WATER POLLUTION CONTROL RESEARCH SERIES • WP-2O-17
Dissolved-Air Flotation Treatment
of
Combined Sewer Overflows
U.S. DEPARTMENT OF THE INTERIOR • FEDERAL WATER POLLUTION CONTROL ADMINISTRATION
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The;Water Pollution Control Research Reports describe
the results and progress in the control and abatement
of pollution in our Nation's waters. They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Federal Water
Pollution Control Administration, in the U. S.- Department
of the Interior (both inhouse and through grants and
contracts with Federal, State, and local agencies, re-
search institutions, and industrial organizations). The
exchange of such data should contribute toward the long
range development of economical, large-scale management
of our Nation's water resources.
Triplicate tear-out abstract cards are placed inside the
back cover to facilitate information retrieval. Space is
provided on the card for the user's accession number and
for additional uniterms.
Water Pollution Control Research Series will be distributed
to requesters as supplies permit. Requests should be sent
to the Office of Research and Development, Department of
the Interior, Federal Water Pollution Control Administration,
Washington, D. C. 20242.
Previously issued reports on the Storm & Combined Sewer
Pollution Control Program:
WP 20-11 Problems of Combined Sewer Facilities and
Overflows - 1967
WP 20-15 Water Pollution Aspects of Urban Runoff
WP 20-16 "Strainer/Filter Treatment of Combined Sewer
Overflows;
WP 20-18 Improved Sealants for Infiltration Control
WP 20-21 Selected Urban Storm Water Abstracts
WP 20-22 Polymers for Sewer Flow Control
ORD-4 Combined Sewer Separation Using Pressure Sewers
DAST-4 Crazed Resin Filtration of Combined Sewer
Overflows
DAST-9 Sewer Infiltration Reduction by Zone Pumping
DAST-13 Design of a Combined Sewer Fluidic Regulator
-------
DISSOLVED-AIR TREATMENT
OF
COMBINED SEWER OVERFLOWS
A DEMONSTRATION PROJECT OF A PROTOTYPE
TREATMENT PLANT DESIGNED TO TREAT WASTES
FOUND AT A COMBINED SEWER OVERFLOW
FEDERAL WATER POLLUTION CONTROL ADMINISTRATION
DEPARTMENT OF THE INTERIOR
by
RHODES TECHNOLOGY CORPORATION
HOUSTON, TEXAS
CONTRACT NUMBER 14-12-11
JANUARY 1970
-------
ABSTRACT
A dissolved-air flotation system was evaluated for primary
treatment of combined sewer overflows. The major pieces of
component equipment were a gyratory screen, hydrocyclones, an
air dissolving tank, and a flotation cell.
The principal aspects investigated were: (1) Performance
of the system during rain events and dry periods; (2) Evaluation
of individual components; (3) Capital costs and operating
costs for utilizing a flotation system for various size
combined sewage overflows; (4) The adaptability of the
system for automation and use in remote location; and (5) The
ability of the system to treat intermittent and highly variable
flows from combined sewage systems. Some chemical aids to
flocculation were also tested.
The system performed comparably to conventional clarifiers.
It appears dissolved-air flotation systems would be economical
for handling combined sewer overflows up to 8 MGD. Automation
of dissolved-air flotation systems appears possible with conven-
tional control equipment. Chemical aids to flocculation appear
to have promise that warrants further study.
The system was unique in that all liquid flow passed
directly through the air dissolving tank with no recycle.
Domestic sewage was studied in lieu of combined sewage during
periods of no rain.
Conclusions, recommendations, and benefit-cost relation-
ships are presented in the report. A description of the
demonstration plant and of the drainage area served by the
flotation system are appended.
This report was submitted in fulfillment of Contract 14-12-
11 between the Federal Water Pollution Control Administration
and Rhodes Technology Corporation, Houston, Texas.
-------
CONTENTS
Section Page
Abstract
I - Results and Conclusions
Recommendations 1
II - Introduction 5
III - Initiation of Investigation and Site
Selection 13
IV - Characteristics of System Components 16
V - Air Flotation Studies and Treatment Plant
Design 32
VI - Design Details of Major Component Parts 38
VII - Sampling and Initial Plant Modifications 43
VIII - Testing and Evaluation 53
IX - Additional Testing of Chemical Aids to
Flocculation 67
X - Component Parts Performance 75
XI - Benefit-Cost Relationships 88
XII - Possibilities for Automation and Other
Potential Applications 95
XIII - Acknowledgements 106
XIV - References and Bibliography 108
XV - Appendices
-------
FIGURES
Figure Page
1 Schematic Diagrams of Various Methods of
Dissolving Air in Waste Water 22
2 Schematic Diagram of an Hydrocyclone 26
3 Drum and Traveling Screens 29
4 Circular Vibrating Screen 30
5 Air Dissolving Tank 33
6 Flow Diagram - Combined Sewer Effluent
Treatment Plant 36
6a Demonstration Pilot Plant 39
7 Pressure Control Valves 41
8 Flotation Tank 42
9 Sample Point Diagram - Demonstration Pilot
Plant 44
10 Distribution Box at the Fort Smith "P" Street
Pollution Control Facility 48
11 Modification of Distribution Box at Fort Smith
"P" Street Pollution Control Facility 49
12 Detail of Flotation Cell Showing the Baffle
Plate Installed to Trap Large Air Bubbles 50
13 Inlet Header as Built; Suggested Design for
Inlet Header 51
14 Relationship Between Suspended Solids Removal
Efficiency and Pressure Differential 81
15 Location and Size of Holes Cut in Exit Weir
of One Cell of Flotation Tank 86
16 Benefit-Cost Ratios 92
17 Automated Standby Combined Stormwater -
Domestic Sewage Treatment Plant 97
ii
-------
TABLES
Table Page
I Air Dissolving Tank, Air Flotation Tank 24
2 Percent Removal of Sewage Components When
Different Equipment Combinations Were Used
(No Chemical Flocculants) 55
3 Percent Removal of the Various Components When
Different Chemical Treatments Were Used. All
Mechanical Separatory Equipment Was On Stream 59
4 Percent Removals of the Various Components
During Rain Events. All Mechanical Separatory
Equipment Was On Stream 62
5 Summary of Removal Rates 64
6 Comparison of Sewage Strengths 65
7 Additional Chemical Tests 68
8 Additional Chemical Tests 69
9 Additional Chemical Tests 70
10 Chemical Treatment Costs 72
11 Oil Removal Test 73
12 Supplementary Data On Pressure Drop Across
Cyclones 78
13 Effect of Air Feed Rate and Pressure Differ-
ential On Total Suspended Solids Removal, Percent 80
14 Effect of Flow Rate On Suspended Solids Removal 86
15 Effective Flotation Depth 87
16 Costs and Benefits, Air Flotation and Conven-
tional Clarifiers 89
17 Physical Sizes and Land Areas Required by
Conventional Clarifiers and Dissolved Air
Flotation Units 93
iii
-------
SECTION 1
RESULTS AND CONCLUSIONS
RECOMMENDATIONS
-------
RESULTS AND CONCLUSIONS
1) The dissolved-air flotation system removed suspended
solids from combined sewage with 12 minutes retention time as
effectively as conventional clarifiers with 4 hours reten-
tion time. During rain events and without chemical aids, the
system removed an average of 69 percent of the suspended solids
passing a gyratory screen installed to removed gross particles.
Injection of alum and a polyelectrolyte into the system in-
creased the removal rate to an average of 84 percent. Alum alone
was ineffective. Without chemical aids, BOD reduction averaged
26 percent. When chemical flocculating aids were injected, BOD
reduction increased to an average of 42 percent.
2) Efficiency during dry weather, was essentially the
same as during rain periods.
3) Automation of dissolved-air flotation systems
appears feasible for the treatment of intermittent, variable,
and high instantaneous flow rates normally encountered with
combined sewage overflow. Surge tanks or retention basins
are unnecessary when dissolved-air flotation is used as a
treatment for combined sewer overflows, provided there are
approximately 2 minutes storage time available in the
sewer system.
-------
4) Total annual costs for dissolved-air flotation systems
are less than costs for conventional clarifiers for flows up
to 8 million gallons per day. For treatment of storm water
flows more than 8 million gallons per day, conventional clari-
fiers show lower total annual costs. As capacities increase,
operation and maintenance costs become very significant in
the dissolved air process. However, dissolved-air flotation
units require only one-tenth as much land area as conventional
clarifiers.
5) The foam collected contained 5 to 7 percent dried
solids of 70 percent volatility. Conventional sludge hand-
ling techniques may be used to dispose of the foam, except
sludge thickeners can probably be eliminated.
6) Evaluations of individual components show the gyratory
screen, the full flow air dissolving tank and the flotation
cell were very effective. Cover against wind and rain was
essential to full efficiency of the flotation cell. The
hydrocyclones used could not be evaluated fully because of
periodic plugging. However, the cyclones did remove the kind
of dense inorganic materials which overload sludge digesters or
form clinkers during sludge incineration.
-------
RECOMMENDATIONS
1) Dissolved-air flotation of combined sewer overflows
should be considered as an alternate to conventional treat-
ment methods.
2) Additional research is necessary to fully evaluate
hydrocyclones. The solids collection pots and pneumatically
operated dump valves should be eliminated from the cyclones
and replaced by adjustable apex valves allowing continuous
cyclone underflow.
3) Consideration should be given the use of screw
conveyors to move foam from the foam collection troughs. This
will permit a drier foam to be produced.
4) Alternate screening mechanisms should be considered.
In future applications of the present dissolved-air flotation
design, a comprehensive study should be made of the
characteristics of suspended solids for each application.
5) Pilot plant studies of chemical aids to flocculation
are recommended to determine costs of producing waters of
secondary treatment plant quality.
6) The efficiency of a total treatment unit consisting
of the dissolved-air flotation system for both primary and final
clarification of trickling filter and activated sludge effluents
and combinations of the following secondary and tertiary treating
systems should be investigated:
-------
a. Rapid sand filtration.
b. High rate trickling filters.
c. Activated carbon filtration.
d. Chlorination or hypochlorination.
7) Additional research and pilot plant work is
recommended to study the applicability of dissolved-air
flotation to the treatment of various industrial wastes
-------
SECTION II
INTRODUCTION
THE COMBINED SANITARY AND STORM SEWER OVERFLOW
PROBLEM IN THE UNITED STATES
PROPOSED SOLUTIONS TO THE OVERFLOW PROBLEM
-------
INTRODUCTION
THE COMBINED SANITARY AND STORM SEWER OVERFLOW
PROBLEM IN THE UNITED STATES
The flooding of basements, low lying buildings, and land
by combined sewage causes immeasurable direct damage and
inconvenience to an estimated 36 million people in over 1300
cities and communities in the United States. Collector sewers
which are too small for the large flows from storm water run-
off are the major cause of the direct damage. The costs of
this direct damage are spread indirectly in the form of higher
costs for goods and services to the entire U. S. population (1).
Reduced water quality is one example of indirect damage
caused by storm water run-off and combined sewer overflows.
Many treatment plants have insufficient capacity to remove the
silt and organic matter flushed from sewers by the surge of
storm waters. It is not uncommon, after an extended dry spell,
for treatment plant operating personnel to bypass the first
waters received after the start of a rainfall to avoid a buildup
of grit and silt in the clarifiers. Receiving waters also
suffer quality reduction when improperly maintained or
inoperative flow regulators permit storm waters to overflow
directly to receiving waters, bypassing all treatment facilities.
Because of these and many similar situations, there is
great need for low-cost and reliable facilities to handle
-------
combined sewage overflows.
The concept of the combined sanitary and storm system is
several thousand years old (2). Originally, sewers were used
for storm drainage only. Domestic wastes were the responsi-
bility of the individual householder and were disposed of in
dry wells, cesspools, and septic tanks. It became necessary
to dump domestic wastes into the streets to be washed into
storm drains when rains occurred. The practice spread. With
increasing urban populations and the advent of industrialization,
true combined sewer systems became a fact through the piecemeal
addition of open channels draining into the storm sewers.
Eventually, closed conduits and pipes were added to the system.
In many instances the old closed facilities still exist and
are in service, but, because they were designed for small
drainage areas and have a limited capacity, they are over-
burdened even in dry weather. During storm periods the
combined waste waters cause local flooding.
Each of these old sewers has its own outfall at a nearby
river or stream. The result is a multitude of outfalls and
evil-smelling areas along water courses. Interceptors have
been constructed to alleviate the situation, but overflows
still course through the outfalls. More than 400 such outlets
are still to be found in Cleveland, for example (3).
The total number of cities having combined sewers has
decreased in recent years. The reduction has
-------
been accomplished by building new interceptors to separate the
wastes and by rebuilding the old sewer systems. Most new
construction involves separate sewers, but occasionally the
separate sewers in new suburbs and residential areas are
connected to the interceptors of a combined system in older
sections of the communities, compounding the existing over-
flow problems. Some combined facilities are still being con-
structed in cities which already have combined sewers (4).
Haphazard additions to sewer systems have led to numerous
overflow and treatment problems. Additionally, lax enforce-
ment of sewer regulations and restrictions plus ambiguous and
conflicting interjurisdictional construction codes have led
to large networks of sewer lines feeding to central treatment
facilities. For example, Cleveland, Ohio, serves 32 govern-
mental units outside its city limits; many of these are
without any form of municipal organization (1).
These problems have not gone unrecognized, and in some
areas sanitary districts or authorities with broad powers and
adequate financial structure have been established to help
combat and correct these problems.
The American Public Works Association in its report for
the FWP.CA, "Problems of Combined Sewer Facilities and Overflows'
(1), states that over 50 percent of the jurisdictions inter-
viewed have problems due to infiltration of ground waters.
The surcharge of sewers due to infiltration of ground waters
-------
is a problem common to both combined and separate sewer systems
The separate sewer systems found in much of the Southwestern
United States are particularly susceptible to infiltration of
ground waters during both wet and dry weather.
Dry weather infiltration occurs when sewers are below the
water table. Infiltration during dry weather is often unre-
cognized until attempts are made to relate water utility
service pumping output to waste water treatment plant flow
records. Dry weather infiltration waters together with the
existing flow of sanitary wastes often approach the capacity
of the treatment facilities, leaving little or no capacity
for rain waters. The infiltration of ground water into the
sewer systems during rain events causes many of the same
problems as occur in combined sewers: Namely, flooding of
basements, overloading of treatment facilities, and dis-
charging of wastes through overflows.
The magnitude of the infiltration problem is illustrated
by an unsigned article in "American City" (5). The author
discusses the methods used by the city of North Miami, Florida,
to dispose of daily treatment plant effluents containing up to
75 tons of salt from salt water constantly infiltrating the
municipal sewer system.
Among the methods used to control infiltration are better
supervision of the installation of the facilities and the
sealing of existing facilities against infiltration.
8
-------
PROPOSED SOLUTIONS
TO THE OVERFLOW PROBLEM
The cost of separating the combined sewers in the
United States has been estimated up to $30 billion plus an
additional $18 billion for plumbing service connections to
private property (1). Other sources list the cost at $10
million per square mile or about $1,000 per family served (3).
The cost in time and inconvenience to the populations involved
is beyond estimation; the need to alter roof and basement
drains alone would entail a tremendous public relations effort
and would provide fertile ground for countless property damage
suits. Peters and Troemper (6) report on the difficulties
encountered by the Springfield, Illinois, Sanitary District
in removing or attempting to remove the rainwater downspouts
from residences. Several "questionnaires, letters, and inspec-
tions were necessary to approach 100 percent compliance with
a long-existing regulation concerning the connection of
downspouts to sanitary sewers. The authors report that
compliance with the regulation eliminated flooding of basements
due to surcharge of sanitary sewers during rainstorms.
Many alternatives to complete separation are available.
Although each situation presents its own problems, there are
enough similarities that solutions can be categorized. The
Chief of the Storm and Combined Sewer Pollution Control Branch,
-------
Federal Water Pollution Control Administration of the U. S.
Department of Interior, in a speech presented at the Spring
1968 Meeting of the New England Water Pollution Control
Association, discussed three basic approaches that can be
utilized to solve combined sewage or storm water pollution
problems:
1) Control.
2) Treatment.
3) Combinations of control and treatment.
It is not the purpose or intent of this report to treat
or discuss all the details and ramifications of each categorical
solution or even to list all the possible solutions. Some of.
the more publicized solutions are listed for illustrative
purposes.
The storage of storm induced overflows in limestone tunnels
deep under Chicago has been mentioned; the sale of excavated
limestone would help to defray some of the construction costs (7).
Two large collapsible rubberized storage tanks, each of 100,00
gallons capacity, to be anchored in the Anacostia River to store
overflows during heavy rainfalls are being constructed and installe
in Washington, D. C. This is not intended to provide complete
relief to the overflow problem; need for ten tanks is estimated.
After the storm ends, stored waste waters will be pumped to
currently available treatment facilities. Similar projects are in
10
-------
process at Cambridge, Maryland, and Sandusky, Ohio.
Chemical additives for waste water have been developed
which reportedly increase the flow in sewer pipes up to 2-1/2
times, thereby achieving peak-flow relief without new
construction (8). A novel plastic sealant to eliminate
excessive sewer flows due to infiltration of ground waters
has been reported by the same source.
Chlorination and hypochlorination of storm waters are
being investigated by the city of New Orleans. Although the
sewers in New Orleans are separated, storm waters pumped into
Lake Pontchartrain carry a tremendous load, necessitating the
closing of some public beaches after major rainfalls. The
City of Boston is studying the use of retention basins for
storage and sedimentation in conjunction with hypochlorination
for the treatment of storm overflows from its combined sewer
system.
Some treatment or control methods have been in use for
some time in various communities throughout the United States
and abroad.
One current test, supported by FWPCA, has been described
as follows.
"One of the Dallas grants in the amount of $828,750, or
75 percent of the total eligible project costs, funds a project
11
-------
which consists of the design, construction, and evaluation of
a facility to treat overflows from sewers carrying a mixture
of domestic waste water plus storm water infiltration.
Physical features include a diversion structure, a pumping
station, flocculation and sedimentation basins, chemical feed
facilities, and a conveyance system for transporting waste
lime sludge from a municipal water plant to the storm water
treatment facility. Unique features of this project include
the demonstration of tube-type clarifiers and the evaluation
of the utilization of waste chemicals from a water-softening
plant to enhance settling in the waste water sedimentation
unit" (9).
12
-------
SECTION III
INITIATION OF INVESTIGATION
AND SITE SELECTION
-------
INITIATION OF INVESTIGATION
AND SITE SELECTION
In mid-1966, the Federal Water Pollution Control
Administration advertised in The Commerce Business Daily for
concepts and new approaches to the solution of the problems
of combined sewer facilities. Engineers of Rhodes Technology
Corporation, Houston, Texas, were convinced that the problems
involved in the treatment of sanitary and combined sewage
were not extremely different from the problems involved in the
treatment of waste water in oil fields and that the techniques
that had been used to clarify waste water in the oil fields
would be directly applicable to the treatment of sewage.
Considerable experience has been gained over the past 15 to
20 years in the use and operation of dissolved-air flotation
units to remove oil and suspended solids from oil field
waste water. Additionally, considerable experience has been
gained in the use of hydraulic cyclones or hydrocyclones for
the removal of heavy materials such as silt, sand, and clay
from water.
A proposal was submitted to FWPCA suggesting the linking
of these pieces of equipment into a single treatment unit of
extremely short retention time (about 10 minutes).
The Federal Water Pollution Control Administration awarded
13
-------
the contract for this study to investigate the concept of
using hydrocyclones, a dissolved-air flotation unit, a low
liquid retention time, and screens as well. Screens were
thought necessary so that particle size entering cyclones
could be limited. Comminutors were not included because
sludge digestion was not a part of this study.
Included in the contract was a provision calling for
the selection of a site for the dissolved-air flotation unit.
Items to be considered in the site selection included the
availability of the following items :
1) Land for the erection of the dissolved-air flotation
plant.
2) storm waters during storm events.
3) Domestic waste to be used in lieu of storm water in
dry weather periods.
4) Fresh water for the dilution of domestic wastes,
should the need arise.
5) Electric power.
6) A laboratory for the analysis of the influent and
effluent waste streams.
7) : The cooperation of the necessary municipal officials
and employees.
Several of the sites inspected included Kansas City and
St. Louis, Missouri; Oklahoma City, Norman,and Stillwater,
Oklahoma; and Fort Smith, Arkansas. Each of the sites offered
many possibilities for the successful completion of the project,
14
-------
Fort Smith, Arkansas, was selected because it was the only
site at which all of the desired items were available.
This report covers results of bench scale tests, design
of a dissolved-air flotation plant, and operation of the
plant from October 1967 to December 1968.
15
-------
SECTION IV
CHARACTERISTICS OF SYSTEM COMPONENTS
-------
CHARACTERISTICS OF SYSTEM COMPONENTS
Dissolved-Air Flotation
Air flotation systems have been used for many years in
the mining industry to concentrate low grade ores by frothing.
Hanson and Gotaas (10) claim the first froth process was
patented in 1860 by Hanes. The froth process does not use
dissolved air; air is injected by several means. In recent
years the trend has apparently been to use air injected
through the shaft of an impeller. The impeller breaks the
air stream into millions of bubbles creating a froth or foam
which rises to the surface, floating the ore or gangue,
whichever is lighter. In most cases frothing is aided by
the use of chemicals such as alcohols, resins, or soaps.
The flotation method of separating ores from overburden
material is discussed in much research literature. Gaudin
(11) mentions that gas flotation was first recognized as early
as 1901. Fromet obtained a British patent on the use of gas
bubbles to remove sulphite minerals from ores in 1903. The
vacuum process is mentioned by one author as being patented
in 1907. Norris (12) was issued a patent in 1907 in which
a pressurized slurry of water and ore was used. Elmore
(13) was granted an English patent in 1905 for the vacuum
separation of ores. Previous systems, according to Elmore,
used frothing aids such as oil, tars, and soaps.
16
-------
Elmore claimed his system would reduce the amount of chemical
aids needed. Another English patent was awarded for the use
of a pressurized flotation system in 1906 to Suhman,
Kirkpatrick-Picard, and Ballot . The patent was also granted
in the United States(14).
All. dissolved air. systems involve injection of air into
a liquid/under pressure followed by transfer of the liquid to
a cell where the air leaves solution in the form of small
bubbles.
D'Arcy (15) claims the modern dissolved-air flotation
system was invented in Norway by Sveen and Pederson. No date
for the invention is given. The Sveen-Pederson process is
widely used in the paper and pulp industry for the clarification
of "white'water". The dissolved-air flotation system is
widely used in.industry, as indicated by many references
throughout the:literature to its various applications. Specific
applications include the removal of oil and suspended matter
from oil field wastes; The use of dissolved-air flotation
systems is discussed in several papers relating to the separation
of oils and fats in the soap, and .detergent industry. Dissolved-
air flotation systems are also used by the food processing,
meat packing, and slaughterhouse industries. Several steel
mills,report the use of dissolved-air flotation for the removal
of grease and oil from water, while both the Santa Fe and the
Union Pacific railroads report the use of dissolved-air
flotation to clean wash water. Chrysler Corporation reports
17
-------
using the dissolved-air flotation system to clean process
water (16).
The operator of a 400-gallon-per-minute dissolved-air
flotation system installed at the Indiana Farm Bureau
Cooperative Association Refinery near Mount Vernon, Indiana,
claims a BOD reduction of 78 percent and a suspended solids
reduction of 93 percent when waste waters with pH 9
are fed through the system. In addition, he reports a 90
percent removal of oil (17).
An extensive literature search reveals a considerable
•>
amount of data relating to the design and use of flotation
cells. Howe (18), in a mathematical derivation of flotation
cell design, recommends that considerable experimentation
/•
with each different waste precede the use of his equations in
determining the exact criteria for flotation cells. He further
states that particle size and density, liquid viscosity, and
liquid density are factors to be considered in designing tank
depth, overflow rate, and retention time. He goes on to state
that bubbles released in the liquid are less than 130 microns
in diameter, smaller than the bubble size in the froth system.
A comprehensive discussion by D'Arcy (15) of the use of
dissolved-air flotation systems to separate oil from waste
waters includes six important general considerations:
1) Dissolving a maximum amount of air in the influent.
