify existing RBC shaft end
walls to accept them.

-------
COVERS
     RBC systems not housed in buildings  are  normally  protected  from  the  ele-
ments by manufacturer-supplied covers.   These covers are  made  of fiberglass
or other reinforced resin plastics  for  durability  and  lightweight  handling.
They all conform to the general  shape of  RBC  media and permit  sufficient
access to the units for observation and minor repairs.  Most are designed in
sections that can be readily dismantled when  major repairs  or  shaft removal
is required.
RBC ARRANGEMENT
                                                                BOD  and
                                                                three  or  four
                                                                stages may  be
                                                                 For small
                                                                installing
     Staging of RBC media is recommended  to maximize  removal  of
ammonia nitrogen (NH3-N).  In secondary treatment  applications,
stages are generally provided for  each  flow stream.   Additional
added for nitrification or for combined BOD and  NH3-N removals.
installations, four stages can be  provided on  a  single shaft  by
three interstage baffles within the tank  and introducing  tne  flow  parallel  to
the shaft.  Installations requiring two RBC units  may be  placed  in series
with a single baffle in each tank, thus providing  four stages.   Four  or  more
units can be placed in series, with each  unit  becoming a  single  stage.   Various
schemes of staging RBC units are shown  in Figure 5.
                    ONE UNIT,
                  FOUR STAGES
               TWO UNITS IN SERIES,
                TWO STAGES EACH
                                         THREE UNITS IN PARALLEL,
                                            FOUR STAGES EACH
                                               LJ    L J
                      MULTIPLE PARALLEL FLOW STREAMS,
                    FOUR OR MORE UNITS PER FLOW STREAM,
                             SINGLE-STAGE UNITS
                    Figure 5. Various schemes of staging RBC units.
                                      10

-------
                                  SECTION 4

                            EQUIPMENT RELIABILITY
 INTRODUCTION
     An EPA-sponsored study (3) of RBC facilities was conducted during the
period of September 1979 through September 1982.  On-site visits and telephone
interviews were utilized to collect pertinent operation and maintenance (O&M)
and equipment-related data from 16 plants equipped with polyethylene media.
A partially updated summary of shaft and shaft-related failures from these 16
plants is presented in Table 2.  For the 153 units included in this evaluation,
41 broken shafts, seven media attachment failures, and 16 bearing failures
were reported.

SHAFTS

     The most serious equipment problem that can impact an RBC plant is a
shaft failure.  A shaft failure involves a structural break in the horizontal
member itself, the loss of the unit, and damage to a portion of the media.
Repair requires that the damaged unit be removed and that a new shaft be
installed along with salvaged and/or new media.  Depending on the site layout,
this may entail removal of protective media covers or relocation of the
entire shaft assembly outside a process building.

     The vast majority of shaft failures can be attributed to known design
and manufacturing defects that have been addressed in the newer generations
of shafts currently being marketed.  In particular, use of a discontinuous
backing strip during the welding of two 16-in. pieces of channel resulted
in a number of defective shafts being fabricated that subsequently failed.
Additional failures were also experienced from structural overloading of
shafts that were designed with inadequate section modulus.

     The wide variation in shaft design procedures has aroused considerable
concern as to the adequacy of RBC shafts to provide 20 yr of reliable
service.  Assuming satisfactory welding practice, the most likely cause
of shaft failure is fatigue.   Extensive experimental  testing of structural
members with applied cyclic loads has shown that the logarithm of fatigue
life is linearly related to the logarithm of the range of applied stress
between certain limits of the cyclic life.   The stress range at which this
curve becomes flat is known as the fatigue limit and represents the theoreti-
cal  value of the applied stress range for which no appreciable fatigue damage
occurs under atmospheric conditions.  A member subjected to rotating/cyclical
loads with a maximum corresponding stress range less than the fatigue limit
will not fail  (within a certain high level  of confidence).  The standards of

                                      11

-------
        TABLE 2.   SUMMARY OF  SHAFT AND SHAFT-RELATED  FAILURES
No. of
Plant Shafts
Cheboygan, Mich.
Cleves, Ohio3
Edgewater, N.J.
Gladstone, Mich.
Gloucester, N.J.
Hamilton Township, N.J.
Hartford, Mich.
Ionia, Mich.
North Huntington, Pa.
Rhinelander , Wise.
Selden, N.Y.
Thermopolis, Wyo.
Voorhees Township, N.J.
Wappingers Falls, N.Y.
Washington Township, N.J.
Winchester, Ky.
8
6
4
6
4
48
2
12
4
10
12
2
6
2
3
24
RBC
System
Startup
Year
1978
1977
1973
1974
1974
1979
1978
1978
1975
1977
1974
1978
1976
1978
1974
19776
Broken
Shafts
3
0
0
1
1
6
0
12
3
6
0
0
3
0
0
6'
Bearing
Failures
0
0
1
0
2
0
0
0
2
0
8
0
0
1
0
2
Media
Support Corrective
Failures Action Taken
0
0
0
0
0
0
0
0
0
0
0
0
0
4
3
0
Replaced

Replaced
Welded in
Replaced
replaced
Replaced

Replaced
Replaced
replaced
Replaced
Replaced
Replaced
Replaced
Replaced
replaced
Replaced
Replaced
3 shafts

bearing
place
2 bearings ;
4 shafts
1 shaftb

12 shafts
all 4 shafts;
2 bearings
all 10 shaftsc
8 bearings
2 shaftsd
6 shafts
bearing;
radial arms
radial arms
2 bearings
a Air  driven RBC units.
b One  shaft replaced as  of  August 1984.  Replacement of the 5 remaining broken shafts
  is underway with completion projected by November 1984.
c All  10 shafts replaced due to anticipated failure of the remaining  4 units.
d Anticipated failure based on performance of similar units elsewhere.
e Plant  upgraded from 4  to  24 shafts in 1980.
f Replacement of broken  shafts under litigation.
                                           12

-------
 the American Welding Society (AWS) include stress curves for structural
 members in several categories (14).

      Geometry, detailing, and fabrication quality can each have a significant
 effect on the fatigue resistance of a member.  Transverse or longitudinal
 welding, welded attachments, bolting, and weld toe treatment are factors that
 influence fatigue behavior.  The weldment fabrication procedures of the AWS
 Structural Welding Code (14) should be adopted as the absolute minimum require-
 ments.  Fatigue resistance can also be affected by potential shaft corrosion
 due to the expected operating environment.

      The estimated fatigue lives for a number of shaft configurations, media
 densities, and biofilm thicknesses were developed by Bowman and Gaunt (15).
 Current manufacturers' shaft designs used in  conjunction with standard density
 media were projected to be resistant to failure from fatigue at biofilm thick-
 nesses ranging from at least 75 mils (0.075 in.) to 150 mils (0.15 in.) or
 greater with the range reflecting the different shaft configurations  employed
 by the various manufacturers.   Further details are summarized elsewhere (1).

 MEDIA

      Six of the 16 plants  visited in the aforementioned survey  (3)  reported
 problems with the media  component,  including  hub failures,  shifting media,
 media brittleness, and media breakage from unspecified causes.   Failures can
 occur due to degradation of the  polyethylene  media from exposure  to heat and/or
 concentrated organic  solvents,  or to UV degradation  if anti-oxidants  are not
 added to the media formulations.   Failures have also occurred  because of poor
 hub design,  poor  media  welding procedures,  and inadequately  designed  radial
 arm systems.

      The method  of forming  a media  pack  and attaching  that pack to the  shaft
 determines  whether an  RBC assembly  must  be removed from  its  tank  for  field
 replacement  of damaged media.  Present  media  pack  attachment designs  of Clow,
 Crane-Cochrane, and Lyco do  not require  shaft  removal  to effect media replace-
 ment.   Envirex and Walker Process designs  require  shaft removal.

 DRIVE  SYSTEMS

      Some of  the  more commonly used RBC  drive  systems  have experienced opera-
 tional  problems.   The most frequently occurring of these is misalignment of
 drive components,  which  contributes to accelerated component wear and opera-
 tional  failures.   Other  concerns  include the maintenance of proper tension  on
 belts in multi-V-belt systems to avoid accelerated wear of drive belts, and
 decreased drive efficiency and wear pn chain-and-sprocket drives.  Two of the
 16 plants evaluated (3) reported broken drive chains.

 BEARINGS

     Six of the 16 plants evaluated (3) reported bearing failures.  Some of
the bearings in earlier designs were tapered roller bearings that required
 lubrication twice a week primarily to purge contaminants from the bearing
race.   Inadequate lubrication contributed to the failures observed.   Current

                                      13

-------
practice is to use self-aligning bearings  with  oversized  grease cups  to
increase lubrication intervals.

