United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600
January 198®
Research and Development
Demonstration of a Closed
Loop Reuse System in a
Fiberglas Textile Plant
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pOllution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
~hiS doc~ment is available to the public through the National Technicallnforma-
tlon Service, Springfield, Virginia 22161.
-------
EPA-600/2-80-040
January 1980
Demonstration of a Closed
Loop Reuse System in
a Fiberglas Textile Plant
by
S.H. Thomas and D.R. Walch
Owens-Corning Fiberglas Corporation
Fiberglas Tower
Toledo. Ohio 43601
Grant No. S801173
Program Element NO.1 BB61 0
EPA Project Officer: Max Samfield
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park. NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
In March 1973, Owens-Corning Fiberglas Corpor ti ded au S
o a on was awar . .
:PA Dem~nstrat1on Grant for the preliminary engineering of a closed loop
1ndustr1al wastewater recirculation system for the fibe 1 textile
o d Tho k rg ass
1n ustry. 1S wor was documented in U.S. EPA publication Industrial
Wastewater Recirculation System: Preliminary Engineering (i), in February.
1977. The report concluded that reclamation of the effluent from the exist-
ing wastewater treatment facilities followed by sand filtration. carbon
adsorption, and disinfection would result in a product water suitable for
reuse in process cooling water, air scrubbing, and washdown water systems.
. Upon completion of the final design. construction of the treatment and
reuse system began in March 1977 and was completed by March 1978. By August
1978. all recycle systems were on-line and the operation of the entire water
reclamation and recycle system became a reality.
This report documents an evaluation of the full-scale system for the
period January 1979 through October 1979. As the evaluation proceeded. numer-
ous design and operational deficiencies were identified. Many of these were
corrected by minor modifications while more complex modifications required
additional pilot study and their impact upon the system was projected using
transformed operational data. The impact of these modifications on attaining
total system closure are presented in this report.
This report was submitted in fulfillment of Grant No. S801173 by Owens-
Corning Fiberglas Corporation under the partial sponsorship of the U.S.
Environmental Protection Agency. This report covers a period from March
1973 to December 31. 1979 and the report was completed as of March. 1980.
ii
-------
CllAl'T ER
I
II
III
IV
V
VI
'fAilLE OF CONTENTS
Abstract
List of Figures
List of Tables
ACKNOWLEDGEMENTS
INTRODUCTION
CONCLUSIONS
RECOMMENDATIONS
PLANT WATER AND WASTE
WATER CHARACTERISTICS
1. Overall Recirculation System Plan
2. Cooling Systems
1. Cooling System Hydraulic Operation
3. Hydraulic Balance Between Recycle
Demand and Wastewater Discharge
4. Cooling System and Reclaimed Wastewater
Quality
OPTIMIZATION AND OPERATIONAL IMPROVEMENTS TO
EXISTING PRIMARY AND SECONDARY TREATMENT SYSTEMS 25
1. System Description 25
2. System Performance 27
3. Performance Improvement Studies 27
4. Operations and Control Summary 36
DESCRIPTION, START-UP, AND OPERATION OF THE
ADVANCED WASTE TREATMENT AND RECYCLE SYSTEMS
1. System Description
1. Sand Filtration
2. Carbon Adsorption
3. Disinfection
2. Hydraulic Considerations for Design of the
Advanced Waste Treatment System
3. System Start-up
1. Biological Quality of Segregated
Industrial Waste Water
Low and High Quality Recycle Systems
Supply System Hydraulic Problems
Inert Gas Scrubber Recycle Problems
Mat Line Recycle Problems
Other System Hydraulic Problems
2.
3.
4.
5.
6.
iii
PAGE
ii
v
vii
viii
1
3
6
8
8
8
8
10
14
37
37
37
40
42
42
43
44
50
51
53
53
55
-------
VII
PLANS
l.
2.
TABLE OF CONTENTS (Continued)
4.
Operation of the Advanced Waste Treatment
System
1. Sand Filtration
2. Carbon Adsorption
Hydraulic Effects of Advanced Waste Treatment
upon Primary and Secondary Treatment Systems
Operations
1. Organic Quality of Recycle System -
"Refractory" Buildup
5.
3.
FOR OVERALL SYSTEM IMPROVEMENT
Overall System Needs
Hydraulic Flow Reduction
1. Raw Wastewater Reduction
2. AWT Recycle Waste Reduction
Raw Waste Primary Treatment Chemical
Coagulants
Fresh Water Makeup at Reclaim Distribution
Tank
4.
VIII
REFERENCES
APPENDICES
A.
Estimation of Total Dissolved Solids
Lost in Partial Discharge
B.
Exerpts from Primary and Secondary
Operations Manual
C.
Special Microbial Procedures
Iv
Page
57
57
60
70
73
75
75
75
75
75
77
78
79
80
82
94
-------
Figure
4.1
4.2a
4.2b
4.3
4.4
4.5
4.6
5.1
5.2
5.3
5.4
5.5
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
6.10
LIST OF FIGURES
Title
Selected Cooling System Water Balance
Overall System Modified Design Water
Balance for Recirculation
Overall System Operational Water Balance
During Partial Recirculation
Weekly Average History of #1 Spray Pond
Cooling System and Reclaim Supply Water
Weekly Average History of #2 Spray Pond
Cooling System and Reclaim Supply Water
Weekly Average History of "D" Process
Cooling System and Reclaim Supply Water
Weekly Average History of "D" Condenser
Cooling System and Reclaim Supply Water
Process Flow Diagram - Existing Wastewater
Treatment Facilities
Weekly Average History of Existing Waste
Treatment Operation - TOC
Weekly Average History of Existing Waste
Treatment Operation - BOD
BOD and TOC Removal Efficience in Secondary
System
Secondary System Performance History -
Suspended Solids
Waste Treatment System Design
Average Weekly Flow in Recycle System
Proposed Mat Line Recycle System without
"Piggyback"
Mat Line Recycle as Presently Operated
Weekly Average TOC for Secondary Effluent,
Lead Carbon Column Effluent, and Lag Carbon
Column Effluent
Weekly Average Secondary and Cargon Column
Effluent BOD
Percent TOC Removal through Carbon Column
System
Percent BOD Removal through Carbon Column
System
Influent and Effluent TOC Summary for a
Complete "Flip-Flop" Cycle for #3 Carbon
Column
Percent TOC Removed per Volume Processed
for a Complete "Flip-Flop" Cycle for 113
Carbon Column
v
Page
11
16
17
21
22
23
24
26
29
30
31
32
38
52
54
56
61
62
63
64
66
67
-------
Number
6.11
7.1
LIST OF FIGURES (Continued)
Title
Page
Cumulative TOC Capacity for a Complete
"Flip-Flop" Cycle for 1/3 Carbon Column
68
Proposed Waste Treatment System Design
76
vi
-------
Table
4.1
4.2
4.3
4.4
4.5
4.6
5.1
5.2
5.3
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
LIST OF TABLES
Title
Major Cooling Systems
Selected Cooling System Summary
Comparison of Projected Wastewater and
Recirculation Flows with 1979 Data During
Recycle
Comparison of Modified Design Projections
with 1979 Data
Cooling System Water Quality Summary
Equilibrium Concentrations in Reclaimed
Wastewater
Past Performance of Primary and Secondary
Systems - Mean Values
Summary Estimated Waste Composition
Nutrient Supplementation Schedule
Plaque Assay of Waste Sample in Primary
Rhesus Monkey Kidney Tissue Culture
Plaque Assay of Wastewater Sample in
HeLa All Culture
Plaque Assay of Stock Virus Solution and
Virus - Seeded Wastewater in HeLa All Culture
Plaque Assay of Stock Virus Solution and
Virus - Seeded Wastewater in Primary Rhesus
Monkey Kidney Tissue Culture
Bacterial Testing Results
Sandfilter Performance Summary
Monthly Average Performance Summary for
Carbon Adsorbers
Hydraulic Effects of AWT System Wastes on
Primary
vii
Page
9
12
13
15
18
19
28
33
35
45
46
47
48
49
58
65
71
-------
ACKNOWLEDGEHENTS
Owens-Corning Fiberglas Corporation acknowledges the efforts of
the following OCF individuals for their assistance in the preparation
of this report. Administrative leadership provided by Mr. S.H. Thomas
proved invaluable throughout the duration of this project. Messrs. W.A.
Candy, R.E. Hardesty, and G.T. Griffith are thanked for their technical
inputs on the various project phases included in this report. Anderson
Plant personnel, especially Mr. M.B. Parker and Ms. G. Tedhams, pro-
vided detailed operational insight and their 'front-line' efforts were
greatly appreciated.
EPA support for this project was greatly appreciated, especially
the overall guidance provided by Mr. Max Samfield, EPA Project officer.
Finally, the invaluable literary and technical contributions pro-
vided by Dr. D.I. Angelbeck regarding the recirculation project and
this report must be acknowledged. His detailed evaluations have pro-
vided much insight regarding the operation of and modified improvements
to this system.
viii
-------
CHAPTER I
INTRODUCT lON
The Owens-Corning Fiberglas Corporation (OCF) in 1968 established
an objective of total water recycle for all its manufacturing facili-
ties. The technology of total recycle was successfully developed and
demonstrated for fibrous glass insulation facilities in 1968 at the
Co~poration's Barrington, New Jersey plant (2). Following the success
in that area, the Corporation set out to develop the technology of
total recycle for textile fibrous glass facilities. After conducting
the necessary background research, OCF submitted an application for
federal funding.
In March, 1973, the Corporation was awarded an Environmental
Protection Agency Demonstration Grant (S801173). Research and
development work (Phase I) for the grant was conducted at the OCF
manufacturing facility at Anderson, South Carolina. This work along
with construction work under Phase II of the grant was originally to
be completed over a three-year period. Research and development work
delays, brought about by interruptions in manufacturing operations at
the Anderson facility, required that the Phase I completion date be
extended to July 1, 1976. The entire preliminary engineering work
that was a result of Phase I was documented in Environmental Protection
Agency Publication, Industrial Wastewater Recirculation System:
Preliminary Engineering (1), in February 1977.
Upon completion of the final design, the construction phase,
Phase II began in March, 1977 and was completed by March 1978. In
April, 1978, all low quality recycle systems were apparently "de-bugged",
and the low quality recycle system became operative. By August, 1978
all high quality recycle systems were apparently "de-bugged" and
operation of the entire reclamation and recycle system became a
reality.
Upon preliminary investigation of the entire operating system
between August, 1978 through December, 1978, it was apparent that total
system closure was not being achieved. During this period several
supply system concepts were revised in the operating system to attempt
system hydraulic balance and system closure. By December 1978, it was
apparent that although these modifications did improve hydraulic
balance, total system closure was not achieved; even though five months
of operational experience had transpired with a wealth of gained
operational knowledge. Thus, the most realistic method to determine all
of the "hidden" problems within the system, was to investigate and
document the entire system in a detailed and organized manner for an
extended period of time. Simultaneously, grant schedule revisions
required a 9-month evaluation of the full scale total recycle system
to be completed by December 31, 1979. Hence, a detailed investigation
of the entire system operation and final design began in January, 1979.
1
-------
As the evaluation preceeded. numerous recycle and treatment system
design and operational deficiencies were identified. Several minor
modifications were conceived, installed, and evaluated during the study;
while more complex modifications required that their impact upon the
system be projected using transformed operational data. These latter
proposed changes have also required additional pilot study and at this
time appear feasible and consistent with the projected impacts
contained in this report, (i.e.) success full implementation will make
major improvements in the system hydraulic balance which should allow
system closure. However, as in any totally closed recycle system,
fresh water inputs must be kept to the minimum amount used to establish
the design. Any excess fresh water entering the system over the design
level will accumulate within the system and most probably result in
hydraulic imbalance and discharge. At this time fresh water control at
the Anderson facility remains a variable which is difficult to assess.
The facility is so large and its manufacturing operations so diverse,
that numerous small insignificant fresh water sources can easily
"infiltrate" the system. In aggregate these sources can account for
very significant levels of fresh water within the recycle system.
However, continued studies developments, and system improvements will
be made, which are aimed at better control of fresh water intrusion and
overall system hardware development to reduce system discharge; since
the goal of zero discharge continues to be a goal of Owens-Corning
Fiberglas Corporation.
2
-------
CHAPTER I I
CONCLUSIONS
Analysis of data obtained from the full scale operating waste
water reclaimation and reuse system at the Owens Corning Fiberglas,
Anderson, S.C. Fiber Reinforcement Manufacturing Plant over a 9-month
evaluation period has produced the following conclusions:
1.
2.
3.
4.
Total system closure has not been achieved because fresh
water use has not been controlled to levels of design
expectations.
a)
Excess fresh water usage of approximately 87 to 100 gpm
should be controllable in the future since it is result
of maintenance related items, treatment of surge
inadequacies and surge control inadequacies present
in the existing system.
Primary, secondary and advanced waste treatment (AWT)
(consisting of sand filtration, activated carbon adsorption
and chlorine disinfection) systems hydraulic patterns should
be and are presently being modified to increase hydraulic
surge (flow) capacity which should decrease excess fresh
water usage by 10 to 25 percent.
Reuse patterns within the manufacturing facility have been
modified and include:
a)
Elimination of fresh water usage and utilization of
reclaimed water to clean combustion gases in the
inert gas scrubbing unit.
b)
While dissolved air flotation without chemical addition
is an effective pretreatment process for removal of
fibrous materials from Mat Line wastewater, the
utilization of reclaimed water as a pressure feed stream,
rather than localized loop effluent feed stream, has been a
requirement dictated by fiber fouling of pressure stream
valving.
The net effects of these changes has been to (1) lower the design
level of fresh water usage by the system and allow a design
closure of the system even during winter months (i.e. better
system design), but (2) increase the design recycle flow rates
which in turn reduces the hydraulic surge capacity within the
primary, secondary and AWT systems.
Coagulation of equalized raw waste water with only organic
polyelectrolytes in full scale trial tests has produced TSS
3
-------
11.
removal efficiences equal to or better than the originally
designed ferric chloride, clay, and cationic polymer system,
while significantly lowering the TDS level within all
recycle streams (a design consideration which is highly
favorable with respect to achieving greater reuse within all
manufacturing systems and ultimate reuse success).
5.
Chlorination has effectively controlled microbes
and viruses within the partially closed system that has been
fully segregated from domestic waste waters.
6.
Reuse of treated industrial wastewaters in process areas
and cooling systems at the Anderson Plant is technologically
feasible with partial discharge.
7.
Reclamation of the effluent from the primary and secondary
waste treatment facilities by sand filtration, carbon adsorption.
and disinfection will result in a product water suitable for
reuse in a partially discharging system.
8.
Based upon current drift loss measurement and original drift
loss estimates; equilibrium concentrations of total hardness,
calcium hardness, silica, sulfate, zinc and organic materials
in the cooling systems may exceed the water quality criteria
for these uses; if drift loss, is the sole mechanism by which
inorganic and residual organic dissolved solids are removed
from the reclaimed wastewater in a totally closed system.
9.
If removal of inorganic dissolved solids is required, it could
be accomplished through treatment by reverse osmosis or lime-
soda softening/anion exchange.
10.
Residual or refractory dissolved organic materials not
removed through chemical, biological, and carbon adsorption
treatments could be of such a nature and could accumulate to
such levels within the totally closed system to reduce
biological treatability which might prove to be significant
upon the ultimate zero discharge system design.
Maintenance of high performance efficiencies with the secondary
systems has been much more difficult following sanitary waste
segregation and requires much closer attention of nutrient
supplementation. Secondary system performance has faltered due to:
(a) nutrient difficiencies. (b) hydraulic overloading
produced by excess fresh water intrusion and treatment
system hydraulic pattern design as described in conclu-
sions 1, 2, and 3 above and (c) possibly refractory
dissolved organic materials resulting in higher than
expected organic concentrations in reclaim water. Even
4
-------
with partial discharge cooling systems have not failed
to date. The only major total system purge to date
has been due to residual sulfide odors within the entire
manufacturing plant water system produced by anaerobic
biological activity within the carbon adsorber due to
high biological suspended solid loadings upon the
adsorber.
12.
Dynamic manufacturing operations resulting in numerous product
changes present a 'moving target' for the design and operation
of reclaim water treatment processes.
13.
Total system success related to inorganic & organic quality
can only be assessed after complete hydraulic closure for
an extended time period.
5
-------
CHAPTER III
RECOMMENDAT IONS
Upon evaluation of the proposed closed loop system over six
months of preliminary operation. plus additional operation over a
nine month extensive testing program documented within this report.
the following recommendations regarding the closed loop system de-
velopment are made:
1.