2) Elimination of all entrained air as the release of
18
-------
entrained air in the flotation system introduces
turbulence and short circuiting.
3) Proper hydrodynamic design of the entire flotation
system, especially the flotation chamber.
4) Selection of proper coagulant and floe-forming
chemicals if they are required--always bearing in
mind that the most economical as well as the most
efficient chemicals should be used.
5) Continuous mechanical removal of oil or floe on the
surface of the water in the flotation chamber.
6) Design of the entire system to produce a unit which
will operate automatically under a wide range of
conditions and which requres the minimum amount of
trained personnel for its operation.
The equipment discussed by D'Arcy is operated by regular
oil field personnel and seldom requires more than 3 man-
hours per day, this time being used for mixing chemicals and
lubrication of equipment. Chemical costs are in the neigh-
borhood of $2 per thousand barrels ($48/million gallons) of
waste treated and have been as low as 80 cents per thousand
barrels ($19/million gallons) when alum or activated silica
are used for treatment.
The fundamental principles of dissolved-air flotation
as applied to industrial wastes were discussed by Vrablic at
the Fourteenth Annual Industrial Waste Conference at Purdue
University (19). Among other things, Vrablic hints that
19
-------
advantage should taken of fact that oxygen is twice
as soluble in water as nitrogen. He further claims
that the flotation system makes use of three funda-
mental processes:
1) Adsorption of air bubbles on the solids.
2) Trapping of air bubbles by the solids.
3) Adhesion of air bubbles on the solids.
Vrablic states that hydrophobic solids will float much more
easily than will hydrophillic ones. He recommends an air to
solids ratio of 0.06 (Ib of air/lb of dry solids).
Eckenfelder, et al, found a ratio of 0.02 most favorable (20)
These researchers report the use of a laboratory scale model
to treat domestic sewage. The scale model consisted of a
stell pressure tank which was filled with waste activiated
sludge, pressurized, and then shaken. The liquid was then
released into a Lucite cylinder for decompression and foaming;
periods of foam formation took up to 20 minutes. They report
excellent suspended solids removal and further indicate that
turbulence must be controlled to reduce the shearing of fine
floe.
Results varying from between 20 and 82 percent removal of
unemulsified oil are discussed by Rohlich (21). A 75 gallon-
per-minute flow of waste water was directed through an air
flotation unit using a retention time of 12 minutes. The tank
had a surface area of 34 square feet and was 3-1/2 feet deep.
20
-------
The air dissolving tank had a capacity of 150 gallons with a
resulting retention time of 2 minutes.
Rohlich feels that a retention time of 2 minutes is
necessary to insure saturation of air in the liquid stream.
Air was dissolved in the waste stream at a pressure of 50 to
60 pounds per square inch. The three types of pressurization
shown in Figure 1 were attempted. These include the diversion
of part of the influent stream through the pressure tank,
recirculation and pressurization of the recycled waste water,
and total pressurization with air injected through a venturi
device. The experiments relating to total pressurization used
a flow rate of 50 gallons per minute.
Prather (22) discusses the reduction in chemical oxygen
demand (COD) in an oil refinery waste using dissolved-air
flotation. In the waste discussed, the COD was due primarily
to suspended solids. The dissolved-air flotation system in
this application was originally designed to remove oil and
suspended material. In order to achieve a significant removal
of COD and suspended solids, pH adjustments were necessary.
Values of pH between 8 and 9 are reported.
Hopper and McGowan (23) report the use of frothing to
purify surface waters in a 1950 experiment using 34 different
surface waters as test media. Nontoxic quaternary ammonium
compounds were used in an attempt to reduce the bacterial
content of drinking water. Bacterial reductions up to
21
-------
SCHEMATIC DIAGRAMS OF VARIOUS METHODS OF
DISSOLVING AIR IN WASTE WATER
AIR FLOTATION
TANK
1
t
^ ( 1 \ /r-ir-tr-
i
PUMP
AIR
COMPRESSOR
(Q) PI VERSION OF PART OF FLOW THROUGH PRESSURE TANK
AIR FLOTATION
TANK
PRESSURE TANK
PUMP
AIR
COMPRESSOR
(b) DISSOLVING AIR IN RECYCLED WASTE WATER
AIR
INLET
I
AIR FLOTATION
TANK
PUMP
VENTURI
TUBE
(c) INJECTION OF AIR THROUGH A VENTURI DEVICE
FIGURE 1
22
-------
99 percent (plate count method) were reported, and
95 percent of suspended solids were removed. Cost of
operating the system is reported as 5 cents per thousand
gallons.
Air bubbles less than 100 microns in diameter were used
for sludge thickening and total solids removal in several units
reported by Katz and Geinapolos (24). The units being studied
used a 50 percent recycle rate in which air was introduced at
the rate of 1 cubic foot of air per 100 gallons of recycle
water. The authors indicate dry solids loading rates of 10
to 20 pounds per square foot per day for activated sludge and
55 pounds per square foot per day for primary wastes. The
units varied from 7 to 12 feet in depth. Katz suggests that
the flotation system includes several types of flotation and
that hindered flotation was one of the phynomena encountered.
Sludge was thickened to a •consistency of 3 to 4 percent
solids.
Bubbles forming in sanitary sewage have terminal velocities
of 0.14 inches per second. This value is the basis of a
design discussed by Masterson and Pratt (25) . They suggest
that free air evolves in the air dissolving tank when over 60
percent of saturation is reached. However, they also state,
"The greater the amount of air dissolved, the greater will be
the (flotation) effect".
23
-------
A summary of the pertinent suggested values reported in
the design of dissolved-air flotation systems are given in
Table 1 below:
TABLE 1
Air Dissolving Tank;
Pressure - 50 to 60 psig
Retention time - 2 minutes
Air feed rate - 1 cu ft/100 gal waste
Recycle rate - 50 percent
Air to dry solids ratio - 0.02 to 0.06 Ib air/lb solids
Air Flotation Tank;
Dry solids loading rate - 55 Ib/sq ft/day
Surface loading rate-- 2.0 to 2.5 gal/sq ft/min
Depth - 3.5 to 12 ft
Hydraulic Cyclones
Storm water overflows and combined sewer systems receive
a variety of dense materials; it is almost impossible to control
the influx and arrival of these materials at treatment plants.
In order to minimize the accumulation of the resulting sediments
on the bottom of flotation tanks, Rhodes Corporation engineers,
on the basis "of their oil field experience, suggested the use
of hydrocyclones for the removal of the dense material.
24
-------
The pneumatic cyclone has long been used in the lumber
and furniture industries for removing wood chips and sawdust
from air streams. The hydraulic cyclone, or hydrocyclone,
uses the same principle as the pneumatic cyclone. A comment
by Pryor indicates that hydrocyclones were first used in the
petroleum industry in 1939. Figure 2 indicates the general
configuration of a cyclone in which an inverted cone has a
cylinder attached to its wide end. Waste water is injected
tangentially into the cylinder and is forced to travel in
a spiral pattern through a shorter and shorter radius toward
the narrow end. Centrifugal force causes the heavier particles
to move to the outer edge of the stream. Upon approaching
the narrow end of the cone, waste water escapes through a
tube called a vortex finder running up the center of the
cone. Solid materials which have been forced to the outside
of the waste stream fall to the bottom of the cone, are
collected, and disposed of.
Leniger states, "When dealing with suspensions of very
fine particles, an obvious measure to be adopted consists
of accelerating sedimentation by centrifugal force. There
has been a choice between hydrocyclones in which a rotation
of the suspension is produced by introducing it tangentially
into a stationary apparatus and centrifuges (here termed
clarifiers), in which the liquid is caused to rotate by
revolving a drum. For hydraulic classification a so-called
25
-------
APEX VALVE
SOLIDS OUTLET
SCHEMATIC DIAGRAM OF AN HYDROCYCLONE
FIGURE 2
26
-------
hydrocyclone is not used with much success. Hydro-
cyclones have the advantage of simplicity and flexibility
so that the results may be modified by altering various
operating conditions. As opposed to other types of
apparatus, they are better for classifying than for
clarifying. The reason is that the high shearing stresses
in the hydrocyclone promote the suspension of particles and
oppose flocculation. A disadvantage lies in the fact that
both the fine and coarse fractions are obtained in suspensions
of relatively high dilution. Furthermore they only operate
well in a medium of low viscosity" (27).
There is abundant literature on the theory and design of
hydrocyclones. Broer (28) discusses efficiency as judged by
the separating capacity and power consumption of the hydro-
cyclone, and Van der Kolk (29) investigates the ability of
cyclones in series arrangements to protect expensive devices
and to collect several grades of bulk material. Van der Kolk
illustrates his discussion with several diagrams showing
different schemes for connecting cyclones in series. He
concludes his article with a discussion of the advantages
of the various ways of linking the cyclones.
A major manufacturer (30) of hydrocyclones uses an adver-
tising brochure for a very enlightening discussion on the use
of hydrocyclones as classifiers. Among other comments, the
manufacturer lists four major cyclone applications.
27
-------
1) Classification or sizing of particles. Cyclones
separate particles according to their relative mass
rather than strictly by particle size. However,
the generally accepted range of cyclone operation
is from 35 mesh to 5 microns.
2) Degritting water or water suspensions of fine solids.
3) Desliming operations.
4) Closed circuit grinding classification.
Screens
The apex valve of the hydrocyclone is generally of much
smaller diameter than the inlet, and large particles which
enter the cyclone can sometimes clog the apex valve. To
prevent this plugging, the manufacturer suggested screening
the waste water entering the cyclone. Of particular hazard
are materials such as sticks, pencils, etc., which can bridge
the narrow apex valve. Once bridgedt the valve easily becomes
further clogged with other materials and, within a short time,
the cyclone must be removed from the system, dismantled,
and unplugged.
The commonly-used bar screen is of no value in this
application; turbulence can cause sticks to twist in such a
way as to pass through the screen. Screens which are available
for the operation envisioned include drum screens and endless
belts illustrated in Figures 3a and 3b and vibrating screens
illustrated in Figure 4. The vibrating screen apparatus
28
-------
SCREENINGS
DISCHARGE
TROUGH
INFLUENT —-<
SPRAY PIPES
SCREEN COVERED
'DRUM
(a) DRUM SCREEN
i EFFLUENT
(b) TRAVELING SCREEN
FIGURE 3
29
-------
FEED
SOLIDS
OUTLET
WASTE WATER
OUTLET
FIGURE 4
30
-------
has several advantages:
1) The screen is easily removed from the apparatus for
replacement and changing of mesh size.
2) There is only one moving part.
3) The screen is self cleaning.
The vibratory screen is found in a variety of applications,
including the removal of stones, rocks, and coarse material in
mining operations; the removal of feathers in turkey and chicken
processing plants; the removal of water from vegetables in the
frozen food industry; and the removal of wastes in the vegetable
and fruit canning industries.
31
-------
SECTION V
AIR FLOTATION STUDIES
AND
TREATMENT PLANT DESIGN
-------
AIR FLOTATION STUDIES AND TREATMENT PLANT DESIGN
Items to be investigated before the dissolved-air
flotation system could be designed included methods of
dissolving air in the waste stream and design of the air
flotation tank.
Several methods are often used for dissolving a gas in
a liquid. Many of the conventional methods suggested in The
Chemical Engineers' Handbook (31) were discarded because of
the possibility of trapping suspended particles in the
dissolving unit. Two designs were selected for trial. One
design included the use of Raschig rings as a packing material,
The other design is illustrated in Figure 5 and consists of
a cylindrical outer tank with a stand pipe in the center of
the tank. Air and the incoming waste stream enter the
bottom of the inner stand pipe; air dissovles in the waste
liquid as the air and waste liquid rise through the stand pipe,
Additional air is dissolved as the waste liquid overflows
the stand pipe and falls through the air gap at the top of
the outer cylinder. Because oxygen is more soluble than
nitrogen, the unit was designed for a constant flow of air
through the air dissolving tank. Air deficient in oxygen
but rich with nitrogen was constantly being replaced with
oxygen-rich air for better dissolving efficiency.
32
-------
INFLUENT
5
o
o
en
UJ
o
o
t
VENT VALVE
, t
LJJ
EFFLUENT
PRESS COh
\AU
AIR DISSOLVING TANK
FIGURE 5
33
-------
The design indicated in Figure 5 was selected after trial
because:
1) The Raschig rings in the alternate design collected
waste solids.
2) Air dissolving efficiency in the alternate design was
low, as evidenced by a lack of bubbles in the flotation
tank.
3) The tank used in the selected design had few places
where solids could become lodged.
4) There were abundant bubbles produced in the flotation
tank when the selected design was tested.
The waste particles entering the flotation cell have two
major velocity components. A horizontal component is imparted
to the particle by the hydraulic flow; a vertical component
results from the buoyant effect of the air bubbles. Therefore,
the critical dimensions of the flotation cell are obviously
depth and length.
A modified version of Stokes Law:
v = g ( P1 - Pd ) D2
18 u
where
v = terminal velocity of particle
g = acceleration of gravity
P^ = liquid density
Pj =* particle density
D = particle diameter
u = liquid viscosity
34
-------
indicates that the vertical velocity is a function of particle
size and particle density. If the vertical travel of the
particle could be decreased, the length of the air flotation
tank could also be decreased.
A model of the air flotation tank was constructed using
a rectangular design 14 inches deep by 2 feet wide by 5 feet
long. The tank was constructed in such a way that the liquid
depth, the length of the tank, and the depth of the influent
stream could be varied. Flow rates with turndown ratios of 15
to 1 were provided. Optimum suspended solids removal rates
occurred for surface loading rates in the neighborhood of 1.5
gallons per square foot of surface per minute. The foam
formed was quite easily removed by means of a scraper and
appeared to be stable.
On the basis of data obtained from the models of the air
flotation tank and air dissolving tank, the demonstration
treatment plant (Figure 6) was designed.
The demonstration plant provided primary treatment only.
No solids treatment facilities were included. Combined waste-
waters first flow over screens to remove the gross debris expected
from storm run-offs. Grit and organic matter are removed by
hydrocyclones. The liquid overflow from the cyclones then
passes through a pressurized air dissolving- tank and on to the
air flotation cells. In the cells dissolved air comes out of
the solution and forms tiny bubbles around the suspended solids
35
-------
INFLUENT
SOLIDS TO
1
P-l
GYRATORY
SCREEN
TANK
DISPOSAL
CLARIFIED WATER
KEY
P-l ...PUMP I
C-l PRIMARY CYCLONE
C-21
C-3| SECONDARY CYCLONES
CA)
PCV. PRESSURE CONTOL VALVE
T\ i
FLOTATION
CELL
AIR
DISSOLVING
TANK
PCV
FLOW DIAGRAM
COMBINED SEWER EFFLUENT TREATMENT PLANT
f
AIR
FIGURE 6
-------
or immiscible liquid microparticles, which act as nuclei. The
bubbler-particles float to the surface and form thin mats which
are removed by scrapers. Dense materials which escape removal
in the cyclones sink to the bottom of the flotation cell and
are scraped into a collection trough. Effluent waters may
be further treated or discharged into a receiving stream or river;
the solids collected may be passed to conventional sludge
equipment.
37
-------
SECTION VI
DESIGN DETAILS
OF
MAJOR COMPONENT PARTS
-------
DESIGN DETAILS
OF
MAJOR COMPONENT PARTS
Figure 6a is a detailed diagram of the dissolved-air
flotation system as contructed at Fort Smith, Arkansas.
Incoming waste water was screened by a 48-inch gyratory screen,
and the screened waste water was then dumped into Tank 1. The
liquid level was controlled by a flow control valve at the
outlet from the Fort Smith sewage distribution box.
A multi-stage vertical turbine pump removed waste water
from Tank 1 and forced it through two banks of hydrocyclones
at a design rate of 350 GPM. The primary cyclone was 12
inches in diameter and was sized to remove particles as small
as 50 microns in diameter. Partially degritted water from
the primary cyclone was directed to a bank of three secondary
cyclones operated in parallel. The secondary cyclones were
10 inches in diameter and were each capable of handling
150 GPM of flow. The secondary cyclones were designed to
remove particles as small as 25 microns in diameter. The
design pressure drop across the secondary cyclones was
approximately 20 pounds per square inch.
The two-stage cyclone design was selected for two reasons:
1) The primary cyclone was included to remove the larger
dense particles, because it was feared that these
38
-------
FROM MAIN
INLET
JUNCTION BOX
...PUMP I
...PUMP 2
..PRIMARY CYCLONE
SECONDARY CYCLONES
FOAM COLLECTION
TROUGH
PCV.., ..PRESSURE CONTROL VALVE
LLC.... LIQUID LEVEL CONTROLLER
DISTRIBUTION
HEADER
BOTTOM DRAIN
FOR SETTLED
SOLIDS
FRESH WATFR SUPPLY
AIR
COMPRESSOR
TO MAIN IN|.ET
JUNCTION BOX
LIQUID COLLECTION
DETAILED DIAGRAM
DEMONSTRATION PILOT PLANT
(FIGURE 6a)
39
-------
particles might tend to overload the secondary
cyclones and clog the apex valves.
2) Experimental flexibility was needed so that various
cyclone combinations could be studied to obtain the
maximum removal of dense particles.
The hydrocyclone overflow passed through the air dissolving
tank and on to a pressure control device consisting of two, 2-
inch diaphragm valves. One valve was operated by a pneumatic
activator; the other valve was manually operated (Figure 7).
This dual operational capacity was included for testing purposes.
For flow rates greater than 350 GPM, both control valves were
necessary to handle the flow. Wastes entered the air flotation
tank through a 6-inch header with 2-inch nozzles evenly spaced
along it and passing through the end wall of the tank (Figure 8).
The air flotation tank was 20 feet wide and 15 feet long and was
divided into two cells, each 10 feet wide, for greater experimental
flexibility. The cell wall height, 29-1/4 inches, was dictated
by the size and availability of the chain sprockets used for
the foam scraper mechanism. The flotation chamber of each cell
was 10 feet wide by 12 feet long; the remaining 3 feet of
length was used as a foam trough and liquid effluent collector.
Solid and liquid effluent wastes from the air flotation
tank were piped into Tanks 2, 3, 4, and 5, for collection,
sampling and disposal. Disposal was accomplished by remixing
and returning the wastes to the Fort Smith sewage distribution
box.
40
-------
THE PRESSURE CONTROL VALVES
FIGURE 7
41
-------
TOP SCRAPER
.£>
N3
SLUDGE
DRAIN
INFLUENT
FROM AIR
DISSOLVING
TANK AND
PRESS. CONTROL
VALVES
. - • - .
'•'.•.'. BOTTGMSCRAPER •.>•.••.
CLEAN WATER
EFFLUENT
DRAIN FOR
SETTLED SOLIDS
FLOTATION TANK
FIGURE 8
-------
SECTION VII
SAMPLING
AND
INITIAL PLANT MODIFICATIONS
-------
SAMPLING AND
INITIAL PLANT MODIFICATIONS
Following the commissioning of the pilot plant and the
initial start-up exercises, a program of sample-point and
sampling-technique evaluation was initiated. Figure 9 is
a schematic diagram of the sampling points finally selected
for the demonstration plant. Note that sampling points are
located so that the efficiency of each major piece of
equipment can be ascertained. Most sampling points were
controlled by a diaphragm valve. In most cases composite samples
were accumulated every 1/2 hour on a 4- or 8-hour schedule.
Grab samples were also used as the need arose. Samples were placed
in gallon jugs and immediately iced to slow chemical and
biological action.
Sampling difficulties with the liquid effluent or over-
flow from the cyclones made it impossible to conduct detailed
material balances for evaluation of performance. In some cases
it was suspected that the cyclones were breaking up part of the
larger or more fragile solids. Initially, sampling was done
from 1/2 inch valves which drained from the center of the pipe
installed immediately down-stream of the cyclone overflows.
Because of the difficulty experienced in obtaining duplicate
samples, it was theorized that the swirling motion imparted
to the liquid as it passed through the cyclones was carried
on by the liquid as it left the cyclone causing the solid
43
-------
FROM
INLET JUNCTION
BOX
KEY
P-l PUMP*I
P-2 PUMP*2
C-l. PRIMARY CYCLONE
f* 9> 1
?.§ SECONDARY CYCLONES
PCV PRESSURE CONTROL VALVE
M-2f"— MIXERS
M-3)
LLC LIQUID LEVEL CONTROLLER
©. SAMPLE POINT
BOTTOM DRAIN
FOR SETTLED
SOLIDS
FRESH WATER SUPPLY
TO MAIN INLET
JUNCTION
LIQUID COLLECTION
SAMPLE POINT DIAGRAM
DEMONSTRATION PILOT PLANT
( FIGURE 9 )
-------
particles to remain near the periphery of the flow rather
than being mixed thoroughly with the liquid as it passed
through the pipe. Additionally, some of the duplication
difficulties were undoubtedly due to rapidly changing waste
characteristics.
Laboratory analysis of the samples were performed to
ascertain the following:
1) pH.
2) Turbidity.
3) Total suspended solids.
4) Volatile suspended solids.
5) Total solids .
6) Total volatile solids.
7) Total nitrogen.
8) Total phosphates.
9) Biochemical oxygen demand (BOD).
The laboratory analyses were accomplished using the
methods outlined in "Standard Methods for the Examination of
Water and Waste Water," 12th Edition, published jointly by
the American Public Health Association, the American Waterworks
Association, and The Water Pollution Control Federation (32).
Laboratory quality control procedures suggested by the Taft
Engineering Center, FWPCA, Cincinnati, Ohio, were used.
Two modifications added significantly to the success of
the demonstration. These included (1) change of the point of
45
-------
influent selection from the Fort Smith sewage distribution
box to eliminate much of the industrial waste, and (2) the
addition of a baffle plate in the flotation cell to eliminate
large bubbles.
The wastes arriving at the distribution box at Fort
Smith's sewage disposal plant consisted of a mixture of
industrial wastes and domestic sewage. Laboratory analyses
showed wide variations in both pH and total suspended solids.
Some of these variations are illustrated in the appendix. The
industries discharging waste into the sewage system at Fort
Smith include a fertilizer plant, packing houses, a slaughter-
house, a major appliance manufacturer, and several metal
plating and fabricating shops. These industries cause Fort
Smith sewage to vary quite drastically from domestic sewage
in both physical and chemical makeup. At times acid wastes
reduced the pH to a value of 3.2. Heavy intermittent loads
of hair, blood ,and animal greases were also noted.
Wastes from two Fort Smith collection systems, Mill
Creek and "P" Street, were mixed in the distribution box.
The Mill Creek sewage main carried primarily domestic wastes,
but the major appliance manufacturing plant and the slaughter-
house also discharged their wastes into the Mill Creek system.
The "P" Street collection system contained a mixture of
domestic wastes and heavy industrial wastes which was charac-
terized by a high percentage of nonvolatile suspended solids
and a widely varying pH.
46
-------
Wastes flowed from the distribution box to the Fort Smith
clarifiers when gates "A" and "D" were opened and also flowed
to the demonstration plant when gates "B" and "C" were opened
(Figure 10). To prevent the "P" Street wastes from entering
the demonstration plant the ends of two 4-inch pipes were
inserted deep into the Mill Creek inlet. The other ends of
the pipes were passed through two flanged holes in a 12-inch^
wide steel plate installed beneath gate "B" (Figure 11).
This plate raised gate "B" so its top edge was 12 inches
above the top of closed gate "A". The mixture of Mill Creek
and "P" Street wastes overflowed gate "A" to the clarifiers
while the hydraulic head thus produced forced Mill Creek wastes
through the 4-inch pipes into the demonstration plant. The
modification was effective in eliminating "P" Street wastes
from the demonstration plant influent.
The air flotation cell as originally built permitted
large bubbles of air to rise through the liquid and disrupt
the mat of floating solids. A baffle plate was installed above
the liquid inlet nozzles to trap and vent these bubbles (Figure 12)
It should be possible to accomplish the venting by inverting
the inlet header to the air flotation cell so that the waste
liquid exits the inlet header from the bottom rather than the
top. The large bubbles of air would then rise to the top of
the inlet header where they could be vented with a 1/2-inch
pipe. Figure 13a shows the inlet header as built; Figure 13b
shows the recommended modification.