SUMMARY

     RBC systems have progressed through several  generations  of design.  Sys-
tems presently being marketed have been improved  compared to  those surveyed.
Only time will tell  whether present systems  will  prove reliable for the con-
templated project design life (normally 20 yr).   Equipment reliability, con-
sequences of failure, equipment  specifications  and warranties,  and operational
flexibility should be considered in all RBC  designs.   Design  considerations are
discussed further in Section 8.
                                      14

-------
                                   SECTION  5

                                ORGANIC  REMOVAL
 WASTEWATER  CARBONACEOUS CHARACTERISTICS

      The  biodegradable  materials  in  municipal wastewaters  are  exceedingly
 diverse,  both  with  respect  to  number  of  components  and  range of particle
 sizes.   If  wastewater  is  filtered  through  glass  fiber filters, which normally
 allow passage  of  particles  up  to about 0.3  to 2  j.im  depending on the particu-
 lar  filter  used,  that material  passing the  filter is generally defined as
 soluble.  Primary effluent  soluble BODs  typically represents from 40 to 60
 percent of  the 8005 loading to a secondary  RBC system.  Where  fine screens
 are  used  in place of primary clarification,  the  soluble 8005 component may
 comprise  as little  as 30  to 40 percent of the 8005  loading.  Where an indus-
 trial  source comprises  a  significant  portion of  the load on a municipal
 system, the ratio of soluble-to-total BOU5  may vary widely from these typical
 values.

      Soluble BOD  loading  is  a  key  design parameter  since smaller molecules
 can  exert a more  rapid  biological  organic demand than larger particulate
 materials (16).   However, an RBC system must not only remove the soluble
 wastewater  components through  a combination of biological  oxidation and cell
 synthesis,  but  must  also  agglomerate, bio-precipitate, biosorb, and/or metabo-
 lize  a substantial  fraction  of  the incoming particulate material  if a clari-
 fied  final  effluent  of  acceptable  quality is to  be achieved.

 MASS  TRANSFER  OF  OXYGEN AND ORGANICS

      Under  most circumstances, mass transfer is  the dominant factor affecting
 organic removal in an RBC system.  Mass transfer resistances associated with
 both  the  liquid phase and the  biofilm result in  significant concentration
 gradients from  the bulk liquid to  biological reaction sites in the biofilm.
 Oxygen transfer becomes limiting and controls the overall  reaction rate in
 heavily loaded  systems.   The importance of mass transfer can be visualized by
 examining the changes depicted in Figure 6.  Here the relative concentration
 of oxygen and organic substrate are shown at different  media locations for
 one hypothetical  loading condition and RBC speed.  These relative values
will, of course, vary for any particular set of design  conditions.   When  the
media are exposed to the atmosphere,  the liquid film boundary  at  the air
 interface immediately becomes saturated with DO as  shown for Point A in
Figure 6.    This saturation in turn results in an  increase  in the  mass  of
oxygen that  diffuses into the biofilm.  When the  media  are submerged,  oxygen
transfer can occur either into or out of the biofilm depending  on the  bulk
liquid DO levels and the degree of mixing of the  liquid  film with the  bulk

                                      15

-------
LU
o:
LU
I
Q.
Q
D
O
D
CD
                         SUBMERGED
                           SECTOR
KS^-SUBSTRATE
                                                     RBC
                                                    MEDIA
         POINT
            D
            LIQUID
             FILM
 Figure 6. Relative concentrations of oxygen and substrate for one
         hypothetical loading condition and RBC rotational speed as a
         function of media location.
                           16

-------
 liquid.   The  substrate  concentrations  within  the  biofilm  are  also  shown  to
 vary with position  on the  media,  with  the  point of maximum  substrate  penetra-
 tion into the biofilm occurring at  the point  of minimum DO  level.

      In  addition  to the oxygen transferred directly  from  the  atmosphere  to
 the nonsubmerged  biofilm,  oxygen  also  enters  the  bulk  liquid  as the result  of
 turbulence generated by the  rotation of the media and  by  the  return to the
 bulk liquid of wastewater  lifted  into  the  atmospheric  sector  that  flows
 freely back across  the  media.  Where organic  removals  in  the  first stage of
 heavily  loaded systems  have  been  limited by DO availability,  it has been
 shown that increasing the  oxygen  content in the atmospheric phase will in-
 crease the organic  removal rate (17)(18)(19).  Gas transfer from the  air
 directly  into the attached wastewater  and  biofilm represents  the major oxygen
 source for the organisms.  The beneficial  effect  on  DO of compressed  air
 injection in  air  drive  RBCs  is offset  to some degree by their lower atmospheric
 oxygenation rates (compared  to mechanical  drive units) resulting from lower
 rotational  velocities.

      The  total  COD  reduction across an  RBC stage  is  a direct  measure  of
 oxygen transfer capability provided that nitrogen and sulfur  species  are not
 oxidized  or reduced  and influent  and effluent DO  levels are the same.  Deter-
 mination  of total COD reduction requires that measurements be made on unsettled
 samples entering  and exiting a stage so that the reductions represent those
 due  to oxidative  reactions only.  COD  balance studies conducted on a  full-dia-
 meter RBC  unit  (20)  indicated a maximum oxygen transfer rate  of approximately
 1.5  Ib 02/day/1000  sq ft.  This maximum transfer rate was achieved at a
 rotational  speed  of  about 1.6 rpm and  is consistent with calculations of
 oxygen transfer based on observed nitrification rates in full-scale units (see
 Section 6).

      The  Autotrol design manual (7) indicates that zero-order removal  of
 soluble BOD5  at a maximum rate of 2.5  lb/day/1000 sq ft is possible for
 mechanically  driven units and 3 lb/day/1000 sq ft is possible for air driven
 units.  Soluble BOD removal represents both the fraction removed through oxi-
 dative reactions  and the fraction converted to new biomass.    In the zero-order
 removal  range, BOD removal  is controlled by oxygen diffusion  (7).   Consider
 the  data  in Table 3 obtained on 11.8-ft diameter discs treating municipal
 wastewater.   A maximum  soluble BOL>5 removal of about 2.6 lb/day/1000 sq ft
 was  achieved.    It is clear, however, that soluble BODs removal was  not constant,
 nor was  it a  single-valued function of bulk liquid soluble BODs concentration.
 The  periods of lowest removal were partially related to periods  of  greatest
 biofi1m thickness.

     The above observations provide a background understanding of  the  key
 factors  affecting substrate removal  and oxygen transfer in RBC systems.   When
 all of these factors are considered, it is  apparent  there  is no  reason to ex-
 pect substrate removal  from the RBC bulk liquid to follow  any  simple mathe-
matical  model.  The  maximum oxygen transfer and substrate  removal rates  indi-
 cated above are paramount in  controlling RBC process  performance and  contribute
 to the basis of the  empirical design procedures discussed  in the following
pages.
                                      17

-------
                        TABLE 3.  FIRST-STAGE RBC DATA REPORTED  BY  HYNEK  AND  CHOU (21)
00
Run No.
V-A
V-B
VI-A
VI-B
VI-C
VII
VIII
IX
X
Waste-
water
Temp.
53
59
73
64
66
65
63
57
54
Total B005
App 1 i ed
/ Ib/day \ /
yitlUU sq tt/ \
8.85
8.98
5.50
-
7.08
11.7
9.16
22.4
-
Mechanical
Applied
Ib/day \ /
Drive Soluble
Removed
Ib/day \ in
BOD5
Reactor
IOW sq tt/ \IOOO sq tt/ (mg/l)
4.25
2.86
2.52
1.29
2.50
4.20
3.54
6.26
5.54
2.41
1.5
1.44
0.76
1.47
0.94
0.76
2.14
1.78
41
40
31
16
32
45
40
44
53
Total BODs
Applied
/ Ib/day \
\ 1000 sq ft/
10.0
10.2
5.39
-
7.45
12.1
9.34
23.4
-
Air
Applied
/ Ib/day
Drive Soluble
Removed
\ I Ib/day
\ 1000 sq tt/ \IOOO sq tt
4.82
3.25
2.47
1.34
2.63
4.31
3.61
6.53
5.65
2.59
1.93
1.65
0.93
1.42
1.16
1.13
2.53
1.59
6005
)in Reactor
(mg/D
44
34
24
12
36
43
35
41
56

-------
 FIRST-STAGE LOADING LIMIT

      A major constraint in the design of any RBC system is limiting the
 organic loading to the first stage(s) to values  compatible with the oxygen
 transfer capability of the system.   With excessive organic loadings, the
 biofilm thickness increases and a white/grey biomass  is frequently  observed.

      Organic overloading is, therefore,  characterized by the  growth of  an
 excessively thick biological film on the media surface and can  be accompanied
 by or result in one or more of the  following process  performance and opera-
 tional  problems (22):

       o  DO deficit in the bulk liquid.

       o  Hindered oxygen transfer from the  bulk  liquid and air  to interfilm
          layers on the RBC discs.  In severe cases, bridging  of biological
          growth between adjacent  contactor  discs  can  occur, effectively
          clogging the  media and virtually stopping both oxygen  and  substrate
          transport.

        Anaerobic conditions  within the biofilm  leading  to the dominance  of
          undesirable microorganisms.   For example,  sulfide formed deep within
          the biofilm will  diffuse outward and can  lead  to the proliferation of
          nuisance organisms  such  as  Beggiatoa, accompanied by potentially
          reduced  BOD removals.

        Development and predominance of microaerophiles  throughout  the  bio-
          film and reduced  BOD  removal  due to their  lower  metabolic rates.