Continue system upgrade and development to reduce uncon-
trolled fresh water addition such that zero discharge might
be achieved on a hydraulic basis.
2.
Modify recycle and reuse patterns to optimize reuse potentials
through utilization of more reclaimed water and the decrease
of designed fresh water usage as presented in Chapter VI.
3.
Modify the original design of the advanced waste treatment
system (AWT) hydraulic flow pattern for backwash wastes as
presented in Chapter VII. to increase surge capacity of the
AWT system and lower the hydraulic loads upon primary and
secondary treatment systems.
4.
In the future. when preliminary design analyses of recycle
system concentration projections are made. include an impact
study of organic residuals escaping biological and physio-
chemical treatment similar to that made for inorganic parameters.
5.
During preliminary design of similar systems. evaluate the
seasonal impact of precipitation upon volumetric increase
of uncovered storage basins. The annual effect may be negli-
gible with evaporation exceeding precipitation. however short
term seasonal variations may be critical to hydraulic balance.
6.
After hydraulic closure is achieved. re-evaluate system opera-
tion with respect to inorganic and residual organic build up.
These impacts should be studied as they affect cooling system
operation and treatment system performance. Furthermore. high
and low quality treatment schemes. tailored to meet demand
quality might evolve which could make overall recycle econo-
mics more favorable.
7.
The following recommendations are made either as (1) a contin-
gency if overall hydraulic closure is never achieved or (2) for
future systems which might be designed'on the basis, of partial
system closure:
a)
Investigate the impacts of polutant concentration and
pollutant mass rate discharges upon receiving enviorn-
6
-------
ments; a partially closed system, with internal advanced
waste treatment, should most certainly reduce the rate of
mass discharged with respect to a corresponding once
through system, however, concentrations of most pollu-
tants may be significantly larger (due to partial closure)
than the corresponding once through system. Thus, dis-
charge standards with respect to concentration limits must
not only be consistant with receiving environment needs,
but a1so be consistant with treatment system concepts re-
lating to partial reuse.
7
-------
CHAPTER IV
PLANT WATER & WASTE WATER CHARACTERISTICS
IV.l.
OVERALL RECIRCULATION SYSTEM PLAN
Based upon numerous hydraulic flow studies made in 1969. 1973
through 1974 and in 1976. it was estimated that flow closure would
be quite reasonable. These data were presented in detail in the
Preliminary Engineering Report (1). Detailed analysis of these
data indicated the following reuse scheme:
1.
Upgrade existing wastewater treatment facilities to enable
production of an effluent of such quality as may be used in
the plant cooling systems.
2.
Utilize this reclaimed water as makeup to the cooling systems.
3.
Cascade blawdowns from one cooling system to another; thus. in
effect. the blawdowns will be part of the makeup to the
systems receiving them.
4.
Final b1awdowns from the cooling systems are to be routed
to the "D" and "E" factory air scrubbers.
IV.2.
COOLING SYSTEMS
There are nine major cooling systems in the Anderson plant which
use water; these are described in Table 4.1. The process cooling
systems require the highest degree of water quality due to extreme
heat loads and small diameter heat exchanger piping. The chillers
possess somewhat liberal physical and chemical water quality require-
ments. but the water used must be free of pathogenic bacteria and
viruses, because the chill water is used to cool the atmosphere
inside each factory. Thus. the sanitary wastes generated within
the plant were completely segregated from industrial discharges to
assure that this requirement would be continuously met.
IV.2.1
COOLING SYSTEM HYDRAULIC OPERATION
Cooling of heated circulating water in evaporative cooling systems
employed at the Anderson plant is accomplished primarily through
evaporation in spray ponds and cooling towers. Additional water is
also lost whrough entrainment of water droplets in the air draft;
this loss is know as "drift". While both evaporation and drift
constitute vapor losses, of the two mechanisms only the drift process
is responsible for dissolved solids removal. The overall effect is
that the dissolved solids concentrate in the remaining liquid. To
prevent a buildup of dissolved solids in the cooling systems (and
associated scaling produced heat transfer problems) a small portion of
8
-------
SYSTEM
"A" Chillers
"E" Chillers
"D" Chillers
II Pond (A" & "E" Condenser Cooling)
\0
12 Pond ("A" , "E" Process Cooling)
"D" Condenser Cooling
"D" Process Cooling
Chemical Cooling Tower No.2
Chemical Cooling Tower No. 1
* Estiaated volume
TABLE /..1
MAJOR COOLING SYSTE1S
Y.1RPOSE
VOLlP.1E (~a1)
Cools water for "A" Factory and Beta Factory air washer Ii
20,000
Cools water for "E" Factory air washers
38,000
.
Cools water for "D" Factory air washers
3S ,000
Cools refrigeration units 1r. "A" Factory, Beta Factory,
and "E" Factory
755,000
Cools bushings, fin shidds, ar.d furnace coils in "A"
Factory, Beta Factory and "E" Factory
505,000
Cools refrigeration units in "D" Factory
38,000
Cools bushings, fin shields, and furnace coils
in "D" Factory
135,000
Cools Checical Factory process ~its: 12 Thinning Tank,
12. 13. and 14 Reactors, and Carrier chill water cond~nsers
6,000
Cools burner in the Chemical Factory Inert Gas manufacturing
operation. 11 Reactor, and Trane chill water condensers
2.000
1,537,500
-------
the circulating water is continuously discharged from the cooling
water system; this loss is termed "blowdown". The end result is that
water is continuously added to each system (makeup) in amounts equal
to the total water lost (blowdown + evaporation + drift - chiller
condensation during summer months).
Makeup, blowdown, evaporation, and drift flows for both summer
and winter operating conditions for all nine major cooling systems
were calculated and presented in the preliminary report (1). Four
of these systems are presented in Figure 4.1 these projections for 111
and 112 Spray Ponds and "D" Process and "D" Condenser cooling towers
are contrasted to those measured during a recent two (2) week winter
period with attempted system closure. The operational data appear in
Table 4.2. These data indicate that all systems were being turned
over through system blowdown to scrubbers at higher rates than in the
original design. Larger blowdown rates were necessary to purge scrubber
systems since scrubbers were accumulating suspended and dissolved solids.
In the future, various chemical dispersants will be tested to achieve
higher levels of suspended and dissolved solids build up within scrubber
systems. In turn this will dec.ease the cascading blowdown requirements
from the process and condenser cooling systems, and correspondingly
slightly reduce overall wastewater discharge and reclaim supply demand
flow rates.
IV.3.
HYDRAULIC BALAliCE BETWEEN RECYCLE DEMAND AND WASTEWATER DISCHARGE
Projected wastewater flows and reuse requirements were developed
from cooling system balances, previously cited process reuse flows
developed in 1974, planned manufacturing reductions in "A" Factory
and Beta Factory, and future manufacturing additions ("E" Factory).
Sanitary wastewaters from the entire Anderson Plant, amounting to
40-60 gpm were segregated from the process wastewater conveyance system,
treated separately in a "package" plant, and discharged to Betsy Creek.
The sanitary waste originates from potable city water uses in the plant
lavatories and cafeterias. Usages requiring high quality water, such
as boiler makeup, binder makeup, and deionized water sprays do not
utilize reclaimed wastewater. Finally, reclaimed wastewater is used
in the process cooling systems, the air scrubbers, ~~t Line, Alloy,
and floor washdown in "A", "D", and "E" Factories. Currently reclaimed
water is also used as input to the IG Scrubber.
Both projected cooling and process reuse flows, together with
projected wastewater discharges, are listed in Table 4.3. Since
cooling system evaporative losses are much greater during the summer
than during the winter, summer reuse requirements were planned to
exceed the amount of reclaimed wastewater available; conversely,
during the winter reclaimed wastewater flows were planned to exceed
reuse requirements. The obvious conservation solution was to store
the excess reclaimed water during the winter for later use in the
10
-------
From "A"
Chillers
From "E"
FIGURE 4.1 SELECTED COOLING SYSTEM WATER BALANCE
S=lO
~
=
W=l
S=9.6
- -
Ch i 11 ers W=4
~=37.4
From "0"
Chi llers
=
5=6.4
Q=2
S=28.6
W=
S=32
W:.16
t
.
.
.
S! W
63 ~ 1
#1 POND
S W
4.5 0
"A" & "E" Condenser
Cooling
4
.
S i W
37 : 25.9
#2 POND
S W
6.1 6.1
"A" & "E" Process
Cooling
,
I
S ' W
28 i 0
:
IID" Condenser
Cooling
4
I
S j w
25 : 14.1
.
"D" Process
Cooling
S
1.4
w
o
S=:E. 5- -+ TO "E" SCRUBBER
W=l
S=3.9
W=Q - -to TO "E" SCRUBBER
S=5.6 - + TO "0" SCRUBBER
W=2
S=5
---..,
W=O
TO "D" SCRUBBER
s t W
2.0 11.9
KEY
.............-... Evaporation
Reclaimed Water Makeup
- - - - -.... Blowdown
S = Summer Conditions, Flow, GPM
... ~ Drift
JI JIll -17
W = Winter Conditions, Flow, GPM
11
-------
TABLE 4.2
SELECTED COOLING SYSTEM SUMMARY
TYPE OF
FLOW
III SPRAY POND
Projected Measured
112 SPRAY POND
Projected Measured
"D"-PROCESS
COOLING
Projected Measured
liD "-CONDENSER
COOLING
Projected Measured
Total Makeup
( gpm)
41
40
39
48
24
24
18
16
Blowdown to 7 8 2 6 2 4 3 4
Scrubbers (gpm)
Drift (gpm) 2 6 * 6 8 * 2 3 * 1 2 *
~
N Evaporation
32 26 ** 31 34 ** 20 17 ** 14 10 **
( gpm)
*
Through Differential Calculation
**
Through Data Measurement and Computation
-------
TABLE 4.3
COMPARISON OF PROJECTED WASTEWATER AND RECIRCULATION FLOWS WITH
1979 DATA DURING RECYCLE
SOURCE
"A" Factory Process
Marble Fact./Binder Room
Boiler Hcuse
Cooling System Filter
Backwash (1)
All Plant Cooling
Systems
Miscellaneous
SUB-TOTAL
"D" Factory PrOC~$s
"D" ScrulJbers l J
SUB-TOTAL
"E" Factory Pr9~~ss
"E" Scrubbers lZ}
SUB-TOTAL
IG ScrubbE;r (3)
Mat Line (4)
SUB-TOTAL
Chemical Factory Process
Alloy
TOTALS
PROJECTED REUSE
DEMAND
SUMMER WINTER
.e
10
o
o
1
174.3
10
195.3
20
.4
20.4
20
3.6
23.6
o
o
...Q..
o
20
~
AVG.
DESIGN
DISCHARGE
gpm
MEASURED
DISCHARGE
FEB. 1979
~
JL. ~
o 35
20 20
(5) (5) ---l5)
(259.3)1.1 (198.6)1.1 (205)1.1 (361) 1.1
=285 =219 =226 =397
(1) Wastewater flow is Chemical Cooling Tower No.1 System Blowdown; reuse
flows are totals for all cooling systems.
(2) Both scrubbers receive cooling systems blowdown in addition to flows
listed in reuse columns.
(3) Originally conceived to be supplied with fresh water and supplied with
fresh water during February, 1979; currently supplied with reclaim water.
(4) Originally conceived to be supplied by IG scrubber discharge (i.e.
"piggyback System"): but supplied by 9 gpm fresh and 10 gpm reclaim
water during February, 1979, currently operated the same as February, 1979.
(5) 110% for flow measurement accuracy limitations.
% INC-
REASE OR'
DECREASE
10
o
o
1
15
2
5
1
89.6
10
1
19
110.6
20
9
29
20
19
39
-
o
o
43
25
11
-1A2.
+ ?4?
~
25
20
45
l&..
+ 111
47
37
19
~
25
10
+ 4.4
+ 115
- 28.6
- 50
59.7
13
-------
summer. However, at the time of design it was not possible to
determine the balance between those days with excess wastewater
and those with a deficit. The "safest" path was to: (1) mak.e
provisions for controlled addition of city water makeup to the re-
claimed wastewater distribution tank. and (2) ensure that the reclaimed
water storage basin was of sufficient volume to hold excess flows
for at least 100 days (the entire winter season).
It is important to note that system modification durina system
startup, involving IG Scrubber conversion from fresh water with
piggyback discharge to matline processes to reclaim supply without
piggyback discharge (see section VI 3.4 and VII 3.5 for details)
does increase both the recycle demand and wastewater discharge flows.
However, this conversion does balance the projected flows during the
winter months (see Table 4.4) and theoretically eliminates the need
for winter holding. Thus, it appears that this system modification,
as compared to that presented in the Preliminary Engineering Report
(1) is much better. Additional data obtained from hydraulic audits in
February, 1979 during partial loop closu~e are also presented in Table
4.3. A daily average discharge on 112 gpm was measured during this
period.
The overall recirculation water balance for the Anderson Plant,
projected and measured, during attempted system closure is given in
Figures 4.2a and b and Table 4.4. Date appearing in parentheses in
Figure 4.2b are 110% average of data during the February hydraulic audit.
It is important to note that the total excess flow, obtained with
110% correction for flow estimates, indicates a surplus raw waste-
water flow of 397-259 - 138 gpm. Measured data at the system over-
flow indicates a 112 gpm x 1.1 - 123 gpm surplus flow. Recycle
demand flow data also appearing in Figure 6.2b additionally indicates
that the average recycle flow was measured at 245 gpm, corrected by 110%,
yields 270 gpm. For complete system balance, 397-123 - 274 gpm of water
should have been recycled rather than that measured, 270 gpm. Thus, a
small system closure error of 4 gpm existed at the time of the audit.
IV.4.
COOLING SYSTEM AND RECLAIMED WASTEWATER QUALITY
As originally conceived, the hydraulic operation (i.e., makeup
and blowdown requirements including fresh water addition to the overall
system) would be governed by dissolved and/or suspended solids accumu-
lation within each recycle loop. As such, solids levels were to be
monitored on a daily basis and all flows, including recycle makeup,
blowdown and any fresh water addition, manually adjusted to maintain
the cooling systems solids at the expected levels presented in
Table 4.5. It is important to note that in the Preliminary Engineering
Report (1) no projections were made regarding organic solids buildup
within the system. This can be seen by inspecting the projected
design equilibrium reclaimed wastewater parameters for the tertiary
system illustrated in Table 4.6. It was anticipated that most biological
14
-------
TABLE 4.4
COMPARISON OF MODIFIED DESIGN PROJECTIONS
WITH 1979 DATA
COMMENT PROJECTED REUSE IWERAGE MEASURED
OR DEMAND DESIGN DISCHARGE %
CORRECTION SUMMER WINTER DISCHARGE FEB. 1979 INCREASE
gpm gpm gpm gpm
Totals from 259 198 205 361
Table 4.3
IG Scrubber 37 37 (37-26)
on Reclaim
Matline on 20 20 20
Reclaim
TOTAL
1.1(255)
=281
1.1(236)
=259
1.1(316)
=:~48
* Includes 37 gpm of fresh water for IG Scrubber
15
1.1(361) *
=397
40
-------
CITY ~1 = 73
----
WATER S - 73
W . 182
S = 157
t-
O'
W - 281
S . 348
FIGURE 4.2a
PROCESS USES
W = 3
S - 30
~
QJ IIJ
4J ~
U QJ
QJ .c
.-I .c
QJ :s
tI) ~
U
o tI)
COOLING USES
W = 95
S = 160
OVERALL SYSTEM MODIFIED DESIGN (1) WATER BALANCE
FOR RECIRCULATION ALL FLOWS CORRECTED BY 110%
W = 258
S = 260 .
W = 1
S - 1
LOSSES TO ATMOSPHERE
W - 259
WASTEWATER
TREATMENT
CONTROLLED
CITY WATER
I
W = 221
S = 8~
I
S - summer flow, gpm
W - winter flow',' gpm
(1)= without piggyback systems and
without mat line sprays on reclaim water
RECLAIMED WATER TO DISTRIBUTION
-------
CITY 73
----
WATER
....
......
{ 37 } I
I
CU.J(2)
WATER
9
{270 }
FIGURE 4.2b
OVERALL SYSTEM OPERATIONAL(l) WATER BALANCE
QijRING PARIIAL~~OLAT~O~ (ALL FLUWS CORRECTED BY 110%)
PROCESS USES
{6}
""
" III
.., ...
u "
"J:j
~i
fI) ...