47
-------
SLIDE GATES
GATE 'D1
TO FT SMITH
CLARIFIERS
GATE 'A'
OVERFLOW TO
ARKANSAS RIVER
/—GATE 'C'
TO DEMONSTRATION
PLANT
GATE 'B1
MILL CREEK
INLET
>" STREET
INLET
>LAr
DISTRIBUTION BOX AT THE FT SMITH V STREET
POLLUTION CONTROL FACILITY
FIGURE 10
48
-------
OVERFLOW TO
ARKANSAS RIVER
GATE 'D1
TO FT. SMITH
CLARIFIERS
GATE A'
GATES
GATE 'C1
TO DEMONSTRATION
PLANT
GATE 'B'
STREET
INLET
MILL CREEK
INLET
PLAN
MODIFICATION OF DISTRIBUTION BOX AT FT SMITH
"P" STREET POLLUTION CONTROL FACILITY
WASTE TO
FT. SMITH
CLARIFIERS
LIQUID LEVEL
WASTE TO
DEMONSTRATION
PLANT
\
STEEL PLATE
MILL CAPPED "P"
CREEK INLET STREET INLET
SECTION A-A
FIGURE 11
49
-------
AIR DISCHARGE
INFLUENT
TOP SCRAPER
;^,'.:':.'.'-ZlLIQUID LEVEL;
^•^i^-: :.'-•••••//.•;:.••.
'"'I'-"'.r^^^-BAFFLE PLATE
BOTTOM
DRAIN
FIGURE 12
50
-------
INFLUENT FROM AIR
DISSOLVING TANK AND
PRESS. CONTROL VALVES
VENT
/INFI
(a) INLET HEADER AS BUILT
INFLUENT FROM AIR
DISSOLVING TANK
AND PRESS. CONTROL
VALVES
(b) SUGGESTED DESIGN FOR INLET HEADER
FIGURE 13
51
-------
During commissioning and start-up activities in the fall
of 1967 it was noted that rain beat some of the particles down
out of the floating mat; extremely high winds had a similar
effect. To protect the foam, a Visqueen cover was installed
over the entire air flotation tank. Location of a dissolved-
air flotation unit so as to take advantage of the protection
offered by already existing walls and cover should be considered.
Photographs of the demonstration plant appear in Appendix
A. Appendix B shows the Fort Smith drainage area, and Appendix
C is a resume of construction costs.
52
-------
SECTION VIII
TESTING AND EVALUATION
-------
TESTING AND EVALUATION
Two groups of tests were scheduled for completion at the
Fort Smith demonstration plant. The first group, called the
basic data collection tests, were selected to perfect operating
techniques and parameters for the plant. The tests included
operation of the plant with various air dissolving pressures,
determination of optimum air feed rates, and the determination
of optimum waste flow rates. Various chemical flocculating
agents were tried in jar tests. Results of the jar tests were
later used in determining the best chemicals for use in the
demonstration plant. During the period of basic data collection,
sampling and laboratory analysis techniques were evaluated.
Retention time studies using tracer dyes were also performed.
Upon completion of the basic data collection tests the
second group of tests were scheduled. The second group of
tests included:
1) Equipment testing.
2) Chemical testing.
3) Rain event testing.
Rain event testing had precedence over all other testing;
arrangements were made so that personnel were on call whenever
a storm event occurred, even if this event occurred after
normal working hours or during weekends.
53
-------
Table 2 lists the removal percentages of the various
waste components when different equipment combinations were
used. No chemical aids to flocculation were used during this
series of tests.
As indicated in Table 2, the equipment combinations in
this phase of operations were: (1) All the separatory
equipment including screen, four cyclones, and air flotation
tank; (2) The screen and air flotation tank; (3) The screen,
primary cyclone, and air flotation tank; (4) The screen, three
secondary cyclones, and air flotation tank; (5) The screen, two
secondary cyclones, and air flotation tank. A one-way analysis
of variance was performed on the results. The computations and
resulting analysis are shown in Tables D-5, D-6, and D-7,
in Appendix, D, pages D-48 through D-58.
These analysis show that:
1) There is a statistically-significant difference
between the suspended solids removal rates. The
best removal rates were obtained when all the
separatory equipment was in use and when two
secondary cyclones and the flotation tank were in use.
2) Any differences between the rates of BOD reduction are
due to chance, and changes in auxiliary separatory
equipment do not significantly affect BOD reduction
rates.
-------
TABLE 2
PERCENT REMOVAL OF SEWAGE COMPONENTS WHEN
DIFFERENT EQUIPMENT COMBINATIONS WERE USED
(NO CHEMICAL FLOCCULANTS)
Equipment Used
All Equipment
Screen &
Flotation Cell
Screen, Primary
Cyclone & Flo-
tation Cell
Screen, 3 Sec-
ondary Cyclones
& Flotation Cell
Screen, 2 Sec-
ondary & Flo-
tation Cell
Removal of
Suspended
Solids
X
62
49
49
53
65
95% C.I,
59 to 65
42 to 57
43 to 55
46 to 70
57 to 73
Reduction
of BOD
X~
26
27
35
36
41
95% C.I.
18 to 34
3 to 65
16 to 54
15 to 57
5 to 80
Removal of
Total Solids
X
17
27
16
23
23
95% C,I.
7 to 27
0 to 70
0 to 65
13 to 33
20 to 26
Removal of
Total
Phosphorous
X
J
:
\
I
.4
95% C.I.
t
8 tc
,
\
3 20
f
Removal of
Total Nitrogen
X
i
L
,
[
i
f
95% C.I.
i
3
,
i
to 6
r
X * Arithmetic Mean
95% C.I. « 95% Confidence Interval
55
-------
Because total nitrogen and total phosphorus content in
the waste treated at the demonstration pilot plant was primarily
due to dissolved solids, a cursory examination of the data is
sufficient to indicate that there is no significant difference
in the removal rates of these components in the various
operational modes. Further examination shows that the total
nitrogen phosphate removal was 14 percent; Table 2 indicates
the 95 percent confidence intervals for the removal rates in
both cases.
The results indicate that the dissolved air flotation
system is capable of removing up to 65 percent of suspended
solids after the influent waste has been screened to remove gross
solids.
The BOD reduction varies from 26 to 41 percent with a
mean of 33 percent. This compares favorably with the removal
by conventional primary treatment plants. The BOD and total
solids removal rates can be attributed to the removal of
suspended solids in the influent waste.
A chemical testing program was accomplished. Letters
of inquiry sent to various chemical companies brought offers
of technical assistance in the initial testing of the chemical
additives. Jar tests were used to reduce the wide field of
possibilities, and the most promising chemicals were tried in
conjunction with each other in attempts to achieve even better
results. Results of the jar , tests were applied to full scale
demonstration plant operation.
56
-------
The chemical companies volunteering to participate in the
program included:
1) Calgon Corporation.
2) Dow Chemical Company.
3) Drew Chemical Corporation.
4) Pennsalt Chemicals Corporation.
5) Tretolite Division of Petrolite Corporation.
Letters of inquiry to several other manufactures of waste
water chemical additives brought no response.
In almost all cases, the chemical additives were used as
"polishing agents" to improve the performance of the alum or
lime used as the primary or main flocculant. The data obtained
are by no means exhaustive. In some cases not enough chemical
additive was available for extensive testing. In other cases
equipment problems, corrosion, and plugging prevented attempts
to inject the chemicals into the waste stream.
Chemical feed rates were usually determined by first
adjusting the alum feed rate to give the least turbid effluent,
then adding increments of polishing agent chemicals to further
decrease turbidity. Effects of feed rate changes were apparent
in the effluent waste stream within 10 minutes of the change.
Because of the extremely variable strength and pH of the influent
waste, it was impossible to maintain optimum chemical feed
rates for longer than 1/2 hour. Operating procedure was
57
-------
to determine optimum feed rates at the start of a test run
and continue the test without changes in this feed rate.
Some polishing agents exhibited a synergistic effect
in combination with alum so that the alum feed rate could be
reduced. Ferric chloride is an example.
Data obtained for ferric chloride and a combination of
ferric chloride, alum, and Tretolite FR-50 (a polyelectrolyte)
are the result of a rather limited testing program. The
extremely corrosive ferric chloride made extensive testing
impossible. Ferric chloride was fed to the system by
siphoning it from plastic-lined drums into Tank 1. When tests
using lime were performed, lime was also introduced into the
waste stream by siphoning into Tank 1. Tests using alum and
the polyelectrolytes indicated that the best point of injection
was after the air dissolving tank. This may well have been
the case with ferric chloride and lime also, and better results
might have been obtained if they had been injected after the
air dissolving tank.
T^ble 3 lists the results obtained when various chemical
flocculation aids were used during periods of no rain. In
all cases all the separatory equipment was in use except as
noted.
Chemical feed rates varied, as previously noted, but
ranged as follows:
58
-------
TABLE 3
PERCENT REMOVAL OF THE VARIOUS COMPONENTS WHEN
DIFFERENT CHEMICAL TREATMENTS WERE USED. ALL
MECHANICAL SEPARATORY EQUIPMENT WAS ON STREAM.
Chemicals Used
No Chemicals
Alum Only
Alum + Tretolite
FR-50
Alum + Dow
SA118.1A
Removal of
Suspended Solids
X
62
64
69
93
95% C.I.
59 to 65
22 to 100
59 to 80
88 to 98
Reduction of
BOD
X
26
47
53
63
95% C.I.
18 to 34
22 to 71
39 to 66
45 to 81
Removal
Total Solids
X
17
19
34
31
95% C.I.
7 to 27
9 to 30
16 to 59
16 to 46
Removal of
Total Phosphorous
X
29
53
43
34
95% C.I.
0 to 100
29 to 77
27 to 59
12 to 56
Removal
Total Nitrogen
x"
13
5
5
7
95% C.I.
10 to 16
2 to 8
2 to 8
0 to 10
"There is insufficient data for full statistical analysis of the following results:
FeCl3 Only
FeCl3 + Alum +
Tretolite FR-50
Alum + Tretolite
FR-50 (Screen, 3
Secondary Cy-
clones & Flota-
tion Tank)
90
95
89
84 to 96
89 to 100
70 to 100
42
86
68
8 to 76
53 to 100
25 to 100
19
58
60
0 to 65
15 to 95
15 to 100
73
80
30 to 100
35 to 100
6
27
0 to 50
0 to 45
X = Arithmetic Mean
95% C.I. = 95% Confidence Interval
59
-------
1) Alum - 15 mg/1 to 175 mg/1.
2) Tretolite FR-50 - 1 mg/1 to 30 mg/1.
3) Dow SA1188.1A - 1 mg/1 to 10 mg/1.
4) Ferric Chloride - 10 mg/1 to 60 mg/1.
Computations and the resulting analyses are shown in
Appendix D, pages D-70 through D-88 and in Tables D-8, D-9 ,
D-10, D-ll.
The analyses show that:
1) Of the chemicals tried, alum plus Dow SA1188.1A
provides the most effective treatment for suspended
solids removal.
2) There is no apparent statistical difference between
the BOD reduction rates, the total solids removal
rates, the total phosphorus and the total nitrogen
removal rates for the chemical treatments listed in
Table 3.
60
-------
The demonstration plant was operated during every rain
event with total precipitation of 0.1 inch or more. Results
of these operations are shown in Table 4. Chemical feed rates
were varied to yield the least turbid effluent and then left
at that rate for the remainder of the storm event or sample
period. Typical feed rates were: (1) alum - 5 mg/1 to 175
mg/1; (2) Tretolite FR-50 - 1 mg/1 to 30 mg/1.
Note that Table 4 shows that treatment with alum resulted
in poorer removal rates than no treatment at all. However,
there were so few rain events that the computed means have a
wide confidence interval (essentially, there can be only a
very low confidence in the answer). The statistical analyses
found in Appendix D, Tables D-l, D-2, D-3 , and D-4, pages D-21
through D-38 bear this out
The analyses show that:
1) Suspended solids removal during storm events is not
a function of the treatment or chemicals used.
2) There was no significant difference in BOD reduction
between treatments using no chemicals and treatments
using alum. However, BOD reduction during the
operations using alum plus Tretolite FR-50 was
significantly better than the other two treatments.
3) Alum plus Tretolite FR-50 was significantly better
in reducing total solids than was treatment with
alum or without chemicals.
61
-------
TABLE 4
PERCENT REMOVALS OF THE VARIOUS COMPONENTS DURING RAIN EVENTS.
ALL MECHANICAL SEPARATORY EQUIPMENT WAS ON STREAM.
Modes of Operation
No Chemicals
Alum Only
Alum + Tretolite
FR-50
Removal of
Suspended
Solids
X
69
56
84
95% C.I.
40 to 98
34 to 78
82 to 86
Reduction
of
BOD
X
40
35
73
95% C.I.
8 to 76
9 to 61
67 to 79
Removal of
Total
Solids
JC
24
36
52
95% C.I.
11 to 57
23 to 48
43 to 60
Removal of
Total
Phosphorous
X
48
30
74
95% C.I.
6 to 80
2 to 72
56 to 92
Removal of
Total
Nitrogen
X
4
6
95% C.I.
0 to 50
0 to 45
X = Arithmetic Mean
95% C.I. = 95% Confidence Interval
62
-------
4) Removal rates of total phosphorus were unaffected by
chemical treatment or lack of chemical treatment.
5) There was insufficient data concerning total nitrogen
removals to make any analysis.
A resume of reductions in suspended solids, BOD,
and total solids appears in Table 5.
Table 6 lists some of the various components of Fort
Smith sewage during both dry weather and rain events and
compares them to the content of a typical medium strength
sewage (33, 34) .
During dry weather the total solids content of Fort
Smith sewage is about 25 percent less than typical waste,
however, the organic content of each waste is nearly the same.
Fort Smith's dry weather sewage contains a greater percentage
of suspended solids, but again the percentages of volatile
content are much the same.
The high phosphate content of dry weather sewage may be
due to the discharge of waste from a fertilizer plant into
the Mill Creek force main. A satisfactory explanation for the
low total nitrogen content could not be found.
During rain events both total solids and suspended solids
content increased. In many other cities solids concentration
decreases during rain events. The organic fraction decreased
for both total and suspended solids. Dilution of phosphate
content was also observed.
63
-------
TABLE 5
PERCENT REMOVAL OF SEVERAL COMPONENTS WHEN
DIFFERENT EQUIPMENT COMBINATIONS WERE USED
(No Chemicals)
Equipment Used
Screen i Flotation
Cell
Screen, Primary
Cyclone & Flo-
tation Cell
Screen, 3 Secon-
dary Cyclones &
Flotation Cell
Screen, 2 Secon-
dary Cyclones &
Flotation Cell
Removal Total
Suspended Solids
X
49
49
53
65
95% C.I.
42 to 57
43 to 55
46 to 70
57 to 80
Reduction
BOD
ic
27
35
36
41
95% C.I.
3 to 65
16 to 54
15 to 57
5 to 80
Removal
Total Solids
X
27
16
23
23
95% C.I.
0 to 70
0 to 65
13 to 33
20 to 26
PERCENT REMOVAL OF SEVERAL COMPONENTS
WHEN DIFFERENT CHEMICAL TREATMENTS WERE USED.
ALL MECHANICAL SEPARATORY EQUIPMENT WAS ON STREAM.
Chemical Used
No Chemicals
Alum Only
Alum + Tretolite
FR-50
Alum + Dow
SA1188.1A
Removal Total
Suspended Solids
X
62
64
69
93
95% C.I.
59 to 65
22 to 100
59 to 80
88 to 98
Reduction
BOD
x"
26
47
53
63
95% C.I.
18 to 34
22 to 71
39 to 66
45 to 81
Removal
Total Solids
X
17
19
34
31
95% C.I.
7 to 27
9 to 30
16 to 51
16 to 46
There is insufficient data for full statistical analysis of the
following:
Fed, Only
FeCl, + Alum +
Tretolite FR-50
*Alura + Tretolite
FR-50
90
95
89
84 to 96
89 to 100
70 to 100
42
86
68
8 to 76
53 to 100
25 to 100
19
58
60
0 to 65
15 to 95
15 to 100
Primary Cyclone not used in this instance.
PERCENT REMOVAL OF SEVERAL COMPONENTS
DURING RAIN EVENTS.
ALL MECHANICAL SEPARATORY EQUIPMENT WAS ON STREAM.
Modes of Operation
No Chemicals
Alum Only
Alun + Tretolite
FR-50
Removal Total
Suspended Solids
X
69
56
84
95% C.I.
40 to 98
34 to 78
82 to 86
Reduction
BOD
X
40
35
73
95% C.I.
8 to 76
9 to 61
67 to 79
Removal
Total Solids
3c
24
36
52
95% C.I.
11 to 57
23 to 48
43 to 60
Tables 2, 3, 4, and 5 contain abbreviations for
several statistical terms. They are(
X-Ex£
TT~
Where X - sample arithmetic mean,
x. - experimental values, and
N « number of values.
95% C.I. - The 95% Confidence Interval. The data
indicates there is a 95% probability
(chance) that the true or population
arithmetic mean lies between these two
values inclusive. Alternatively, there
is a 5% probability that the true mean
lies outside the given set of values.
64
-------
TABLE 6
COMPARISON OF SEWAGE STRENGTHS *
Component
Total Solids
Total Volatile Solids
Total Volatile Solids x 100
Total solids
Suspended Solids
Volatile Suspended Solids
Suspended Solids v inn
Total Solids
Volatile Suspended Solids Y -\ nn
Suspended Solids
BOD
Total Nitrogen
Total Phosphate
pH
Turbidity
Fort Smith Sewage
Dry
Weather
mg/1
621
349
56%
272
195
44%
72%
174
18
40
7.0
180.0
J.U.
Rain
Events
mg/1
880
396
45%
534
273
61%
51%
212
16
28
7.0
231.0
J.U.
Typical Medium
Strength Sewage
(33, 34)
mg/1
880
420
52%
200
135
25%
68%
210
40
10
*From the above table, it can be seen that wet weather
solids content is higher than the dry weather content. This
fact does not support opinions that bypassing during rain events
constitutes but a minor pollution problem because wastes are
weak and diluted. If the above data is typical of storm weather
flows from many municipalities, the importance of controlling
excess flows, rather than bypassing, becomes more apparent.
65
-------
Because the demonstration plant was contracted to test
the feasibility of operation during storm events, several tests
were made to determine the time necessary for start-up. These
tests were performed during storm events as well as during dry
weather operation.
Starting activities included:
1) Starting the air compressor to build up enough
pressure to close the dump valves on the cyclones.
2) Closing the drain valves at the bottom of the air
dissolving tank.
3) Closing Gate "A" and opening Gate "C" in the Fort Smith
sewage disposal plant distribution box.
4) Starting Pump P-l.
Average time between arrival of operating personnel at
the plant site and start-up was 2 minutes. In all tests,
Tank 1 was partially filled. If Tank 1 were empty at the
start of the tests, one minute additional time would be
necessary.
66
-------
SECTION IX
ADDITIONAL TESTING OF CHEMICAL
AIDS TO FLOCCULATION
-------
ADDITIONAL TESTING OF CHEMICAL
AIDS TO FLOCCULATION
Tables 7, 8, and 9 show the results of testing of
additional chemical aids to flocculation. The data collected
for inclusion in these tables are insufficient for inclusion
in Table 5. However, the data are indicative of the ability
of these various chemicals to aid the removal of suspended
solids and to reduce BOD. In many of the tests included in
the following tables, grab samples were used as opposed to
the composite samples which were used to compile the data
for Table 3. Some rain events occurred during the tests. The
data which include percent reduction in BOD are the result of
composite sampling.
Chemical feed rates varied widely and are a function of
influent waste strength and pH. Reduction of turbidity was
initially used as the basis for chemical feed rate adjustment.
However, little correlation could be found between turbidity
and suspended solids due to the widely varying influent waste
characteristics.
Justification for chemical treatment depends largely upon
effluent water quality specifications. Data obtained during
the demonstration of the dissolved-air flotation unit indicate
67
-------
TABLE 7
ADDITIONAL CHEMICAL TESTS
Chemical Aids
75 mg/1 Alum + 25 mg/1 Dow SA1188.1A
75 mg/1 Alum + 50 mg/1 Dow SA1188.1A
75 mg/1 Alum + 50 mg/1 Dow SA1188.1A
75 mg/1 Alum + 1/4 mg/1 Dow A23
75 mg/1 Alum + 1/2 mg/1 Dow A23
75 mg/1 Alum = 1 mg/1 Dow A23
75 mg/1 Alum + 2 mg/1 Dow A23
75 mg/1 Alum + 2 1/2 mg/1 Dow A23
75 mg/1 Alum + 8 mg/1 Drew Floe 400
60 mg/1 Alum + 4 mg/1 Drew Floe 410
Suspended
Solids
% Removal
97
96
86
73
56
62
88
90
78
91
BOD
% Reduction
--
—
76
—
--
—
--
57
68
68
-------
TABLE 8
ADDITIONAL CHEMICAL TESTS
Treatment
Lime (mg/1)
150
150
150
150
150
150
0
75
100
50
100
100
100
100
???
???
Drew Floe
Number
400
400
400
400
410
410
410
410
410
410
410
410**
410**
410
410
mg/1
0
2.5
1
5
2
5
5
5
5
5
1
2
2
10
2
1
Suspended Solids
Plant
Influent
mg/1
260
226
206
268*
242*
205*
32
116
80
56
66
98
200
274
322
264
Plant
Effluent
mg/1
100
58
54
126
126
94
26
54
66
54
64
60
90
88
78
70
Removal , %
62
74
74
53
48
54
19
53
18
4
3
39
55
68
76
74
* Blood present in the influent stream.
** Drew Floe injected before air dissolving tank.
??? Measuring equipment inoperative, feed rate unknown, data
is included to indicate the potential of the polishing
chemical.
Lime was injected immediately before the hydrocyclones except
for the two cases marked ??? .
Drew Floe was injected immediately after the air dissolving
tank, except as noted.
69
-------
TABLE 9
ADDITIONAL CHEMICAL TESTS
Alum
mg/1
0
0
0
75
50
75
100
100
125
125
Calgon
ST 25*
mg/1
40
50
50
0
0
0
0
0
0
30
Calgon
St 266*
mg/1
25
25
30
0
10
10
15
20
20
25
20
Turbidity
Removal ,%
52
44
50
17
7
-
41
45
72
64
-
Suspended
Solids
Removal , %
61
58
73
48
42
21
42
45
56
65
83
* ST 266 is an anionic polyelectrolyte; ST 25 is a clay.
70
-------
that its effluent waters were often of secondary treatment
plant quality when chemical aids were used. The possible
savings in equipment and construction costs made possible by a
dissolved-air flotation system and chemical aids suggest their
consideration as alternatives to secondary treatment.
Costs for various chemical treatments are listed in Table
10. The chemicals are listed alphabetically, and the suggested
feed rates are those which gave best removal rates under the
influent waste conditions existing during the test. No
conclusions have been made. Freight expenses have not been
included. Unit costs vary with quantity ordered; the minimum
order varies from single 55 gallon drums to 5,000 pound lots.
An average specific gravity of 1.01 for waste water was
used in calculating the costs in Table 10. Rates are given in
terms of cost per million gallons of waste rather than in
cost per pound of dry solids.
Interviews with several filling station operators in the
Fort Smith area led to the conclusion that used crankcase oil
is often disposed of (illegally) by pouring it into the floor
drains in the service station or into nearby storm water catch
basins. Several chance observations bore out this fact. In
order to determine the effectiveness of the dissolved-air
flotation system in removing the oil washed through the combined
sewers during the first surge of a rain event, it was necessary
to inject oil directly into the flow stream of the demonstration
71
-------
TABLE 10
CHEMICAL TREATMENT COSTS
Chemical Used
And Feed Rate,
mg/1
Dry Alum — 75
Liquid Alum - 75
Calgon ST266-20
+ Calgon ST25 - 30
Dry Alum - 75
+ Dow SA1188.1A - 25
Dry Alum - 75
Dow A23 - 2-1/2
l
Dry Alum - 60
Drewfloc 410 - 4
Dry Alum - 75
Drewfloc 400 - 8
Dry Alum - 30
Anhydrous Ferric
Chloride - 30
Tretolite FR-50 - 4
Anhydrous Ferric
Chloride - 56
Alum - 75
Tretolite Fr-50 -
15
Unit Cost
/lb
4
1.9
9. 75
50
4
21
4
31
4
20
4
30
4
10
13.7
10
13.7
Removal
Total
Suspended
Solids, %
64
64
83
97
90
91
78
95
90
69 to 84
Cost $ per
Million Gallons
of Waste
25.01
11.88
141.30
68.77
31.47
26.67
45.01
39.58
46.68
110.66
72
-------
plant. This was done by using both chemical feed pumps and
by pouring oil into the suction stream of pump P-l.