      9  Creation  of septic  odors from both  the bulk  liquid and  the  biofilm.

        Elimination of  the  ability  to control biofilm  sloughing and thickness
          via  liquid shearing forces.

         Increased energy  requirements.

         Unbalanced structural  loading,  structural overloading,  and  structural
          failure  of the  media  support framework and central shaft.

     A  survey of  23 RBC  installations conducted for EPA related  the presence
of sulfur oxidizing organisms  to  overloading caused by high hydraulic loading
and/or  high  influent BOD concentrations.   The survey results are summarized
in Figure 7  (1).   It can be seen that a  first-stage loading limit in excess
of approximately 6.4 Ib  BODs/day/lOOO sq  ft was associated with  the presence
of sulfur oxidizing organisms.

     Based on these findings, first-stage loadings in  the range  of  6 to  8 Ib
total BOD5/day/1000 sq ft or 2.5 to 4 Ib  soluble  BOD5/day/1000 sq ft are
considered to be acceptable  (2).  Loadings in the higher end of  these ranges
will increase the likelihood of developing problems such as heavier-than-normal
biofilm thickness and nuisance organism growth,  depletion of DO, and deteriora-
tion of overall  process performance.   The structural capacity  of the shaft,

                                      19

-------
500





400

^
1
z
o
(-
(E 300
z
LU
O
z
o
o
D
O
CD
z 200
LU
D
Lt.
Z

o
m
cc

100




0
l_l O.
LEGEND:
O Plants reporting absence
of nuisance organism
growths
n Plants reporting
problems with nuisance
organism growths
^ 	 Represents an organic
loading breakpoint of 6.4
I Ib BOD5/day/1000 sq ft
1 34 n
\ D33
\NOTE: Numbers correspond to
plant identifications
from survey.
\

\
\ 036
\ 35
10r\ D3
6O . 31
\ o7
D 1
30 \ 11
140  D D15 29
^^. a 12
016 *>>^>^..
2004 '- ^ 	
	 ! 	 | 	  i i 	 1 	 1 	 1 1 	 1
2 4 6 8 10 12 14 16 18 20
                 FIRST STAGE HYDRAULIC LOADING (gpd/sq ft
Figure 7. DO limiting conditions related to influent organic concentration
        and hydraulic loading [from Brenner et al. (1)].
                                  20

-------
 provisions  for  stripping  biofilm,  consistently  low  influent  levels  of  sulfur
 compounds to the  RBC  units,  the  media  surface area  required  in  the  remaining
 stages,  and the ability to  vary  the  operating mode  of  the  facility  may justify
 choosing a  loading  in the high end of  the  range,  but close attention must  be
 given  to process  operations  to minimize  the  possibility  of structural  over-
 loading  and/or  operational  problems.

 DESIGN APPROACHES

     Procedures for predicting organic removal  relationships  in the RBC
 process  continue  to evolve  as additional information becomes  available.
 Relationships that were recommended  a  few years ago have subsequently  been
 modified or replaced  with alternative  methodologies.   Reported  experimental
 and  pilot and field data  exhibit sufficient  variation  such that support for
 conflicting design approaches can  usually  be found.

 Deterministic RBC Models

     Mathematical models  that take a completely deterministic approach by
 attempting  to incorporate all of the factors affecting RBC performance
 provide  considerable  insight into  those  variables and  ranges  of variables
 that impact the process.  Employing  this approach in design,  however,
 entails  making  certain assumptions about the wastewater film  thickness in
 the atmospheric portion of the cycle,  mixing of this wastewater film with
 the bulk liquid,  the  effect of RBC surface shape on mixing and surface
 biofilm  depth and uniformity, biofilm  density, diffusion coefficient(s)
 within the  biofilm and possible variation with depth and/or the type of
 organisms that  predominate, and biochemical kinetic parameters.  A mathe-
 matical modeling  approach has been applied in some design situations in
 conjunction with  pilot plant studies carried out on the wastewater  in question.
 Calibration of  the model  coefficients  can be incorporated in the pilot plant
 program and is  a  necessary step in the use of complex mathematical models for
 design purposes.

 Pilot Plant Studies

     The best source  of RBC design information is a comprehensive on-site
 pilot plant evaluation.  A common design practice in the past has been to
 scale up small pilot  plant results to  full-scale applications by setting
 the media perpheral  speed at the same value for  both size units, i.e.,
 normally about 60 fpm for  mechanically driven units.  More recent studies
 have shown  that  less-than-full-diameter units can exhibit greater removal
 capacities  per unit surface area (23)(24).   The  use of full-scale diameter
media is recommended  in any pilot study to avoid scale-up problems.   Insuffi-
cient information  is presently available to accurately predict appropriate
scale-up factors.    If small-diameter units must be operated  to collect
design  data, it  is important that each stage be  loaded below  the oxygen
transfer  capability of a full-diameter unit to minimize scale-up problems.

Manufacturers' Design  Procedures

     RBC  manufacturers have  adopted guidelines related  to various  design

                                      21

-------
parameters based on field experience.   For example,  most manufacturers have
standardized tank volume-to-media surface area at 0.12 gal/sq ft based largely
on studies (6) that indicated higher ratios did not  improve BOD removal.   The
manufacturers have also found that loading variations  do not adversely impact
process performance at peak-to-average flow ratios of  2.5 or less.   Two design
manuals (7)(9) recommend utilizing peak flow and load  conditions for design
at ratios of 2.5 or less and flow equalization for ratios above 2.5.  All
manufacturers contend that wastewater  temperature does not affect organic
removal above wastewater temperatures  of 55F.  Below  this temperature,
various correction factors are recommended as illustrated in Figure 8.

     The number of stages utilized in  an RBC design  depends on several factors
including effluent requirements, first-stage and overall process loading cri-
teria, wastewater composition, and wastewater temperature, among others.
General staging recommendations developed by three manufacturers are summarized
below.
           Envirex (7)
      Target
     Effluent
   Soluble BOD5
      (mg/D
       >25
       15-25
       10-15
Recommended
  Mi nimum
  No.  of
  Stages
    1
  1 or 2
  2 or 3
  3 or 4
                Clow (9)
At least
four stages
per flow path
recommended
                       Lyco (12)
  Target
Total  BOD5
Reduction
                  up to 40%
                  35 to 65%
                  60 to 85%
                  80 to 95%
Recommended
   No.  of
   Stages
                    1
                    2
                    3
                    4
                                                  Minimum of four stages
                                                  recommended for combined
                                                       and NH-N removal
     The depth of design information provided on various aspects of RBC
organic removal differs considerably among the various manufacturers.  To
provide perspective on the range of predicted performance that results from
using these methods, predicted effluent quality was examined for a range of
loadings for the case where both influent and effluent soluble-to-total BODs
ratios were assumed to be 0.5.  The results are presented in Figure 9.  The
three Lyco predictions are based on total influent BODs concentrations of
100, 150, and  225 mg/1.  Both Envirex and Clow predict identical relationships
for the loading ranges examined.  Published design procedures for Crane-
Cochrane and Walker Process are not available.  In lieu of establishing
published design procedures, both firms have indicated they prefer to evaluate
wastewater characteristics and process requirements of potential clients on an
individual basis before recommending specific loading criteria and projecting
anticipated performance.
                                       22

-------
                           SURFACE AREA CORRECTION FACTOR
  31
  

3?

33 5T
O O
(D CD
213
o -
  8
  CD

  O

  D
  01

  CD

  3
  o

  0>
         ro
         tn
      CO
      H
      m
~ 0)  >*
IV) ~  H
-H! ^  m
 CD  33

  o  H
  o  m

  CD  ?
  o  T>
  ~.  m
  O  3)
         tn
      C

  s   S
  CO   ^^
  o  3  g
                                       ro

                                       T"
                                                      co
                                                                                  en
                                                                            o
                                                                            3
                                                                            o
                                                                       Jr?

                                                                       81
                                                                          X
                                                                          0}
                                                                          Q.
                                                                          O
                                                                          O

-------
               40 r
ro
o)  30


O
O
CO
_i
<
r5  20
           UJ
           D
           u.
           LU
               10
                                                            ENVIREX
                                                              AND
                                                              CLOW
                                                                              LYCO(150)
                     NOTE: Soluble BOD5/total BOD5 = 0.5
                                            LYCO(100)
                                                                                 ,LYCO(225)
                                                                        NOTE: Numbers in (   ) are
                                                                               influent total
                                                                               BOD5 concentrations.
                                           2345

                                       ORGANIC LOADING RATE (Ib total BOD5/day/1000 sq ft)
                                 Figure 9. Effluent BODs as a function of organic loading for selected RBC
                                          manufacturers design techniques.