U
oen
fo4
COOLING USES
91
I 108
I
I
CITY(3)
WATER
LOSSES TO ATMOSPHERE
{397 }
WASTEWATER
TREATMENT
{ 123}
DISCHARGE
TO CREEK
{} . gpm measured in 2/79
(1) . without piSgyback systems end
without matline sprays on reclaim water
(2) - IG Scrubber on city water
(3) . Miscellaneous unplanned city water
including emergency peak additions at A
(4) . Balance error of 4 gpm
RECLAIMED WATER TO DISTRIBUTION
EMERGENCY
CITY WAXER
I
I {A} (3)
I
I
274 (4)
-------
TABLE 4.5 COOLING SYSTEM WATER QUALITY SUMMARY
$U/1I'U::1t ---- - . - . - . ---.---.!!!!!~ -.------
nHll.lNL I'AIlAMETEI( EXI'ECl'IW ~.!!!
-------
TABLE 4.6
EQUILIBRIu~ CONCENTRATIONS IN RECLAIMED WASTEWATER
SUMMER WINTER
Projected Limits Measured Projected Limits Heasured
Equilibrium mg/1 May - Sept 1979 Equilibrium mg/l Jan - April 1~79
mg/1 mg/1 mg/l mg/l
Avg Max Avg Max
Total Dissolved
Solids (TDS) 736 422 512 1300 744 190 508 1150
Total Hardness 94 84 95 62
Calcium Hardness 87 76 88 55
....
\0 Silica 42 42 42 30
Sulfate 186 115 188 85
Zinc 6 5 6 5
Chloride 320 570 285 500
-------
refractory organic material would be removed by activated carbon
absorbers within the tertiary system and/or subsequent treatment through
the activated sludge secondary system with wastewater recycle. It has
been found through system operation that this is not the case. Sub-
sequently. the accumulation of dissolved organic solids may be a key
operational limit to the system which might prevent ultimate total
recycle with zero discharge on a continual basis. This specific
problem is examined in greater detail in Chapter VI of this report.
The anticipated inorganic solids levels originally thought
critical t,) the recycle system are also compared to the actual values
measured during system operation between January and September 1979
and are presented in Tables 4.5 and 4.6. These data indicate that
all parameters measured within cooling systems and reclaim wastewater
were far below the expected levels. This finding is in agreement
with the hydraulic operational findings presented in Section IV.2.1.
The cooling systems were being turned over far too much and as a re-
sult. were not developing high enough concentrations of various para-
meters within the cooling systems. Thus. the net effect of increased
drift. decreased evaporation and increased systems blowdown (Table 4.2)
was to establish lower equilibrium concentrations within the cooling
loops. Furthermore. large masses of residual inorganic and organic
dissolved solids have been continuously removed from the overall
system through the continual average partial discharge of approxi-
mately 86 gpm* of reclaimed waste water.
For example. the estimated rate of total dissolved solids input
through fresh water and manufacturing losses (after treatment) was
estimated in Appendix A to be 728 lbs/day while the estimated TDS loss
rate through partial discharge alone was 527 lbs/day. Assuming that
the TDS levels were pseudo steady. this would indicate a need of
201 lbs/day less through cooling system drift. However. an estimate
of the total system drift loss. if all cooling systems were drifting
as estimated in the preliminary report (.1) would be 425 lbs/day.
Thus. increased cooling system drift (Table 4.2) plus partial
system discharge could easily produce lower than expected equilibrium
concentration:levels within the overall system. Nevertheless. Figures
4.3. 4.4, 4.5, and 4.6 contain graphical documentation of cooling
system chemical parameter weekly variations.
In summary. no conclusion can be made regarding either system
tolerances and failures related to solids build up or the need for
additional solids removal treatment within recycle systems. Only
after operation of a completely closed system over an extended time
period can such information be obtained.
*123 gpm discharge includes 37 gpm freshwater on IG Scrubber
during 2 week audit. Therefore. over 9 month period average discharge
estimated at 123-37 - 86 gpm.
20
-------
-
;r. ~... 139
~ ::
i ~
;: - 106
< :r.
= '"
~ ::.
... :<
; E
.... ~
N
....
...
... 1580
E
~
II:
~ 1040
750
~
...
......
It 650
.!
'r.
~
550
450
350
FIGURE 4.3
WEEKLY AVERAGE HISTORY OF #1 SPRAY POND
COOLING SYSTEM AND RECLAIM SUPPLY WATER
7.5
7.0
6.5
"1 Spray Pone
aecirculated Cooling Systeo
73
::: Sp:-ay Pen!
2660
\~'
'~,
VI 5;>ra;; Pond
~ecirculated Cooling Syste~
2120
Reclaited Sup;>ly
fv-J'
, iv'
, I ,
' I I
, I I '
I
I
~ .,.
:!2 N
.. ... ......
...... ...... ...... ". 0-
co co ...
... - - :::: ;:; '! .., ~ '" "" ;:::
:e N '" N .,. N ....
...... ...... ...... '::: ...... ...... ...... ...... ...... ...... ...... ...... .... ......
- - N " ... "" .. "" '" '" '" '" .... ....
I.'EEK DiDING 1979
-------
'E.
"'-
"'- 100
:.;
z C
S ~
<: 0: 85
= r:
:;; - 70
...
..
.:. 55
N
N
- 11'Jt;..
-
...
'"
~
~
800
750
-
-. 650
It
~
....
!Ii 550
~
450
350
FIGURE 4.4
WEEKLY AVERAGE HISTORY OF #2 SPRAY POND
COOLING SYSTEM AND RECLAIM SUPPLY WATER
i.5
~,..,'~,
7.0
6.5
Recirculated Cooling System
"2 Spray pon~
Recirculate~ Cooling
S)'S :e:
1400
"2 Spray Po"d
Recirc~ated Coc:ir.~
Syste!ll
I~/
Reelaiced Supply
oD
-.
C
N
-.
...
.... ~ ac N ", .... ~ ~ on 0-
N .... N '" ::: N
- N N 0- .... .... .... .... -.
-. ... .... .... .... .... ....
... '::; .... .... ... ... 0- 0- 0-
.... ~ "'" on on ... ...
."EEX ENDING 1979
....
...
...
-.
N
, .
lv'
, \
, \
, \
, I
,
,
-------
; .5
7. C.
-
6.;
-
t 96
~ -
:;:
< "!: ."
:;: - 5:
;: ~
2140
J -'0:'
,
N ..
UJ .! I~~O
~
10~0
FIGURE 4.5
WEEKLY AVERAGE HISTORY OF "0" PROCESS
COOLING SYSTEM AND RECLAIM SUPPLY WATER
, .
I '
I
"D" Process
~ecir:~lat~: Cooling Syste~
7>0
~ecirc~latE: :ooling Syste~
"D" Proc~ss
Recirculat~; Cooling System
Reclair:ed Sup,ly
6;;0
-
.....
r
~ 550
450
350
c ,... ,...
.:; '" ,.. -
..... ..... ..... ,... ~ ~
- - .... '"
!\ '\
I V\
, "
, , I
'" I ',' I
'..I~'~'~ ~
...... ...... ...... ...... ......
co CI) 0'1 0\ 0'1
.....
~
N
.....
~
, cA I
N
.....
oD
. I .!.. .
.... N
..... .....
.... ....
-
.....
'"
~~EK ESDI~G 1979
-------
FIGURE 4.6
WEEKLY AVERAGE HISTORY OF "0" CONDENSER
COOLING SYSTEM AND R~CLAIM SUPPLY WATER
"D" Condenser
Recirculated Coo1iD8 System
7.~
~ 7.0
6.5
~
OIl ....
"'0
!~ 16
~ : 89
...~
~....
~! 62
,.
, ,
3160 "D" Condenser Recirculated Coo1in8 "
Syate.. , ,
, ,
2950 , \
N ~
... 2440
s:- ....
~
IS 1930
..
1420
910
750
650
~
...
....
! 550
~
450
350
"D" Condenser
Rec1a1:oed Supply
I
,
\
,
I
lv,
, I
, I I
I I ,
I I I
o ~ ~ <: OD ~ '" :!: ,. '"
'" N "" .... .... N N ~ N ... N <: ~ - N
.... :::: .... .... ~ .... .... .... .... .... .... .... .... .... .... .... .... ....
- N N .... r 4 <: .,.. .~ .., -= ... ... at at ~ f:'
WED< ESDlJiG 19;9
-------
CHAPTER V
OPTIMIZATION AND OPERATIONAL l}IPROVID-1ENTS TO EXISTING
PRIMARY AND SECONDARY TREATMENT SYSTEMS
V.1
SYSTEM DESCRIPTION
An extensive backlog of historical development and data has been
published regarding the waste treatment systems at the Owens Corning
Fiberglas, Anderson Plant, including:
1.
Thomas, S.H., and Walch, D.R., "An Industrial Wastewater
Recirculation System for the Fibrous Glass Textile
Industry", Textile Industry Technology Conference,
1978. (3).
2.
"Industrial Wastewater Recirculation: Preliminary
Engineering": EPA-600/2-77-043. February 1977. (1).
3.
West. A.W.. '~lant Performance at the Owens-Corning
Fiberglas Corporation Wastewater Treatment Facility.
Anderson, South Carolina". EPA. December 1973, (4).
4.
Pharis and Monaghan, '~iological Treatment of
Textile and Sanitary Wastes from a Fiberglas Plant"
WPCF Conference. 1965, (5).
The Preliminary Engineering Report (1) described waste treatment
operational activity in detail up to 1977. The existing waste treatment
facility process flow diagram prior to the wastewater recirculation
project appears in Figure 5.1.
25
-------
CH
FACTORY
N
0-
FIGU8E 5.1
PROCESS FLOW DIAGRAM- EXISTING WASTEWATER
SURGE
TANKS
"A","B","D"&"E"
FACTORIES
P'tlMARY
SEDIMENT AT I
BAR SCREEN
BASKET
ACTIVATED
SLUDGE
a::
I&J
:IE
- >
->
ctU
..J ..
UIL.CL
FACILITIES
FLAS
MIX
FLOCCULATION
8 NEUTRALIZATION
BETS Y
CREEK
..
-------
V.2
SYSTEM PERSORMANCE
A performance summary of the existing treatment facilities
from 1972 up until mid-1976 is presented in Table 5.1. However,
during the winter of 1976-1977, plant treatment efficiency began to
decrease. Review of the operational data at that time indicated
that loss of performance could be associated with an abnormally
cold winter season. Thus, a preliminary investigation involving
temperature control through the use of heaters and/or an air sup-
ported dome covering the biological waste process of the existing
facility was undertaken. This investigation was found to be
extrem1y cost prohibitive. Furthermore, by early spring of 1977 treatment
efficiency began to dramatically recover. The treatment performance
is typically illustrated in Figures 5.2, 5.3, 5.4 and 5.5 in the form
of historical plots of weekly average TOC & BOD data during the period
of+June thro¥gh August 1977. The mean effluent BOD and TOC were
37-15 and 56-15 mg. 1. The mean TOC removal within the primary system
was 62.5 percent and the mean TOC removal within the secondary system
was 71.2 percent.
V.3
PERFORMANCE IMPROVEMENT STUDIES
Beginning in October 1977, performance again faltered; however, these
winter temperatures were not as low as the previous winter and other
problems associated with poor performance were explored. This led to:
evaluating waste treatment biokenetics through pilot studies by contracting
external consultants, a detailed "in-house" investigation of changes in
waste composition, and a detailed "in-house II appraisal of nutrient effects
upon waste treatment. All three studies were undertaken simultaneously
to improve system performance shortest possible time. The results of the
materials inventory-waste composition study are pres61ted in Table 5.2.
A dramatic difference in waste composition between 1973. during the EPA
West study (4), and September 1976, a period of equally good waste treat-
ment plant performance, was noted. Epoxy and PVA components dramatically
increased and starches and polyester components dramatically decreased.
yet the bio-treatment system apparently handled these changes. Further-
more, these changes were consistent with manufacturing trends within the
facility, when in early 1976 the Anderson plant changed from intergrated
glass textile and reinforcements manufacture to glass reinforcement
manufacture only. Comparison of September 1976 to November 1977, a time
period of plant upset, revealed no significant difference in the composi-
tion of wastewater. Composition was further evaluated during the period
of treatment plant recovery in three successive weeks in March 1978.
Again, no significant changes in waste composition were observed.
The results of a comprehensive pilot study
biological kinetics on the waste indicated that
kinetics could not be obtained using either BOD
limiting carbon. During these studies nutrient
justify these results; however, a detailed mass
undertaken to determine
Monod or modified Monod
or TOC as a variable of
limitation appeared to
balance of nitrogen and
'-7
-------
TABLE 5.1
PAST PERFORMANCE SUMMARY OF PRIMARY AND SECONDARY SYSTEMS
MEAN VALUES
~ FLOW BOD 1 COD1 'ruC1 TSS1
(gpm) 5
PI2 PE3 SE4 PI PE SE II ~ SE PI PE g
-
1972 335 369 14 1352 614 141 440 94 23
1973 374 450 348 18 1712 847 220 553 143 35
N 1974 333 528 19 1975 764 226 617 114 18
cc
1975 296 313 11 1233 477 124 376 150 32 396 56 8
19765 303 255 10 452 139 36 604 59 13
1. Concentration in terms of mg/l.
2. PI. primary clarifier influent.
3. PE. prfmary clarifier effluent.
4. SE. .econdary clarifier effluent.
5. Values are for five months operation.
-------
800
7(10
600
~ I
500 1 1\ 1
l/; I \. J
I ' , I
I ~f
400+
to.)
\0
:::
r
~ 300
200 ;h
100
,',
j '~"-'
;
.....,',.,'-
~
"
I,
II
I I
1
I I
, ,
I I
I
I
I
,
1
FIGURE 5.2
r
'I
I
I I i I A
I I i I ./ I
J I I 1 I
'I I I 1
'I I I ~ I'
J", I I {I r I
,I' , J I I I "
}. ,If I I f I I 1 ,
V \ I ~ ' ,.J I ' , ~ I 1 . A
V f V I) I \1\( I ",1 II
,," \d I I,ll :~ r
1 \' 'I I ,I'",
~ r 1 I f I , I ,
I I I " 1 I
\ 1 l, I I
V " II
Recycle... t
Bega:l
WEEKLY AVERAGE HISTORY OF EXISTING WASTE TREATMENT OPERATION - TOC
; 1 II
i ':
,I II
(I t : PAW TOC
I I~
I I
I
I
I n
: " A
I I I Ii
: ~\' 1 I I i
~ ~ I, , 'I
, ~ I , \
i I
\ I \ ~!
, I " 'I r
\, I ~ 1\.\'
~ 'v
1 ' V
,
I /'-1
II
I,
~
r I
I A',IT SVdte...
r.~n.."
~
N:.1"'ri.f'r:"
Adc!it ion
Be~an
.....'
I!
J
II ~
I I 1\ r
J' n I
I' -I I
I II I '
I ,I I
r I I
, I, 1 I
~ I I
r I I
1 I
I :
\ I
'I
~
,(r~0\ /\ t-
V ~ I! \J 'J
! I .':
- ,
I. I
) ", I
Primary Effluent
.
E f flueI)jt
"
" I
,.' \ I
'.'
, \
I '.
. \ / I I
,,/ \.
I
~ "
(i Ii I
,i 'I I
I I
!: 'I ~I !
'I) , , \ I
,.1 ; ! 'I,
I i I ~ I I
I I I . I r I , \ I \." I
'I ,I ' '" I oJ
I ,! i ;. I j "
, i I j I I
I "" I I I
I' t I I '
I I
I: l, I I,
\1 I,
J I I
U
--
".
1\ ,. '0"".". ,
, ,
,"0 '
."
~ '
.'1
. ,
,. \ I
,. ,
;",..' ! ../
\' .
, '.
/' ,
I
,
t
~ ~
! I ;'.
i' ,. "
.' . ,
. r I ,1\ . i' '"
, " . \..., . I
-. ...J '. LI
n
I'
I
~
i'
I'. ! i
" ~.\ " ...
'oJ
,"
.'
" .
"......' .- '''..
. , ~' I 'D" . , , , ' I '~I " ',' I . /', ' , "', ' " I . I . I' 1'1 '/ ' I ' I ' 1'1 ' : 'i' J . I " . , , , 'I . I ' , . I . , " I' : ' I ' I 'I ' r . . ' , , , , I ' , . , . .. I ' I . 1'1 . I ' I
:~"'=~=~S;; 00 ~".~.., ~;:::::;:=~=~'" ;::oD~ ,.,~-~ ~;::~ co;::: "'N"~",~~~""~..,~-:!:....,,, ... - . "'...