Several tests were performed in which oil was injected
into the waste stream or dumped into Tank 1. Little or no
oil was visible in the effluent from the air flotation tank,
Tests in which analyses were performed confirmed the visual
observations. Results are shown in Table IT.
TABLE 11
OIL REMOVAL TEST
mg/1 Oil Injected mg/1 Oil in the
into the System Plant Effluent
Blank 0.6
100 0.6
200 0.6
300 0.6
The oil used for this test was SAE 30 motor oil which
had previously been used as a break-in oil for motor vehicles,
The oil was injected into an influent waste stream containing
slaughterhouse wastes as indicated by the presence of paunch
wastes and blood. Table 11 indicates that all the injected
oil was removed from the system. It is probable that the oil
which was not removed was emulsified or dissolved oil and
grease from the slaughterhouse operation. The analytical
73
-------
procedures used for the determination of oil involved use of
toluene as an extractant. Colorimetric methods were used to
analyze the toluene bearing the extracted oil.
74
-------
SECTION X
COMPONENT PARTS PERFORMANCE
-------
COMPONENT PARTS PERFORMANCE
S creen
The initial design of the pilot demonstration plant called
for the evaluation of a 3-mesh (1/4 inch) and a 6-mesh (1/8 inch)
screen. Removal rates of suspended solids by these screens
varied from 6 percent to 49 percent. Removal rates were
dependent more upon time elapsed since cleaning of the screen
than upon screen size. Solids removal was also a function of
the time of the day and waste characteristics. Almost no
screenings were collected in the early morning hours. The
solids discharge volume increased during the day, reaching a
peak in the late afternoon. Frequent manual cleaning, often
after 12 to 16 hours of operation, was necessary to prevent
total clogging. Clogging was caused by the stapling of fibers
and hair over the wires of the screen.
The hair and fiber load was so heavy that addition of
plastic cleaning rings recommended by the screen manufacturer
proved ineffective. The rate of stapling was so rapid that
the cleaning rings became entangled and immovable soon after
the cleaned screen was placed in operation.
The manufacturer supplied a 32-mesh screen at his own
expense for evaluation to replace the 3- and 6-mesh screens.
The 32-mesh screen provided markedly improved solids removal
and was in use for one month during which time it was never
75
-------
cleaned. When the screen was removed for return to the
manufacturer, it was still clean with no evidence of
stapled hair or fibers.
The 32-mesh screen removed between 13 percent and
61 percent of the suspended solids. There are insufficient
data to state a statistically significant difference between
these removal rates and those for the 3- and 6-mesh screens.
However, visual observations indicated that the 32-mesh
screen was far superior to the 3- and 6-mesh screens. Also,
the 32-mesh screen removed a volume of the solids so great
that screenings had to be continuously shoveled from the solids
collection box and wheelbarrowed to Fort Smith's grit disposal
system.
The removal of this tremendous quantity of solids
improved the capacity of Pump P-l to the point where it was
necessary to bypass part of the flow to maintain a flow rate
of 350 GPM.
The screen is essential for the treatment of combined
sewage; proper mesh size is important. For other applications,
the screen may eliminate need for the cyclones.
76
-------
Cyclones
The demonstration pilot plant was designed with one
primary cyclone and three secondary cyclones. Results as
indicated in Table 12 show that a flow of 350 gallons per
minute, two 10-inch cyclones are sufficient. Cyclones provide
maximum efficiency in removing total suspended solids when
liquid flow is near capacity. Provision should be made in
the design of future systems so that additional cyclones can
be readily added if the through-put of the system is increased.
Table 12 lists the pressure differentials across the
cyclones for varying flow rates and for several combinations
of flow paths through the cyclones.
Cyclone efficiency is a function of pressure differential
across the cyclones and optimum pressure differentials indicated
by Tables 2 and 12 appear to be in the neighborhood of 20 psi.
All four cyclones were equipped with air-operated dump
valves signalled by an electrical impulse coming from a timer.
During several storm events, fine clay silt accumulated at
such a high rate that solids collection pots on all three
secondary cyclones filled and became plugged. The plant had
to be shut down and the cyclones dismantled and cleaned. If
cyclones are retained they should be designed to permit
continuous solids discharge.
77
-------
TABLE 12
SUPPLEMENTARY DATA ON PRESSURE
DROP ACROSS CYCLONES
Flow Thru
Plant,
GPM
350
350
350
350
350
350
350
385
300
250
200
200
200
Back
Pressure,
psi
50
40
30
20
50
50
50
50
50
50
50
50
50
Pressure Differential
Across Cyclones,
psi
Primary
Cyclone
22
20
20
21
20
not in use
not in use
24
14
9
3
5
not in use
Secondary Cyclones
1
10
10
11
10
not in use
20
10
12
6
4
3
15
15
2
10
10
9
7
not in use
20
10
9
6
2
0
not in use
not in use
3
10
10
10
12
not in use
not in use
7
11
8
5
3
not in use
not in use
78
-------
AIR DISSOLVING TANK
The literature indicates that a liquid retention time of
one to three minutes is desirable for dissolving tanks. The
air dissolving tank dissolved air by three separate mechanisms:
1) As the air bubbles through the liquid in the inner
stand pipe.
2) As the liquid falls through the air cap in the outer
stand pipe.
3) In the turbulence produced as the falling liquid
strikes the liquid surface at the bottom of the air
dissolving tank.
The efficiency of the air dissolving tank was tested by
comparing suspended solids removal rates with changes in air
pressure in the tank and with changes in air feed rate. A
scattergram of the results is shown in Table 13. Values in
the body of the table are percent removal rates of suspended
solids .
Additional equipment testing was scheduled to determine
the proper liquid level in the air dissolving tank. Reference
to Figure 5 shows the liquid level controller was located in
such a way that liquid level control in the upper half of the
sight glass was not possible. The results of the tests were
inconclusive; very little difference in the suspended solids
removal rate was noticed for all controllable levels of liquid
in the air dissolving tank.
79
-------
TABLE 13
THE EFFECT OF AIR FEED RATE AND PRESSURE
DIFFERENTIAL ON TSS REMOVAL, PERCENT
Pressure psi
Air Feed Rate,
cf m
20
25
30
40
Mean Values of
Suspended
Solids Removed
60
73
58
65
50
64, 43
62
69
54
40
58
67, 51,
50, 28
51
30
53, 34
43
20
43
43
Mean Values of
Suspended
Solids Removed
54
54
51
69
fin
-------
The Relationship Between Suspended Solids Removal
Efficiency and Pressure Differential
Suspended
Solids
Removal,
Percent
70
60
40
3O
10 2O 30 40 50 60 70
AP
psi
FIGURE 14
Figure 14 illustrates that suspended solids removal
efficiency approaches a maximum in the neighborhood of 50
to 60 psig. Table 13 shows that there is no relationship
between air feed rates tested and suspended solids removal
rate. This is to be expected, since a surplus air feed rate
was designed into the system. Low air feed rates were not
tested.
The chemical feed system was arranged so that chemicals
could be injected into the waste stream before the cyclones,
immediately after the cyclones, or immediately after the air
dissolving tank. Various tests were run to determine the
81
-------
optimum point of injecting chemicals. In practically all cases
it was determined that best results were obtained when chemicals
were injected immediately after the air dissolving tank.
Little or no effect was noticed when chemicals were injected
before the cyclones.
There were occasions when the feed lines to the chemical
feed pumps were clogged. The addition of sight glasses or
rotameters in the chemical feed lines to show when chemicals
were being pumped might be helpful. Feed pumps capable of
handling relatively thick slurries such as might be
encountered in the feeding of lime are also desirable.
Two pressure control valves were included in the system
(see Figure 7). One valve was automatically operated by a
liquid level controller; the other was regulated manually. The
automatically operated valve worked very well. It was also
relatively easy to maintain pressure in the air dissolving tank
using the manually operated pressure control valve. Manually
operated valves should be adequate for most applications
except in remote or automatic operations.
Air Flotation Cell
Two sets of scrapers were installed in the air flotation
cell. The scrapers on the bottom of the cell were used to
scrape dense materials deposited on the bottom to a collection
channel. The upper scrapers were used to remove the floating
82
-------
foam. Both scrapers were driven by a 1/2 hp variable speed
motor which was included in the design to determine the optimum
rate of scraper travel. The bottom scraper was activated by a
chain drive and fitted with an air-operated clutch and timer to
permit intermittent operation. However, when the bottom
scraper was not kept in continuous operation, sediment deposited
on the bottom of the cell was picked up by turbulence and
carried over the exit weir with the effluent stream.
Scraper travel of 6 to 8 feet per minute yielded a foam
that was sufficiently thin to flow readily in the foam collection
hopper. If other means of removing the foam are used, such as
an endless belt or an auger, slower foam scraper speeds can be
used.
The amount of water in the foam was dependent upon two
factors related to foam scraper speed:
1) The scraper blades extended below the foam into the
waste water in the flotation tank. As the blades
moved up the foam collection ramp, water was pushed
along. At low scraper speeds (4 ft/sec or less),
water was able to trickle past imperfections in the
blades.
2) At low scraper speeds, the water in the interstices
between the foam particles had time to drain away;
a dryer foam resulted.
During most of the demonstration, foam consistency was
deliberately kept thin to avoid having to wash it from the foam
83
-------
collection trough. The mean total solids content of the foam
was 0.43 percent and varied from 0.08 percent to 3.4 percent.
During experiments to determine the maximum foam solids concen-
trations, scraper speeds of 2 to 3 feet per second were used
to yield foam consistencies of 5 percent to 7 percent. All
foam samples were collected by sampling from Tank 4 with the
mixer in operation.
Volatile foam solids varied from 24.7 percent to 83.4
percent, with a mean of 70.3 percent volatility. After standing
for several hours, the foam broke and the dense material sank
to the bottom of the sample bottle. The less dense material
floated. A layer of relatively clear water separated the two
fractions. The high volatility of the foam suggests incineration,
after dewatering, as a possible method of sludge disposal.
The air flotation cell was 29-1/4 inches deep by 20 feet
wide by 15 feet long and was divided into two cells, each 10
feet wide. The exit weir was adjusted to a liquid depth of
19 inches. The inlet nozzles entered the tank 5-3/4 inches
from the bottom so the bubble rise was 13-1/4 inches. The
effective flotation length of each cell was 12 feet. The
remaining length was used for foam collection and effluent
liquid collecting troughs (see Figure 8)- The theoretical
hydraulic retention time of each air flotation cell was 8.2
minutes. Assuming that the 5-3/4-inch layer of liquid below
the inlet nozzles is relatively quiescent, the theoretical
hydraulic retention time is 5.6 minutes. This agrees closely
84
-------
with the value of 5 minutes indicated with tracer dye tests.
The bottom scraper moved counter to the liquid flow and probably
set up a circular flow pattern with the hydraulic effect of a
still layer. The exit baffle was modified in an attempt to
increase the retention time. Modification consisted of
moving the exit baffle nearer to the exit weir and extending
it to within 2 inches of the bottom of the flotation tank.
Although the modification did not noticeably affect the
retention time, it did increase efficiency of suspended solids
removal approximately 5 percent, apparently by decreasing some
hydraulic short-circuiting in the tank.
Valves installed in the inlet leaders were adjusted to
direct all flow through one cell in an attempt to determine
the optimum flotation cell flow rate. The pumping rates were
varied from 125 GPM to 380 GPM. Design through-put per cell
was 175 GPM. Pressure differential and air feed rates were
held constant at 50 psig and 30 cfm, respectively; no chemicals
were used. Results of this test shown in Table 14 lead to
the conclusion that the flotation tank had greater capacity
than the design value; there was little or no difference
between the rates of suspended solids removal for the flow
rates used.
85
-------
TABLE 14
THE EFFECT OF FLOW RATE ON SUSPENDED SOLIDS REMOVAL
Suspended Solids
Removal
Pumping Rate Effective Flotation
Tank Throughput Rate
GPM GPM %
350-380 700-760 60
200-225 400-450 65
175* 350 62
150 300 61
125 250 66
* Design Flow Rate
To determine the minimum depth of the flotation tank, a
series of holes were cut in the exit weir of one cell of the
flotation tank. The location and size of the holes is
indicated in Figure 15.
0>
vj
t
36"
V|
L : 1 — 1
36"
H
t
36"
Normal Liquid Lew
Vil ±
=«=
t
FIGURE 15
86
-------
The lower holes were covered and sealed in order to test
the effect of lowering the liquid depth to 14-1/4 inches
(three-quarters of the designed depth of 19 inches). The lower
set of holes was used to test the efficiency of suspended solids
removal at one-half design depth. The results shown in Table
15 show a sharp drop in efficiency for the shallower
cells and fix the minimum depth in the neighborhood of 19
inches. A simultaneous test was performed using the unmodified
cell for comparison purposes.
TABLE 15
Liquid Depth, in.
EFFECTIVE FLOTATION DEPTH
Distance of
Bubble Rise, in
19(unmodified cell)
14.25
9.5
Suspended
Solids
Removal,%
13.25
8-. 5
3. 75
92.1
73.0
71.3
To aid in suspended solids removal in this test, 100 mg/1
alum and 20 mg/1 Tretolite FR-50 were used as flocculating
aids. The tests were performed during a period when there
was little variation in the influent waste stream; influent
pH was 7.1 and the influent suspended solids content was
784 mg/1.
87
-------
SECTION XI
BENEFIT-COST RELATIONSHIPS
-------
BENEFIT-COST RELATIONSHIPS
A hypothetical community was considered in order to
obtain an analysis and comparison of costs and benefit-cost
ratios. 75 rain events, averaging 4 hours each occur
annually in this community. Run-off disposal is by means of
combined sewers which overflow directly into a nearby river.
The suspended solids content of the overflow averages 534 mg/1,
which was the average value at Fort Smith.
It was assumed that the city needed to provide primary
treatment of the overflow to comply with effluent waste water
quality standards. Conventional clarifiers and dissolved air
flotation were chosen for comparison. The costs and benefits
of each method are presented in Table 16 for flow rates varying
from 1 MGD to 20 MGD.
To aid in the analysis, it has been assumed that an
overflow outfall already exists above the high water line of
the river, so the cost of delivering the waste overflow to the
treatment plant need not be considered. Land is available at
$100 per acre. Twenty-year, 5.5 percent bonds will be used
for financing. The expected life of both treatment plants is
50 years.
Evans, et al., (35) in their study of the treatment of
urban storm water run-off suggest that during storm events,
conventional clarifiers with four hours retention time can
88
-------
TABLE 16
COSTS AND BENEFITS, AIR FLOTATION ANP CONVENTIONAL CLARIFIERS
Capacity
MGD Treatment
Air
1
Conv
Air
2
Conv
Air
4
Conv
Air
8
Conv
Air
16
Conv
Air
20
Conv
Flotation
. Clarifier
Flotation
. Clarifier
Flotation
. Clarifier
Flotation
. Clarifier
Flotation
. Clarifier
Flotation
. Clarifier
Total
Installed
Cost
Including
Land
$26,380
44,520
43,270
69,435
73,085
108,225
123,440
168,950
208,495
263,550
253,135
308,465
Total
Interest
(3 5.5%
20
$29
48
47
76
80
119
135
185
229
290
278
339
yrs .
,020
,960
,600
,380
,400
,160
,800
,900
,420
,020
,460
,320
Annual
Amortized
Cost
$2,
4,
4,
7,
7,
11,
12,
17,
21,
27,
26,
32,
770
675
545
290
675
370
960
745
895
680
580
390
Annual
Operating &
Maintenance
Costs
$2,
1,
3,
1,
5,
2,
8,
2,
14,
4,
17,
5,
990
500
970
770
720
140
750
750
890
600
600
050
Lb
Suspended
Total Solids
Annual Removed
Benefit
Cost
Ratios
Cost Annually *
$5
6
8
9
13
13
21
20
36
32
44
37
,760 38,920
,175
,515 77,840
,060
,395 155,690
,510
,710 311,380
,495
,785 622,750
,280
,180 778,440
,440
6.8
6.3
9.1
8.6
11.6
11-. 5
14.3
15.2
16.9
19,3
17.6
20.8
* Lb suspended solids removed/$.
89
-------
remove 70 percent of the suspended solids. These values were
used as a basis for the design of the clarifiers, since the
removal rates approximate those attained during storm events
using dissolved-air flotation at Fort Smith.
The bases for calculating operating and maintenance costs
are:
1) Electricity @ IC/hr/H.P.
Horsepower
Capacity Flotation Conventional
MGD Units Clarifiers
1 50 1
2 95 1
4 180 1
8 350 2
16 680 4
20 840 4
2) Labor costs are the same for both air flotation units
and conventional clarifiers of equal capacity at the
rate of 4 hours per rain event for the 1-, 2-, 4-, and
8-MGD plants and 8 hours per rain event for the larger
plants. Cleanup activities are responsible for most
of the labor charges.
90
-------
3) Maintenance
Air Flotation Conventional Clarifiers
Labor 4 hours per week 1 hour per week
@ $3.00 per hour @ $3.00 per hour
Parts and 5% of initial cost 1% of initial cost
Supplies
Cost analyses are provided for the installed clarifiers
and flotation units only and do not include the costs of treat-
ing the separated solids and sludge or the effluent waste waters.
These cost factors must be included in any comprehensive cost
analysis (36).
Table 17 shows that only 0.1 as much land area is needed by
dissolved-air flotation units. This could be important at
overflow points.
The benefit-cost ratios in pounds of suspended solids
removed per dollar of annual cost, shown in Table 16, favor
dissolved-air flotation for capacities less than 8 MGD.
Figure 16 illustrates the data of Table 16 in graphical form.
If dissolved-air flotation is used for treatment,
additional savings are realized because the floated foam has
a solids content of 7 percent and a thickener will probably be
unnecessary. Flotation cell underflow, containing solids which
sink to the bottom, can be controlled to yield a low volume
sludge of 1 to 2 percent consistency. Mixing the solid screenings
with the underflow will increase the solids content of the
foam-underflow mixture. This sludge should be amenable to
91
-------
24 -
•co-
•o
01
>
o
6
o>
•o
•o
01
tJ
c
to
3
(A
20
16
12
(0
O
m
o
Dissolved-air flotation
Conventional Clarifiers
-------
TABLE 17
PHYSICAL SIZES AND LAND AREAS
REQUIRED BY CONVENTIONAL CLARIFIERS
AND DISSOLVED AIR FLOTATION UNITS
Conventional Clarifiers Air Flotation
Tank Size
Capacity
MGD
1
2
4
8
16
20
Diameter
ft
50
70
100
140
200
150
Depth
ft
11
11
11
11
12
11
Number
Required
1
1
1
1
1
2
Area
Needed
sq ft
3600
6400
12100
22500
44100
50400
Cell
Length
12
12
12
12
12
12
Size
Width
10
10
10
40
40
40
Number
of Cells
2
4 *
8 *
4 *
8 *
10 *
Area
Required
sq ft
350
350
700
2000
4000
5000
* Cells can be stacked two high to conserve space.
93
-------
direct vacuum filtration with an expected filter cake moisture
content of 70 percent (36). Final disposition of the filter
cake is dependent upon local conditions. Some typical options
include:
1) Incineration.
a . On site.
b. Trucking to off-site incinerator.
2) Burial.
3) Composting.
Alternatives to dewatering the sludge on site include
digestion on site and pumping to an existing treatment plant
for treatment. On-site digestion appears to present more
problems and is the less attractive.
94
-------
SECTION XII
POSSIBILITIES FOR AUTOMATION
AND
OTHER POTENTIAL APPLICATIONS
-------
POSSIBILITIES FOR AUTOMATION
AND
OTHER POTENTIAL APPLICATIONS
Automation Possibilities
Results of the tests performed and observations obtained
at the plant site indicate that the operation might well be
automated. In very few instances was the operator necessary.
With the use of standard, easily available equipment, the
entire operation could be automated from start-up to shut., down.
An automated unit would also be adaptable for use at remote
locations.
Detailed design of an automated unit is somewhat dependent
upon conditions and location. However, the basic premises
will remain fairly constant. The modification of the existing
Fort Smith plant will be used as an example in the discussion
of the design of an automated dissolved-air flotation system.
The design includes a method of disposing of the foam
and the solid wastes from the screen, the cyclones, and the
bottom of the flotation cell. Some of the design modifications
recommended earlier in this report have been included in this
design.
The following items are considered essential for the
modification. Figure 17 is a detailed diagram of this design.
95
-------
1) Select start up mechanism. The criteria is the
detection of increased flow to the Fort Smith municipal
treating plant.
a. A liquid level sensor set at a predetermined
level in the 12 ft concrete "P" Street interceptor
will detect rising water and will signal the start
of a sequential start-up, washing, and shut-down
cycle .
b. Manual start of the sequence. (For check out of
the unit.)
2) Revise the existing flow rate measurement and control
sys tern.
a. A flow controller signaled by the existing orifice-
flow recorder system will:
1. Actuate a pneumatic flow control valve (FCV-1)
upstream of the air dissolving tank.
2. Regulate air flow to the air dissolving tank
(T-7) by means of a pressure control valve
(PCV-3).
3. Open and close a by pass valve (BPV-1)
controlling flow to the primary hydrocyclone (C-l)
b. Control of the by pass valve (BPV-1) may be
provided through the use of signals from a liquid
level controller (LLC-1) in the liquid to influent
Tank 1.
96
-------
SCREEN
FROM MAJN
DISTRIBUTION
BOX
J
-------
3) Design a sequential turn-on, turn-off, clean-up, and
shut-down cycle.
a. An electric signal will start the air compressor
and energize all electrical control circuits.
b. Air will open shut-down valve (SDV-1) and start
the screen.
c. Air will close all dump valves.
d. A rising liquid level in Tank 1 will close a
pump switch to :
1. Start Pump P-l.
2. Start the foam and bottom scrapers.
3. Start Chemical Feed Pump 1.
e. A high-low liquid level safety shut-down will
control SDV-1 and prevent overflow of Tank 1.
f. The flow rate will be controlled by a signal from
a flow controller connected to the orifice 3 pen
recorder system (FM-1), from liquid level controller
(LLC-1), or from both. The signals will:
1. Provide throttling through flow control valve
(FCV-1).
2. Control flow through the primary cyclone by
opening by pass valve (BPV-1) when flow rates
exceed 350 GPM.
3. Control flow to one flotation cell by opening
by pass valve (BPV-2) when flow rates exceed
350 GPM.
98
-------
4. Regulate air flow to the air dissolving tank
(T-7) by means of the pneumatically operated
pressure control valve (PCV-3). Start Feed
Pump-2 .
g. An air operated valve signaled by a timer on the
chain drive turning the bottom scrapers will permit
periodic dumping of the bottom sludge in the
flotation cells.
h. A liquid level controller (LLC-2) will stop the
liquid effluent pump (P-2) in the liquid effluent
Tank 2.
i. Fort Smith Sewage Department personnel pump the
sludge hoppers to the existing clarifiers every
2 hours. Volumes of sludge and foam accumulated
in 2 hours are not expected to exceed the capacity
of the storage facilities (Tanks 3, 4 and 5).
Mixers will be controlled by liquid level
controllers.
j. When the liquid level in the 12 ft concrete inter-
ceptor falls below the predetermined height or at
the discretion of the Superintendent of the Fort
Smith Sewage Disposal Facility, the clean-up and
shut-down sequence will begin.
k. The shut-down valve (SDV-1) will close.