-------
 Alternative Design Approach

      In an attempt to devise an improved method of organic removal  estimation,
 Opatken (26) evaluated soluble organics interstage data by using the following
 equation for a second order reaction (27):
                                      4kt
                                   _
                      n            2t

 where Cn is the concentration of soluble organics  in  the nth  stage  (rng/1),  k
 is the second-order  reaction  rate constant  (1/mg/hr),  t  is  the  average  hydrau-
 lic residence time in the  nth stage (hr), and  Cn_i  is  the concentration  of
 soluble organics entering  the nth stage  (mg/1).   Interstage and final effluent
 soluble BODs values  predicted using this expression were compared against
 measured values obtained at nine full-scale  air  drive  plants.   A k  value of
 0.083 was used for all  installations.  The predicted  and measured values were
 in good agreement  for seven of the nine  plants as shown  by  tne  representative
 results in  Figure  10.   The lack  of close agreement at  one plant  could be
 explained by inadequate oxygen transfer  in the first  stage  to handle the nigh
 influent organic load;  the lack  of close agreement at  the other  plant could
 not be explained.

      The second-order  predicted  values in general more closely  approximated
 measured soluble BODs concentrations  than did values predicted  by Envirex's
 empirical organic  removal  design  method  for  air  driven RBCs.  These results
 indicate that  second-order kinetics may  offer an improved basis  for predicting
 interstage  soluble organic removals  in RBC systems tnat  are not  oxygen transfer
 limited.  As with  the manufacturers' empirical techniques,  the second-order kin-
 etics  approach  only addresses  the  impact of  soluble organics on  effluent quality.

 SECONDARY CLARIFICATION

      The  concentration  of unsettled suspended solids (SS) leaving the last
 stage  of an RBC  train treating municipal  wastewater will  normally be less than
 300 mg/1  if primary clarification  is not  provided and less than  200 mg/1 where
primaries are used.  A number   of investigators have undertaken studies  of sur-
 face overflow rates of secondary clarifiers  used in conjunction with RBCs to
achieve certain  desired effluent qualities.   The following is  a  summary  of the
findings of some of those studies:


Surface Overflow          Performance
Rate (gpd/sg ft)           Objective                    Reference

   550-650          <30 mg/1 SS                  Scheible and  Novak  (20)
     740            >80% SS removal              Srini vasaraghavan  et al. (28)
    <600            Maximize solids removal       Murphy and  Wilson  (24)
     800            30 mg/1 SS (t.f. upgrade)    Smith et al.  (29)
     500            15 mg/1 SS (t.f. upgrade)    Smith et al.   29
     800            30/30 effluent               Clow  (9)
    <500            10 mg/1 SS                   Clow  (9)
  1000-1200         Peak hydraulic rate           DeCarlo  (30)

                                     25

-------
a>


in
O
o
CD
UJ

m
O
CO
DODGEVILLE, WISC.
                  2            4
                    TIME (hr)
                                                   80
                               60
                             D)
                             E.

                             in
                             O
                             O
                             m 40
                             LLJ
                             _l
                             CD
                             D
                             _l
                             O
                             CO

                               20
                 WOODBURN, WASH.
LANCASTER, WISC.
                                                                2          4
                                                                 TIME (hr)
                                                             100
                                                                    WEST DUNDEE, ILL.
                      TIME (hr)
 LEGEND:
 O	Measured values
 A	Predicted by 2nd-order
       kinetics
                                                                          TIME (hr)
                     Figure 10. Disappearance of soluble BOD5 with time at
                               four representative RBC plants.
                                             26

-------
                                 SECTION 6

                               NITRIFICATION
 INTRODUCTION
      The factors  affecting nitrification  in  the  RBC  process  include  influent
 organic concentration,  influent  nitrogen  concentration  and  composition,
 wastewater  temperature,  DO concentration,  pH and alkalinity,  and  influent
 flow and load  variability, among others.   Because of the varying degrees  of
 hydrolysis  that can  occur  in  sewer  lines,  RBC nitrification  designs  should
 not  be  based on influent ammonia nitrogen  concentration alone.  To do  so
 risks serious  undersizing  if  substantial  amounts of  organic  nitrogen are
 present.  One  manufacturer (31)  recommends basing design on  influent soluble
 TKN  (mg/1)  minus  0.10 x  soluble  BOD5  removed (mg/1)  to account for heterotro-
 phic nitrogen  metabolism.   A  less empirical  approach is to base design on
 influent  total TKN (mg/1)  minus  1.0 mg/1 refractory  soluble  organic nitrogen
 minus 0.055 x  soluble BODs removed, which  assumes a yield factor  of 0.6 and a
 cell  nitrogen  content of 9.2  percent  to estimate heterotrophic nitrogen metab-
 olism (1).

      The  impact of flow and mass  loading variations  is usually more severe on
 the  nitrification efficiency  of  RBCs  than on  their organic removal performance.
 Nitrifiers  have long generation  times compared to heterotrophs.   Further,
 external  accumulation and/or  internal storage of ammonia nitrogen  for delayed
 metabolism  does not take place with nitrifiers.  Consequently, influent
 surges  in flow or unoxidized  nitrogen concentrations will  be accompanied by
 similar,  delayed  (roughly  equal to reactor detention time) spikes of unoxidized
 nitrogen  in the effluent unless adequate RBC  surface is provided to compensate
 for  expected variations.   Where consistently  low effluent  ammonia nitrogen
 residuals are required, flow  equalization may be cost effective.

 INFLUENCE OF ORGANICS

     The normally  staged configuration of an  RBC system, which promotes
 sequential or  plug flow removal of substrate, is conducive to the development
 of a nitrifying bacterial population.   The degree of  this  development in  any
 stage depends  primarily on  the soluble organic concentration in the stage
 bulk liquid.  Population dynamics dictate that heterotrophic bacteria will
predominate  in  an  RBC biofilm when the organic concentration is high.   As the
organic  concentration decreases to a level  where the  growtn  rate  of the
nitrifiers is  greater than  the rate  of active biofilm sloughing,  the  percentage
of nitrifying  bacteria  in the biofilm  will  increase  to  a point where  efficient
nitrification  is possible.   Selective  predation of nitrifying bacteria  by
higher life  forms  has been  observed  in tail-end RBC  stages where  high DO

                                      27

-------
concentrations  (6 to 8 mg/l) and low soluble 8005  levels  (<5 mg/l) can  coexist.
Sullivan  (31) recommends limiting tail-end DO's to 3.5 mg/l through reduced
media rotational velocity, if possible, to discourage predation of nitrifiers.

     Most empirical design procedures  (7)(9)(12) are based on the assumption
that significant nitrification doesn't begin in an RBC system until bulk
liquid soluble  BODs has been reduced to 15 mg/l.   In combined carbon oxidation-
nitrification units, BODs values of this magnitude may first be encountered
in the second,  third, or fourth stages, depending on influent strength, organic
loading rate, and wastewater temperature.  In separate-stage nitrification
applications, the soluble BODs concentration of the wastewater entering the
RBC reactor is  usually well below 15 mg/l and substantial nitrification is
evident in the  first stage.  Analysis of field data indicates that while
incipient nitrification is generally observed in RBC stages with soluble BODs
concentrations  of about 15 mg/l, maximum nitrification rates are not attained
until bulk liquid soluble BODs drops to 10 mg/l or less  (1).

NITRIFICATION RATES

     Extensive  testing and data evaluation were conducted by Autotrol to
model ammonia nitrogen oxidation in RBCs.  The current Envirex procedure for
full-diameter RBC units operating at wastewater temperatures of 55F or greater
is based on prior Autotrol  testing and is summarized graphically in Figure 11.
This curve projects first-order removal (oxidation) of ammonia nitrogen at
concentrations  in the stage liquid below about 5 mg/l.  Above 5 mg/l NH3-N,
removal is claimed to proceed at a zero-order rate of approximately 0.3 Ib
NH3-N/day/ 1000 sq ft.  It is interesting to note that a nitrification rate
of 0.3 Ib NH3-N/day/1000 sq ft represents an oxygen demand of about 1.4 Ib/
day/1000 sq ft, which corresponds closely with the estimates made by Scheible
and Novak (20)  (see Section 5).  Hence, the nitrification rate of 0.3 Ib
NH3-N/day/1000  sq ft in full-scale systems is apparently the result of oxygen
transfer limitations.
 LU
 (-
 <
 tr
 if
 tu >>
 cc w
  m~

 z
0.4


0.3


0.2


0.1


  0
                                    WASTEWATER TEMPERATURE = 55F
               1.5     3.0     4.5    6.0     7.5    9.0    10.5    12.0

                          STAGE NH3-N CONCENTRATION (mg/l)
                                                            13.5    15.0
         Figure 11. Second-generation Envirex ammonia nitrogen removal rate curve
                 for full-scale RBCs [from Antonie (6)].
                                     28

-------
      Designs for nitrification using RBCs are often based on pilot-scale
 testing.  Although some pilot studies have reported nitrification rates as
 high as 0.7 Ib NH3-N/day/1000 sq ft (32)(33)(34),  it should be noted that
 rates this high are not attainable with full-diameter units rotating at 1.6
 rpm.  As with organic removals, higher inherent rates of atmospheric surface
 renewal and oxygen transfer associated with increased rotational  velocity
 explain the observed differences between small-diameter and full-diameter
 units rotated at the same tip speeds.