~~~~"'~"""""""""""'-""~""~,,,,,~,N-N~N~N.---N
-= .c po.. .... .... C) (I) C'" '" 0 0 - - N N N - - (".,; ="'- M ~ ..:r ..:: 11"'1 -... ..c .., :--.. - !"o. X = ::- =- 0 a - - N N ~ - N N C'"'" "" CO') ~ ~ ~ ::-:a ")' ....... ........ fiD:::' ............. .........
-......--.....-- ------, ............ ,..o-.O"oV\
+ ~ ,j
r
.,'\ ;
. .
\.
,
\
WEEK ~~ING 1977
\lEEK E:,;JI:;C
1978
-ED~ E\DI~.( IS- 79
-------
\,oJ
o
so
400
JOC
""
~
.....
,
.....
on 200
c
o
..
Secondary
~,
,,". \ ,
1" \.... I
--.,1 '...'
FIGURE 5.3
WEEKLY AVERAGE HISTORY OF EXISTING WASTE TREATMENT OPERATION - BOD
AJo"T Systes
Startup
Nut rien .
Add i tior.
Bellin
lec' c1..
Began
Primary Effluent
" ;1
!I n
- 'III
III' t
I 'I
. I I t:
f\ I J "
" ";'! \ I i"';
C .' \ f ~ .
, ~ I 'i! I I
I ., .' ; .". i
\ . \. .' \.
I :.. :,
i'\.' \
'1 ;
II
,I
, {
I i (
" I
I .
,.. I I 1
~I
,1
I ~'.,\ II
i: \".1 ~
-' ':' \'.i
If
'j
,.
. ,.
, .
,.'..
I'
'".i ,
,
,
~ I .
I I II
I'
!
,:
Secondary Effluent ,--~ ,
~ /,' \~. , ~
\.. ...... ""._-,."..,'...,,..,- '\. -..,' \'
.,., '-----"-----.-.. -..--...--............:!.. ......"
"' ~ ~.~ ~ ~ ~ ~ N !: ~ ~
\iD""""""''''''''''''''''''''''''''''''''O..........
~ ,...,....caCl)O"O'"""'O"'"
~
Week Endir.g 197i
,
;'. '"
,. '. .'\ i ., ;
... . .'
,
N"""N"""N""":"'~~c::...;...~. .,....
;;<~ '" -;;~~~;:: ;:~~ ~~S
Week Ending 1978
.:: ~ ~ 0 .... - ~ .... - - . -" " N!::: :1", ~::! ~=
N N ':::- ~ ;;- ~ ~ ;: ~ :e: ~ -.:; ~ \b ",.... ....... . ....... . "
- - ... ... 4' -. .. ... . .....
~ee< !~~i~~ l~-~
-------
50
~
0
~ 40
~
iI-e
30
90
80
70
60
20
10
FIGURE 5.4
BOD AND TOC REMOVAL EFFICIENCY IN
SECONDARY SYSTEM
,
,
,
I
BOD
, ~, ,--
" , ....
I , I
; , I
.,
Nutrient Addition Began
Toe
co
-
o
.-I
N 0'1
N 11'\ .... M
-- --
0"" .... N
........ .... ....
...... ....
.... M
- -
N N
.... ....
~ co ....
.... N ....
- --
.... .... ('oj
11'\ ....
('oj ....
- -
('oj M
11'\ ('oj
N co ('oj
- --
c-\ ~ -:t
o
\D ('oj M
- --
11'\ 11'\ ~
r Week Ending 1977+
...... 11'\
.... .... ....
- --
\D ...... ......
C7'I ('oj
('oj ....
- -
...... co
1
Week Ending 1978
31
-------
40
r-i
-
00
s
tI)
Q
~ 3
0
tI)
Q
~
Q
Z
~
p..
tI)
~
tI)
~
z 2
~
3
~
~
~
~
Q
es ~
u
~ 1
N
00 N 11"1
- --
o 0 r-i
r-i r-i r-i
FIGURE 5.5
0\ .....
r-i ('t\ r-i
- --
r-i N N
r-i r-i r-i
t- Week Ending 197 7 "I~
SECONDARY SYSTEM PERFORMANCE HISTORY
SUSPENDED SOLI DS -
r-i ~
M r-i
--
N r-i
r-i
, Nutrient A~tion Began
00 r-i
N r-i
- -
r-i N
11"1
NOON
- --
('t\ ~ ~
N r-i
- -
..... CIO
~
Week Ending 1978
32
-------
TABLE 5.2
S~UffiY ESTLv~TED WASTE COMPOSITION
Waste Treatment Plant Peforwance
EPA Study ~ Poor Intermediate Bio P£rforrnance
A. Uest Group Report Sept. Nov. Week of Week of Week of
1973 1976 1977 3-1-78 3-8-78 3-17-78
Polyvinyl Acetate 26% 32.2 27.7 33.6 35.8 37.1
Polyester 27% 1.0 4.2 8.6 5.4 4.1
Starch 14% 0 0 0 0 0
Epoxide 5% 41.9 42.0 36 31 31
Silicone 5% 6.9 5.9 4.9 4.7 4.2
011 4% 1.1 .9 1.7 1.7 2.1
Aromatic Po1yether 5%
w Po1yo1efin 2% .2 .8 0 0 0
w
Acetic Acid 1% 1.4 1.3 1.5 1.3 1.3
Polyurethane 2% 2.1 1.3 .82 1.14 3.5
Po1y(oxy)ethy1ene Glycol 3% 9.6 13.2 8.9 11.5 12.3
Po1yviny1proyo11idone 2% .9 .6 .45 .35 .28
Dispersants .04 .05 .11 .103 .104
Acetone .6 0 0 0
Citric Acid .0003 .03 0 .07 .07
Formic .0004 .04 0 .1 .1
Chromic Compounds .5 .9 2.52 3.07 3.42
Ammonium Chloride & Hydroxide 1.1 .3 .88 .77 .76
TOTAL % 96% 99.9 100.5 98.4 95.8 99
-------
phosphorous in the full-scale system indicated that they were not
limiting since total phosphorous, nitrates, and ammonia routinely appeared
in the full scale plant effluent at approximately 0.5 mg/l as P, 0.02
mg/l as NO-3' and 0.9 mg/l as NH ,respectively. Initially, there was
reluctance to evaluate increase~ nutrient addition in the full-scale
system since the nutrient residual values in plant effluent would
most probably increase to even higher levels than those previously
cited. However, during the first week of February 1978, nutrient
supplementation was increased to levels outlined in Table 5.3. The
treatment plant responded by a dramatic recovery as illustrated in
historical plots contained in Figure 5.4. By May 24, 1978, the
treatment plant appeared to be in a full state of recovery and the
typical nutrient supplementation program outlined in Table 5.3 was
adopted. Furthermore, a nutrient supplementation program would more
likely be needed after segregation of sanitary wastes from the
industrial waste treatment system.
Prior to system recovery, the advanced waste treatment system
was constructed and started up. However, sanitary segregation and
reclaim supply systems were not completed until March 15, 1978 and
April 16, 1978, respectively. Thus, during the winter of 1977-78 the
tertiary system was operated with the total plant effluent being
discharged. As a matter of record, the tertiary treatment systems
did help overall treatment efficiency during the winter upset; however,
even with tertiary treatment, effluent discharge parameters were
quite often above those specified in the discharge permit. It became
apparent that if nutrient addition did not permanently solve the
'winter upset condition", even with the operation of the tertiary
treatment system during these winter upsets, closed loop recycle plans
not only could be jeopardized, but also discharge standards would not
be routinely achieved!
Presently only one winter has transpired since the nutrient
system was adopted. Treatment plant performance data for the entire
year of 1978 and the 9-month system evaluation period from January
through September also appears in Figure 5.2, 5.3 and 5.4. As
illustrated, the winter upset condition was avoided. As a matter of
fact, winter 1978-79 performance appeared equal to or better than
any other previous winter. These data indicate that the nutrient
program changes have helped avoid winter upset. However, only time
and experience over several successive future winters will increase
the certainty that winter upset within the primary and secondary
system has been controlled. As previously noted, permanent control
of winter upset within the primary and secondary systems is a
mandatory prerequisite for system recycle. This and other problems
related to requirements for recycle will be discussed in further
detail in Chapter VI of this report.
In addition to the changes in nutrients previously note, modifi-
cations were made to the inlets and outlets of the secondary clarifiers.
34
-------
TABLE 5.3
NUTRIENT SUPPLEMENTATION SCHEDULE
* SECONDARY EFFLUENT
DATE GPD OF NUTRIENT LEVEL
FEED TOTAL P NO - NH3
3
mg/l mg/l mg/l
February 9, 1978 5 0.6 0.10 1.5
March 1, 1978 10 0.7 0.15 1.8
March 28, 1978 15 1.0 0.20 2.0
TYPICAL NUTRIENT SUPPLEMENTATION
Final schedule based upon maintenance of 1.0 mg/l total P and
**
2.0 mg/l NH3 in secondary effluent.
*
Nutrient Peed:
A 9% (W/W) SolutIon of Po~3
A 24% W/W Solution of NH3
**2.0 mg/l NH in secondary effluent resulting in 1.6 mg/l
in tertiary3effluent discharge as specified by NPDES permit.
35
-------
These modifications consisted of distributing the clarifier influent
evenly over the entire width of the tank to improve plug flow and
installation of target baffles to dissipate inlet energy and thus
provide a greater area for quiescent settling. Additionally greater
weir length was provided for each secondary clarifier outlet to
minimize suspended solids carry-over.
As a result of the biokinetic studies, it was decided that a
sludge age of approximately 15 days was optimum for summer operation
and approximately 30 days was optimum for winter operation of the
secondary system. Since the overall waste treatment system is rela-
tively small, it was believed that better, more accurate, control of
sludge age could be obtained by direct mixed liquor wastage from the
aerator rather than through sludge wastage from the return-sludge line.
In addition, sludge age would be relatively independent of sludge
concentration. Thus, a variable speed positive displacement pump was
installed in the spring of 1978 when sludge age control through wast-
age from the aeration tanks was adopted. These improvements have
allowed accurate control of sludge age. Control is achieved by
pumping between five to twenty gpm of sludge from the aeration tanks.
The wasted sludge is returned to the aerated equalization tanks,
which in effect gives additional degradation of incoming waste
materials, acting as a roughing biological pretreatment unit. The
sludge is eventually removed along with raw sludge in primary clari-
fiers and is then pumped to aerated digestor.
V.4
OPERATIONS AND CONTROL SUMMARY
As part of the final activity related to the grant, a complete
mffi1ual was generated to aid operations for the proper control of the
primary and secondary systems. A summary of these operational pro-
cedures is presented in Appendix B of this report. It also contains
detailed descriptions of all primary and secondary waste treatment
water processes.
36
-------
CHAPTER VI
DESCRIPTION, START-UP, AND OPERATION OF THE ADVANCED WASTE
TREATMENT AND RECYCLE SYSTEMS
VI.l
SYSTEM DESCRIPTION
A schematic diagram of the advanced wastewater treatment system
as designed is illustrated in Figure 6.1.
The facility includes three major unit processes: sand filtration,
activated carbon adsorption, and chlorination. Other facilities
included in the system, which are pertinent to wastewater reclaim,
include: off specification basin, distribution tank and recycle
supply pumps, and reclaimed wastewater storage basin. It should be
noted that certain differences exist between the design presented in the
Preliminary Engineering Report (1) and the final design as presented
in this report. Thus, the final design and operation of the system
will be presented and contrasted to the preliminary design in this
chapter.
VI.l.l
SAND FILTRATION
Sand filtration is employed to remove suspended solids remaining
in the secondary clarifier effluent of the existing facility. The
effectiveness of this process is especially important for two major
reasons: (1) the effluent eventually enters the plant process and
cooling systems in production areas, and (2) high suspended solids
levels impair the operation of the carbon adsorption system.
Sand filters were originally evaluated in Phase I of the Grant in
both upflow and downflow configurations. These pilot test data were
fully documented in the Preliminary Engineering Repor~ (1). At that
time, downflow pressure filters operating 1t 4 gpm/ft were recommended.
Filters were to be backwashed at 20 gpm/ft with 5 cfm/ft2 air scour.
Upon cost effective evaluation during later stages of design, it
was decided that single and/or multi-media pressure filters would be
cost prohibitive. Several design alternatives became apparent: one being
upflow sand filters and the other being an innovative gravity fed single
media-automatic intermittent air scoured filter. Even though pilot data
indicated that the upflow sand filter could produce acceptable solids
removal efficiencies, past experiences indicated that positive control
of suspended solids breakthrough would be difficult, if not impossible.
If suspended solids would periodically break through the filter, it
was feared that carbon adsorbers would be routinely plugged and fouled.
In other words, depth filtration was desired from a filtration process
efficiency; however, the low probability of solids breakthrough feature
common to surface filtration was a must, (i.e., surface filtration filters
will usually plug and/or blind providing prohibitive flows and head loss
37
-------
FIGURE 6.1
WASTE TREATMENT SYSTEM DESIGN
o - 60 Reclaimed Wastewater Return
~
SAND
FILTERS
4
307
307
~
PRDfARy(2) AND
SECONDARY
SYSTEM
313
,
259 (1)
-.-
---,,- CARBON
ADSORBERS
j
-.
....
6 30
By-Pass During
Filter Backwa~h --.I MUD' 1
-1 WELL
30
16
~ACKWASH -.
LSUPPLY ~
46
54 36 Adsorber
\D
Backwash .-.
~ Effluent
00 ~ OFF-
SPECIFICATION 2
BASIN Drain Water
Off-Specification Return
J,
Emergency
City Water
,
..
Motive Water
.-J CHLORINE
......., AD}) ITI ON
I 261
I
4
DISTRIBUTION
TANK
0-60
o
ChlorinE
Solution
,
To Duck
Pond
N
.
RECLAIMED
WASTEWATER
STORAGE
-
....
,
261
259
-
-r
To Process and Cooling Uses
NOTE: _(lL All numbers are average flow. gpm
-------
prior to suspended solids breakthrough). Thus, another alternative, a
uniform bed sand filter utilizing automatic intermittent hydraulic and
air scour to redistribute accumulated solids was selected. Two 68
square feet filter cells were designed for an average hydraulic flux
of 2.5 gpm/ft2 with a maximum of 4 gpm/ft2. Design criteria for
suspended solids included a maximum of 25 mg/1 for influent water and
filtered effluent not to exceed 5 mg/1. In addition, the final filtrate
was specified to contain not more than 2 mg/1 suspended solids when the
influent contained less than 10 mg/1 suspended solids.
During a typical filter cycle, secondary effluent enters the
filter cell through proportioning weirs. The effluent cascades to a wash
water trough and is distributed by a trough to V-notch weirs. The
effluent flows out through the V-notch weirs, impinging on a splash
plate, and then to the filter media surface. As the effluent reaches
the filter media surface, all but the very fine particles are retained
on the surface of the quartz sand media have an effective size
of 0.45 mm and a 1.5 uniformity. Fine particles enter the interstices
and are trapped in the media. In time, large particles may completely
cover the filter media surface, causing the liquid level to rise over
the media surface.
A rising liquid level actuates an air mix system, causing low
pressure air to be distributed to a tubular air diffuser. Fine air
bubbles leaving the diffuser create a gentle rolling motion over the
filter surface. This gentle rolling motion over the filter surface
entrains the large particles previously trapped on the filter surface
and holds these particles in suspension allowing for continued
filtering with minimal head loss accumulation. Depending upon the
amount of solids trapped on the filter surface, this action may cause
the liquid level to drop in the cell.
The incoming and previously collected suspended solids eventually
create higher head losses and in time the liquid level (headloss) will
rise to a point at which a pulse mix is energized. The pulse mix system
immediately closes the underdrain outlet valve, trapping atmospheric
air in the underdrain, and simultaneously energizes the backwash pump.
The backwash pWlIp causes the liquid level to rise in the underdrain.
This rising liquid, acting as a piston, compresses the previously
trapped air which now passes up through the media to the surface,
dislodging particles previously trapped in the interstices. These
solids expelled from the interstices are now entrained in the gently
moving admixture above the media surface.
After 20 to 30 seconds, the pulse mix is terminated and the
backwash pump shuts down. The admixture level over the filter media
surface drops because the particles previously trapped in the upper
interstices of the filter media have been driven out. The previously
described air mix system then keeps these particles in suspension
above the media rather than immeshed in the media, thus extending
39
-------
the filter run.