1. The pump switch will stop Pump P-l when the
liquid level falls in the liquid influent Tank 1.
99
-------
An override switch will keep the screen and
scrapers in operation.
m. A signal will trigger an air operated valve,
dumping fresh rinse water through the screen into
Tank 1.
n. Pump P-l will start, flushing the system with
rinse water.
o. A timer will close the fresh water valve and stop
the screen. Pump P-l will stop at the low level
signal.
p. The air compressor and scrapers will stop.
q. Dump valves will drain all lines and the flotation
cells to the liquid effluent Tank 1. When Tank
2 is empty a timer will turn off power to the
electrical control circuits and reset the air
compressor switch to repeat the sequence on
signal.
r. An electrical lockout will prevent SDV-1 from
opening when the wash-out cycle is in operation.
4) Replace the 3- and 6-mesh screens with screens of
32 mesh or smaller. During these tests, the 32-mesh
screen exhibited little or no tendency to blind
because of stapling.
5) Change the solids discharge system on the hydro-
cyclones to continuous blowdown. The use of smaller
screens will permit the removal of the automatic
dump valves and the solids pots on the hydrocyclones
100
-------
and the installation of apex valves with small
diameters. This will permit continuous blowdown of
solids with a bypass of approximately 3 percent
of the liquid flow.
6) Modify the hydrocyclone flow sequence.
To accommodate a flow rate of 700 GPM, the primary
cyclone (C-l) will be placed in parallel with the
bank of secondary cyclones (C-2, C-3, C-4). C-l will
be cut out of the circuit at flow rates of 350 GPM
or less.
7) Select a discharge system for the flotation cell
liquid effluents.
The present demonstration plant remixes all the solid
and liquid effluents. Modification of the discharge
system will permit the discharge of the separated
solids and liquids. The existing liquid effluent
pump (P-2) will be moved to the liquid effluent Tank 2.
Liquid effluent from the flotation cells will be
pumped to the exit cell of the Fort Smith distribution
box, decreasing the hydraulic load in the Fort Smith
sewage disposal plant and increasing the efficiency
of solids removal during storm events.
8) Select a disposal system for the solids collected.
a. A gravity flow line will run from mixing Tank 3
to an existing line upstream of the Fort Smith
sludge pump.
b. A stop valve would prevent back flow.
101
-------
c. A pump (P-3) installed in mixing Tank 3 will pump
the accumulated sludge and foam to the Fort Smith
sludge thickeners. The solids removed by the
screens, cyclones, and flotation cells would not
pass through the Fort Smith clarifiers. The
automated unit thereby will decrease the solids
load and further increase the efficiency of the
Fort Smith sewage disposal plant.
9) Redesign the sludge collection trough on the bottom
of each flotation cell.
The slotted pipe in the collection trough will be
removed and replaced with an inclined plane to improve
the bottom sludge collection efficiency, provide for
a more positive hydraulic sweeping action, and minimize
channeling.
10) Design a spray jet system to wash the chains and
sprockets during the clean-up cycle.
Other Potential Applications
Combined sewer and storm water overflows are not the
only source of pollution in the nation's receiving
waters. The research project discussed in this report
has suggested answers to industrial waste pollution
problems as well. Some of these are discussed below.
102
-------
One application includes use by the meat processing
industry. Fort Smith sewage contained quantities of
feathers, hair, paunch wastes, and blood. These
materials were easily removed by the plant. Most of
the blood was removed, and a very clear liquid effluent
was obtained. However, it was difficult to remove
all of the color.
Some present applications of the dissoIved-air
flotation system were discussed in Section IV.
Similar systems have been used in the petroleum
industry, and the design discussed in this report
seems particularly adaptable to oil field applications.
Low retention time plus extreme compactness make
dissolved-air flotation very suitable for use on
offshore production platforms.
Dissolved-air flotation systems are currently being
used by several of the food processing industries.
Some canneries use the air flotation system to remove
suspended solids from their process wastes. In most
cases, the air flotation cells being used are of the
old design in which retention times are approximately
one hour or longer. The design demonstrated at Fort
Smith is unique in that:
1) Air is dissolved in the entire waste flow, and
2) The retention time in the air flotation tank is
extremely short.
103
-------
Future plants might well be designed around a basic
unit with dimensions the same as a single cell of the
air flotation tank located at Fort Smith; approximately
10 feet wide with an effective inner length of 12
feet. An entire plant (0.5 MGD) could be contained
on a skid, trailer, or pad with dimensions of 10 feet
by 25 feet. If additional hydraulic capacity is
necessary, tanks could be paralleled or stacked one
above another. The area of the pad not occupied by
the air flotation tank would be used for the ancillary
equipment.
The same concept is sound for larger or smaller
flotation cell dimensions and capacities.
Recommendations for future development of dissolved-
air flotation include:
1) Further investigation using specific industrial
wastes from the ferrous and nonferrous metal
industries, packing houses, rendering plants and
slaughterhouses, and the petrochemical and
petroleum industries.
2) Design and construction of a completely automatic,
in-line plant to be used in one or more of the
above applications.
3) The construction and operation of a pilot plant
in which the specific goal would be to test various
104
-------
chemical aids to flocculation both singly and
in combination, attempting to reduce suspended
solids, BOD, total phosphates, and total nitrogen.
4) Use of the dissolved-air flotation system as the
primary treatment device in combination with
various high rate secondary devices to produce
a very high quality effluent waste water. Suggested
secondary devices include high rate trickling
filters or a rapid sand filter (to remove the
remaining suspended materials) followed by an
activated carbon unit (to remove BOD and dissolved
chemicals). This step, in turn, could be followed
by chlorination or aeration or both.
5) Use of the dissolved-air flotation unit with
chemical aids as a replacement for both primary
and secondary treatment plants.
105
-------
SECTION XIII
ACKNOWLEDGEMENTS
-------
ACKNOWLEDGEMENTS
The research team is most appreciative of the aid and
assistance given them by a host of people and organizations.
The demonstration was carried to a successful completion by
their willing and unselfish support.
Gene Andrews, Superintendent,
Water Pollution Control Facilities - Springdale, Arkansas
Robert F. Andrews - Pennsalt Chemical Corporation
Darrel L. Cornelius - Drew Chemical Company
V. Bruce Dorsett - Drew Chemical Company
Dr. Robert A. Gearheart, Professor of Sanitary Engineering,
University of Arkansas - Fayetteville, Arkansas
Everett H. Janssen - Calgon Corporation
Dr. Edwin H. Klehr, Professor of Water and Sanitary Chemistry,
Civil Engineering and Environmental Science,
University of Oklahoma - Norman, Oklahoma
Webb Minor, Superintendent,
Water Pollution Control Facilities - Russellville, Arkansas
Zack Mouradian - Southwestern Engineering Company
Carl Reames, Superintendent, and his staff,
npii street Sewage Pollution Control Facility -
Fort Smith, Arkansas
Professor George W. Reid, Director
Civil Engineering and Environmental Science
University of Oklahoma - Norman, Oklahoma
Ray A. Sierka, Graduate Student
Civil Engineering and Environmental Science
University of Oklahoma - Norman, Oklahoma
G. Wade Spencer - Pennsalt Chemical Corporation
106
-------
ACKNOWLEDGEMENTS - CONTINUED
Max I. Suchanek - Dow Chemical Company
Robert W. Taylor
Tretolite Division - Petrolite Corporation
107
-------
SECTION XIV
REFERENCES AND BIBLIOGRAPHY
-------
REFERENCES
(1) USDI, FWPCA; "Problems of Combined Sewer Facilities and
Overflows, 1967." Water Polution Control Research Series,
WP-20-11, xii - xx, 2-5, 30-39, 74-86, 163-173.
(2) Weiner, D. L., "Understanding the World of Sewage." Can-
Tex Industries, Mineral Wells, Texas, (1963), 3.
(3) "Restoring the Quality of Our Environment." Report of
the Environmental Pollution Panel, President's Science
Advisory Committee, 1965, 158-167.
(4) "Roundtable: Wastes, Separate and Combined Sewers."
Water and Waste Engineering, 5, 7, 26. (1968).
(5) Anon, "Salt Water in the Sewers." American City, 80, 12,
112 (1965).
(6) Peters, G. L. and Troemper, A. P., "Reduction of Hydraulic
sewer Loading by Downspout Removal." Journal Water
Pollution Control Federation, 41, 1, 63-81 (1969).
(7) Koelzer, Victor A., Bauer, William J., and Dalton, Frank E.;
"The Chicagoland Deep Tunnel Project - A Use of the Under-
ground Storage Resource." Paper presented at the 41st
annual meeting of the Water Pollution Control Federation,
September 23, 1968, Chicago.
(8) Overfield, J. L. , Baxter, J. K. , Crawford, H. R., and
Santry, I. W., "Increasing Sewage Flow Velocity by Using
Chemical Additives." Paper presented at WPCF Annual
Meeting September 23, 1968, Chicago.
(9) Putnicki, G. J., "FWPCA's Research and Development Program."
Water, Southwest Water Works Journal, 50, 12, 25 (1969).
(10) Hanson, C. A. and Gotaas, H. D., "Sewage Treatment by
Flotation." Sewage Works Journal, 15 2, 242-252 (1943).
(11) Gaudin, A. M., "Flotation." 1st edition, McGraw Hill
Book Co., Inc., New York, N. Y., 1932 .
(12) Norris, U. S. Patent #864, 856, September 3, 1907.
(13) Elmore, F. E., U. S. Patent #826, 411, May 11, 1907.
(14) Suhman, Kirkpatrick-Picard, and Ballot, U. S. Patent #793,
808. July 4, 1906
108
-------
(15) D'Arcy, N. A., Jr., "Dissolved Air Flotation Separates
Oil From Waste Water." Proceedings, American Petroleum
Institute, 31M, 3, 34-42 (1951).
(16) Osterman, J., "Chrysler Plant Pretreats Oil Wastes by
Flotation." Wastes Engineering, 26,2, 69, (1955).
(17) Jacobsen, S. E. and Meinhold, T. F., "Removes Oil by
Dissolved-Air Flotation." Chemical Processing, 12-14,
(February 1955).
(18) Howe, R. H. L., "Mathematical Interpretation of Flotation
for Solid-Liquid Separation." in "Biological Treatment
of Sewerage and Industrial Wastes." Reinhold Publishing
Co., Inc., 2nd edition, New York, N. Y., 1958, 241-250.
(19) Vrablic, R. R., "Fundamental Principles of Dissolved
Air Flotation of Industrial Wastes." Proceedings.
Industrial Waste Conference, Purdue University, (1959)
743-779.
(20) Eckenfelder, W. W., Jr., Rooney, T. F., Burger, T. B.,
and Gruspier, J. T., "Dissolved Air Flotation of
Biological Sludges." in "Biological Treatment of Sewerage
and Industrial Wastes,"Reinhold Publishing Co., Inc.,
2nd edition, New York, N. Y., 1958, 251-258.
(21) Rohlich, G. A., "Application of Air Flotation to
Refinery Waste Waters." Industrial and Engineering
Chemistry, 46, 3, 304 (1954) .
(22) Prather, B. V., "Will Air Flotation Remove the Chemical
Oxygen Demand of Refinery Waste Water?v Petroleum Refinery,
(May 1961), 177-180.
(23) Hopper, S. H., and McGowan, M. C., "A Flotation Process
for Water Purification." Journal of the American Water-
works Association. 44, 8, 719-726 (1952).
(24) Katz, W. J. and Geinapolos, A., "Sludge Thickening by
Dissolved Air Flotation." Paper presented at Ohio Water
Pollution Control Conference, (June 16, 1967).
(25) Masterson, E. M. and Pratt, J. W. , "Applications of
Pressure Flotation Principles to Process Equipment Design."in
"Biological Treatment of Sewerage and Industrial Wastes,"
Reinhold Publishing Co., Inc., New York, N. Y., 2nd
edition, 1958 , 232-240.
(26) Pryor, E. J., "Mineral Processing." Elsevier Publishing
Co., Amsterdam, Holland 1965 222-232.
109
-------
(27) Leniger, H. A., "General Remarks on Phase Separations
and Classification." in "Cyclones in Industry." Elsevier
Publishing Co., Amsterdam, Holland 1961 12-13.
(28) Broer, L. J. F., "Flow Phenomena in Cyclones." in "Cyclones
in Industry." Elsevier Publishing Co., Amsterdam,
Holland 1961, 33-45.
(29) Van Der Kolk, H., "Linking Cyclones in Series and Its
Effect on Total Separation," in "Cyclones in Industry."
Elsevier Publishing Co., Amsterdam, Holland, 1961,
77-88.
(30) Technical Bulletin no. 3301, "Krebs Cyclones" Krebs
Engineers, Palo Alto, California.
(31) Perry, J. H., "Chemical Engineers"Handbook." McGraw-
Hill Book Co.. New York, N. Y., 4th edition 1963, 14-1 -
14-69, 18-1 - 18-59.
(32) American Public Health Assn., Inc., "Standard Methods
for the Examination of Water and Wastewater." 12th edition,
New York, N. Y., 1965 .
(33) Sawyer, C. N., and McCarty, P. L., "Chemistry for
Sanitary Engineers," McGraw-Hill Book Co., New York, N. Y.,
2nd edition, 1967, 466-472.
(34) Steel, E. W., "Water Supply and Sewerage," McGraw-Hill
Book Co., New York, N. Y., 1960, 444-462.
(35) Evans, L. S., Geldreich, R. S., Weibel, S. R., and
Robeck, G. G., "Treatment of Urban Stormwater Runoff."
Journal Water Pollution Control Federation, 40, 5,
R162-R170 (1968).
(36) Burd, R. S., "A Study of Sludge Handling and Disposal."
USDI, FWPCA Publication WP-20-4, 130-159 (1968).
(37) Freund, J. E., Livermore, P. E., and Miller, J., "Manual
of Experimental Statistics." Prentice-Hall, Inc.,
Englewood Cliffs, N. J., 1960, 54, 123, 131.
(38) Kennedy, J. B., and Neville, A. M., "Basic Statistical
Methods." International Textbook Co., Scranton, Pa., 1964,
125, 307, Table A-6.
110
-------
BIBLIOGRAPHY
Chase, E. S., "Flotation Treatment of Sewage and Industrial
Wastes." Sewage and Industrial Wastes. 30, 6, 783-791 (1958).
Clark, J. W. and Viessman, W., Jr., "Water Supply and Pollution
Control." International Textbook Company, Scranton, Pa., 1965.
Eckenfelder, W. W., Jr., "Industrial Water Pollution Control."
McGraw-Hill Book Co., New York, N. Y., 1966.
Fahlstrom, P. H., "Studies of the Hydrocyclone as a Classifier."
Proceedings of the International Mineral Processing Congress,
Pergamon Press, London, 1963, 87-114.
Harding, J. C., and Griffin, G. E., "Sludge Disposal by Wet
Air Oxidation in a Five MGD Plant." Journal of the Water
Pollution Control Federation, 37, 8, 1134,(1965).
Hay, T. T., "Air Flotation Studies of Sanitary Sewage."
Journal of the Water Pollution Control Federation, 28, 1,
100 (1956).
Kalinske, A. A. and Evans, R. R., "Comparison of Flotation and
Sedimentation in Treatment of Industrial Wastes," in "Flotation
in Waste Treatment," "Biological Treatment of Sewerage and
Industrial Wastes," Reinhold Publishing Co., Inc., New York,
N. Y. 2nd edition, 1958, 222-231.
Marson, H. W. "The Disposal of Sewage Sludge by Combustion
with Special Reference to Fluidization Methods." Journal and
Proceedings, Institute of Sewer Purification, Part 4, 320
(1965).
McGraw, H. A., "The Flotation Process." McGraw-Hill Book Co.,
Inc., New York, N. Y., 1918, 81.
McKinley, J. B., "Wet Air Oxidation Process." Water Works and
Wastes Engineering, 2, 19, 97 (1965).
Nemerow, N. L., "Theories and Practices of Industrial Waste
Treatment." Addison-Wesley Publishing Co., Inc., Reading,
Mass., 1963.
Prather, B. V., "Development of a Modern Petroleum Refinery
Waste Treatment Program." Journal of the Water Pollution
Control Federation, 36, 1, 96-102 (1964).
Ill
-------
Rebhun, M., and Argaman, W., "Evaluation of Hydraulic Efficiency
of Sedimentation Basins." Journal of the Sanitary Engineering
Division, Proceedings of American Society of Civil Engineers.
91, SA 5, 37 (1965).
Rich, L. G., "Unit Operations of Sanitary Engineering." John
Wiley and Son, New York, N. Y., 1961.
Rich, L. G., "Unit Processes of Sanitary Engineering." John
Wiley and Son, New York, N. Y., 1963.
Simpson, G. D. and Curtis, L. W., "Treatment of Combined
Sewer Overflows and Surface Waters at Cleveland, Ohio." Paper
presented at the 41st Annual Conference Water Pollution Control
Federation, Chicago, 111., September 23, 1968.
USDHEW, Public Health Service; "Modern Sewage Treatment Plants,
How Much Do They Cost?" PHS Publication no. 1229, (1964) 14-28.
USDI, FWPCA; "Storm Water Runoff From Urban Areas, Selected
Abstracts of Related Topics." (1966).
Van der Kolk, H., "Linking Cyclones in Series and Its Effect
on Total Separation," in "Cyclones in Industry," Elsevier
Publishing Co., Amsterday, Holland, 1961 , 77-88.
Vilentin, F. H. H., "Absorbtion in Gas-Liquid Dispersions:
Some Aspects of Bubble Technology," E. & F. N. Spon, Ltd.,
London, England, 1967.
Weibel, S. R., Anderson, R. J., Woodward, R. L., "Urban Land
Runoff as a Factor in Stream Pollution." Journal of the Water
Pollution Control Federation, 36, 7, 914, (1964).
Wine, R. L., "Statistics for Scientists and Engineers."
Prentice-Hall Inc., Englewood Cliffs, N. J., 1964.
112
-------
SECTION XV
APPENDICES
A. Photographs of the Demonstration Plant
B. The Fort Smith Drainage Area
C. Construction Costs
D. Data and Calculations
E. Typical Data Obtained During Plant
Shake Down in 1967
-------
APPENDIX A
PHOTOGRAPHS OF THE DEMONSTRATION PLANT
-------
Two views of the demonstration pilot plant showing the major pieces
of equipment. The visqueen cover has been removed from the air-
flotation tank.
-------
A view of the four hydrocyclones used in the demonstration pilot
plant.
Comparison of different treatments. The first jar on the left con-
tains untreated influent waste; the remaining jars contain plant
effluents, reading left to right, no chemical treatment, 50 mg/1
Alum, 75 mg/1 Alum, 100 mg/1 Alum, 125 mg/1 Alum, 125 mg/1 Alum +
15 mg/1 Tretolite FR-50, and tap water. A heavy load
present in the influent stream.
of blood was
-------
APPENDIX B
THE FORT SMITH DRAINAGE AREA
-------
The Fort Smith drainage area is approximately 12,000
acres. About 10 percent of this total has water-impervious
covers such as streets, parking lots, houses, etc. The city
proper covers an estimated three-fourths of the area.
The 1960 U. S. Census listed the Fort Smith population
as 52,991; the 1962 population was 63,309; and the 1968
population of Fort Smith has been estimated in the neighborhood
of 70,000. The Fort Smith sewer department had an average
of 16,300 non-industrial customers in 1968 and 222 industrial
customers. Three major industries dispose of their wastes
directly to the Arkansas River.
Of the nine million gallons of potable water produced
per day, about 70 percent is used for domestic and residential
purposes. The total daily waste volume of 5.3 million gallons
is treated in two plants - North "P" Street, the location of
the demonstration plant,and Massard Creek. The Massard Creek
facility provides both primary and secondary treatment for
1.8 million gallons per day of waste estimated to be of 95
percent domestic origin. The Massard Creek treatment plant
was built as the result of an engineering study submitted to
the City of Fort Smith in 1962. The plant has a daily capacity
of ten million gallons to provide for future expansion of the
city in the Massard Creek area. The plant's current flow is
sufficient to operate only one of the two trickling filters.
There are no sludge treatment facilities other than a vacuum
-------
filter; filter solids are buried.
The North "P" Street treatment plant consists of bar
screens, primary clarifiers, degritters, sludge thickeners
and vacuum filters. Total daily waste flow averages 3.5
million gallons which is estimated at 77 percent industrial
waste. The clarifiers have a retention time of one hour and
forty-five minutes and have a design surface loading rate of
700 gallons per square foot per day.
The sewage collection system in the city consisted of
170 miles of combined and separate sewers as of January 1,
1969. At that time, there were seven pump stations in operation
with five more in various stages of construction. Upon
completion of these pumping stations, there will be two sewage
outfalls for the city, one for each of the treatment plants.
As of January 1, 1969, the City of Fort Smith had no
municipal restrictions or regulations pertaining to sewer
connections or sewage discharge rates and strengths. The State
of Arkansas Water Pollution Control Regulations are being used
in lieu of city laws.
The Fort Smith drainage area is described in Table B-l,
and a map of this area is shown in Figure B-l.
-------
TABLE B-l
THE FORT SMITH DRAINAGE AREA
The area covers about 36 square miles and includes:
T 10 N, R 27 E, Sections 9*. 10, 15, 16*, 21 *, 22
TUN, R 27 E, Sections 34* in the State of Oklahoma, and;
T 8 N, R 32 W, Sections 2, 3, 4, 5*. 8*, 9, 10, 11, 14, 15, 16, 17, 20,
21, 22, 23, 26, 27, 28, 29, 32, 33, 34, 35
T 9 N, R 32 W, Sections 21*, 22*, 26, 27, 28, 33, 34, 35 in the State
of Arkansas .
*Part of the section.
B-3
-------
0 I/?
I Mile
SCALE: MILES
C.I.= EOFT.
THE FORT SMITH DRAINAGE AREA
FIGURE B-l
-------
APPENDIX C
CONSTRUCTION COSTS
-------
CONSTRUCTION COST RESUME
FOR THE
DISSOLVED AIR FLOTATION DEMONSTRATION
PILOT PLANT AT FT. SMITH, ARKANSAS
CONSTRUCTION
Subcontractor's Fee:
Section I Civil Work
Section II Mechanical Work
Section III Electrical Work
SUBTOTAL:
Material Furnished by Rhodes
Concrete, Gravel, Sand
Reinforcement Steel,etc.
SUBTOTAL:
I - TOTAL
$ 19,700.00
$ 6,800.00
8,400.00
4,500.00
$ 1,860.25
$ 21.560.25
908.38
951.87
II MECHANICAL EQUIPMENT
(1) Flotation Cell
(1) Air Dissolving Tank
(1) Screen
(4) Cyclones
(1) Electric Control Panel
(1) Motor Control Center
(2) Sewage Pumps
(2) Chemical Feed Pumps w/100 gal. tank
(1) Air Compressor
(5) Mixers
(7) Liquid Level Controllers
(1) 3-Pen Recorder
(1) Instrument Air Dryer
(1) Flow Meter w/40" open flow nozzle
(1) Air Flow Meter
9,955.55
2,948.00
3,673.00
4,068.50
2,401.90
3,794.00
2,371.70
1,408.00
1,351.00
1,900.00
1,175.00
453.70
107.85
814.50
99.25
C-l
-------
II Mechanical Equipment (Continued)--
(4) Slide Gates $ 2,210.00
(9) Air Regulators 135.00
(1) Liquid Level Gauge 142.02
(2) Temperature Indicators 1,046.00
(24) Pressure Indicators 1,072.77
(4) Pressure Controllers 357.85
Electric Supply Material 1,856.15
(I) Flow Tube 343.20
(2) Flow Controller 102.00
(106) Valves 7,062.85
Pipe Fittings 1,736.13
II TOTAL $ 51,585.79
GRAND TOTAL $ 73,146.04
C-2
-------
APPENDIX D
DATA AND CALCULATIONS
-------
Influent pH During Dry Weather
3.2 1
6.2 1
6.4 2
6.5 4
6.6 7
6.7 5
6.8 12
6.9 5
7.0 26
7.1 18
7.2 10
7.3 11
7.4 3
7.5 2
7.8 1
8.3 1
8.7 1
TOTAL 110
X = 7.0
S = 0.3
Median = 7.0
95% Confidence Interval
6. 9 < M ±1.1
X = Sample Arithmetic Mean
S = Standard Deviation
M. - True Population Arithmetic Mean
D-l
-------
Influent Turbidity During Dry Weather
Turbidity
Jackson Units £
500 - 549 1
450 - 499 1
400 - 449 1
350 - 399 3
300 - 349 7
250 - 299 6
200 - 249 26
150 - 199 17
100 - 149 33
50 - 99 10
0-49 8
TOTAL 113
X = 180 J.U.