      Nitrification data obtained at three full-scale facilities when waste-
 water temperatures were 55 +_ 2F are shown in Figure 12.   Each point repre-
 sents data for 1 day for a given stage.   For  the combined carbon  oxidation-
 nitrification systems (Gladstone, Michigan, and Cleves, Ohio), ammonia  nitrogen
 removal data prior to the stage in the train  where  maximum nitrification  rates
 were observed were omitted on the assumption  that organic removals were influ-
 encing the nitrification rates.  The curve shown in this  figure essentially
 duplicates the curve used in  the current Envirex procedure.   The  zero-order
 removal rate above bulk liquid ammonia nitrogen concentrations of 5  mg/1
 corresponds to 0.3 Ib NH3-N/day/1000 sq  ft.

      If all  other  factors are not rate limiting, e.g.,  pH,  DO, and/or oxygen
 transfer,  an increase in temperature above 55F would  be  expected to increase
 the  NH3-N  oxidation  rate observed in RBC systems.   Field  data  from five nitri-
 fying plants operating  at higher  wastewater temperatures  (65  + 5F)  are plotted
 in Figure  13.   While the data are more scattered than  in  the "5*5 _+ 2F plot,
 the  average removal  rate at concentrations  above 5  mg/1 NH3-N  is  actually  no
 higher  than the 0.3  Ib  NH3-N/day/1000  sq  ft rate estimated  at  55F.   Compari-
 son  of  Figures  12  and 13 supports  the  manufacturers' contention that wastewater
 temperatures above 55F do  not  result  in  higher nitrification  rates  on  full-
 scale units.   The  probable  explanation for  this phenomenon  is  that above  55F
 oxygen  transfer  capacity  rather  than biological growth  rate controls the maxi-
 imum nitrification rates  attainable  in full-scale systems.

      Low wastewater  temperature  (47  +  2F) nitrification data were available
 for  Gladstone  (35).   Evaluation  of tFese  data indicated that an estimate of
 the  amount  of additional  media  surface area required to achieve an effluent
 ammonia nitrogen level  of 1.5 mg/1 at 47F compared to  55F was identical  to
 or within  10 percent  of  that  predicted by the temperature correction curves of
 Envirex, Clow,  and Lyco  illustrated  in Figure 14.    These data indicate that
 as wastewater temperature drops increasingly below  55F, biokinetic response
 displaces oxygen transfer capacity as the dominant   factor controlling full-
 scale nitrification rates.

      Nitrification rates can  be significantly  affected by pH, particularly
 in unacclimated systems.  One study  (32) where pilot RBC units were  allowed
to acclimate to each pH level  tested reported  a constant nitrification rate
 between pH 7.1 and pH 8.6, 25 percent of this  constant rate at pH  6.5, and
zero nitrification at pH 6.0.   On another RBC  pilot  study (34), increasing
nitrification rates were observed as pH was increased from 6.5 to  8.5  Nitri-
fication enhancement with increasing pH values above the typical pH  range  for
raw wastewater of 7.0 to 7.5 has never  been demonstrated,  however, on full-
scale RBCs.  As with wastewater temperature, oxygen  transfer limitations may

                                     29

-------
                  CT
                  if)
                       0.8 r
                       0.6
                                    LEGEND:
                                       O Gladstone, Mich. (35)
                                       D Guelph, Ontario (36)
                                       A Cleves, Ohio (37)
CJ
O
                  03
                  o
.Q


LU
                  IJJ
                  DC
                   CO
                  I
                       0.4
                       0.2
                                                         O
                                           o
                                                    o
                                                             o
                                                                    o
                                               O
                                                                  _L
                                                                              O    D
                                                        u
                                                                                     -o
                                                                                                         D
O
                                                                                      D
                                                                             NOTE: Temperature - 552F
                                                                                J_
                                                     4             6            8             10

                                                    STAGE NH3-N CONCENTRATION (mg/l)
                                                                                         12
                                 Figure 12. Full-scale RBC nitrification rates at design wastewater
                                          temperature (55 F).

-------
U.8
'day/1 000 sq
o
b>
-.

UJ
$ 0.4
tr
0
 0.2
z
i5
0
LEGEND:

O Gladstone, Mich. (35)
D Lancaster, Wise. (37)
A Lower East Fork, Ohio (37)
 Columbus, Ind (37)
 Indianapolis, Ind. (38)

.
O 
**
-  o  o o0
0 A Dn
<%

 0 
*  0 
A 0 o
0  o
0 o  0
o o
0 0 0
0 NOTE: Temperature = 655F
 .
                                              10
12
14
16
18
                      STAGE NH3-N CONCENTRATION (mg/l)
Figure 13. Full-scale RBC nitrification rates at high wastewater temperatures.

-------
                         SURFACE AREA CORRECTION FACTOR
                                                 co
        ro
        tn
  Tl
  10'

CD C
O -i

^ CD
N) 3

-2-0
- CD
  O
  O
o_

5'


aT

o
^


cT
-r
2

^;
^
o'


6'
  O
  3
        tn
        O
Crt
H
m
m
      m
      m
      33
      C
      33
      m
         tn
         tn
         ro
         tn
         tn
         tn
         o
                                                                ii
                                                                ni
                                                                n
                                                                2
                                                                o
                                                          r-Om
                                                          o o <
                                                          o g =

-------
obscure pH optimization of nitrification in 12-ft diameter field units rotated
at 1.6 rpm.

     The process of nitrification consumes alkalinity.   In poorly buffered
RBC nitrifying systems, alkaline chemicals such as lime,  soda ash,  or sodium
hydroxide may have to be added to the wastewater to maintain sufficient
alkalinity to prevent a precipitous drop in pH and nitrification rate.
                                    33

-------
                                SECTION 7

                           ENERGY REQUIREMENTS

INTRODUCTION

     Energy consumption in mechanically driven systems  results from the power
used to overcome internal  resistances and  losses  in  the motor, friction
losses in gear reduction and drive belts or chains,  friction losses in shaft
support bearings, and the drag forces resulting from rotation of the media
and attached biofilm through the wastewater.   The drag  force, which constitutes
the major portion of the energy required for  shaft rotation, is affected by
the amount of media surface area, the shape of that  surface area, rotational
speed, wastewater viscosity (temperature), and the type and amount of biologi-
cal growth.

     With air driven systems, compressed air  is discharged beneath the RBC
media as previously shown in Figure 4.  The rising air  is captured by the air
cups, and the resulting buoyant forces provide the torque necessary for media
rotation.  Energy is required in air driven systems for losses in the motor,
compressor, air headers, control valves, and diffusers, and also to overcome
the static head of wastewater in the RBC tank.

MECHANICAL DRIVE SYSTEMS

     Manufacturers' estimates of energy requirements for mechanical drive
units rotating at approximately  1.6 rpm are as follows (39):

                                 	Media Type	
                                 Standard Density     High Density
     Manufacturer                    (kW/shaft)         (kW/shaft)

     Clow                             2.1                2.4
     Envirex                          1.55(1.8)a         2.3(2.6)*
     Lyco                              -       .          1.7  -  2.1
     Walker Process                   2.0(2.6)b          2.3(3.0)&
      a  First  number  is  mean  of  field measured values; second is 99.9
        percent  confidence  level.
      b  First  number  is  average  value;  second is maximum value.


      Environmental  Resources Management,  Inc.  (ERM),  conducted an  extensive
 EPA sponsored field  investigation  to determine actual energy requirements  on  a
 total of 105  mechanically  driven  units (4).  A variety of  shaft lengths, media
                                                            i
                                       34

-------
  surface areas, and manufacturers' equipment were included in the study  A
  summary of all field measured data is given below:

           Motor Size              	kU/Shaft	
           	  (nP)                AverageRange

               5                    2.02        1.05 - 3.76
              7-5                   2.05        1.32 - 2.99


      Data from 80 of the units tested by ERM were selected as representative of
 facilities with typical  shaft lengths,  media diameters,  and  rotational  speeds
  These data were combined with data from 12 units  evaluated by EPA staff (40)
 to ascertain  the  impact  of media surface area  on  energy  requirements.   Of
 these 92 units,  a  total  of 55 were equipped with  media having a  surface area
 in the range  of 100,000  to 128,250 sq ft,  a range generally  regarded in the
 industry as representing standard density  media.   The  media  surface  area of
 the other  37  units  varied from 138,000  to  180,000 sq  ft,  a range characteristic
 or high  density media.   The average measured energy consumption  for  the
 standard density units was 2.09  kW/shaft with  a standard  deviation of 0 46
 kW/shaft.   For  the  high  density  units,  the average power  requirement recorded
 was 2.40 kW/shaft with a standard deviation  of 0.59 kW/shaft.  These values
 agree  well with the manufacturers' estimates of mechanical drive energy re-
 quirements.

     Excessive  energy consumption  is generally  indicated when  field-measured
 power  levels exceed by one  to  two standard deviations  the means  indicated
 above  for  standard and high density media.   Higher-than-expected energy con-
 sumption can be caused by  equipment problems, heavier-than-normal biofilm
 growth, or both.  Potential equipment problem areas include improper belt align-
ment, inadequate lubrication, excessive rotational speed, excessive  belt ten-
sion or belt slippage, and  general wear and deterioration of the drive com-
ponents.