After a period of time, fine particles will again be trapped in
the interstices of the filter media, causing the liquid level (headloss)
to rise, energizing the pulse mix system and repeating the procedure
previously described. Three (3) to five(5) pulse mix cycles are used
prior to a major backwash. The frequency and duration of the pulse
mix cycles can be varied, depending upon the quality of the secondary
effluent. There are two options for energi~ng the pulse mix system,
which are based either on the quality of the secondary effluent, a
manual option, or on the liquid level in the filter cell, an automatic
option.
In time the pulse mix system register reaches the pre-set number
of pulses and is deactivated. The admixture level rises in the filter
cell to the point of maximum head loss and the backwash punp is
energized. The backwash trough outlet valve opens automatically permit-
ting the mixture of high solids to drain over the wash water trough into
the mudwell. Drain period is automatically controlled by an adjustable
timer. The air diffuser stays on during this sequence and the subsequent
backwash cycle, functioning as an air wash mechanism.
Each cell is backwashed for 3 to 5 minutes at a flux of approximately
12 gpm/ft2. Both cells cannot be backwashed simultaneously. During
backwashing of a cell, incoming wastewater to this cell is collected
with the spent filter backwash volume. Thus, during a cell backwash,
approximately only one-half of the secondary effluent feed is processed.
Spent backwash and unfiltered feed are diverted to a 9000 gallon
capacity mudwell and pumped to the 1.5 million gallon off-specification
basin for return to the equalization basin for further processing through
the primary and secondary system. Filtered water continues to drain
through the filter underdrain system to a 12,000 gallon clear well.
The clear well serves as a wet well for pumps to feed the carbon adsorbers.
VI. 1. 2
CARBON ADSORPTION
The carbon adsorption system contains three pressurized fixed bed
carbon adsorbing columns. The column tanks are filament wound glass
fiber reinforced polyester tanks equipped with appropriate inlet, outlet,
and backwash hardware, capable of working pressures between 55 psig
pressure and 1 psig vacuum. The columns were designed to withstand these
pressures on either or both sides of underdrain media supp~rt plates.
Each column has a cross sectional surface area of 78.5 ft. , a depth of
10 feet, and contains approximately 200,000 lbs. of activated carbon. The
design allows for two columns to operate at any given time while the third
column is a standby.
Activated carbon adsorbs materials either from liquids or gases
because it has a highly porous structure. Each carbon granule contains
a vast interconnecting pore network of various sizes. The smaller pores
4u
-------
are Dearly the same size as the molecules being adsorbed. This great
porosity provides a very large surface area for adsorbing molecules
and hence a very large adsorptive capacity. Almost always, adsorption
onto activated carbon is a result of Van der Waals, or dispersion forces.
These forces exist among all molecules or atoms whether or not
they are chemically combined and are related to the forces responsible
for condensation or liquefaction of vapors. Generally, molecules
of higher molecular weight are attracted more strongly by carbon than
lower weight molecules. Hence, activated carbons have a preference
for high molecular weight substances, provided they are small enough
to enter the carbon pore structure. Additionally, activated carbon
prefers non-polar substances. Thus, quite importantly, there is an
affinity for the adsorption of non-polar organic molecules from
polar solvents such as water. Thus, activated carbon is especially
effective in removing dissolved organic pollutants from water.
An important concept in the design of granular activated carbon
processes is the adsorption column breakthrough curve. If a solution
containing an adsorbable substance passes through a granular activated
carbon bed or column, a plot of adsorbable substance concentration
in the effluent versus the solution volume passed through yields a
breakthrough curve. The adsorbing operation of the fixed bed system
can be based on the concept that at breakthrough through the lead
column of two series operating columns, the column farthest upstream,
is taken out of service. The lag column which is partially loaded to
capacity is then placed into the lead position and a fresh column is
placed into service as the lag column at the downstream end of the
series. In this so-called "flip-flop" operation, the carbon taken out
of service is in nearly complete equilibrium or saturation with the
incoming stream. Thus, the adsorptive capacity is more fully utilized,
reducing the quantity of carbon to be replaced or regenerated as
compared to a single fixed bed system. This type of operation improves
the economics, especially for high pollution loads. When the adsorptive
capacity of the carbon in the lead adsorber is reached and that adsorber
is taken out of service, as discussed above, the "spent" carbon is
then hydraulically conveyed to the spent carbon tank, constructed
similar to the aforementioned fiberglas carbon vessels.
Upon economic appraisal of carbon regeneration, it became
apparent that on-site regeneration would be more costly than carbon
services purchase through major supply vendors. Thus, fresh carbon
is transported to the plant by bulk trailer, where it is hydraulically
transferred into the empty carbon adsorber vessel. After the transfer,
the used carbon contained in the spent carbon tank is hydraulically
conveyed to the empty bulk trailer for transport back to the vendor's
regional off-site thermal regeneration facility.
All design criteria were based upon extensive bench and pilot
scale treatability studies outlined in the Preliminary Engineering
Report (1). The design hydraulic flux for the carbon columns
arranged in series was 4.5 gpm/ft2. based upon average flows. This
41
-------
loading provided in average carbon contact period of 33 minutes,
since two series adsorber vessels, each 10 ft. diameter x 10 ft.
deep were to be used. Organic carbon removal efficiencies were planned
at approximately 50% based upon 34 mg/1 TOC influent and 17 mg/l TOC
contained in the effluent. Average design conditions also provided
for 667 lbs. of carbon to be exhausted per day which would provide
an average carbon replenishment need of one carbon adsorber vessel
recharge, 20,000 pounds, every 30 days.
Every two to three days, with a maximum of once per day, the
on-line series carbon adsorbers can be backwashed in parallel with
reclaim water at a flux of approximately 15 gpm/ft2 for 20 minutes.
Spent backwash water is collected in a 1.5 million gallon capacity
off-specification basin for eventual return to the equalization basin
for further processing through the primary and secondary system.
VI. 1. 3
DISINFECTION
The chlorination system was designed to provide disinfection of
treated industrial wastewater prior to reuse within the production
facilities. Carbon adsorber effluent flows by gravity to a 1920 gallon
capacity chlorine flash mix tank to allow a contact period of approx-
imately 6 to 7 minutes. based upon average flow. Gaseous chlorine,
fed from one ton storage cylinders into a recycled water stream. is
then mixed with carbon adsorber effluent prior to the flash mix tank.
In the flash mix tank the chlorine solution and treated wastewater
intermix. The mixture is pumped to the 190,000 gallon capacity
distribution tank for further contact and disinfection. The distribution
tank provides a hydraulic retention period of approximately 10 to 11 hours
during average recycle flow rates. Treated water containing a free
chlorine residual is supplied to plant manufacturing areas by high
pressure centrifugal pumps. The pH of carbon adsorber effluent should
be between 6 and 7, slightly acidic, such that most of the free chlorine
is in the form of HOC1 which is about 40 to 80 times more effective for
disinfection than OC1. With the free ammonia level in the range of 0.5
to 1 mg/1, and an average organic chlorine demand of 1 to 2 mg/l
chlorine, the total chlorine demand through the system ranges from 2.5
to 5 mg/l. Thus. between 3.5 and 6mg/1 chlorine can be routinely
supplied through the system.
VI.2
HYDr~ULIC CONSIDERATIONS FOR DESIGN OF THE ADVANCED WASTE
TREATHENT SYSTEM
The design flow for the advanced wastewater treatment system was
set at 285 gpm in the Preliminary Engineering Report (1). This figure
was determined through anticipation of an average 205 gpm net discharge
of wastewater from the manufacturing facility. A correction of
anticipated values by 10% to accommodate flow measurem~1t inaccuracy.
and a need to handle an additional 54.3 gpm of water during the summer
months. which was stored in the reclaimed water storage basin during
winter months, would produce (205 gpm + 54.3 gpm) 110% or 285 gpm flow
42
-------
which was allowed
treatment system.
be processed only
for in the preliminary design of the advanced waste
The flow of 5~ gpm of stored reclaimed water was to
through the advanced waste treatment system.
In the preliminary design, it was anticipated that sand filter
backwash and activated carbon backwash would amount to 43,120 gpd and
47,120 gpd, respectively. These sources would produce an average
discharge of 30 gpm, respectively and should have been included into
the design flow rate for the AWT system, since they were planned for
recycle back through the treatment process. Upon final design, these
backwash flows were included into the hydraulic design; thus, the
primary and secondary system was planned to be hydraulically loaded
at approximately 327 gpm. This flow was arrived at by considering
226 gpm waste water flow from manufacturing (with all originally
conceived "piggyback systems "), an additional 33 gpm of manufacturing
waste water generated by segregation of "piggyback" systems (see
Table 4.4 and Section VI.3.4 and VI.3.5), and an additional 68 gpm
wastewater flow spent advanced waste treatment backwashing operations
and waste treatment cleaning, carbon transport and general operations.
The effects of this hydraulic flow of 337 gpm through the primary
and secondary treatment system were not considered to be unreasonable.
Review of past operational data summarized previously in Table 5.1
indicated that the primary and secondary system had the necessary
hydraulic flow capacity of 300 to 330 gpm. Latter sections in this
chapter will address hydraulic effects of the advanced waste treat-
ment system upon the primary and secondary treatment systems. It is
also noted that during the operational performance over the five month
period in 1976, waste character as previously presented in Table 5.2
was that which was expected during operation of the recycle system.
Finally, it should be noted that the raw waste composition
observed during the early pilot trails, 1973, see Table 5.2, was
also expected to be representative of the waste at the time of recycle
system operation. However, waste composition did not remain constant,
shifting more towards PVA's and epoxy's with fewer starch materials.
VI.3
SYSTEH START UP
The major units of the advanced treatment system, filtration,
carbon adsorption and chlorination, began operation in October 1977.
At this time sanitary segregation was not completed. Thus, between
October 1977 and March 1978 combined sanitary and industrial wastes
were treated through the advanced treatment system and discharged.
It is important to note that a definite prerequisite, based upon
potential health hazards, for recycle was the removal of all sanitary
wastes from the industrial waste collection system. During March 1978,
the segregated sanitary waste collection and treatment system
became operative. Between March 1978 and April 1978, biological studies
were performed to assess viral and bacterial contamination of segregated
industrial wastes. These studies began after a minimum of three weeks
43
-------
of continuous operation of the entire industrial waste treatment system,
(i.e., primary, secondary, and advanced).
VI.3.1
BIOLOGICAL QUALITY OF SEGREGATED INDUSTRIAL WASTE WATER
Totally free of sanitary-domestic discharges, it was believed that
the three week period of time would allow the treatment system to
be purged of any residual pathogenic organisms that would enter the
proposed recycle system through sanitary waste collection systems. Thus,
a nationally recognized consultant specializing in virus and bacterial
evaluations was contracted to perform the evaluation. Virus sampling
was performed during the week of April 17, 1978 and bacterial sampling
was performed during the week of April 26, 1978. The procedures used
for sampling and analysis were similar to those outlined in Standard
Methods for the Analysis of Water and Wastewater (6). A detailed
description of the procedures used is presented in Appendix C.
Results of the virual plaque assay on primary rhesus monkey kidney
cells and HeLa cells for a wastewater sample along with those of a
control tissue culture plate are provided in Table 6.1 and 6.2. These
results indicated no plaques on HeLa cells for the wastewater sample
or the control. Plaques were found on the primary rhesus monkey kidney
cell (PMK), see Table 6.1. These PMK plaques are of Simian origin
due to a foamy type virus which is indigenous to the animals and
frequently found in PMK cells. In summary. no virus plaques related
to the wastewater sample were recovered in either primary rhesus
monkey or HeLa cell cultures.
Results of the field efficiency run are provided in Tables 6.3
and 6.4. The stock virus solution was found to contain 2.5 x 102
plaque forming units (PFU)/ml on PMK cells and 3.03 x 102 PFU/ml on
HeLa cells. Assay of the process wastewater, which was seeded with
the stock virus solution, provided a total virus recovery of
4.5 x 104 PFU/ml on PMK and 3.26 x 102 PFU/ml on He La cells. The range
and efficiency of recovery between 9% and 53% was not a result of
variation in the field concentration procedure, but was a result of
tissue culture response to the virus during laboratory incubation.
It was believed that the titration performance of the HeLa cell cultures
was doubtful and they were repeated to eliminate the possibility of a
technical error during the inoculation. Thus, HeLa cell efficiency was
found to be incorrect and the 9% efficiency as found on PMK was accepted
as the efficiency of recovery and is a value typically found in waters
of industrial origin with complex organic constituents.
Results of the bacterial testing are provided in Table 6.5. No
fecal coli forms or fecal streptococci were found in sample sizes of
up to 250 mI. Total plate counts were found to range between 2,650 cells/
liter and 4,450 cells/liter depending on the technique and medium
utilized. Obli~ate aerobes, facultative organisms, and obligate
anaerobes were found in wastewater samples as well as staphylococcus.
44
-------
TABLE 6.1
PLAQUE ASSAY OF WASTE SAMPLE IN PRIMARY
RHESUS MONKEY KIDNEY TISSUE CULTURE
WASTEWATER SAMPLE *
Plate Number of Plaques Virus Identification
1 0
2 2 2 Simian foamy type
3 3 3 Simian foamy type
4 0
5 0 .
6 2 1 Simian foamy type
1 passed to new cell
culture (on test)
7 0
8 0
9 1 1 passed to new cell
culture (Simian foamy type)
10 0
CONTROL **
Plate Number of Plaques Virus Identification
1 1 1 Simian foamy type
2 3 3 Simian foamy type
3 2 2 Simian foamy type
*
0.4 ml/plate of undiluted sample
**
0.4 ml/plate of tryptose phosphate broth
45
-------
'Jl\BLE 6. 2
PLAQUE ASSAY OF WASTEWATER SAMPLE IN
BeLa CELL CULTURE
WASTEWATER SAMPLE *
CONTROL **
Plate
Number of Plaques
Plate
Number of. Plaques
1
2
3
4
5
6
7
8
9
10
o
o
o
o
o
o
o
o
o
o
1
2
3
o
o
o
*
0.4 ml/p1ate of undiluted sample
** 0.4 ml/p1ate of t~tose phosphate broth
46
-------
TABLE 6.3
PLAQUE ASSAY OF STOCK VIRUS SOLUTION AND VIRUS -
SEEDED WASTEWATER IN BeLa CELL CULTURE
STOCK VIRUS SOLUTION
No. Plaques No. Plaques No. Plaques No. Plaques
Dilution Plate 1 Plate 2 Plate 3 Average PFU/ml
0.2 ml/p1ate undiluted 2
19 19 7 13.33 .66 x 10
101 2
4 3 No Test 3.50 1. 75 x 10
102 2
1 3 0 1.33 6.67 x 10
103 0 0 0 0
.J:-
..... 104
0 0 0 0
VIRUS-SEEDED WASTEWATER
No. Plaques No. Plaques No. Plaques No. Plaques
Dilution Plate 1 Plate 2 P1ftte 3 Average PFU/ml
0.2 ml/p1ate undiluted 12 13 No Test 12.50 2
.625 x 10
101 4 6 5 5.00 2
2.50 x 10
102 3 1 0 1.33 2
6.65 x 10
103 0 0 0
*Stock and seed virus titrations on test for verification of recovery.
Avera.ge
PFU/ml
*
.2
3.03 x iC
Average
PFU/ml
*
2
3.26 x 10
-------
TABLE 6.4
PLAQUE ASSAY OF STOCK VIRUS SOLUTION AND VIRUS-SEEDED
WASTEWATER IN PRIMARY RHESUS MONKEY KIDNEY TISSUE CULTURE
STOCK VIRUS SOLUTION
No. Plaques No. Plaques No. Plaques
Dilution Plate 1 Plate 2 Plate 3
0.2 ml/plate undiluted Total CPE * Total CPE Total CPE
101 TNTC ** TNTC !NTC
102 TNTC TNTC TNTC
103 TNTC TNTC TNTC
~
co 104
6 7 2
No. Plaques
Average
PFU/ml
5
5
2.5 x 10
VIRUS-SEEDED WASTEWATER
No. Plaques No. Plaques No. Plaques No. Plaqu~s
Dilution Plate 1 Plate 2 Plate 3 Average PFU/11l
0.2 ml/plate undiluted Total CPE Total CPE Total CPE
101 TNTC TNTC TNTC
102 TNTC TNTC TNTC
103 17 9 4
5 5 4.5 x 10
* Cytopathic effect
** Too numerous to count (or very large and plaques ran together)
-------
TABLE 6.5
BACTERIAL TESTING RESULTS
1M!