S = 99 J.U.
Median = 177 J.U.
95% Confidence Interval (37)
178 * M £ 182
X = Sample Arithmetic Mean
S = Standard Deviation
Jil - True Population Arithmetic Mean
D-2
-------
Influent Suspended Solids and Volatile Suspended Solids Concentrations
During Dry Weather
Suspended Solids
mg/1
Volatile Suspended Solids
mg/1 f
900-999
800-899
700-799
600-699
500-599
400-499
300-399
200-299
100-199
3
1
3
2
4
2
20
34
25
350
325
300
275
250
225
200
175
150
125
100
75
2
3
0
1
8
11
5
9
4
2
0
4
TOTAL
112
TOTAL
49
X = 272. 3 mg/1
S = 201 mg/1
Median = 239 mg/1
95% Confidence Interval
269. 7 £ JLL ^ 274. 9
X = 195. 1 mg/1
S= 14.5 mg/1
Median = 202 mg/1
95% Confidence Interval
194. 0 £JUL ± 196. 2
The table does not include suspended solids for two isolated events during which
the concentrations were 1297 mg/1 and 1140 mg/1.
D-3
-------
Influent BOD Concentrations During Dry Weather
320
300
280
260
240
220
200
180
160
140
120
100
80
60
40
20
0
BOD
mg/1
- 339
- 319
- 299
- 279
- 259
-239
- 219
- 199
- 179
- 159
- 139
- 119
- 99
- 79
- 59
- 39
- 19
TOTAL
1
2
4
4
6
5
3
10
9
11
7
1
3
2
3
4
1
76
X = 174.5 mg/1
S = 75.8 mg/1
Median = 174 mg/1
95% Confidence Interval
168.3 ±JUL ± 180.7
X = Sample Arithmetic Mean
S = Standard Deviation
JH = True Population Arithmetic Mean
D-4
-------
Influent Total Solids and Total Volatile Solids Concentrations During
Dry Weather
Total Solids
Total Volatile Solids
mg/1
900-999
800-899
700-799
600-699
500-599
400-499
300-399
200-299
100-199
TOTAL
F
5
8
8
17
17
8
1
2
3
69
mg/1
601-650
551-600
501-550
451-500
401-450
351-400
301-350
251-300
201-250
151-200
TOTAL
F
1
1
1
1
5
10
11
7
7
3
47
X = 621. 0 mg/1
S = 189. 5 mg/1
Median = 623 mg/1
95% Confidence Interval
617. 8^JJ 1 624.2
X = 348. 9
S = 18. 8
Median = 340
95% Confidence Interval
347. 7 ±JJL £ 350. 1
The table does not include total solids for several isoladed events in which
concentrations were as high as 2136 mg/1.
X = Sample Arithmetic Mean
S = Standard Deviation
JU. = True Population Arithmetic Mean
D-5
-------
Total Influent Phosphate Concentration During
Dry Weather
Total Phosphate
mg/1
2
2
11
13
9
15
21
6
2
TOTAL 81
80
70
60
50
40
30
20
10
0
- 89
- 79
- 69
- 59
- 49
- 39
- 29
- 19
9
X = 39.8 mg/1
S - 25.5 mg/1
Median » 38 mg/1
95% Confidence Interval
38.7 ±JUL * 40.9
X = Sample Arithmetic Mean
S = Standard Deviation
JLl = True Population Arithmetic Mean
D-6
-------
Influent Total Nitrogen During Dry Weather
Total Nitrogen
mg/1
1
0
2
2
6
22
20
9
2
8
4
4
2
2
TOTAL 84
30.
28.
26.
24.
22.
20.
18.
16.
14.
12.
10.
8.
6.
4.
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
31.9
29.9
27.9
25.9
23.9
21.9
19.9
17.9
15.9
13.9
11.9
9.9
7.9
5.9
}T = 17.7 mg/1
S = 5.0 mg/1
Median = 19.3 mg/1
95% Confidence Interval
17.Z ±M ± 18.2
X = Sample Arithmetic Mean
S = Standard Deviation
M - True Population Arithmetic Mean
D-7
-------
Influent pH During Storm Events
PH
7.4
7.3
7.3
7.3
7.2
7.2
7.0
7.0
7.0
6.9
6.9
6.9
6.8
6.7
6.6
TOTAL 1055
T = 7.0
S = 0.2
Median = 7.0
95% Confidence Interval
6.8 *M ± 7.2
X = Sample Arithmetic Mean
S = Standard Deviation
JLL - True Population Arithmetic Mean
D-8
-------
Influent Turbidity During Storm Events
Turbidity
Jackson Units
340
332
326
252
238
238
230
223
210
210
210
192
186
50
TOTAL 3237
X = 231 J.U.
S = 73 J.U.
Median = 226
95% Confidence Interval
226 ±11 ± 236
X = Sample Arithmetic Mean
S = Standard Deviation
jLL = True Population Arithmetic Mean
D-9
-------
Influent Suspended Solids and Volatile Suspended Solids Concentrations
During Rain Events
Suspended Solids
og/1
987
788
775
730
665
520
484
438
425
405
386
385
377
329
317
Volatile Suspended Solids
mg/1
377
333
333
282
263
231
220
144
X = 534 mg/1
S = 197 mg/1
Median = 438 mg/1
95% Confidence Interval
526 ±JUL = 542
X = 273 mg/1
S = 75 mg/1
Median =272
95% Confidence Interval
387 =>U = 405
X = Sample Arithmetic Mean
S = Standard Deviation
M. - True Population Arithmetic Mean
D-10
-------
Influent BOD Concentrations During Rain Events
BOD
mg/1
440
282
245
242
202
202
200
190
165
160
148
147
139
X = 212 mg/1
S = 80.8 mg/1
Median = 200 mg/1
95% Confidence Interval
207 ±J1< £ 217
X = Sample Arithmetic Mean
S = Standard Deviation
Jd = True Population Arithmetic Mean
D-ll
-------
Influent Total Solids and Total Volatile Solids Concentrations During
Rain Events
Total Solids
mg/1
1282
1216
1190
1073
1061
929
820
812
777
777
753
750
721
677
647
602
X = 880 mg/1
S = 209 mg/1
Median = 794 mg/1
95% Confidence Interval
872 ±JLL ± 888
X = Sample Arithmetic Mean
S = Standard Deviation
~U- = True Population Arithmetic Mean
Total Volatile Solids
mg/1
588
410
410
406
360
303
294
X = 396
S = 98
Median = 406
95% Confidence Interval
387 ^ JUL ± 405
D-12
-------
Influent Total Phosphate Concentrations
During Rain Events
Total
Phosphate
mg/1
68
50
35
29
26
24
24
24
23
21
18
17
16
12
X = 2 7..6 mg/1
S = 14.3 mg/1
Median = 24 mg/1
95% Confidence Interval
25.4 £ JLL ± 29. 8
X = Sample Arithmetic Mean
S = Standard Deviation
JU. = True Population Arithmetic Mean
D-13
-------
Influent Nitrogen During Storm Events
Nitrogen
mg/1
25.2
24.0
21.8
20.3
17.9
17.9
15.7
15.4
14.8
14.2
13.8
11.4
10.4
8.2
TOTAL 231.0
T = 16.5 mg/1
S = 5.4 mg/1
Median = 15.6 mg/1
95% Confidence Interval
15.2 ± JUL ± 17.8
X = Sample Arithmetic Mean
S = Standard Deviation
AL = True Population Arithmetic Mean
D-14
-------
CALCULATIONS
Computations of the mean removal rates of suspended solids,
BOD, total solids, total phosphate and total nitrogen during
rain events using various chemical treatments. All mechanical
equipment on stream.
Waste Flow Rate = 350 GPM
AP = 50 psi
Air Feed Rate = 30 cfm
95% confidence interval was calculated using values of "t" found
in Table II in the Manual of Experimental Statistics (37).
/(X-X)2
Where
X = Removal Rates
X~ = Mean Sample Removal Rate
01 = Mean Population Removal Rate
n = Number of Observations
X - t S. -s P ;= t S. + x
Krf {ri
D-15
-------
Suspended Solids Removal- During
Rain Events Using No Chemical Treatment
X
79
71
56
206
69
X-X
10
2
-13
(X-X)
100
4
169
273
X" = 69
= 11.7
If Q ^ 4.303, Reject 56 **
t = 13(1.732) = 1.924 cannot reject 56
11.7
69 - 4.303(11.7)
1.732
69 - 29.1 *
39.9 ± » ±
± 4.303(11.7) + 69
1.732
69 + 29.1
98.1
** Application of Chauvenet's Criteria; critical values found
in Table A-6, Basic Statistical Methods. /OON
(.JO)
D-16
-------
309
62
X-X
85 23 529
71 9 81
66 4 16
44 18 324
43 19 361
Suspended Solids Removal During Rain Events
Using Alum Only
(X-X)2
1311
S = ,1311 = 18.1
If t >. 2.571, Reject 43 *
t = 19 ^ = 2.35
18 .1
Cannot Reject 43
If t ^ 2.571, reject 85
t = 23 |(F = 2.841
18.1
Reject 85
X-X
224
56
X = 56
(X-X)2
71 15 225
66 10 100
44 -12 144
43 -13 169
638
S -»/ 6^8
3
If t
t = 15
16.03
•• 16.03
2.776, Reject 71
= 1.871, Cannot Reject 71
56 - 2 .776(16.03)
2
^ 2.776(16.03) + 56
2
56 - 22.2 ^ p. 4. 56 + 22.2
33.8 ^ /i 78.2
11 Application of Chauvenet's Criteria; critical values found in
'able A-6, Basic Statistical Methods.
D-17
-------
Suspended Solids Removal During
Rain Events Using Alum and Tretolite (Continued)
X-X (X-X)2
85 3 9
85 3 9 S - \j 71 = If 17.75 = 4.213
84 2 4 4
82 0 0
75 -7 49 If t ^. 2.776, reject 75 **
411 71
82 t = 7 /5~ = 3.727, Reject 75
4.2
X X-X (X-X)
85 1 1
85 1 1 S =\fT =/~2 = 1.414
84 0 0 3
82 -2 4
336 If t i. 3.182, Reject 82 **
84
t - 2 \/4" = 2.828
1.414 Cannot Reject 82
84 ; S - 1.414
X = t s ^ p 4. X + Ju s_
n n
84 - 3.182(1.414) £ /i ± 84 + 3.182(1.414)
2 2
84 - 2.25 ^ ju ^ 84 + 2.25
81.75 ^ ju ^ 86.25
** Application of Chauvenet's Criteria; critical values found in
Table A-6, Basic Statistical Methods.
D-18
-------
Suspended Solids Removal During Rain Events
Using Alum and Tretolite
X-X
(X-X)
85
85
84
82
75
75
74
66
626
78
X
85
85
84
82
75
75
74
560
80
X
85
85
84
82
75
75
486
81
7
7
6
4
- 3
- 3
- 4
-12
X-X
5
5
4
2
-5
-5
-6
X-X
4
4
3
1
-6
-6
49
49
36
16
9
9
16
144
328
— 2
(X-X)
26
25
16
4
25
25
36
156
— 2
(X— X)
16
16
9
1
36
36
114
328 = \/46.86
7
6.85
If t ^ 2.365, Reject 66 **
for X = 66, t = 12 i/T = 4.95
6.85
Reject 66
= / 156 = \/2T = 5.099
If t ^ 2.447, Reject 74 **
for X = 74, t = 6 iTf = 3.116
5.1
Reject 74
S = \/114 = \/22.8 - 4.775
5
If t ^ 2.571, Reject 75 **
for X = 75 ; t
Reject 75
6 \/6~
4 .8
2.811
** Application of Chauvenet's Criteria; critical values found in
Table A-6, Basic Statistical Methods.
D-19
-------
ONE WAY ANALYSIS OF VARIANCE
SUSPENDED SOLIDS REMOVAL DURING RAIN EVENTS
Rain Events
No Chemicals
Alum
Alum + Tretolite
Totals
T = 766
2
r. . T2 = (766)
N"~ 11
SSB= s: T±2 - C
n
2
i
SSE= SST - SSB =
MSB= SSB = 1572
k-1 2
MSE= SSE - 6542
N-k 8
F - MSB - 0.96
MSE
T T 2 X 2
Ti Ti Xi
206 42436 14418
224 50176 13182
336 112896 33855
766 205508 61455
- 586756 - 53341
11
54913 - 53341 - 1572
61455 - 53341 - 8114
8114 - 1572 = 6542
- 786
- 817.75
n T12/n
3 14145
4 12544
4 28224
11 54913
Source of
Variation
Between
Samples
Error
Total
X
79 k =
56 k-1
84 N -
N-k
CX - C
F
Degrees of
Freedom
2
8
10
3
- 2
11
- 8
0.05
4.46
Sum of
Squares
1572
6542
8114
Mean
Squa
786
818
D-20
-------
TABLE D-l
One way analysis of variance of suspended solids removal
during rain events using various chemical treatments. All
mechanical equipment on stream. Chemical treatments include
(1) no chemicals, (2) alum only, and (3) alum plus Tretolite
Fr-50.
Null Hypothesis
HQ: There is no significant difference between the mean rates of
TSS removal for the modes of operation listed above.
Alternate Hypothesis
Ha: There is a significant difference between the mean rates of
Suspended Solids removal for the modes of operation listed above.
**• = 0.05
Fot = 4.46
Criteria: Reject HQ if F >• F , reserve judgement if F < Fo«,
Result: F = 0.134
Decision: F is less than F oe. , therefore cannot reject HQ .
There is no apparent significant difference between the mean
suspended solids removal rates for the modes of operations listed
above.
To determine where the difference between these mean exists,
a modified version of Duncan's Multiple Range Test (39) was used.
D-21
-------
119
X-X
39.7
X = 40
BOD Removal During Rain Events
Using No Chemical Treatment
(X-X)2
69 29 841
35-5 25
15 -25 625
1491
1491
= \/745.5
80 *
27.3
T.S. Removal During Rain Events
Using No Chemical Treatment
X X-X (X-X)2
45
29
29
103
34.3
X =
11
- 5
- 5
34
121
25
25
171
9.25
34.3 - 9.25(4.303) ± ju ^ 34.3 + 9.25(4.303)
34.3 - 23.0
£. 34.3 + 23.0
57
* 95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-22
-------
BOD Removal During
Rain Events Using Alum Only
X-X
(X-X)2
59 23 529
31 -4 16
30 - 5 25
20 -15 225
139
795
34.75
X = 35
s • 1/795 = 1/265 = 16.3
' 3
Reject 58 if Q i. 1.53
**
23 = 1.41, Cannot Reject 58
16.3
35 - 16.3(3.182)
35 + 16.3(3.182)
35 - 25.9 ^ u -j. 35 + 25.9
9 ^ ju ^ 61
I
36
T.S. Removal During
Rain Events Using Alum Only
X
48
36
35
23
142
X-X
12
0
- 1
-12
(X-X)2
144
0
1
144
289
35.5
17. = 9.82
3
35.5 - (9.82) (3.182)
\TT
^ 35.5 + (9.82) (3.182)
35.5 - 12.4
23
35.5 + 12.4
48
** Application of Chauvenet's Criteria, critical values found in
Table A-6, Basic Statistical Methods.
D-23
-------
BOD Removal During Rain Events
Using Alum + Tretolite
X-X (X-X)2
X
82
77
74
72
70
64
439
X-X
9
4
1
- 1
- 3
- 9
(X-X)2
81
16
1
1
9
81
189
73
82 14 196
77 9 81
74 6 36 S = v/1438 - \|239.7 = 15-5
72 4 16 6
70 2 4
64 - 4 16 Reject 35 if Q ;> 1.80 **
35 33 1089
474 1438 Q = .33 - 2.13, Reject 35
67.7 15.5
S - ,189 \D7.8 = 6.15
73 - 6.15(2.571) ^ ju ^ 73 + 6.15(2.571)
73 rg" vTg-
73 - 6.5 ^ » ^ 73 + 6.5
67 ^ ju ^ 79
** Application of Chauvenet's Criteria, critical values found in
Table A-6, Basic Statistical Methods.
D-24
-------
ONE WAY ANALYSIS OF VARIANCE
BOD REDUCTION DURING RAIN EVENTS
n
/n
No Chemicals
Alum
Alum + Tretolite
Tot
C =
SSB
SST
SSE
MSB
MSB
F -
als
T2 -
N
• Z-
= SST
" SSB
k-1
" SSE
N-k
MSB
MSB
(697)2
13
i! - c =
n
.2 - C =
- SSB =
- 4300
2
= 2474
10
= 2150
247
119 14161 6211
139 19321 5624
439 192721 32309
697 44144
= 485809 = 37370
13
41670 - 37370 = 4300
44144 - 37370 = 6774
6774 - 4300 - 2474
= 2150
= 247
= 8.70
3 4720
4 4830
6 32120
13 41670
Source of
Variation
Be tween
Samples
Error
Total
40 k = 3
35 k - i = 2
73 N = 13
N\r — TO
— K — -LU
ex - 0.05
F = 4.10
Degrees of Sum of
Freedom Squares
2 4300
10 2474
13 6775
Mean
Square
2150
247
D-25
-------
DUNCAN'S MULTIPLE RANGE TEST (39)
(MODIFIED VERSION)
BOD REDUCTION DURING RAIN EVENTS
1. Alum
2. Alum + Tretolite
X
35
73
.=/ (MSE)2 _= 1(247) 2 =Y49.4 = 7.05
10
= 0. 05; N = 10
r & r = Sample Sizes
1 2
P = 2
SSR
LSR =
(SSR)S-
x
3. 15
22.21
Ranked Means
Means
201
1.
2.
Sv =,i
Diff
P
1
35
LSR
38 2 22.2
No Chemicals
Alum + Tretolite
(MSE)2 =/
(247) 2
2
73
De
X
40
73
= 1
Decision: Difference is significant.
9
r & r? = Sample Sizes
= 0. 05; N = 10
Lt
P = 2
SSR
LSR =
(SSR) S-
.X.
3. 15
23. 31
Ranked Means 1
40
Means
201
Diff
33
P
2
LSR
23.3
2
73
- r
Decision: Difference is significant.
D-26
-------
DUNCAN'S MULTIPLE RANGE TEST
(MODIFIED VERSION)
BOD REDUCTION DURING RAIN EVENTS
1.
2.
s,-r =,,
Alum
No Chemicals
/ (MSE) 2 =1(24
X
35
40
7) 2
rl + r2
r, & r = Sample Sizes
•*• £A
= 0. 05; N = 10
P = 2
SSR
T GT3 — fQCt?\C;
j_joxv — ^ oo x\ /OY
3. 15
26.46
Ranked Means
1 2
35 40
Mean
201
Diff
5
P
2
LSR
26.5
Decision: Difference is not significant.
D-27
-------
DUNCAN'S MULTIPLE RANGE TEST
(MODIFIED VERSION)
BOD REDUCTION DURING RAIN EVENTS
1.
2.
Alum
Alum + Tretolite
J(2150)2
10
0.05; N2 = 10
P = 2
SSR
LSR =
(SSR)S-
A
3.15
65.33
Ranked Means
1
35
Means
201
Diff
38
P
2
LSR
653
X
35
73
V43~0
rl & r2
2
73
20.74
Sample Sizes
Decision: Difference is not significant
1. No Chemicals
2. Alum + Tretolite
S- = (MSE)2
tf (2150)2
' 9
0.05; N2 = 10
2
X
40
73
SSR
LSR =
Ranked
(SSR)S-
3.15
68.86
Means
1
40
Means
2-1
Diff
33
P
2
LSR
6886
1/3 V4300 = 1/3 (65.57)
r-, & r_ = Sample Sizes
21.86
2
73
Decision: Difference is not significant
D-28
-------
1. Alum
2. No Chemicals
DUNCAN'S MULTIPLE RANGE TEST
(MODIFIED VERSION)
BOD REDUCTION DURING RAIN EVENTS
X
35
40
i/ (MSB) 2 = .[(2150)2 =i/655.7 = 24.79
V n+ ro V 7
r2 = Sample Sizes
0.05; N2
10
1 2
Ranked Means 35 40
Means
2-1
Diff
5
P
2
LSR
78.1
Decision: Difference is not significant,
D-29
-------
TABLE D-2
One way analysis of variance for BOD reduction during rain
events using various chemical treatments. All mechanical equip-
ment on stream. Chemical treatments include (1) no chemicals,
(2) alum only, and (3) Alum plus Tretolite FR-50.
Null Hypothesis
HQ: There is no significant difference between the mean rates of
BOD reduction for the modes of operation listed above.
Alternate Hypothesis
Ha: There is a significant difference between the mean rates of
BOD reduction for the modes of operation listed above.
ex: = 0.05
Foe = 4.10
Criteria: Reject H if F > F , reserve judgement if F £ Foe
Result: F = 8.70
Decision: F is greater than F«: , therefore reject HQ. There is
an apparent significant difference between the mean rates of BOD
reduction as listed above.
The application of a modified version of Duncan's Multiple
Range Test indicates that a difference exists between the mean re-
duction rate of BOD when alum and Tretolite FR-50 are used and
the other treatments.
D-30
-------
Total Solids Removal During Rain Events
Using Alum + Tretolite FR-50
538 = \tf6.9 = 8.77
51.7 - 8.77(2.447) ^ ;u ^ 51.7 + (8.77) (2.447)
s/T
51.7 - 8 .5 4. jn ^ 51. 7 + 8.5
43 ^ 60
Total Phosphate Removal During
Rain Events Using Alum + Tretolite FR-50
X
100
91
83
81
58
53
52
513
X* -
X-X
26
17
9
7
-16
-21
-22
74
(X-X)2
676
289
81 S = ./2276 - J/379.3 = 19.47
49 6
256
441 74 - (19.47) (2.447) ^ ju ^ 74 + (19 . 47) (2 . 447 )
484 VT~ V 1
2276
74-18 ^ u ^ 74+18
56 ^ p ^ 92
D-31
-------
ONE WAY ANALYSIS .OF VARIANCE
TOTAL SOLIDS REMOVAL DURING RAIN EVENTS
Ti Ti2
No Chemicals 103 10609
Alum 142 20164
Alum & Tretolite FR-SO 362 131044
TOTALS 607
T = 607
C - T2 = 6072 = 368449 = 26318
J~ 14 14
SSB =y Ti - C = 27298 - 26318 = 980
n
SST -SIZIXi2 _ c - 28319 - 26318 = 2001
SSE - SST-SSB = 2001 - 980 = 1021
MSB - SSB = 980 = 490
k-1 2
X 2
3707
5354
19258
28319
Source of
Variation
Be tween
Samples
Error
Total
T 2 /
n TI /n
3 3536
4 5041
7 18721
14 27298
Degrees of
Freedom
2
11
13
X
34 k = 3
36 k-1 = 2
52 N - 14
N-k = 11
F
Sum of
Squares
980
1021
2001
.05
= 3.98
Mean
Square
490
92.8
5.28
MSB - SSE , 1021 = 92.8
MSB - 490
MSB 92.8
5.28
D-32
-------
TABLE D-3
One way analysis of variance of total solids removal during
rain events using various chemical treatments. All mechanical
equipment on stream. Chemical treatments include: (1) no
chemicals; (2) alum only; (3) Alum plus Tretolite FR-50.
Null Hypothesis
H0: There is no significant difference between the mean rates
of total solids removal for the modes of operation listed above.
Alternate Hypothesis
H : There is a significant difference between the mean rates of
3
total solids removal for the modes of operation listed above.