     Power factor  is an important parameter in  determining the energy cost
associated with operating an induction motor.  Many electric  utilities  have
demand charge  schedules that penalize  customers with  power factors less  than
u.9.  The power factors for mechanically driven RBC units  are typically  auite
low as indicated by  the following data from the 92 units  summarized  above-


        Surface Area  Range    Motor  Size           Power Factor
       	(sq  ft)	       (hp)           Mean        Range    ~

        100,000 - 128,250         5           0.51(28)3  0.26  - o.7l
        100,000 - 128,250        7.5          0.38  27 a  0.18  - 0.82
        138,000 - 180,000         5           0.49  20 a  0.26  - 0  68
        138,000  - 180,000        7.5          0.38  17 a  0  16  - 0  63
       a Number of shafts.
                                     35

-------
     The Upper Mill  Creek (Cincinnati,  Ohio)  treatment  plant  utilizes  mechani-
cally driven RBC units equipped with power  factor  correction  capacitors.
Measurements made on seven units with power factor correction indicated that
the capacitors were increasing power factor from about  0.5 to 0.99 (1).  The
resulting savings in apparent power are worth $17.30/month/shaft in lower
electrical costs at an assumed demand charge of $6.92/month/kVA.  Based on
an approximate installed capacitor cost of  $200, the payback  period in this
case would be about 12 mo.

AIR DRIVE SYSTEMS

     The energy consumption for an air drive RBC system can only be accurately
estimated once the motor and blower characteristics, line losses, and operating
range have been addressed.  The sizing and  design of the air  distribution
system will directly affect the line losses and, therefore, the discharge
pressure at which the compressor must operate.  The usual industrial practice
is to design for headlosses in the range of 0.1 to 0.6 psi/100 ft of line.
It is important that the design conditions  stipulated for the centrifugal
compressors match actual operating conditions as closely as possible.  Com-
pressor operation at conditions other than  those for which the compressor was
designed can result in a substantial loss of efficiency.  For those facilities
in which significant variations in air flow requirements are anticipated,
consideration should be  given to using multiple compressors with different
design capacities,  both  for air flow and discharge pressure.

     The air flow requirements to  operate an air-driven RBC system are variable.
Envirex  (31) recommends  that  350 cfm of capacity be provided for each  shaft.
The  maximum normalized air  flow at ambient conditions  (68F) with 4-in.  air
cups for  various rotational  velocities is presently indicated by the manufac-
turer to  be as  follows  (41):

                                       Air  Flow  (acfma) at  Indicated
                                           Rotational  Speed	
          Rotational  Speed  (rpm)          1.0         1.2         1.4

          Standard  Density  Media          100         150         220

          High  Density  Media              145         225         350
          a acfm = actual  cfm


 The actual air flow requirements  to operate a  shaft  at  a  given  speed are also
 reported to vary by 40 to 80 acfm,  and the mean  requirement  over  a range of
 1.0 to 1.3 rpm will vary more than  this (42).

      Results of an EPA study (40) that evaluated the relationship between
 rotational speed and air flow requirements for one process train  at the Lower
 East Fork (Batavia, Ohio) plant are presented  in Figure 15.   Also shown are
 the manufacturer's design curves  (41) for the  range of  conditions applicable

                                       36

-------
 500
 400
E
t3
 300
 200
 100
                  LEGEND:
                     Autotrol curve for high
                         density media, heavy
                         growth, 6-in. cups, and
                         50 F
                   Autotrol curve for high
                         density media, normal
                         growth, 4-m. cups
                         53.6
        "Letter designations from Table 4
       	1	i
                 0.5               0.75               1.0

                          ROTATIONAL SPEED (rpm)
                                                           1    '   '   _'
                                                                        1.25
Figure 15.Summary of air flow versus rotational speed measurements made
         at Lower East Fork-Little Miami River Regional Wastewater
         Treatment Facility, Claremont County, Ohio.
                                 37

-------
to the measurements made at this site.   The operating conditions under which
the data represented by curves A to F were generated are summarized in Table
4.  It can be seen that air flow requirements for a given rotational  speed are
highly variable and unpredictable (compare curves B and C).   The specific
energy requirements for an air driven RBC facility can only  be determined on a
case-by-case basis.


       TABLE 4.  SELECTED OPERATING CONDITIONS OF LOWER EAST FORK PLANT


Curve Designation   Air Cup Depth         Date Measured  Biofilm Thickness
    in Figure	(in.)     Stage     (1982)	(In.)

      A                   6          1        2/11             0.050
      B                   6          1        3/15             0.044
      C                   6          1        3/17             0.042
      D                   43      3/15,3/17       0.038,0.037
      E                   45        2/23             0.045
      F                   49        2/23
                                      38

-------
                                  SECTION 8

                           PLANT DESIGN CONSIDERATIONS
 INTRODUCTION
      Proper design of an RBC facility entails,  among other  things,  accurate
 determination  of influent and sidestream loadings,  proper media  sizing,
 staging and equipment selection  to meet  effluent  requirements, and  selection
 of an overall  plant layout that  provides sufficient flexibility  to  promote
 good operation and maintenance practices.   A number of  design considerations
 (in addition to those discussed  in the preceeding sections)  are  germane  to
 the quality construction and successful  operation of an RBC  installation.

 PRETREATMENT

      Raw municipal  wastewater  should  not be  applied to  an RBC system.  Effec-
 tive preliminary treatment is  essential  for  good  operation.  Grit removal and
 either  primary clarifiers  or  screens  are necessary  to remove materials that
 may settle  in  the  RBC  tankage  or plug  the RBC media.  High influent grease
 loads require  the  use  of primary clarifiers  or an alternate  system  for grease
 removal  prior  to the  RBC units.                                              :

      The impact  of  sidestreams from other unit processes must be considered.
 Anaerobic digesters increase ammonia nitrogen loadings, and sludge conditioning
 processes such  as  heat treatment contribute  increased organic and ammonia
 nitrogen  loadings.  Where  septage  addition is to  be allowed, separate facili-
 ties  for  its acceptance  and controlled feeding to the RBC system should be
 considered.  Septic tank waste addition  or a separate unit process located
 ahead of the RBC system  that has the potential for  sulfide production, e.g.,
 clarigesters,  produces an  incremental  oxygen demand associated with the
 sulfide  that must be considered in system design.

 EQUIPMENT SPECIFICATIONS, MAINTENANCE, AND RELIABILITY

     The suggested ranges for first-stage organic  loading rates  as presented in
 Section  5 are based on both process and structural considerations.   First-stage
 loadings in the upper end of the  range may result  in a variety  of operational
and process problems, of which the most critical  is the structural  overloading
 of the RBC shaft.  Over the life  of a  facility,  organic loadings  will  vary
considerably and excessive biofilm growths can frequently occur.   Therefore,
the specifications for the load bearing capacity  for each shaft  snould consider
the maximum anticipated biofilm growth, the  capacity to strip excess biofilm,
and an adequate margin of safety.   In  addition,  structural designs  of  RBC
shafts should be based on appropriate  AWS stress category curves  (14)(see

                                     39

-------
Section 4) modified as necessary to account for  the expected corrosive environ-
ment.

     Full building enclosures normally provide more convenient access to RBCs
for routine maintenance and visual  observation than the use of individual
fiberglass covers.  Housing RBC units within a building is undesirable, how-
ever, because of the corrosive atmospheric conditions associated with the
presence of hydrogen sulfide and high humidities.   Condensation problems
have been encountered on interior building walls in cold climates,  arid the
accompanying high humidities and ventilation requirements increase  heating
costs.  Where buildings are chosen, the building design must meet the addi-
tional requirements for ventilation, heating, and humidity control, and pro-
vide for removal of a shaft/media assembly should repair or replacement
prove necessary.

     In all RBC designs, access to individual shafts for repair or  possible
removal must be considered.  Bearings should also be accessible for easy
removal and replacement.  The weight of a 27-ft long shaft and media assembly
may  be expected to range from 18,000 to 25,000 Ib for clean media depending
on whether standard or high density media is specified.  A fully-loaded,
100,000-sq ft shaft with a 0.1-in. thick biofilm has a dead weight of about
70,000 Ib.  Some manufacturers assemble their units in such a way that the
media can be removed from the shaft while the shaft remains in the RBC tank,
whereas  others  do not  (see Section 3).  Where all units  in a  large instal-
lation are physically  located closely together, it has been necessary to
utilize  large off-the-road cranes for shaft removal.  Crane reach, crane
size, and the impact of being able to drain RBC tankage  and dry a unit
prior to  shaft  removal should all be considered when designing the RBC
layout.

      Equipment  warranties can be negotiated with the manufacturers,  and,  in
some cases, extended equipment warranties have been obtained.  This  possi-
bility should be  thoroughly  considered  in equipment specifications.   Speci-
fications can set strict warranty requirements.  Careful thought needs to
be  given  to the acceptability of field  repairs versus  new  factory  replace-
ment equipment,  as well as  other areas  of potential conflict.  Considering
the  history of  equipment failures,  the  specification of  a  rigorous performance
and  replacement guarantee may offer  the best  protection  available  to the
treatment plant owner.