Total Plate Count
(membrance filter technique)
Fecal Streptococcus
Staphylococcus
Fecal Coliform
Obligate Aerobes
(nutrient agar - pour-plate
technique)
Facultative Organisms
Obligate Anaerobes
Total Bacterial Count
(nutrient agar - pour-plate
technique)
4~
ORGANISM CONCENTRATION
(cells/liter)
2650
o
3.5
o
(total areobes - total
faculatative)
1550
1950
(total anaerobes -
total facultative)
950
4450
-------
None of the concentrations were in a range that would cause health
concerns for nonpotable use of this water in industrial applications.
In summary, it was concluded that virus, fecal streptococcus and
coli forms , and staphylococcus levels were so low that it was not
anticipated that these organisms would cause health problems within
the reuse system. Furthermore, similar conclusions were obtained
pertinent to broad general classifications of bacteria as obligate
aerobes, facultative organisms, and obligate anaerobes. Thus, these
results indicated that biologically, the finished water would be
acceptable for a reclaim supply.
VI. 3.2
LOW & HIGH QUALITY RECYCLE SYSTEMS
The recycle scheme, as previously discussed in Chapter IV,
described that reclaimed wastewater would be supplied to plant
process cooling systems and plant process cleaning and washdawn
operations. These potential demand areas were classified into
two groups: cooling system-high quality demands and washdawn
operation - low quality demands.
Beginning on April 16, 1978, isolated low quality demand areas
were brought "on-line" with reclaimed wastewater. The program was to
gradually increase recycle systems systematically by adding more and
more low quality water demanding areas into the recycle system.
High quality water demanding areas of process cooling were
avoided until pilot cooling loops similar to those used and documented
in the Preliminary Engineering Report (1) were operated on reclaim water.
The pilot cooling trials using reclaim water were conducted from June
through July 1978. The pilot evaluation of the effects of reclaim water
upon the high quality demanding process cooling systems indicated no
significantly different results than those obtained and documented in
the pilot cooling studies outlined in great detail in the Preliminary
Engineering Report (1). With a successful outlook, high quality
water demanding process cooling systems were systematically connected
to the recycle system one by one beginning with the smallest and
ending with the largest during July 1978.
A historical graphical presentation of the average weekly hydraulic
flows: wastewater processed, wastewater discharge and treated waste-
water recycled is presented in Figure 6.2. In summary, the important
events were:
(1)
March 1978 sanitary segregation with total discharge;
(2)
April 16. 1978 beginning of low quality demand recycle:
(3)
The end of July 1978. the beginning of high quality demand
recycle.
50
-------
As can be seen in the figure, sanitary segretation did create a
measurable reduction in wastewater discharge. Between January 1, 1978
and March 11, 1978, the weekly average wastewater flow was 0.382 mgd.
During this period combined sanitary and industrial wastes were
treated through the advanced waste treatment system and discharged.
Between March 11, 1978 and April 15, 1978, the weekly average flow
was 0.332 mgd. Thus, sanitary segregation accounted for an average
flow reduction of approximately 0.05 mgd, as expected.
VI. 3.3
SUPPLY SYSTEM HYDRAULIC PROBLEMS
On April 15, 1978, low quality recycle began. As illustrated in
Figure 6.2, the recycle flows increased as each low quality demand was
incorporated in the system. During July and August, high quality
demand addition to the system produced a continued increase in recycle
flow. It is noted that discharge dramatically dropped during the
initial recycle startup. However, periods of time have existed when
the average daily total treated water flow has been in excess of the
recycle demand; e.g., see periods 7/8/78 - 9/16/78 and 11/18/78 -
5/12/79. This condition creates surplus wastes within the system which
must be stored and/or discharged. If water is stored in the reclaim
storage pond, it must be reprocessed through the AWT system prior to
use in the reclaim system, see Figure 6.1. If waste water flows are
high, use of stored reclaim water to supply peak "instantaneous" recycle
demands can hydraulically overload the AWT system. Thus, during these
periods fresh water makeup was used to supply peak "instantaneous"
demands.
It is important to note that wastewater processed flows do not
necessarily equal the sum of recycled water flow and discharged waste-
water flow. This inequality is due to the addition of fresh water
into the system at the recycle distribution tank to maintain an
adequate supply to the plant. As previously stated, periods of time
have existed when the treatment system count not process wastewater
and/or stored reclaim water at high enough flow rates to satisfy
demand flow rates. This problem of hydraulic balance has been due to
several factors; some of which have been easily corrected and others
which are presently being evaluated. By the end of July. after all
major high quality demands were on line in the recycle system, it
became apparent that the system was being hydraulically overloaded
with fresh makeup water at the distribution tank, see Figure 6.1.
Huch of this overload was created by large peak demands. It is
noted that all flows represented in Figure 6.2 are weekly averages
of average daily flows.
The supply system was equipped with pumps capable of delivering
70 to 80 psi pressure at all points within the distribution system.
The 70 to 80 psi was designed to aid cleaning in washdown operations
and to encourage use, from a psychological point of view, of reclaim
water within the manufacturing plant. Daily results indicated that,
51
-------
U.
N
FIGURE 6.2
AVERAGE WEEKLY FLOW IN RECYCLE
SYSTEM
Flow Excluding AWT System
Backwash
Wastes
0.5
2
Ii
, \
, \
"'4 \
,
¥' \;\
y A '
, I \ ,
. . }-!\ AfJ
I ~,;' \Vt\il'~ i
....~I ~! . 'J Hilii'. .,
.:. "" QU-;i~\. {
=..:.1 ~ \ ( I Re~;~~~ B~ga.' ~. '\
~I ~ \ I I. \. t. .\4-Discharge From Total System
~ ~, , /'.
U)~ (' '. !.,! i.
..."'''w , E 1\ .i ':\ .'~ I" .' \
~ t I . i
it ~! ~ Ii! ~,.\ ; '. . ~ \ f
:t . I I '.", .1'1' /\ 1\ 1
~I ~ I ~ (\ : \ ..' ',.,' .". "-", ; \. I . \ I '.' . \
J \.. ! \' \,.-',,.; \! '!. '.1 \. p. ! " !
" \f \ ,.. '. i ~ ~ ~ . \ I \. .'
. -'.'I'I'I';'I'j'I'I'I'~-'"'1'1'1'1'1'1'1'111'1'''11~'i"-i .\..t'I'
.... CI) co ".... ='" r-. ,.... C':: N C\ . 0 ..:: co - In M 0 ,.... ,...,....... CD N \D C""I ~ aD "" 0\
""'N..:r - ~ -- .- ~ - N - NQ) N ""... N"" C""'\ - N- NO\ N \CtN ~.... t""\ ...."", ""'1'1"" 1'10\ N"- N 4""'''''''''1'1
.............. ................ ...... ............... ...... ...... ........ ........ ......... ................. ........ ........ ...... .............. ........ ....... .............. ................ ........ ................ ........ ................ ................ ....................... ............... ................ ........ ........ ................ ................
......... .... N f'*'. ~ ~ ..:: ..:: &I'" "" \iC .:1"'" ,..., CC' co C\ 0\ CJ\ 0 C - ~ N N .... ... N N C""\ .... ~ -4' <' W"'\ '" \D \C ,.... ,.... GO co 0\ 0\ 0\
- ......... .......... .....
- ,
/ ' J. \
\ ,- I
J \ 1
\ ~ ,/
\ , \ " J \-1'\
\/\J ~ 1\ I V
\ , \ )
\ I \, '---
' )
..j Recycle
Demand
0.4
?,0. ~
i
....
0.2
0.1
~
~
1~
Week Endinl 1971
\!~ek E::dir.g 197E
-------
not only was reclaim supply used, but the high energy condition
of 70 to 80 psi also increased instantaneous flow demands such that
significant volumetric deficiencies existed in the distribution
tank. Since the volume of the distribution tank supply was relatively
fixed, it was decided to reduce the pressure head by approximately
40% down to 40 to 50 psi within the plant. This was accomplished
by reducing the speed of the belt driven supply pumps.
The effect of supply pressure reduction on lowering the overall
demand of recycle water can be seen in Figure 6.2, beginning at the
end of September 1978. Not only was the average daily demand for
recycle reduced by approximately 0.1 mgd, but also peak demands were
considerably reduced.
VI. 3.4
INERT GAS SCRUBBER RECYCLE PROBLEMS
During August and September of 1978, a hydraulic audit was done to
identify other unplanned sources of fresh water additon to the recycle
system. Upon inspection, it was found that the inert gas scrubber was
not connected to the mat line in a "piggyback" fashion as planned and
documented in the Preliminary Engineering Report (1), in order to
minimize distribution piping. However, the scrubber and the mat line
were both operating on recycle water at approximately 40 to 30 gpm,
respectively. By November 1978 it became apparent that the reclaim
supply pressure to the inert gas scrubber was too low. Thus, the
scrubber was operated on fresh water for an interim time period until
March 1979 and during the February hydraulic audit, when appropriate
modulation valving was received and installed. Presently, no problems
related to the use of reclaim water in the inert gas scrubber have
been observed.
As an added advantage, utilization of reclaim water as feed for the
inert gas scrubber eliminates approximately 37 gpm of fresh water
addition into the overall system. As such, the average design discharge
during winter becomes less than the minimum design recycle demand, see
Tables 4.3 & 4.4 and Figure 4.2. This is highly desirable since the
1.5 million gallon reclaim basin should not be needed for winter
storage, i.e. the system design is hydraulically balanced during winter
and summer months, provided excess fresh water uses are controlled.
From a practical point, storage would only be needed to accomodate
occasional discharge surges that the reclaim supply tank cannot
accomodate.
VI.3.5
MAT LINE RECYCLE PROBLEMS
Figure 6.3 illustrates the process flow diagram for the mat line
and dissolved air flotation (DAF) system as planned. The high pressure
chain wash sprays were to be operated on reclaim water. Wastewater
from the mat lines, laden with fibers was to be treated through a
dissolved air flotation unit. A pressurized 70 psi, 10 gpm recycle
stream was to supply the necessary dissolved air for fiber separation.
53
-------
High pressure
chain wash
10 ~pm
10 gpT:l
from
reclaim
supply
Fiberous
Sludge
Float
Net Effects:
FIGl'RE 6.3
PROPOSED MA1~INE RECYCLE SYSTEM
WITHOUT "PIGGYBACK"
MAl LINE PROCESS
20 gpm
30 gpm
DISSOLVED
AIR
FLOTATION
30 gpm
10 gpiR
20 9pm
To Wa.~te Treatment
20 gpm Recycle Demand
20 gpm Wastewater Discharge
Fibers Removed
54
PRESSURE
STAGE
-------
Upon installation and start up in the fall of 1978, although good
fiber separation was achieved, the DAF effluent was unacceptable for
recycle to the pressurization vessel. Residual fibers caused plugging
problems in the DAF eductors. Strainers were installed in the pressure
recycle line to attempt residual fiber removal. Although the strainers
did improve the operation, intermittent eductor fouling remained.
Thus, the system was modified as illustrated in Figure 6.4. Reclaim
supply water is currently used for pressurization. Although operational
problems have been minimized using this scheme, 20 gpm more are
recycled in this area than originally planned.
A major problem was encountered during startup of the mat line
processes with reclaim supply water in September 1978. Reclaim supply
water created major plugging problems in the high pressure
chain wash spray nozzles. After several trials, the high pressure
chain wash system was connected back to fresh water. After several
months, during the spring of 1979, another attempt was made to bring
the spray wash system onto reclaim supply water. Again, nozzle
fouling and pluggage was encountered. Thus, spray washes are
currently supplied by city water, creating an additional 10 gpm of
unplanned fresh water makeup to the overall recycle system.
At the time of writing of this report, a third set of trials
were made and indicated reclaim water can be used to supply the high
pressure mat line wash system.
VI.3.6
OTHER SYSTEM HYDRAULIC PROBLEHS
Results from a detailed hydraulic water audit conducted from
February 5 through February 28, 1979, see Table 4.3, 4.4 and Figure
4.2 (a and b), indicated that the raw waste flow rate averaged 361 gpm,
which was 40 percent larger than the expected modified design of 259.
Furthermore, the recycle flaw averaged 270 gpm or 24 percent larger
than the average winter reuse flow of 218 gpm expected without system
modification as presented in VI.3.4 and VI.3.5. Thus, the recycle
flow was 52 gpm larger, with Matline & IG Scrubber separation.
(Table 4.4), which would also make the raw waste flaw 52 gpm larger
than the expected design flow, totaling approximately 257 to 278 gpm.
Therefore, the real excess fresh water flow implied is between
(361-257) and (361-278) or between 104 and 83 gpm. This extra flaw
must be extra fresh water entering the system over and above the
amount of planned city water input of 73 gpm (Figure 4.2). The addi-
tional fresh water flow of 83 to 104 gpm was found to be caused by:
leaks in condensate return lines through the plant ~20 gpm; leaky
valves throughout the plant ~20 gpm; and the aforementioned recycle
problems related to the inert gas scrubber system ~40 gpm during
audit, the matline operations ~10 gpm, and ~10 gpm due to miscellaneous
washing operations. All of the maintenance related problems are in
the process of being corrected. Furthermore, excess fresh water is
introduced into the system at the reclaim distribution tank to meet
55
-------
10 GPM
10 GPM
Fiberous
Sludge
Float
FIGURE 6.4
MATLIHE RECYCLE AS PRESENTLY OPERATED
High pressure
chain wash
from fresh watpr
. .
.
- MAT LINE
from reclaim
supply PROCESS
20 GPM
~ 10 GPM
~ 4
30 GPM
.
...
DISSOLVED
AIR PRESSURE
FLOTATION STAGE
4J.
10 GPM
. 30 GPM From Reclaim
Supply
't
To Wastewater Treatment
Net Effects:
20 gpm Recycle Demand
10 gpm Freshwater Intrusion
30 gpm Wastewater Discharge
Fibers Removed
56
-------
recycle demands, a point that will be later developed in Sections
VI.4 and VII.4.
VI.4
OPERATION OF THE ADVANCED WASTE TREATMENT SYSTID1
VI.4.1
SAND FILTRATION
A summary of the performance of the sand filtration process
over the 9-month test period appears in Table 6.6. Influent sus-
pended solids monthly average concentrations ranged between 10 mg/1
and 55 mg/1. These averages were in the upper range of those
anticipated during design. The average hydraulic flux ranged between
2.93 and 3.09 gpm/ft2, which is also higher than the design average
of 2.5 gpm/ft2, originally expected. The backwash volume expressed
ad a percent of processed volume ranged from a monthly average of 8.2
to 16.8 percent. Again, this is slightly larger than the 8 to 12
percent range expected in the design. However. this increase in
backwash requirement is consistent with higher hydraulic flux and
suspended solids loadings.
During the operation of the filters. effluent quality has been
acceptable and has averaged between 2 and 5 mg/l suspended solids.
However. during a two-day period in late July 1979. effluent sus-
pended solids rose to over 10 mg/l. which produced biological growth
with subsequent anaerobic activity within the downstream carbon
adsorbers. During mid-August. this activity resulted in the generation
of volatile sulfides which produced objectiOPable orders within all
plant recycle systems. Subsequently. all water recycle systems were
purged and the carbon adsorbers recharged and cleaned. The effects
of the major system purge which occurred in mid-August can be
observed in Figure 6.2. These purges required major demand increases
which maximized at 0.678 MGD during the week of August 18. 1979. It
is noted that system discharge did not dramatically increase during
this perio~ because excess water was stored within the system off-
specification basin.
In summary. the filters have operated very well. considering that
the applied hydraulic and suspended solids flux levels have been larger
than anticipated. Some of these problems have been due to the effects
of the tertiary system wastes (i.e.. spent sand filter and activated
carbon backwash waters) which are recycled back to the entire treatment
system via the mudwell and off-specification basin (see Figure 6.1).
During the preliminary design, it was estimated that the combination
of spent sand filter and activated carbon backwash wastes would account
for a continous average flow of 30 gpm and 33 gpm. respectively. In
the preliminary design, treatment of these flows were not considered.