«*• = 0.05
F-c = 3.98
Criteria: Reject H if F > F , reserve judgement if F < F«
Result: F = 5.28
Decision: F is greater than Foe , therefore reject HQ . There
is a significant difference between the mean rates of Total Solids
removal during rain events.
An analysis using a modified version of Duncan's Multiple
Range Test indicates the difference exists between the mean
removal rate of total solids when alum and Tretolite FR-50 is
used and the other treatments.
D-33
-------
Total Phosphate Removal During
Rain Events Using Alum Only
X
48
11
59
X-X
18.5
-18.5
(X-X)2
342.25
342.25
684.5
29.5
V684.5
26.16
71 *
X « 30
* 95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-34
-------
Total Phosphate Removal During Rain Events
Using No Chemical Treatment
83
12
95
X-X
(X-X)2
35.5
-35.5
1260.25
1260.25
2520.5
47.5
X = 48
S = V 2520.5 = 50.2
80 *
Total Nitrogen Removal During Rain Events
Using No Chemical Treatment
X X-X (X-X)2
S
10
4
1
0
15
6
0
3
4
36
0
9
16
61
3.75
i/61 = V 20.3 = 4.51
50% *
* 95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-35
-------
Total Nitrogen Removal During
Rain Events Using Alum + Tretolite FR-50
Total Nitrogen
mg/1
X
15
12
10
8
0
0
0
45
— o • —
x-x
8.6
5.6
3.6
1.6
6.4
6.4
6.4
(X-X)2
73.96
31.36
12.96
2.56
40.96
40.96
40.96
243.72
6.4
0 %
= y60.7
<£ 45%
7.79
Suspended Solids removal during an isolated rain event using alum
+ Tretolite FR-50, waste flow rate » 200 GPM, p = 50 psi, air feed
rate =25 cfm.
Suspended Solids
mg/1
In Out %R
775
79
90
Suspended solids removal during an isolated rain event using 150
mg/1 lime and 5 mg/1 Drew Floe 410 waste flow rate =350 GPM,
p=50 psi, air feed rate =30 cfm.
Suspended Solids
In Out % Removal
220
175
21
* 95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-36
-------
ONE WAY ANALYSIS OF VARIANCE
TOTAL PHOSPHATE REMOVAL DURING RAIN EVENTS
It V
Su'11"1"1' H IHI llll
Alum + Tretolite FR-50 518 268325 40608
Totals 672 50066
T = 672
C - T2 = 6722 = 451584 = 41053
N 11 11
SSB - T±2 - C • 44584 - 41053 = 3531
n
SST =52X.2 - C - 50066 - 41053 - 9013
SSE = SST - SSB = 9013 - 3531 - 5482
MSB = SSB = 3531 = 1766
k-1 2
MSE - SSE = 5482 = 685
N-k 8
F - MSB - 1766 = 2.58
MSE 685
n Ti2/n X
2 4512 48 k - 3
2 1740 30 k - 1 - 2
7 38332 74 N = 11
N- V = 8
— K. ^ o
11 44584 ex = 0.05
F = 4.26
Source of Degrees of Sum of
Variation Freedom Squares
Between
Samples 2 3531
Error 8 5482
Total 10 9013
Mean
Square
1766
685
D-37
-------
TABLE D-4
One way analysis of variance for the removal of total
phosphates during rain events using various chemical treatments.
All mechanical equipment on stream. Chemical treatments include
(1) no chemicals, (2) alum only, and (3) alum plus Tretolite
FR-50.
Null Hypothesis
HQ: There is no significant difference between the mean rates of
total phosphate removal for the modes of operation listed in
Table 4.
Alternate Hypothesis
H : There is a significant difference between the mean rates of
a
total phosphate removal fro the modes of operation listed in
Table 4.
00 = 0.05
Foe =4.26
Criteria: Reject HQ if F •>• F , reserve judgement if F < F«.
Result: F = 2.58
Decision: F is less than F«* , therefore reserve judgement.
There is no significant difference in the TS removal rates during
rain events.
D-38
-------
Computations of The mean rates of suspended solids ,
BOD, total solids, total phosphate, and total nitrogen removal
using The various combinations of the screen, cyclones and
flotation cell.
Waste Flow Rate = 350 GPM
AP = 50 psi
Air Feed Rate - 30 cfm
95% confidence interval was calculated using values of " t " found
in Table II in the Manual of Experimental Statistics.
n-1 Where
JC = Removal Rates
It = Mean Sample Removal Rate
n = Number of Observations
M = Mean Population Removal Rate
X - t _s_ ^ ju ^ t_s_ +X
/n" ~
D-39
-------
Removal of Suspended Solids Using Screen and Flotation Tank
X
58
49
48
48
41
244
X-X (X-X)2
9
0
- 1
- 1
8
Totals
81
0
1
1
64
147
49
49 - 6.06(2.776)
5
49-7.5
41.5 ^
ju
6.06
49 + 6.06(2.776)
5
ju ^ 49 + 7.5
z 56.5
D-40
-------
Removal of Suspended Solids Using Screen,
Primary Cyclone and Flotation Tank
X-X (X-X)2
60 11 121
55 6 36 S =\j 399 = 19.975 = 7.5
53 4 16 7 2.646
52 3 9
51 2 4 49 - (2.365) (7 .5) ^ >u ^ 49 + (2.365) (7.5)
42 -7 49 /3~
41 -8 64
39 -10 100 49 - 6.3 ± ju. ^ 49 + 6.3
393 Totals 399
49.1 42.7 ^ ja ^ 55.3
X = 49
D-41
-------
Removal of Suspended Solids Using Two Secondary
Cyclones and Flotation Tank
_X X-X (X-X)2
67 24 S = \/21 - 4.583 - 3.24
66 1 1 2 1.414
61 - 4 16
194 2_1 64.7 - 3.24(4.3) < fi < 64.7 + 3.24(4.3)
64.7 V3~ VT
64.7 - 8.0 ^ » < 64.7 + 8.0
X = 65
56.7 ^ ju ^ 72.7
D-42
-------
Removal of Suspended Solids Using Screen,
Three Secondary Cyclones and Flotation Tank
X-X (X-X)2
60 7 49
59 6 36
54 1 1 S = J202 = 14.213 = 6.35
53 0 0 5 2.236
49 4 16
43 10 100 53 - (6.35) (2.571) *. p ^ 53 + (6.35) (2.571)
318 _ 202 6 6
53
53-6.7 ^ /u ^ 53 + 6.7
X = 53 46.3 ju 59.7
D-43
-------
ONE WAY ANALYSIS OF VARIANCE
SUSPENDED SOLIDS REMOVAL USING VARIOUS COMBINATIONS OF SEPARATORY EQUIPMENT
n
Screen
Screen &
Screen &
Screen &
TOTALS
T = 11
C = T2 =
N
ll
SSB - 2,
SST - 2
SSE -
MSB -
MSB -
244
Primary Cyclone 393
3 Secondary Cyclones 318
2 Secondary Cyclones 194
49
(1149)2
22
3' - c
n
c v 2 r
> X-JL - L
SST - SS
SSB =
k-1
SSE -
N-k
1149
- 1320201 = 60009
22
= 60612 - 60009 - 603
= 61348 - 60009 = 1339
B = 736
603 - 201
3
736 = 40.9
18
59536 12054 5
154449 19703 8
101124 17056 6
37636 12535 3
61348 22
Source of
Variation
Between
Samples
Error
Total
11907 49
19306 49
16854 53
12545 65
60612
Degrees of
Freedom
3
18
21
k = 4
k-1 - 3
N = 22
N-k = 18
— — _. - _ r\ r\ c
cXi = U . U j
F = 3.16
Sum of Mean
Squares Square
603 201
736 40.9
1339
MSB - 201 - 4.91
MSE 40.9
D-44
-------
DUNCAN'S MULTIPLE RANGE TEST
(MODIFIED VERSION)
SUSPENDED SOLIDS REMOVAL USING VARIOUS COMBINATIONS OF SEPARATORY
EQUIPMENT
Non-Rain Events
1. Screen, primary cyclones, flotation cell
2. Screen, 3 secondary cyclones, flotation cell
Sx
(MSE)2
rl+r2
\/ (40.9)2
14
5.843 = 2.42
r2 = Sample sizes
0 .05 ;
18
P = 2
SSR
LSR = (SSR)S-
X
2.97
7.19
1 2
Ranked Means 49 53
Means
B-A
Diff
4
P
2
LSR
7.19
Decision: The difference between the
mean removal rates is not
s ignifican t.
D-45
-------
DUNCAN'S MULTIPLE RANGE TEST
(MODIFIED VERSION)
Suspended solids removal using various combinations of Separatory
equipment.
Non-Rain Events
X
1. Screen, flotation cell, 3 secondary cyclones 53
2. Screen, flotation cell, 2 secondary cyclones 65
S- = (MSE)2 ^\/(40.9)2 = (2 . 13) (1. 414)
+r \T 9
.05; N2
18
rl & r2
3.014
Sample sizes
SSR
LSR
= (SSR)S-
A
2.97
8.95
1
Ranked Means 53
Means
B-A
Diff
12
P
2
LSR
8.95
D
2
65
Decision:
There is a significant difference
between the two mean rates of
suspended solids removal.
D-46
-------
BOD Removal Using Screen and
Flotation Tank
X
47
26
9
82
X-X
20
1
-18
(X-X)2
400
1
324
725
27.3
V362.5
19.04
Y = 27
75
Total Solids Removal Using Screen and
Flotation Tank
X-X
(X-X)2
35 8
30 3
15 12
80
26.7
= 27
64
9
144
217
75
95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-47
-------
TABLE D-5
One way analysis of variance for the suspended solids removal rates using
the various combinations of the screen cyclones and flotation cell. No chemicals
in use.
Null Hypothesis
Ho: There is no significant difference between the mean total suspended
solids removal rates for the five modes of operation shown in Table 2.
Alternate Hypothesis
Ha: There is a significant difference between the mean total suspended
solids removal rate for the five modes of operation shown in Table 2.
<=*- = 0.05
F^ = 3.16
Criteria: Reject HQ if F> F ; reserve judgement if F ^ F,^
Result: F = 4.91
Decision: F is greater than Fo< , therefore reject H ;
there is a significant difference between these
mean total suspended solids removal rates.
D-48
-------
BOD Removal Using Screen,
Primary Cyclone, and Flotation Tank
X X-X (X-X)2
57 22 484
36 1 1
36 1 1 S = \/93I = ]]237 .8 = 15.25
33 - 2 4 4
14 -21 441
176 931 35 - 15.3(2.776) < ju ^ 35 + 15.3(2.776)
35.2 /5~ /F
X" = 35 35 - 19 ^ ,u < 35 + 19
16 ^ w < 54
D-49
-------
BOD Removal Using Screen,
Two Secondary Cyclones and Flotation Tank
X X-X (X-X)2
75 34 1156
29 -12 144 S = ^1784 - i/892 = 29.9
19 -22 484 2
123 1784
41
* - 41 5 ^ ^ ^ 80 *
95% confidence interval obtained from Table V, Manual
of Experimental Statistics.
D-50
-------
BOD Removal Using Screen,
Three Cyclones and Flotation Tank
X
44
44
40
16
144
X-X
8
8
4
20
(X-X
64
64
16
400
544
36
X
36
S = x/544 = \Tl81. 3 = 13.5
3
36 - 13.5(3 .182) .£ ju < 36 + 13.5(3 .182)
^ p. < 57
Total Solids Removal Using Screen,
Three Secondary Cyclones and Flotation Tank
X
41
'29
28
21
20
13
11
163
X-X
18
6
5
2
3
10
12
(X-X)2
324
36
25
4
9
100
144
642
23.3
S = \/642
6
10.3
= 23
Z3 - 10.3(2.447) ^ ju < 23 + 10.3(2.447)
/T
23-9.5 ^ >u ^ 23+9.5
13 ^ ju < 33
D-51
-------
ONE WAY ANALYSIS OF VARIANCE
BOD REDUCTION USING VARIOUS COMBINATIONS OF SEPARATORY EQUIPMENT
All Equipment
Screen & Flotation Cell
Screen ,
Screen ,
Screen ,
Flotation
Flotation
Flotation
Cell,
Cell,
Cell,
Primary Cyclone
2
3
S econdary
Secondary
Cyclones
Cyclones
130
82
176
70
144
16900
6724
30976
4900
20736
3556
2966
7126
1636
5728
5
3
5
3
4
3380
2241
6195
1633
5184
26
27
35
23
36
TOTALS
602
21012 20 18633
k - 5
k-1 - 4
N = 20
N-k - 15
o< = 0.05
F_^ - 3.06
602
C » T
N
SSB =
SST
SSE
MSB
MSB
F -
1 = 362404 • 18120
20
T±2 - C - 18633 - 18120 - 513
n
= 2£Xi2 - C - 21012 - 18120 = 2892
= SST - SSB - 2379
- 513 - 128.25
4
- 2379 - 158.6
15
MSB - 128.25 - ' 0.807
MSB 158.6
Source of
Variation
Between
Samples
Error
Total
Degrees of Sum of Mean
Freedom Squares Square
4 513 128.25
15 2379 158.6
19 2892
D-52
-------
TABLE D-6
One way analysis of variance for the BOD reduction rates using the various
combinations of the screen, cyclones and flotation cell. No chemicals were used.
Null Hypothesis
HQ : There is no significant difference between the rates of BOD
reduction for the operational modes listed in Table 2
Alternate Hypothesis
Ha ; There is a significant difference between the rates of BOD
reduction for the operational modes listed in Table 2.
«=»« = 0.05
F^ = 3.06
Criteria: Reject HQ if F > F ; reserve judgement if F-^
Result: F = 0.807
Decision: F is less than
; reserve judgement.
This one-way analysis of variance indicates that any difference
between the rates of BOD reduction in Table is due to chance and
that changes in auxiliary equipment do not significantly affect
BOD reduction rates. Apparently flotation produces the major reduction
in BOD.
D-53
-------
Total Solids Removal Using Screen,
Primary Cyclone and Flotation Tank
X
27
23
13
3
66
X-X
10.5
6.5
3.5
13.5
(X-X)2
110.25
42.25
12.25
182.25
347 .00
V/116
10.8
0%
65% *
16
* 95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-54
-------
70
Total Solids Removal Using
Two Secondary Cyclones and Flotation Cell
X X-X (X-X)
24 1 1
24 11 S = \I~T~ = 1.225
22 1 1 V 2
23 23 - (1.225) (4.303) ^ ;u ^. 23 + (1.225) (4.303)
3 3
23-3
-------
Total Solids Removal Using Screen,
Three Secondary Cyclones, Flotation Tank With
Alum and Tretolite FR-50
X-X (X-X)2
61 0.5 .25 S = \ /0.5 0 = 0.707
60 -0.5 .25 V~T
121 0.50
60.5
_ 15% £ M < 100% *
X = 60 ~
*
95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-56
-------
ONE WAY ANALYSIS OF VARIANCE
TOTAL SOLIDS REMOVAL USING VARIOUS COMBINATIONS OF SEPARATORY EQUIPMENT
All Equipment
Screen Flotation Tank
Screen, Primary Cyclone , Flo ta t ion Tank
Screen, 3 Secondary Cyclone, Flotation Tank
Screen 2 Secondary Cyclone , Flo tation Tank
Totals
TI
134
80
66
121
70
471
'I2
17956
6400
4356
14641
4900
^
X
3390
2350
1436
7321
1636
1 £1 T 1
n
7
3
4
2
3
1 Q
T^/n
2565
2133
1089
7320
1633
i /. 7 /. n
X
19
2 7
16
60
23
k
k-1
N
N-k
F
« 5
« 4
= 19
- 14
= 0.05
= 3.11
471
I!
N
SSB =
SST
slii
n
= 471'
19
- C
- C
SSE
MSB
MSE
SST -
= SSB
k -1
= SSE
N-k
SSB
3064
4
1393
14
221841
19
14740
11676
11676
3064
16133 - 11676 = 4457
= 4457 - 3064
766
9.95
1393
Source of Degrees Sum of Mean
Variation of Freedom Square Square
Between
samples
Error
Total
4
14
18
3064
1393
4457
766
9.95
F = MSB
MSE
776 = 78
9.95
P-57
-------
TABLE D-7
One way analysis of variance for removal of total solids using the
various combinations of the screen, cyclones and flotation cell. No chemicals
were used.
Null Hypothesis
HQ: There is no significant difference between the mean removal rates
of total solids for the operational modes listed in Table 2.
Alternate Hypothesis
Ha: There is a significant difference between the mean removal rates
of total solids for the operational modes listed in Table 2.
Foe ; reserve judgement if F-£ F^ .
Result: F = 78
Decision: Cannot reject HQ; there is a significant difference
between the rates of total solids removal. The difference exists between
the otal olids removal rate for the screen, 3 secondary cyclones, and
flotation cell and the other treatments.
D-58
-------
Total Phosphate Removal Using the Various
Combinations of The Screen, Cyclones and Flotation Cell
X
37
36
33
21
21
20
17
13
10
8
8
7
6
5
4
2
2
2
252
X-X
23
22
19
7
7
6
3
-1
-4
-6
-6
-7
-8
-9
-10
-12
-12
-12
(X-X)2
529
484
361
49
49
36
9
1
16
36
36
49
64
81
100
144
144
144
2332
14
14
14 - 2.11(11.7) ^ » .<. 14 + 2.11(11.7)
VT8~
14-5.8 ^ M ^ 14+5.8
8 < yu < 20
D-59
-------
Total Nitrogen Removal Using the Various
Combinations of The Screen, Cyclones and Flotation Cell
X X-X
(x-x)2
*32
*21
10
8
8
8
6
6
5
5
5
4
1
0
0
0
0
66
5.6
3.6
3.6
3.6
1.6
1.6
0.6
0.6
0.6
0.4
3.4
4.4
4.4
4.4
4.4
31.36
12.96
12.96
12.96
2.56
2.56
.36
.36
.36
.16
11.56
19.36
19.36
19.36
19.36
144.24
S = y/144.24 \/10.3 = 3.21
14
4.4 - (2.145) (3.21) * p. < 4.4 + (2 . 145) (3 . 2 1)
~
4.4-1.8 ^ » ^ 4.4+1.8
2.6 < 11 ^ 6.2
4.4 * Values disregarded in computation
o £ mean.
D-60
-------
Computations of the mean removal rates of suspended solids, BOD,
total solids, total phosphate and total nitrogen using various
chemical treatments. All mechanical separation equipment on
stream.
Waste Flow Rate = 350 GPM
£P = 50 psi
Air Feed Rate = 30 cfm
95% confidence interval was calculated using values in Table II
as found in the Manual of Experimental Statistics. ( )
Where
N-l X = Removal rates
X = Mean removal rate
N - Number of observations
ji = Mean population removal rate
X - t S
/TT
D-61
-------
Removal of Suspended Solids Using No Chemical Treatment
X
68
63
61
61
61
60
58
34
27
493
55
X
68
63
61
61
61
60
58
34
466
58
X
68
63
61
61
61
60
58
432
x-x
13
8
6
6
6
5
3
-21
-28
Totals
X-X
10
5
3
3
3
2
0
-24
Totals
X-X
6
1
-1
-1
-1
2
-4
(X-X)2
169
64
36
36
36
25
9
441
784
1600
(X-X)2
100
25
9
9
9
4
0
576
732
(X-X) 2
36
1
1
1
1
4
16
60
62
S = \/1600 = \/200 = 14.14
If Q > 1.91, Reject 27 **
= 28 1.98 Reject 27
14.14
S = 1/732 = 27.055 = 10.22
* 7 2.646
If Q ^ 1.860, Reject 34**
= 24 .
10.22
= 2.348, Reject 34
62 - 3.162(2.447) ^ p
\rr
62-2.9 < /a ^ f. 2
59.1 ^ M ^ 64.9
62-3.162 ( 2 . 4|
-*• ? . 9
** Application of Chauvenet's Criteria; critical values found in
Table A-6, Basic Statistical Methods.
D-62
-------
Removal of Suspended Solids Using Alum
X-X
(X-X)2
S =1/2050 =/683 = 26.1
92
69
68
29
258
64
X =
28
5
4
35
64
784
25
16
1225
2050
3
Reject 29
Q = 35
26.1
Cannot re1
64 - 26.1 (3.
if Q > 1.53 **
- 1.34
ect 29
182) . „ . <
2 2
64 - 41.5
ju
64 + 41.5
22
100
BOD Removal Using Alum
27
237
47.4
X-X
73 26
59 12
50 3
28 -19
-20
= 47
(X-X)2
676
144
9
361
400
1590
S = 1/1590 = /398
' 4
19.95
47 - 19.95(2. 776) -£ p. < 47 + 19.95(2. 775)
/T
47-24.8
22
< 47 + 24.8
. 71
** Application of Chauvenet's Criteria; critical values found in
Table A-6, Basic Statistical Methods.
D-63
-------
Removal of Suspended Solids Using Alum and Tretolite FR-50
X-X
(X-X)
100.0
88.4
78.8
77.5
73.0
71.0
58.2
*55.4
54.1
51.3
*42.4
39.6
*12.4
691.9
30.8
19.2
9.6
8.3
3.8
1.8
-11.0
--
-15.1
-17.9
-29.6
948.6
368.6
92.2
69.2
14.4
3.2
121.0
228.0
320.4
876.2
3041.8
69.2
Eliminated because of blood
present or low pH (3.2).
Eliminate 39.6%
X
88.4
78.8
77.5
73.0
71.0
58.2
54.1
51.3
552.3
69.0
X-X
19.4
9.8
8.5
4.0
2.0
-10.8
14.9
17.7
and 100
(x-x)2
376.4
96.0
72.2
16.0
4.0
116.6
222.0
313.3
1216.5
S=\/3042 = ^304 = 17.4
10
% as outliers.
S = 1/1216.5 = V173.8 = 13.2
V 8
69.0 - 13.2(2.365) ^ p < 69.0 + 13.2(2.3651
1 8 . /~8~
69.0 - 11.0 *z p ^ 69.0 + 11.0
58.0 ^ u , 80.0
69
D-64
-------
Removal of Suspended Solids Using Alum and Dow 1188. 1A
Chemical feed rate adjusted to give least turbidity.
= 93
X-X (X-X)2
97
96
92
92
87
466
93
.2
.5
.9
.3
.6
.5
.3
3
3
0
1
5
.9
.2
.4
.0
.7
15
10
1
32
59
.21
.24
.16
.00
.49
.10
s =\/59.lO
4
93.3 - 3.84(2. 776)
/5~
93.3-4.8 /
88.5 <
/14.77 = 3.84
u ^ 93.3 + 3.84(2.776)
: 93.3 + 4.8
< 98.1
D-65
-------
Removal of Suspended Solids Using Alum and
Dow SA1188.1A
Chemical feed rate varied by pattern
4 mg/1 SA 1188.1A + 75 mg/1 and 100 mg/1 alum
8 mg/1 SA 1188.1A + 75 mg/1 and . 100 mg/1 alum
X
*77
71
65
63
61
339
67
.4
.6
.1
.6
.8
.5
.9
X-X
9
3
-2
-4
6
.5
.7
.8
.3
.1
Totals
(X-X)
90
13
7
18
37
167
2
.25
.69
.84
.49
.21
.48
=\A67.48 = \/41.87 = 6.47
V /l I
67.9 - 6.47(2.776) * P ± 67.9 + 6 . 4 7 (2^7^
68 ^"
67.9-8.0 ^ p ^ 67.9+8.0
59.9 < u < 759
* Heavy blood load in the influent stream.
One experiment was performed in which all flow (350 GPM)
was forced through one cell of the flotation tank. This
in effect halved the retention time in the flotation tank.
75 mg/1 alum and 50 mg/1 Dow SA1188.1A were used as
flocculation aids.