FLOW CONTROL

      Whenever multiple process  trains  are  employed,  provision for  positive
and  measureable flow  control  to  individual  trains  is essential.  Use of
single,  long  influent  cnannels  with  slide  gate  control  for  individual  trains
makes it difficult  for the  operator  to locate flow maldistributions  and
 implement appropriate corrective procedures.   Splitter  boxes  and/or  weirs
are low  cost  solutions to  this  problem.  Provision of  adequate flow  control
equipment is  especially  important  if individual  trains are fed from  a single
 channel  with  rotation  of  some trains with  and other  trains against the
 direction of  plant  flow.
                                      40

-------
      Feeding and discharge flexibility should be considered in RBC design.
 Step  feed capability can relieve overloaded first stage(s) and potentially
 decrease or eliminate excessively tnick biofilm growth.  Removable baffles
 may also be effective in this regard.  In underloaded plants, the final
 stages  frequently must be operated to keep suspended solids in suspension
 rather  than to provide additional treatment to meet effluent standards;
 capability to bypass these terminal units would result in energy and O&M
 savings.  If sufficient flow flexibility is available, loss of an individual
 unit  need not result in shutting down an entire process train.  Adjusting the
 distribution of total flow and/or loading may be an operational necessity,
 again emphasizing the need for positive and measurable flow control  and/or
 splitting.

      The use of deep channels leading to and exiting from RBC tanks often
 results in solids deposition and subsequent accumulation.  Providing for
 channel aeration (3.5 scfm/linear ft) or employing channel configurations
 that  promote adequate scouring velocity should overcome this problem.

      Most RBC designs are based on the units operating at a submergence of
 approximately 40 percent based on total media surface area.  To avoid possible
 shaft overstressing and inadequate media wetting, the RBC manufacturers
 strongly recommend against dropping the liquid operating level below 35-percent
 submergence.  They also recommend a clearance of 4 to 9 in. between the tank
 floor and the bottom of the rotating media to maintain sufficient bottom
 velocities to prevent solids deposition in the tank.

 BIOMASS MONITORING AND CONTROL

      Organic loading limits and detrimental  impacts resulting from organic
 overloads were previously presented in Section 5.  Since a higher percentage
 of carbonaceous material  is removed in the initial  stages, the first and
 second stages can be expected to experience the thickest biofilm growth and
 heaviest shaft loads.  Load cells (see Section 3) located at these and other
 strategic locations can provide the operator with advance notice of a gradual
 buildup in biofilm thickness.

      DO monitoring is another operational  tool that can be used advantageously
 for process control.  Falling stage liquor DO levels,  particularly in the
 lead  stages, may forewarn of conditions conducive to the growth of undesirable
 organisms and the development of excessive biofilm thicknesses.  However,
 increasing DO levels in the bulk liquid by itself will  not necessarily overcome
process performance difficulties that may  have developed and other actions to
 strip excess biofilm may  be required.  Two of the plants surveyed by Roy F.
Weston, Inc., had nuisance organisms (Beggiatoa) present in the first stages
despite bulk liquid DO levels of 1.5 to 2  mg/1 (3).   A similar observation
was also made at the RBC  facility in Edgewater,  New Jersey (20).   A  major
advantage of the air drive RBC system is  claimed to be related to the increased
turbulence and stripping  of excess  biofilm induced  as  a portion of the air
bubbles rise through the  media.

     Sufficient operational  flexibility must be  provided to allow effective
response to organic overloads.   Possible corrective action includes  increasing

                                     41

-------
rotational  speed to enchance oxygen  transfer  and  biofilm  shearing  force,
periodically reversing the direction of  RBC rotation  to promote  biofilm
stripping,  using supplemental  aeration to increase  bulk liquid DO  concentra-
tion and induce biofilm stripping,  increasing media surface  area in  the
affected stage by removing baffles  to reduce  overall  loading, staging  feed  of
incoming flow, and temporarily adding chemicals such  as hydrogen peroxide or
chlorine.

MISCELLANEOUS CONSIDERATIONS

     As discussed in Section 5, influent soluble  BOD  is a critical parameter
in the design of RBC units.   In the  design of any facility,  soluble  BOD
levels should be verified by influent sampling whenever possible.

     Depending on whether the media  formulation includes  carbon  black, media
strength can be severely degraded by exposure to  sunlight (ultraviolet degra-
dation).  When RBC units are stored  on-site for an  extended  period of  time
prior to installation, provisions must be made to ensure  that they remain
protected from direct sunlight.  Media can also be  severely  impacted by  high
wastewater  temperatures (>95F); this is a potential  problem in  some industrial
applications or in municipal installations that receive large industrial  flows
or are located in desert-like environments.

     High density media has been used advantageously  in the  middle and latter
stages of RBC trains where decreased availability of  organic carbon  results
in biofilm thicknesses of <50 mils  (0.05 in.). A recommended general  practice,
subject to site specific adjustment, is  not to employ high density media
before the third stage of a sequentially-staged RBC module.

     Air drive installations should be provided with  positive air  flow meter-
ing and control to each RBC unit.  Operating  an air drive facility without
these controls adds to the difficulty of appropriately responding  to opera-
tional problems that may arise.

     Stop motion detectors, rpm indicators,  and clamp-on  ammeters  are  poten-
tially useful monitoring devices for individual RBC shafts.

     Energy requirements for RBC rotation increase  exponentially with  rota-
tional speed.  Previous tests with clean media have shown that  power consump-
tion increases as a cubic function of rotational  velocity (3).   Other  energy-
consuming factors exist independent of  speed, however. Accordingly, the
cubic relationship should not be used indiscriminately.

     Nitrification is slow to develop in cold temperatures,  and  8 to 10  wk  may
be required before a nitrification system approaches  equilibrium conditions.
Where seasonal standards are required for the final design,  the  transition
time and temperatures necessary to develop an adequate nitrifying population
must be considered.

     O&M requirements associated with biofilm control, drive train and radial
support arm maintenance and repair, and  media and shaft repairs  and  replace-
ments should  be considered  in  laying out the RBC  process  configuration and

                                     42

-------
selecting equipment.  The O&M manual  should specify a  schedule for  reading
load cells, visually inspecting biofilm growth and media  integrity,  and
determining the status of mechanical  and structural  components.   The manual
should also outline procedures for dealing with identified  problems.
                                    43

-------
                                 REFERENCES


1.  Brenner, R. C., J. A. Heidman, E. J. Opatken, and A. C.  Petrasek, Jr.
    Design Information on Rotating Biological  Contactors.  EPA-600/2-84-106,
    NTIS PB84-199561, June 1984.

2.  Rotating Biological Contactors (RBCs) - Checklist for a  Trouble-Free
    Facility.  USEPA, Office of Water Program Operations, Washington, D.C.,
    May 1984 (rev.).

3.  Review of RBC Design and Process Operation and Maintenance, Equipment,
    and Power Performance.  Final report prepared for USEPA, Municipal
    Environmental Research Laboratory, Cincinnati, Ohio, under Contract
    Nos. 68-03-2775 and  68-03-3019 by Roy F. Weston, Inc., July 1984,
    (Publication pending).

4.  MacGregor, A. and  K. Raghavan.   Evaluation of the Energy Requirements
    for Rotating Biological Contactors  (RBCs).   Final report prepared for
    USEPA, Office of Water Program Operations, Washington, D.C., under
    Contract No. 68-01-6622 by  Environmental Resources  Management,  Inc.,
    May 1984,  (Publication pending).

5.  Hynek, R. J. and H.  lemura.   Nitrogen and Phosphorus Removal with
    Rotating Contactors.   In:   Proceedings of the 1st National Symposium/
    Workshop on  Rotating Biological  Contactor Technology, Vol. I,  EPA-
    600/9-80-046a,  NTIS  PB81-124539,  June  1980.   pp. 295-314.

6.  Antonie, R.  L.   Fixed  Biological  Surfaces -  Wastewater Treatment:   The
    Rotating Biological  Contactor.   CRC  Press, Cleveland, Ohio,  1976.

7.  Autotrol Wastewater  Treatment Systems  Design Manual.  Autotrol  Corpora-
    tion,  Bio-Systems  Division,  Milwaukee, Wisconsin, 1978.

8.  Dallaire,  G.   U.S.'s Largest Rotating  Biological Contactor Plant to Slash
    Energy Use  30%.   Civil  Engineering,  ASCE, 49(l):70-72,  January 1979.

9.  Clow  Envirodisc Rotating  Biological  Contactor Systems Catalog.  Clow
    Corporation,  Florence, Kentucky, 1980.

10.  Personal  communication from R.  W. Hankes, Crane Company,  Cochrane Environ-
    mental Systems, King of  Prussia, Pennsylvania,  to R. C.  Brenner, USEPA,
    Municipal  Environmental  Research Laboratory, Cincinnati,  Ohio, August 25,
     1982.
                                      44

-------
 11.  Personal  communication from R. A. Sullivan, Envirex, Inc., Waukesha,
      Wisconsin, to R. C. Brenner, USEPA, Municipal  Environmental  Research
      Laboratory, Cincinnati, Ohio, May 1984.