During the final design these flows were recycled back through the
treatment system as indicated in Figure 6.1. The calculated average daily
flows for the sand filter backwash only, over the 9-month trial period are
tabulated in Table 6.6. These flows ranged from 6 gpm to 195 gpm with an
57
-------
TABLE 6.6 - SAND FILTER PERFORMANCE SUMMARY
MONTH SUSPENDED SUSPENDED SOLIDS HYDRAULIC(1) BACKWASH VOLUME(l) SAND FILTER BACKWASH
SOLIDS lOAD FLU.X gpm/ft 2 PERCENT OF FLOW RATE
mg/l . lb/Day PROCESSED PROCESSED gpm (Q!l)
-
i (2) a (3) H(4)l(5)x (J H L X t1 H l X a H. L X a H L
21. 3 3'f:4 98 "3 93.3 ill 372 -- 3.93 2.26 ---- 32 . 1 19. 3 79.9 6.7
Jau. 15 3.02 .40 8.22 4.87 23.4 1.49
Feb. 32 39 104 4 144 174 403 10.5 3.03 .33 3.48 2.21 8.93 4.65 21.5 2.69 35.1 18.9 86.1 12.5
Mar. 12.3 6.3 20 5 57.7 28.5 93.1 24.3 3.09 .42 3.46 1.90 9.82 4.14 21.5 5.21 38 . 1 11. 2 56.7 22.1
\It Apr. (6)12.8 8.4 27 5 60.4 45.3 133. 6 18. 3 2 . 93 .68 4.28 1.87 12.5 6.03 22.0 4.48 47.3 27.4 107.6 7.93
0)
May (6)18 10.1 35 5 82.7 53.4 163 21.6 3.11 .53 4.24 1. 79 16.8 10.2 56.4 3.3 64.4 35.6 1~3 11.9
June(7)54.3 100.9 302 6 227 406 1221 29 3.03 .49 3.77 1.55 15.5 10.8 58 6.3 58 37.7 195 14.2
July(7) 10.6 5.6 18 7 54.1 30.9 107 37.5 3.26 .60 4.53 1.41 15.2 9.5 45.5 3.0 62.7 40.4 181 10.2
Aug. 10.4 9.4 31 2 49.9 43.1 119' 9,5 3.08 .60 4.28 2.28 8.78 4.25 16.8 2. 68 34. 3 15. 7 69.7 11. 3
(1) Based upon One Carbon Adsorber Backwash per Day
(2) Monthly Mean
(3) Monthly Standard Deviation
(4) Monthly High Value
(5) Monthly Low Value
(6) Filters Temporarily Converted to
Dual Med ia .
(7) Filters with Anthrafilt media only.
-------
overall daily average during the 9-month test period of 46 gpm, which is
16 gpm or 53 percent larger than originally expected for the sand
filter alone.
By March 1979, it became apparent that the sand filters were being
operated over design conditions (see Section VI.l.l), and more capacity
would be needed to meet peak demand flow rates. At that time two'
alternatives were considered: (1) increasing the size of the filters
through addition of another filter cell; or (2) changing filter media
to a coarser, more porous, grade or a combination of the existing grade
and a new grade (i.e., converting the filters to a modified dual media
type) .
The second option was selected for evaluation since it would not
require a major capital expenditure. Therefore, in early April through
Hay of the test period, filter cells were loaded with a combination of 50
percent design media, .45 mm quartz sand with a 1.5 uniformity. and 50
percent of a 1:1 mixture of anthrafilt having 0.7 mm and 1.5 mm effec-
tive sizes. Reference to the performance summary, Table 6.6, during
these two months of April and May, no significant improvement in average
hydraulic flux was obtained through change of the media, although peak
hydraulic flux was improved as indicated by the larger daily deviation
and the increase in the monthly high hydraulic flux to 4.28 and 4.24a ,
gpm/ft2.
During the months of June and July, media was again replaced to a
single media, anthrafilt, in an attempt to further increase the filter
porosity and increase hydraulic flux capacity. Although the average
hydraulic flux and peak flux as indicated by a, and the monthly high
flux did increase (see Table 6.6), filter suspended solids breakthrough
and resultant loading on the carbon adsorbers produced unacceptable
anaerobic odor problems within the adsorbers and the entire recycle sys-
tem as previously noted. Thus, it was decided that if anthrafilt media
was to be used, it could only be used in conjunction with the finer
sand media in a dual media configuration. However, since the effects of
major suspended solids breakthrough were observed and resulted in major
purges of all plant recycle systems, the single media sand was selected
for continued use as a precautionary measure.
Throughout the entire period, media has been occasionaly replaced
when vigorous and repeated backwashing would not throughly cleanse the
sand. Close inspection of the media during these occasions has revealed
that the media was relatively clean and rapid headloss accumulation was
being caused by a media clumping phenomenon. When media was placed on a
flat surface clumps or aggregates of media ranging from 1/32 to 3/32
inch in size would remain as if the media particle were statically
charged. When lightly touched, these clumps would fall apart, indicating
that the binding forces present were relatively small. Qualitative
analysis through UV and IR spectroscopy of solvent extractions of
clumped media indicated the presence of chromophoric bodies related to
polyester & polyvinyl acetate (PVA) residues that had escaped coagulation
59
-------
and/or degredation within the primary and secondary treatment systems.
Numerous commercial cleaning and/or dispersing agents and also acid and
caustic solutions have been tried. with limited success. to prevent
media clumping. Currently routine application of a dilute caustic wash.
once per week apparently has extended media live from 4 to 6 months. It
is important to note that these problems may be intensified with complete
loop closure. It is also important to note that although significant
clumping has been observed in the sand filter. very little clumping has
been observed in the activated carbon absorber. However, carbon is
routinely changed every l~ to 2~ months as it is exhausted, and sand media
clumping has usually been observed after l~ to 2!~ months. Thus, these
same materials could be coating the carbon surface which may also
reduce the absorption rate and/or capacity as the result of a blinding
or coating action. Furthermore, the TOC values are a190 high due to
concentrating effects of system closure. These effects will be discussed
in a later section. The poor performance of the secondary system is
believed to be partially related to the hydraulic overloads produced by
wastes generated by the adv&1ced waste treatment system and the raw
manufacturing waste.
VI.4.2
CARBON ADSORPT ION
A summary of the weekly average performance of the carbon adsorbers
over the 9-month evaluation period appears in graphical form in Figures 6.5
and 6.6. These data represent weekly averages of daily total organic carbon
and BOD measurements. The percent removal of TOC and BOD by the lead and
lag adsorbers. based upon this data, appears in Figures 6.7 and 6.8. The
monthly data appears in Table 6.7. Over the entire test period TOC
removals averaged 40 percent, based upon secondary effluent
TOC feed. This value is below the 50 percent removal designed and expected.
However. the average daily influent concentration was 122 mg/l or 3.6 times
higher than the expected design level of 33 mg/l TOC. Furthermore, the
average hydraulic flux and contact time were 5.3 gpm/ft2 and 27.6 min.,
respectively. These values are 17.7 percent higher and 16.3 lower than the
anticipated design values of 4.5 gpm/ft2 and 33 min. Thus, the organic and
hydraulic loadings experienced are considerably higher than those expected
in the preliminary report and are certainly responsible for the observed
decreased removal efficiency. The BOD removals average 49.4% based upon
secondary effluent feed. The average daily influent concentration (second-
ary effluent) was 27 mg/1 vhich was 2.7 times larger than that expected
(i.e.. 10 mg/1 based upon past operation, see Table 5.1). These data
indicate that the primary and secondary system performance has decreased
since the recycle system became operational.
The operational history for a typical 'flip-flop' bycle for one carbon
vessel is presented in Figures 6.9, 6.10 and 6.11. This cycle was taken
from the daily operation of a fresh carbon load in the lag position between
March 5, 1979 through April 10, 1979 and in the lead position between
April 10, 1979 and May 4, 1979. The average hydraulic flux during these
time periods were 5.3 gpm/ft2 &1d 5.07 gpm/ft2, approximately the same as
60
-------
r- 140
........
0'1
E
..
u
~ 120
lLJ
(,!)
c:(
a:::
lLJ
~ 100
:>
-.J
~
lLJ
~ 80
.EliiUR£ 6.5
WEEKLY AVERAGE TQC FOR SECQNPARY EFFLUENT, LEAD CARBON
COllJMN (FFIIJENT, AND I Art CABBON COLUMN EFFLUENT
200
180
160
A
I , I
~ , ,
l /\/ ~
'1 ! 1\\ I r:.\
, , . I I ,
~ ,1 ".. :1 ~. :.\~
i , "., " , ,~ I ' I "
A 'I "',, 'If I
h :~ \ : t: :: p;' \ r \
~ rt/ r \ ,/ I . \ ~ \!. i \
~, ;J '\i \ ~:v' l.J /1 (\
'~-_.I . V
60
40
20
~
Secondary Effluent
Lead Carbon Column Effluent
Lag Carbon Column Effluent
\0 0
~ N M
................ -
~ ~ N
------
---
,....
~ C""'I
- -
N C""'I
,.... ~ ...r co N \0
~M~N~N Q\
-------
M M ...r ...r &t"\ an \0
WEEK ENDING, 1979
61
C""'I
N ,....
- -
\0 ,....
00
~ ~
- -
co 0\
an 0\
~ N
- -
Q\ Q\
~
N ...r
- -
,.... co
-------
r-I 25
-
00
a
Q
~ 20
FIGURE 6.6
WEEKLY AVERAGE SECONDARY AND CARBON COLUMN EFFLUENT BOD
50
Secondary Effluent
45
40
35
30
15
10
Carbon Column Effluent
5
o
\0 N C"'I
- --
- - N
~ co N
- N -
- - -
~ ~ u,
"
- ("',
- -
N M
" -
- C"'I
- -
M C"'I
\0 C"'I
N Q\ N
- --
U, \D \D
- co '" Q\
"N~-- - N
-------
" " co co Q\ Q\ Q\
Week Ending
62
-------
40
~ 30
~
~
u
0
H 20
a-e
60
50
10
o
o
\0 N M
- --
.-4 .-4 N
FIGURE 6.7
PERCENT TOC REMOVAL THROUGH
CARBON COLUMN SYSTEM
N 0\ N
- --
11"\ \0 \0
Week Ending
63
0\
.-4 .-4 N
- --
0\ 0\ 0\
-------
40
~
~ 30
~
g
IQ
IN! 20
80
70
60
50
10
\0 N
- -
.... ....
FIGURE 6.8
PERCENT BOO REMOVAL THROUGH
CARBON COLUMN SYSTEM
"'"
M .... M
- --
N N M
,... .... .q
.... M ....
- - -
M M .q
co N ~
N .... N 0\
- - --
.q -n -n ~
Week Ending
b4
M ....
N "'" N .q
- - --
~ ,... ,... co
-n 0\
.... N
- -
0\ 0\
co
.... ....
- -
co 0\
-------
TABLE 6.7
---
MONTHLY AVERAGE PERFORMANCE SUMMARY FOR CARBON ADSORBERS
tJONTH TOC INFLUENT PERCENT TOC BOO INFLUENT PERCENT BOD HYDRAULIC
REMOVED REMOVED FLUX HYDRAULIC
~ CONTACl .
mg/l 1 b / day % !!!9L! 1b/day % gpm/ft ]mE (mr:.)
Jan. 83 412.6 41.5 11 57. 1 43.3 5.22 28.0
Feb. 101 503.7 23.6 24 121.2 41.5 5.25 27.9
Mar. 93 469.1 45.9 23 118.0 51.6 5.35 27.4
Apr. 125 600 . 3 37.4 23 113.3 55.2 5.07 28.9
May 115 587.8 45.3 21 108.7 43.8 5.39 27.2
(J\ June 174 .860.9 47.5 19 97.0 57.7 5.25 27.9
V1
JuTy 140 745.4 37.5 41 218.7 44.7 5.64 26.0
Aug. 150 754.6 41.3 49 249.6 57.4 5.34 27.6
-------
140
105
0'.
C
r-f
"-
OIl
S
g 70
f-4
35
FIGURE 6.9
INFLUENT AND EFFLUENT TOC SUMMARY FOR A
COMPLETE 'FLIP-FLOP' CYCLE FOR 13
CARBON COLUMN
It.
I ,
, \
I \
I \
I \
,
,
,
I
I
,
,
I
Influent ;
-- ;
.-....---- ---, "
..... "
v
c - 72 mg/l TOC
_0 2
HF . 5.35 gpmlft
Effluent
65
130
195
I,
I ,
I \
I ,
I .. '
, ,
I ','
I
,
I
I
I
I
I
\ -'"
",
Column in Column in
Lag Position Lead Position
4\
, \
I \
"'- I \
... ,
~J
c . 126 mg/l TOC
o
HF . 5.07 gpm/ft2
260
325
390
455
VOWME PROCESSED, THOUSAND GAL/FT2
-------
FIGURE 6.10
PERCENT TOC REMOVED PER VOLUME PROCESSED
FOR A COMPLETE 'FLIP-FLOP' CYCLE FOR 13
CARBON COLUMN
Column in
Lag Position
40
~
Co = 72 mg/l TOC
30 HF .. 5. 35 2
gpm/ft
~
~ i
......
u
0 20
~
~
:z:
f&J
u
~
f&J
p.,
10
65
130
195
260
I Column in
I Lead POSi~on
I
I
I
I
I
(
I
Co - 126 mg/l TOC
2
HF = 5.07 gpm/ft
325
390
455
VOLUME PROCESSED, THOUSAND GAL/FT2
-------
FIGURE 6.11
CUMULATIVE TOC CAPACITY FOR A COMPLETE 'FLIP-FLOP'
CYCLE FOR 13 CARBON COLUMN
0.300 Icoluam in
~ ColWln in
~ Lag Position I Lead Position
.. C ~
.......
g
~ 0.225
..
(J'I ><
CD ~
H C - 72 mg/l TOC
~ 0
~ - 2
u HF - 5.35 gpm/ft
u 0.150
0
~
r;
H C
S - 126 mg/l TOC
o
! HF .. 5.07 2
0.075 gpm/ft
u
65
130
195 260 325
VOLUME PROCESSED, THWSAND GAL/FT2
390
455
-------
those previously noted over the evaluation period. Influent TOC concen-
trations to the fresh absorber in the lag position averaged 72.4 mg/1 TOC;
while influent TOC concentration averaged 126 mg/l in the lead position
(see Figure 6.9). While in the lag position the maximum capacity reached
was 0.0975 lb TOC/lb Carbon (Figure 6.11), which should have been relatively
saturated with an equilibrium concentration of 100 to 105 mg/1 TOC
(Figure 6.9). It is believed that relative saturation was reached since
the percent removal at the end of this lag cycle (see Figure 6.10) approached
zero.
An ultimate capacity of 0.285 lb TOC/lb Carbon (Figure 6.11) was
reached by the end of the lead cycle while an average TOC concentration of
126 mg/l. was applied to the absorber. By the time this capacity was reached,
the percent TOC removal dropped to less than 10 percent. This indicates that
most of the ultimate capacity was obtained for the equilbrium condition of
123 to 140 mg/1 TOC concentration applied in the absorber feed. The ultimate
capacity records of .285 lb TOC/lb Carbon approached that obtained during
the overall test period, 0.333 lb TOC/lb Carbon as calculated by monthly TOC
removal and carbon inventory.
Since the absorber influent TOC concentrations have been 3.6 fold
larger than those expected in the design absorber, capacity expressed as
the mass of TOC removed per mass of activated carbon should also be larger
(i.e., the activated carbon should be more efficiently utilized since it
should be in equilibrium with higher concentrations of influent TOC at
exhaustion). The average accumulative mass of TOC removed over the 8-
month period between January and August 1979 was 59,891 lbs. TOC. Over
this time period, nine (9) recharges of carbon at approximately 20,000 lbs.
each were used. Therefore, the TOC capacity of spent carbon onver the 8-
month period was 0.333 lbs. TOC/lb carbon. As noted, this value is
larger than that expected, 0.087 lbs. TOC/lb. carbon, when the influent
equilibrium TOC concentration was anticipated to be only 33 mg/l TOC.
If, however, the absorber has been coated with polymeric materials as
previously postulated in Section VI.4.1, the increased equilibrium
organic level appears to be more significant in raising the ultimate
capacity than any inhibitory effects due to polyester or PVA, coating
and/ or blinding.
As previously noted, during mid-August significant anaerobic odors
were observed emanating from the carbon adsorbtion vessels. Prior to
this, carbon columns were backwashed at a frequency of approximately
once every 48 hours. These odors were caused by organic growth within
the adsorber vessel and believed to be stimulated by previous second-
ary system solids breakthrough across the filters during single media
anthrafilt trial as previously discussed. During this period adsorb-
ers were backwashed more frequently at approximately once per 24 hours.
After sand filter media was replaced, no major odor problems have
been encountered with carbon backwash frequency reduced to once every
48 hours.
69
-------
In summary, the adsorbers have performed as good as can be expected.
High effluent TOC and BOD levels have o~~u uue to high organic and
hydraulic flux loadings which in turn are related to secondary system
performance.