Input was 207 mg/1 T.S.S.; output T.S.S. was 30 mg/1 for
a removal rate of 85.5%. The test was of 4 hours duration,
D-66
-------
Removal of Suspended Solids
Using FeCl3 only
S = \/0 .50 = 0. 707
1
x-r
90. 7 0.5 0.25
89 .8 -0.5 0.25
180.5
90.2
90
0.50
90.2 - (.71) (12.7)
<£, 90.2 + (.71)(12.7)
\T~2
90.2 £ 6.4 ^ AI ± 90.2 + 6.4
83.8 < AI < 96.6
Using FeCl3 + Alum + Tretolite FR-50
X
92. 1
97.4
96.6
X =
X-X (X-X)2
3.3 10.89
2.0 4.00
1.2 1.44
95
S = J16.33
V 3 '
95.4 - 2.33(4.3) <
3
95.4-^.8 ^ Ai ±
89.2 ^ Ai ± 100
= \/ 5.44
1^ ^*
95.4 +
= 2.33
95.4 + 2.33(4.3)
3
D-67
-------
ONE WAY ANALYSIS OF VARIANCE
SUSPENDED SOLIDS REMOVAL DURING NON-RAIN EVENTS
Ti Ti2
No Rain
No Chemicals 432 186624
Alum 258 66564
Alum + Tretolite 552 304704
Alum + Dow SA1188.1A 467 218089
Totals 1709
T = 1709
C = T2 = 2920681 = 116827
N 25
SSB » T.2 - C
n
SSB = 120776 - 116827 = 3949
SST ^^^X-L2 - C
SST = 128212 - 116827 = 11385
SSE - SST - SSB = 11385 - 3949 -
MSB = SSB = 3949 - 1316
k-1 3
MSE - 7436 - 354
21
F - MSB = 3.72
MSE
29
XM rri *• y "v"
^ n 1 £ /n X
26720 7 26661 62 k = 4
18690 4 16641 64 k - 1 - 3
39320 9 33856 69 N = 25
43482 5 43618 93 N - k = 21
<=>< = 0 .05
128212 25 120776 F = 3i07
Source of Degrees of Sum of Mean
Variation Freedom Squares Square
Between
Samples 3 3949 1316
Error 21 7436 354
7436 Total 24 11385
D-68
-------
DUNCAN'S MULTIPLE RANGE TEST
(MODIFIED VERSION)
Suspended solids removal using various combinations of chemicals
Non-Rain Events
No Chemicals
Alum
1. Alum + Tretolite FR-50
2. Alum + Dow SA1188.1A
X
69
93
S- =
= \/(MSE)2 -y/(354)2
Vn + r, V 14
0.5; N2
21
50.6
7.11
rj. &
- Sample sizes
1 2
Ranked Means 69 93
Means
2-1
Diff
21
P
2
LSR
20.9
Decision: There is a significant
difference between the rates of
suspended solids removal for alum
and Dow SA1188.1A. Inspection
shows that the difference between
the suspended solids removal Bate
for alum and Dow SA1188.1A and the
other treatments would also be
significant.
D-69
-------
TABLE D-8
One way analysis of variance for suspended solids removal rate
as indicated below. The chemical treatments to be analyzed include:
(1) No chemicals; (2) Alum only; (3) Alum plus Tretolite FR-50;
and (4) Alum plus Dow SA1188.1A.
Null Hypothesis
HQ: There is no significant difference between the total suspended
solids removal rates for the chemical operations listed above.
Alternate Hypothesis
Ha: There is a significant difference between the total suspended
solids removal rates listed for the chemical operations listed above,
ex = 0.05
F^ - 3.07
Criteria: Reject HQ if F > F ; reserve judgement if F ^ F.<
Result: F = 3.71
Decision: F is greater than F«< , therefore reject H0. There is
a significant difference between these mean total suspended solids
removal rates.
The application of the modified version of Duncan's Multiple
Range Test indicates that a difference exists between the chemical
treatment using alum plus Dow SA1188.1A and the other chemical treat-
ments .
D-70
-------
Total Solids Removal
Using No Chemical Treatment
X-X (X-X)2
40 21 441
26 7 49
24 5 25
17-2 4 S = 1/825 = 1/13775 = 11.8
10 - 9 81 6
10-9 81
7 -12 144 19.1 - 11.8(2.365) ^ ju ^ 19.1 + 11.8(2.365)
134 825
-------
43
41
31
19
335
X-X
9
7
-3
-15
34
Total Solids Removal Using Alum
(X-X)
81
49
9
225
364
S =./364 = 1/121.3 = 11
33.5 - (11)(3.182)
2
33.5 - 17.5 ^ /i
16 ^ ^
11(3.182) + 33.5
2
33.5 + 17.5
51
D-72
-------
Total Solids Removal Using Alum and Tretolite FR-50
X
34
31
29
26
13
13
4
2
2
154
(X-X)
17
14
12
9
-4
-4
-13
-15
-15
(X-T)2
289
196
144
81
16
16
169
225
225
1361
17-1
70.1 = 13.0
17.1 -
17.1 -
17
13(2.306)
3
9.9 A »
7.2 #1
17.1 + 13(2.306)
3
17.1+9.9
27
D-73
-------
ONE WAY ANALYSIS OF VARIANCE
Total Solids Removal Using Various Chemical Treatments
n
No Chemicals
Alum
Alum + Tretolite FR-50
Alum + Dow SA1188.1A
Totals
134
134
171
185
624
17956
17956
29241
34225
3390
4862
3996
6707
18955
7
4
9
6
26
2565
4489
3249
5704
16007
19
34
17
31
k
k
N
N
ex
F
- 4
-1 = 3
- 26
- k - 22
c; = 0.05
= 3.05
624
C = T_ = (624) - 389376 - 14976
N 26 26
SSB= v-T±2 - C - 16007 - 14976 = 1031
SST=^
SSE =
MSB =
MSE =
F -
ElXi^
SST
SSB
k-1
SSE
N-k
MSB -
MSE
- C =
- SSE =
- 1031
3
= 2948
22
344 -
134
18955 - 14976
3979 - 1031
- 344
= 134
2.57
3979
2948
Source of
Variation
Be tween
Samples
Error
Total
Degrees of
Freedom
3
22
25
Sum of Mean
Squares Square
1031
2948
3979
344
134
D-74
-------
TABLE D-9
One way analysis of variance of removal of total solids with the various
chemical treatments listed below. The chemical treatments to be analyzed
include: (1) no chemicals; (2) alum only; (3) alum plus Tretolite FR-50; (4)
alum plus Dow SA1188. 1A.
Null Hypothesis
H : There is no significant difference between the rates of removal of total
solids using the various chemical treatments listed above.
Alternate Hypothesis
H : There is a significant difference between the rates of removal of total
cl
solids using the various chemical treatments listed above.
= 0. 05
oc = 3. 05
Criteria: Reject H if F - F^ ; reserve judgment if F <. Foe .
Result: F = 2. 57
Decision: F is less than Foe , therefore, reserve judgment. There is no
significant difference in the total solids removal rates using the chemical
treatments listed above.
D-75
-------
BOD Removal Using No Chemical Treatment
X
59
36
27
27
21
19
189
X-X
27
4
-5
-5
-11
-13
(X-X)2
729
16
25
25
121
169
1085
31.5
= 1/2l7
S = 14.7
If Q >. 1.73, Reject 59**
Q 27 = 183, Reject 59
14.7
X
36
27
27
21
19
130
X-X
10
I
1
-5
-7
(X-X)2
100
1
1
25
49
176
26
S = ^176 - 1/44" - 6.633
4
26
26 - 6.63(2.776) <; u <_ 26 + 6.63(2.776)
i/T 7?
26 - 8.2 < /i ^ 26 + 8.2
18 ^ p < 34
** Application of Chauvenet
Table A-6, Basic Statistical
's Criteria;
Methods.
critical values found in
D-76
-------
BOD Removal Using Alum and Tretolite FR-50
X-X
(X-30
89
77
71
70
58
56
53
51
36
31
28
18
638
36
24
18
17
5
3
0
-2
-17
-22
-25
-35
1296
576
324
289
25
9
0
4
289
484
625
1225
5146
53.2
= V4 6 7 . 8 = 21.6
53 - 21.6(2.201) ^ ju < 53 + 21.6(2.201)
53-13.7 ^ » ^ 53 + 13. 7
39.3 ^ u ^ 65.7
X = 53
D-77
-------
BOD Removal Using
Alum and Dow SA1188.1A
X
76
74
71
49
46
316
63.2
X-X
13
11
8
-14
-17
63.2
(X-X) 2
169
121
64
196
289
837
S = 1/837 = 1/209
V 4 '
63 - 14.5(2.776) ^ p
/5~
14.46
-------
BOD Removal
Using Fed- Only
X
22
62
84
X-X
+20
-20
(X-XK
400
400
800
42
42
S = /800 = 28.28
< ju < 80 *
Using FeCl3 + Alum + Tretolite FR-50
X X-X
88
83
171
85.5
-2.5
(X-X)2
6.25
6.25
12.50
S = yi2.50
1
3.54
86
86 - 3.54(12. 7) < p < 86 + 3.54(12. 7)
rr VT
85.5 - 32 <. p
54 < /i 100
85.5 + 32
* 95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-79
-------
ONE WAY ANALYSIS OF VARIANCE
BOD REDUCTION USING VARIOUS CHEMICAL COMBINATIONS
T. Tt2 X.2
No Chemicals 130 16900 3556
Alum 237 15169 12823
Alum + Tretolite 638 407044 39066
Alum + Dow SA1188.1A 316 99856 20810
TOTALS 1321 76255
T = 1321
C = T2 = 1745041 - 64631
N 27
SSB= yTt2 - C = 68505 - 64631 = 1874
n
SST = £Zxi2 - C - 76255 - 64631 = 11624
SSE= SST - SSB - 9750
MSB= SSB - 1874 » 625
k-1 3
MSE= SSE - 9750 • 424
N-k 23
F = MSB - 625 = 1.74
MSB 424
n Ti2/n Y
5 3380 26 k
5 11234 47 k
12 33920 53 N
5 19971 63 N
27 68505 F
Source of Degrees of
Variation Freedom
Between
Samples 3
Error 23
Total 26
- 4
•1 = 3
= 27
• k = 23
= 0.05
= 3.03
Sum of Mean
Squares Square
1874
9750
11624
624
D-80
-------
TABLE D-10
One way analysis of variances of BOD reduction using all
separatory equipment with various chemical treatments as
indicated below. The chemical include: (1) no chemicals,
(2) alum only, (3) alum plus Tretolite FR-50, and (4)
alum plus Dow SA1188.1A.
Null Hypothesis
HQ: There is no significant difference between the rates of
BOD reduction for the chemical treatments listed above.
Alternate Hypothesis
H : There is a significant difference between the rates of
a
BOD reduction for the chemical treatments listed above.
ex: = 0.05
Foe = 3.03
Criteria: Reject H if F> F ; reserve judgement if F ^ F
Result: F = 1.47
Decision: F is less than Fix. , therefore the null hypothesis
cannot be rejected.
The one-way analysis of variance indicates that there is
apparently no significant difference between the rates of
BOD reduction for the chemical treatments listed above.
D-81
-------
Total Phosphate Removal
Using No Chemical Treatment
X X-X (X-X)2
67 38 1444
52 23 529
21 - 8 64 S = 1/3088 = \J617 .6 = 24.84
20 - 9 81 V 5
8 -21 441
6 -23 529 29 - 2.571(24.84) -c ju ^ 29 + (2.571) (24.84)
174 3088 2.236 2.236
29
29-28.6 -i M <: 29+28.6
X~ = 29
0 -: u -c 58
Total Nitrogen Removal
Using No Chemical Treatment
X-X (X-X)2
16 3.4 11.56 S = 1/23.20 = /5T8 = 2.408
14 1.4 1.96 4
12 - .6 .36
11 -1.6 2.56
10 -2.6 6.76 12.6- 2.41(2.776) ^ ju <. 12.6 + 2.41(2.776)
63 23.20
12.6 12.6 - 3.0 ^ » ^ 12.6 + 3.0
= 13 9.6^^<.15.6
D-82
-------
Total Phosphate Removal Using Alum
X
78
69
48
39
31
265
53
X
X-X
25
14
-5
-14
-22
= 53
(x-x)2
625
196
25
196
484 53
1526
S = 1/1526 = \/381.5 = 19.53
' 4
- (19.53) (2.776) * » ^ 53 + (19 . 5 3) (2 . 7 76)
2.236 2.23o
53-24.2 ^ jj ^ 53+24.2
29 ^ M ^- 77
Total Nitrogen Removal Using Alum
X
21
17
7
4
4
4
19
4. 75
X-X
2.25
. 75
. 75
. 75
(X-X)2
5.0625
.5625
.5625
. 5625
6. 75
= 5
S =i/6.75 = 1/2.25
4.75 - (1.5)(3.182)
2
4.75 - 2.75 ^ ^u .
2.0 ^ u .
1.5
4.75 + (1.5) (3. 182)
2
4.75 + 2. 75
7.5
D-83
-------
Total Phosphate Removal Using Alum and Tretolite FR-50
X
75
64
60
42
38
37
31
25
14
386
42.9
X
X
X-X
32
21
17
-1
-5
-6
-12
-18
-31
- 43
Total
X-X
(X-X)2
1024 S = 1/3245 = 1/405.6 = 20.14
441 ' 8
289
1 43 - (20.14) (2.306) ^ u ^ 43 + (20 . 14) (2 . 306 )
25 3 3
36
144
324 43 - 15.5 ^ u < 43 + 15.5
961
77 , ?» - SQ
3245 +
Nitrogen Removal Using Alum and Tretolite FR-50
(X-X)2
30 Eliminated as Outliers
24
8
8
7
7
2
2
1
35
5
T =
3
3
2
2
-3
-3
-4
5
9
9
4 S = i/60 - \/10" = 3.162
4 V 6
9
9
10 5 - (3.16X2.447) , u j. 5 + (3.16X2.447
60 f7~ ^7-
5-2.9 ^ M ^ 5+2.9
2 ^ u ^ 8
D-84
-------
Total Phosphate Removal Using Alum plus Dow SA1188.1A
X X-X (X-X)2
57 23 529
44 10 100
37 3 9
21 13 169
12 22 484
171
34.2
X = 34
1291
\/322.7 = 17.96
34.2 - 17.96(2.776)
2.336
34.2 + (17.96) (2.776)
2.336
34.2 - 22.3 *. ju *. 34.2 + 22.3
12 56
Total Nitrogen Removal Using Alum plus Dow SA1188. 1A
X
17
16
0
0
0
33
X-X (X-X)2
10
9
7
7
7
6.6
X = 7
100
81
49
49
49
328
S =
328
\/82
9.06
6.6 - (9.1)(2.776)
6.6 + (9.1)(2.776)
6.6 - 11.3 ± u
0 ^ 11
6.6 + 11.3
17.9
D-85
-------
Total Phosphate Removal
Using FeCl3 Only
73
73
146
"75"
X =
S = 0
30 ^ ju < 100 *
73
Using FeCl3 + Alum + Tretolite FR-50
X
90
71
161
X-X
9.5
9.5
(X-X)2
90.25
90.25
180.5
80.5
V180.5
13.4
80
35%
100% *
Using FeCl3 Only
9
3
12
X-X
3
-3
(X-X)2
9
9
18
Total Nitrogen Removal
S = 1/18 = 4.242
0% ^ AI -c 50%
FeCl.
Using FeCl3 + Alum + Tretolite FR-50
2.
X~
X
5
0
5
5
_
X
2
2
2
-X
.5
.5
(X-X)2
6.
6.
12.
25
25
5
12.5 = 3.46
45% *
* 95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-86
-------
ONE WAY ANALYSIS OF VARIANCE
TOTAL PHOSPHATE REMOVAL USING CHEMICAL TREATMENT
No Chemicals
Alum
Alum + Tretolite FR-
Alum + Dow SA1188.1A
Totals
C = T2 = 996 =
N 25
SSB =yT..2 - C
n
SST ^^Xi2 - C =
SSE = SST - SSB
MSB - SSB = 1813
k-1 3
MSE = SSE = 8090
N-k 21
F = MSB = 604 =
MSE 385
T± Ti2 Xi2 n
174 30276 8134 6
265 70225 15631 5
50 386 148996 18680 9
171 29241 7139 5
996 49584 25
992016 - 39681
25
= 41494 - 39681 = 1813
49584 - 39681 = 9903
9903 - 1813 = 8090
= 604
= 385
1.57
Ti2/n X
5046 29 k = 4
14045 53 k - 1 = 3
16555 43 N = 25
5848 34 N - k - 21
Cy; = 0.05
41494 F,^ = 3.07
Source of Degrees of Sum of
Variation Freedom Squares
Between
Samples 3 1813
Error 21 8090
Total 24
Mean
Square
604
D-87
-------
TABLE D-ll
One way analysis of variance of removal of total phosphate
with the chemical treatments listed below. The chemical treat-
ments to be analyzed include (1) no chemicals; (2) alum only;
(3) alum plus Tretolite FR-50; and (4) alum plus Dow SA1188.1A.
HQ: There is no significant difference between the phosphorous
removal rates using the various chemical treatments listed above.
Alternate Hypothesis
Ha: There is a significant difference between the phosphorous
removal rates using the various chemical treatments listed above.
ex = 0.05
Fo< - 3.07
Criteria: Reject Ho if F >• F«. , reserve judgement if F ^ F0
Result: F = 1.57
Decision: F is less than Fo< , therefore reserve judgement.
There is no significant difference between the total phosphate
removal rates with various chemical treatments listed above.
D-88
-------
ONE WAY ANALYSIS OF VARIANCE
TOTAL NITROGEN REMOVAL USING CHEMICAL TREATMENTS
No Chemicals
Alum
Alum + Tretolite FR-50
Alum + Dow SA1188. 1A
Totals
Ti
64
19
35
33
151
V
4096
361
1225
1089
X 2
xi
818
97
235
545
1695
n
6
4
7
5
22
T±2/n
683
90
175
218
1166
X
11
5
5
7
k =
k-1
N =
- N-k
=><
F~
4
- 3
22
= 18
- 0.05
= 3.16
151
Il
N
SSB =
n
22801 = 1036
22
= 1166 - 1036
SST
SSE
MSB
MSE
F =
= 52Xi2
= 529
• SSB
k-1
= SSE
N-k
MSB =
MSE
- C
= 130 =
3
• 529 =
18
43.3 =
29.4
1695 -
43.3
29.4
1.47
130
659
Source of
Variation
Between
S amples
Error
Total
Degrees of
Freedom
3
18
21
Sum of Mean
Squares Square
130
529
659
43.3
29 .4
D-89
-------
TABLE D-12
One way analysis of variance of the removal of total
nitrogen with the various chemical treatments listed below.
Chemical treatments to be analyzed include: (1) no chemicals,
(2) alum only, (3) alum plus Tretolite FR-50, and (4) alum
plus Dow SA1188.1A.
Null Hypothesis
HO 5 There is no significant difference between the mean nitrogen
removal rates using the chemical treatments listed above.
Alternate Hypothesis
Ha: There is a significant difference between the mean nitrogen
removal rates using the chemical treatments listed above.
•=x =0.05
F« = 3.16
Criteria: Reject HQ if F >• Foe ; reserve judgement if F ^ Fo< .
Result: F = 1.47
Decision: F is less than F«< , therefore reserve judgement. There
is no significant difference between the total nitrogen removal
rates using the various chemical treatments listed above.
D-90
-------
Computations of mean rates of suspended solids, BOD, total solids,
total phosphates, and total nitrogen removal using all separatory
equipment, alum and Dow SA1188.1A.
Waste Flow Rate = 350 GPM
A P = 50psi
Air Feed Rate = 30 cfm -
X = % Removal
95 percent confidence interval was calculated using
values found in the Manual of Experimental Statistics
(38).
Where
X = Removal Rates
N-l ~X = Mean Removal Rate
N = Number of Observations
M = Mean Population Removal Rate
X - tS ^ u £_ tS_ + X
N N
D-91
-------
Removal of Suspended Solids Using Screen,
Three Secondary Cyclones and Flotation Cell + Alum
and Tretolite FR-50
X X-X
90
88
178
89
1
- 1
89
(X-X)2
1
1
S =\ = 1.414
1
89 - (1.414) (12.7) £. >u < 89 + (1.414) (12.7)
/T
89 - 12.7 ^ » ^ 89 + 12.7
75% ^ u < 100%
Removal of BOD Using Screen,
Three Secondary Cyclones and Flotation Cell + Alum
and Tretolite FR-50
80
57
137
X-X
(X-X)2
68.5
11.5
11.5
68
132.25
132.25
264.5
S =/ 264.5
25
16.2
100 *
* 95% confidence interval obtained from Table V, Manual of
Experimental Statistics.
D-92
-------
APPENDIX E
TYPICAL DATA OBTAINED DURING
PLANT SHAKEDOWN IN 1967
-------
Data accumulated during the commissioning and equipment
shakedown exercises in late 1967 indicated that the waste influent
contained a widely varying load of industrial solids: dissolved
and suspended.
Graphs of some of the data obtained illustrate the hourly
and daily variations. The relatively low suspended solids content
on November 18th and 19th clearly suggests a weekend with little
industrial activity. These graphs are included on the following
pages .
E-l
-------
VARIATION OF TOTAL SOLIDS
WITH TIME WEDNESDAY DEC. 13,1967
o
24OOi-
5
o.
tO
(O
*
p
20 OO-
1600 -
1200
800
4OO
9:00
AM
IOOO
I COO
12-00
NOON
100
200
300
PM
TIME
E-2
-------
VARIATION OF SUSPENDED SOLIDS WITH
TIME WEDNESDAY DECEMBER 13,1967
2500
2000
(O
Q
O
at
UJ
0.
to
CO
1500
1000-
500-
ro
I
W
1
6:00
AM
1
900
1
IO-.00
1
IIOO
1
12:00
NOON
1
I:OO
1
^00
1
3:00
1
4-OO
PM
TIME
-------
VARIATION OF SUSPENDED SOLIDS
WITH TIME WEDNESDAY DEC. 6 ,1967
4000
9
E
tn
Q
CO
o
UJ
UJ
Q.
CO
CO
g
3000
2000
1000
12.00
NOON
COO
2-00
3:00
PM
E-4
-------
TOTAL SOLIDS mg/£ AS A FUNCTION OF TIME
WED. NOV.I5 THROUGH SUNDAY NOV. 19,1967
6000
5000
o< 4000
>v
9
E
o
CO
3000
2OOO
1000
J L
.1
_L
J_
_L
12
8 4
NOV. 15
12
8 4
NOV. 16
12
8 4
NOV. 17
12
8 4
NOV. 18
12
8 4
NOV 19
TIME AND DATE
-------
ACCESSION NO.I
KIT WOlDft
dual com-
1 <5)*the
Combined Sowar*
Storm Hater
Overflow!
Flotation.
Primary Tr*atB«nt
•tori tuaoff
ACCeSSIOR ».:
* fl.
The |
*y*tem for various • !** combined aewege
Of th« ayatem for automation and ua* la
overflows; (4) Tha edaotabll
remote location; and (S) Tb«
Combined Sawera
•tor* Vator
Ovorflova
Diaoolv*d-»Alt
Plotatloa
ayat**a appaaro po»albl* with conventional control cquipatnt. Cha*ieal
aida to flocculatlon appear to b»a promlaa that warrant* further atady.
TroaCBant M*thod«
Primary TroatB*Rt
Ploccttlant Aid*
Storm lunoff
ACCESSION 10.i
Tha principal aapacta in»*«tls«ted waia: (1) P«rfor»«nc« of tha »7«e««
durloc rain evant* and dry period*; (2) Evaluation of individual com-
ponent*; (3) Capital coata and operating coata for utlllting a flotation
•yatem for varloua al*a combined **w*ge ovatflowa; (4) The adaptability
of Che aystcm for cutomatlon »nd oac in remote location; end (5) The
ability of tha *yat*m to treat Intermittent and hlfhly variable flow*
Combined Sawcn
Storm Water
OvertIowa
Dlaaolved-Alr
Flotation
Infiltration
dlaaolved-air flotation ayatem* would be economical for handling com-
bined eewer overflow* up to 8 HCD. Automation of dl**olved-alr flotatioi
ayatem* appaara poeaible with conventional control equipment. Chemical
aide to flocculatlon appear to have promlae that warranta further atudy.
Toe *y*tam wa* unique in that all liquid floi
air dloBolvlns tank with no recycle. Domaat
lieu of combined **wase during period* of no
pa**ed din
Sueponded Solid*
Storm Knnoff
------- |