 12.  Lyco Wastewater Products - RBC Systems Catalog.   Lyco Division of
      Remsco Associates,  Marlboro, New Jersey,  1982.

 13.  Walker Process BioSpiral  Rotating Biological  Contactors  Brochure.   Walker
      Process Corporation,  Aurora, Illinois, 1979.

 14.  American  Welding Society.   Structural  Welding  Code-Steel.   5th Edition,
      ANSI/AWS  Dl.1-81,  1981.

 15.  Bowman, M. D.  and J.  T.  Gaunt.   Fatigue Behavior  of  Rotating  Biological
      Contactor  Shafts.   Internal  report  prepared for  USEPA, Municipal  Environ-
      mental  Research Laboratory,  Cincinnati, Ohio,  under  Purchase  Order  No
      C2159NAST, April  1982.

 16.   Balmet, J. L.   Biochemical Oxidation of Various  Particulate Fractions  of
      Sewage.   Sewage Works  Journal,  29:757,  1957.

 17.   Torpey, W.,  H.  Heukelekian,  A.  J. Kaplovsky, and  L.  Epstein.   Effects  of
      Exposing  Slimes  on  Rotating  Discs to Atmospheres  Enriched  with  Oxygen.
      Presented  at the 6th  International  Conference on  Water Pollution Research,
      Jerusalem,  June  18-23,  1972;  Available in:  Advances in Water  Pollution
      Research,  Permagon  Press,  New  York  City,  1973.   pp.  405-415.

 18.   Huang, J.  C.   Operational  Experience of Oxygen Enriched Rotating Biologi-
      cal Contactors.  In:   Proceedings of the  1st National Symposium/Workshop
      on Rotating Biological Contactor Technology, Vol.  I, EPA-600/9-80-046a,
      NTIS PB81-124539, June 1980.  pp. 637-659.

 19.   Binitanjia, H.  H. J., J. J.  Brunsmann, and C.  Boelhouwer.  The  Use of
      Oxygen in  a Rotating Disc Process.  Water Research,  10(6):561-565, 1976.

 20.   Scheible,  0. K. and J. J. Novak.  Upgrading Primary  Tanks with  Rotating
      Biological Contactors.  In:  Proceedings of the 1st  National  Symposium/
      Workshop on Rotating Biological Contactor Technology, Vol.  II,  EPA-
      600/9-80-046b, NTIS PB81-124547, June 1980.  pp.  961-996.

21.   Hynek, R.  J. and C.  C. Chou.   Development  and  Performance of Air-Driven
      Rotating Biological  Contactors.  In:  Proceedings of the  34th  Industrial
     Waste Conference, Purdue University, West  Lafayette,  Indiana,  May 8-10,
      1979;  Ann Arbor Science, Ann Arbor, Michigan,  1980.   pp  805-815.

22.  Evans, F.  L., III.   Consideration of First-Stage  Organic  Overloading in
     Rotating Biological  Contactor Design.   Draft report prepared  by USEPA,
     Municipal  Environmental Research Laboratory, Cincinnati,  Ohio,  (Publi-
     cation pending).
                                      45

-------
23.   Reh,  C.  W.,  T.  E.  Wilson,  and R.  Srinivasaraghavan.   An  Approach to De-
     sign  of  RBC's for  Treatment  of Municipal  Wastewater.   Presented at the
     ASCE  National Environmental  Engineering  Conference,  Nashville,  Tennessee,
     July  1977.

24.   Murphy,  K.  L. and  R.  W.  Wilson.  Pilot  Plant  Studies  of  Rotating Biologi-
     cal  Contactors  Treating  Municipal  Wastewater.   Report SCAT-2,  Environment
     Canada,  Ottawa, Ontario, Prepared for  Canada  Mortgage and Housing Corpora-
     tion, July  1980.

25.   RBC  DIS  Subtask on RBC Design Approaches.  Internal  report prepared for
     USEPA, Municipal  Environmental Research  Laboratory,  Cincinnati, Ohio, under
     Contract No. 68-03-3019 by Roy F. Weston, Inc., March 5, 1982.

26.   Opatken, E.  J.   Rotating Biological  Contactors - Second  Order  Kinetics.
     In:   Proceedings  of the First International  Conference on Fixed-Film
     Biological  Processes, Vol. I, EPA-600/9-82-023a, Kings Island,  Ohio,
     April 20-23, 1982.  pp.  210-232.

27.   Levenspiel,  0.   Chemical Reaction Engineering.  John Wiley and Sons, New
     York  City,  1972.

28.   Srinivasaraghavan, R., C. W. Reh, and S. Liljegren.   Performance Evalua-
     tion  of Air Driven RBC Process for Municipal  Waste Treatment.   In:
     Proceedings of the 1st National Symposium/Workshop on Rotating Biological
     Contactor Technology, Vol. I, EPA-600/9-80-046a, NTIS PB81-124539, June
     1980.  pp.  525-552.

29.   Smith, D. A., C.  P. Poon, and R.  D. Miller.   Upgrading DA Trickling-
     Filter Sewage Treatment Plants.  Technical Report No. N-102, USA Corps
     of Engineers, Construction Engineering Research Laboratory, Champaign,
     Illinois, May 1981.

30.   Personal communication from D. A. DeCarlo, Burgess and Niple, Ltd.,
     Columbus, Ohio, to J. A. Heidman, USEPA, Municipal Environmental Research
     Laboratory, Cincinnati, Ohio, April 14,  1982.

31.   Personal communications from  R. A. Sullivan, Autotrol Corporation, Mil-
     waukee, Wisconsin, to J. A. Heidman and  R. C. Brenner, USEPA, Municipal
     Environmental Research Laboratory, Cincinnati, Ohio, April 14,  1982.

32.  Borchardt, J. A.,  S. J. Kang,  and T. H.  Chung.  Nitrification of  Secondary
     Municipal Waste Effluents by  Rotating Bio-Discs.  EPA-600/2-78-061,
     NTIS  PB-285  240/8BE, June 1978.

33.  Pano, A., E. J. Middlebrooks,  and J. H.  Reynolds.  The Kinetics  of Rota-
     ting  Biological Contactors Treating Domestic Wastewater.  Water  Quality
     Series  UWRL/Q-81/04, Utah State Unviersity, College  of  Engineering,  Logan,
     Utah, September 1981.
                                      46

-------
 34.  Stratta, J. M. and D. A. Long.  Nitrification Enhancement  Through  pH
      Control with Rotating Biological Contactors.  Final report prepared for
      USA Medical Research and Development Command under Contract  No.  DAMD17-
      79-C-9110 by Pennsylvania State University, September  1981.

 35.  Letter (data) communication from R. A. Sullivan, Autotrol Corporation,
      Milwaukee, Wisconsin, to R. C. Brenner, USEPA, Municipal Environmental
      Research Laboratory, Cincinnati, Ohio, March 12, 1982.

 36.  Crawford, P. M.   Use of Rotating Biological Contactors for Nitrification
      at the City of Guelph Water Pollution Control  Plant, Guelph, Ontario,
      Canada.  In:  Proceedings of the 1st National  Symposium/Workshop on
      Rotating Biological  Contactor Technology, Vol.  II,  EPA-600/9-80-046b
      NTIS PB81-124547,  June 1980.  pp 1247-1273.

 37.  Letter (data)  communication from R. J.  Hynek,  Autotrol Corporation,
      Milwaukee,  Wisconsin, to E.  J. Opatken, USEPA,  Municipal  Environmental
      Research  Laboratory,  Cincinnati,  Ohio,  July 22,  1980.

 38.   Advanced  Wastewater  Pilot Plant Treatment Studies.   Report  prepared for
      City of Indianapolis  by  Reid,  Quebe, Allison,  Wilcox,  and Associates,
      Inc.,  Indianapolis,  Indiana, January 1975.

 39.   Gilbert,  W.  G., J. F.  Wheeler,  and  A. MacGregor.  Energy  Usage  and  Other
      Considerations in  the Operation and Design  of  Rotating Biological Contac-
      tor  Facilities.  Paper presented at 57th  Annual  Conference  of the Water
      Pollution Control  Federation,  New Orleans,  Louisiana,  October 3,  1984
      (Publication pending).

40.   Heidman, J.  A., W. W.  Schuk, and A.  C.  Petrasek, Jr.   Field Measurements
      of Power Consumption  and  Air Flow at RBC  Installations.   Internal report,
      USEPA,  Municipal  Environmental  Research Laboratory, Test  and  Evaluation
      Facility, Cincinnati, Ohio,  May 5,  1982.

41.  Aero-Surf Energy  Requirements.  Autotrol Corporation internal report
     Milwaukee, Wisconsin, April  1,  1981.

42.  The Latest Word on RBC Reliability:  Envirex.  Envirex,  Inc.,  internal
     RBC status report,  Waukesha, Wisconsin, undated.
                                      47
 U.S. Government Printing Office  1984 -421-545/11830

-------

-------