VI.5
HYDRAULIC EFFECTS OF ADVANCED WASTE TREATMENT UPON PRIMARY AND
SECONDARY TREATMENT SYSTEHS OPERAT ION
As previously indicated in Section IV.4.2, VI 4.1, VI.2, VI.2.,
VI.4.1, and VI.4.2, the advanced waste water treatment system generates
wastes in the form of sand filter and carbon adsorber backwash waters
which, in the final design, are returned to the primary and secondary
and AWT systems via the mudwell and/or off-specification basins (see
Figure 6.1). In addition, much of the forward process flow rate to the
sand filter is by-passed to the mudwell during the sand filter backwash
period. This by-passed flow also appears in the return flow to the
primary system. These return AWT waste flows appear in Table 6.8,
and are continuous monthly averages for eaca return source. The main
monthly raw waste water flow rates an~ the corresponding monthly stan-
dard deviations also appear in Table 6.0. The average total processed
monthly flow rates, when the carbon adsorber is bac~vashed at frequencies
of every other day and once perday, are also included in the table. As
illustrated in the tabulation, the cor.1billed AWT recycle waste flows
produced between 18.9 percent and 42.9 percent higher flow rates to all
treatment systems, depending upon carbon adsorber backwash frequency,
than the existing raw process waste flow rate. It is noted that the
final design (see Figure 6.1) allowed for an average AWT waste return
flow of 68 gpm with a raw waste water flow of 259 gpm, or a 26.3%
increase in processed flow rate over the raw waste water base flow.
Thus, the total processed flow, the swmnation of raw manufacturins, and
AWT return wastes, was anticipated to be 327 gpm, a flow rate which
should have been treatable by the existing primary and secondary
system as previously discussed in Section VI.2.
The average raw manufacturing waste flow, anticipated to be 259 gpm
with design modifications, ranged between 296 and 370 gpm, indicating 100
gpm more fresh water usage as previously discussed in Section VI.3.6. As
the raw waste flow rate increases, the AWT backwash flow rate should also
increase through increased backwash needs of the AWT system. The effect
of the combination of these larger flows not only places a hydraulic burden
upon the AWT system, but also a severe hydraulic burden upon the existing
primary and secondary system. The average processed flows through the primary
and secondary system presented in Table 6.7 ranged between 393 and 457 gpm
with an average of 413 gpm. These flows are between 66 and 130 gpm larger
than the expected design flow of 327 gpm. The hydraulic overload effects
upon decreased hydraulic residence periods for any primary, secondary, or
AWT systems can be formulated as:
70
-------
TABLE 6.8
HYDRAULIC EFFECTS OF AWT SYSTEM WASTES ON PRIMARY
AND SECONDARY PROCESS FLOW RATE
- -
AVERAGE SAND AVERAGE CARBON
FILTER BW & ADSORBER AVERAGE AVERAGE PERCENT
RAW WASTE SECONDARY BW FLOW PROCESS FLOW INCREASE IN PROCESSEO
MONTH WATER FLOW BY PASS gpm gpm gpm FLOW BY AWT WASTFS
...
One AC One AC One AC One AC
Q gpm ::Q gpm BW By Pass Total OW/day EW/2 day BW/Day BW/2 Day % Max. % Min.
Jan. 329 47 32 16 48 33 16 410 393 24.6 19.4
Feb. 318 75 35 26 61 33 10 412 395 29.6 24.2
..... Mar. 331 58 38 18 56 33 16 420 403 26.8 21.8
~
Apr. 296 67 47 22 69 33 10 398 381 34.4 28.7
~Iay 296 62 64 30 94 33 10 423 406 42.9 37.1
June 295 69 58 26 84 33 16 412 395 39.6 33.9
July 317 62 03 30 93 33 16 443 426 39.7 34.3
Aug. 370 89 34 20 54 33 10 457 440 23.5 18.9
-------
o
Design
¥ /Q Existing
¥/Q Design
Q , Design
o Existing
-
.
,
Q Existing
where: 9 - hydraulic residence period of a unit (¥/Q) ;
% reduction in hydraulic residence period - 0 Design - 0 Existing x 100%
o Design
-
1 - Q Design
Q Existing
x 100%.
The minimum, maximum and mean reductions in hydraulic residence
periods in all primary, secondary, and AWT systems, based upon the excess
flow rates and the above relationships were 16.8, 28.4, and 20.80 percent,
respectively. It is obvious that such reductions in residence periods
will produce the following conditions in primary and secondary systems:
1.
Primary System
a)
Lower suspended solids and organic removal efficiency
which will produce organic overloads on secondary
system, through hydraulic carryover.
2.
Secondary System
a)
Biological Process
i) organic overload carryover from primary system,
ii) decreased organic removal efficiency due to
reduced residence periods and organic overloads.
b)
Secondary Settling
i) decreased suspended
producing increased
to AWT systems.
solids removal efficiency
suspended solids carryover
The net effects to the AWT system are:
flux upon the sand filter, as evidenced in
organic loadings upon the activated carbon
Section VI.4.2; and 3) increased hydraulic
and adsorption systems.
1) increased suspended solids
Section VI.4.1; 2) increased
adsorbers, as evidenced in
loadings on both filtration
The increased hydraulic and mass loadings upon AWT filtration
increases backwash water requirements which: 1) create higher flow rates
through the primary secondary and AWT systems; and 2) create larler demands
of reclaimed water which decreases the available supply of reclaim water
for manufacturing recycle. Thus: 1) the sand filter is the overall
72
-------
system "bottleneck". having difficulty processing the total flow. let
alone the extra flow of stored processed reclaim water to be used during
summer months (see Figure 6.1); 2) excess fresh water is added at the
distribution tank. to supply recycle demands. and adds unwanted volume
to the entire system; and 3) additional reclaimed water which accumu-
lates in the system and is stored in the storage basin. (see Figure 6.1).
cannot be reused since it must first be processed through the AWT system
before it can be supplied to the distribution tank. The end result is
that water accumulates in the system and must be eventually discharged.
Over the winter months during the test period, the 1.5 million
gallon storage basin has been routinely filled over a 3-week period
which indicates a winter system hydraulic imbalance of approximately
50 gpm excess flow. The design winter imbalance was estimated at
approximately 10 gpm excess flow. indicating that the basin storage
would last for approximately 14~ weeks.
VI.5.1
ORGANIC QUALITY OR RECYCLE SYSTEM - I~FRACTORY" BUILDUP
As previously noted. the hydraulic overloads to the secondary system
produced lower organic removal efficiencies and higher than expected BOD
and TOC levels in the secondary effluent, larger than expected BOD and
TOC loadings upon the carbon adsorbers, and larger than expected BOD and
TOC levels in the adsorber effluent, reclaim water, (see Section VI.4.2) .
The adsorber effluent BOD level was expected to be zero and the TOC
level was set at a maximum of 17 mg/l. The 17 mg/1 TOC described in the
Preliminary Engineering Report (1) was thought to be "refractory ".
It is important to note that refractory TOC could have several definitions
such as: 1) organic material that could never be adsorbable on
activitation carbon under any conditions (an absolute or true refractory).
or 2) organic material not removable in the flowing column process
condition considering constant applied hydraulic flux. contact periods
and/or column length and a given applied TOC concentration level (an
apparent refractory). If definition (1) is used. then the refractory
TOC level should rise in a completely closed system through evaporative
concentration, similar to predictions made for inorganic constituents in
the Preliminary Report (1) and presented in Table 4.6*. As such. the
predicted net TOC input through city water and reclaimed wastewater
would have been:
City water input + reclaimed wastewater input - Net Total Input
1440 ~
D
1 MG
106 gal
73 gpm (0 mg/l) + (186 gpm) (17mg/l)
8 . 34 lb
MG
mg/l
- 38.0 lb TOC
D
* assumes no further biochemical degredation within process coalings and
waste treatment systems.
73
-------
Since, the Toe input = drift losses for summer or winter conditions,
the predicted design equilibrium TOe level of reclaim water calculated in
the Preliminary Report (1) ~hould have been:
WSummer 2}.1 " 45.2 mg/1 Toe refractory equilibrium
0.840
Wwinter 23.1 " 45.7 mg/1 Toe refractory equilibrium
0.831
rather than 17 mg/l.
If the expected refractory carbon level of 17 mg/1 is defined as in
Definition 2 above, the system equilibrium Toe could even be larger than
the 45 mg/1 Toe previously calculated. As adsorber effluent TOC, which
is adsorbable, increases in the system through evaporation concentration,
and the system hydraulic flux and contact periods remain constant,
effluent Toe values should rise over the once through value of 17 mg/l
Toe, which was the input value used in the above calculation. Thus,
the Toe values measured in the adsorber effluent should have been at
least 45 mg/l. even if all treatment systems were hydraulically loaded at
design condition. It is no wonder that adsorber effluent TOe values
approached a mean value of 73 mg/l Toe when all systems were hydraulically
overloaded.
Not only is it presently not kn~ what fraction of the 73 mg/1
Toe contained in the adsorber effluent is truly refractory, but also it
is not known what fraction of the 17 mg/1 Toe design level was truly
refractory. This may be important since the concentration of these
materials would surely build up in the system. Furthermore, it is not
known at this time whether any of either the true or apparent refractory
TOC material can cause or did cause any decrease in biological activity
within the secondary activated sludge system during this study. Certainly
the hydraulic and hydraulically promoted organic overloads to the
secondary system are responsible for high BOD and TOe values in the
secondary effluent. However, it is not known at this time whether or
not the high BOD and TOe values recorded during the study were solely
promoted by the hydraulic overloads or promoted by a combination of hydr-
aulic overloads and reduced bio-activity resulting from refractory build
up. The effect, if any, of refractory build up can only be evaluated
if secondary system hydraulic overloading can be reduced and system
closure is maintained for extended periods of time.
74
-------
CHAPTER VII
PLANS FOR OVERALL SYSTEM IMPROVEMENT
VII.!.
OVERALL SYSTEM NEEDS
In general, all flaws of (1) raw waste water, (2) AWT recycle
wastes, and (3) fresh water make-up must be reduced in order to achieve
extended total system closure. If system closure is achieved, inorganic
and organic dissolved solids may present problems. the scope of
these potential problems cannot be accurately predicted until extended
hydraulic closure of the system is achieved. The justification
for these potential problems has been fully described in Sections
IV and VI.5.l. It is certain that if the inorganic chemical coalgulants,
FeC13 and bentonite clay as described in Appendix 2, were reduced and/or
eliminated and emulsion breaking performance within the primary system
maintained or bettered; the total dissolved solids level within recycle
systems would be more favorable during extended system closure. Thus,
the following plan for system improvement with optimistic system hydraulic
closure has been either formulated and/or adopted.
VII. 2.
HYDRAULIC FLOW REDUCTION
VII.2.1
RAW WASTEWATER REDUCTION
Maintenance, repair, and waste collection system modification,
as described in Section VI.3.6 are presently underway to reduce
extraneous raw wastewater and fresh water intrusion within the manu-
facturing plant.
VII.2.2
AWT RECYCLE WASTE REDUCT lOtI
---
As presented earlier, AWT recycle waste reduction can be achieved
through reduction of hydraulic anu/or suspended solid loadings upon
the AWT systems. This could be accomplished through treatment system
modification as illustrated in Figure 7.1. The modification essentially
involves isolating one of the five (5) parallel existing primary
clarifiers from the raw waste flow and processing segregated AWT
backwash recycle wastes through the isolated clarifier. To achieve
necessary low suspended solids concentrations from the isolated
clarifier, a separate, new and effective coagulant system may be
developed. If effective suspended solids removal can be achieved
and the raw waste flow can be effectively treated by the remaining
clarifiers, then the isolated clarifier effluent could be directly
reused for backwash water supply to the entire AWT system. Excess returned
backwash water generated through secondary by-pass during sand filter backwash
operations (see Section VI. 1. 1) could be diverted to the adsorber
feed sump via a gravity overflow located at the backwash supply
sump. This flow could range from 16 to 30 gpm (Table 6.7). It
most probably would be even lower, since the magnitude is directly
related to sand filter backwash frequency and the frequency would
75
-------
"
(J\
City Wa
0(22-87)
NOTE:
FIGURE 7 1 PROPOSED WASTE TR ATM NT SYSTEM DESIGN
o - 60 Reclaimed Wastewater Return
259 (1) PRIMARY (2) 259 256 '~1 'l61 -
AND SECONDARY SAND CARBON
Raw Waste r r r r-..
SYSTEM FILTERS ADSORBERS
3 20 ~ 5 ~ ~
By-Pass 20 16
During ~
Filter Backwash MUD I I BACKWASH I
""'-1 .
~ PRIMARY 41 t Adsorber
CLARIFIER . Backwuh
23 Effluent 16
41 ~ OFF-
SPECIFICATION
BASIN Drain Water 2
Off-Specification Return J,
.
ter
~
CHLORINE 261 ,.
.... ADDITION -- DISTRIBUTION
TANK
~ 0
0
,
,
To Duck Pond 2 RECLAIMED
Chlorine WASTEWATER .
Solution STORAGE
, 261 259 (281-346) -
-r
(1) All nUlllbers are averaBe flow. gpm
(2) Primary system includes four parellel clarifiers
To Process and Cooling Uses
-------
be lower since the hydraulic and suspended solids loading upon the
secondary and AWT systems would be reduced. Hydraulically, the
overall modification would account for a flow reduction of 64 to
127 gpm to the secondary and AWT system process flow rate (Table
6.7). The modification would also increase the hydraulic residence
periods in all processes within the secondary system by as much
as 21% and AWT system by as much as 20% for the sand filter and
18% for the carbon adsorber (based upon design flows in Figures
6. 1 and 7. 1) .
A pilot trial program is currently underway which involves:
(1) clarifier isolation; (2) identification of coagulant systems
to effectively remove suspended solids in both segregated AWT system
backwash wastes and the segregated raw waste water flow; (3) the
effects of the processed segregated flow upon AWT filter and adsorber
backwashing operations, and (4) the effect of the hydraulic excess
flow, created by secondary bypass during sand filter backwash, which
is only treated by carbon adsorption, upon reclaim water quality.
The program involves the isolation of the clarifier and AWT
recycle waste feed and evaluation of full-scale clarifier perform-
ance. Simultaneously. the performance of the remaining clarifiers
with segregated raw waste will be evaluated. Upon successful evalua-
tion of the isolated clarifier effluent, through bench scale tests,
further full-scale trials will be undertaken. These trials will
involve the full-scale segregation of the isolated clarifier effluent
to the backwash supply sump, and full-scale backwash testing while
the entire secondary and AWT system response is monitored with respect
to hydraulic, organic, and inorganic parameters.
If this entire program proves successful, permanent piping for
clarifier-backwash treatment segregation will be installed. Further-
more, the modification will certainly improve the recycle system.
However, only total system operation over an extended time will pro-
vide data which will demonstrate that the modification produced an
improvement approaching total recycle with zero discharge.
VII. 3.
RAW WASTE PRIMARY TREA~ffiNT CHEMICAL COAGULANTS
Since the recycle system began operation, several time periods
have existed when the chloride concentration of reclaim water rose
to unusually high levels (500 to 600 mg/1 Cr-). Surely these con-
centrations would be even larger if the entire system was completely
closed with no partial discharge. Large chloride concentrations are
unwanted, since they can accelerate corrosion within cooling and
process water system hardware. In addition to manufacturing chemical
losses, chlorides are added to the wastewater through the use of the
existing FeC13 - Bentonite Clay-anionic polymer primary treatment
coagulant system (see Appendix B).
77
-------
If a new primary coagulant system consisting of only polyelectrolytes
was developed which, (1) destabilized emulsified chemicals contained
in raw wastewater, equally or better than the FeCl) system, and (2)
was cost effective; a distinct advantage with respect to minimizing
chloride accumulation within the recycle system would be achieved.
Additionally, such a coagulant system could possibly result in more
compact primary suspended solids floc having larger terminal settling
velocity which would allow a larger hydraulic flux upon the existing
four primary system clarifiers (see Section VII.2.2). Presently,
an apparently effective polyelectrolyte primary coagulant system has
been developed, and pilot testing of this system is in progress.
Results indicate that the performance of the polymer coagulants, are
equal to or better than that of the FeCI) - bentonite-polymer system.
VII.4.
FRESH WATER MAKEUP AT RECLAIM DISTRIBUTION TANK
When the recycle demand flow exceeds the AWT process flow, a
demand surge, the supply volume in the reclaim distribution tank de-
creases. The entire cooling and process water needs of the plant are
supplied by this system. If any of the waste treatment systems fail
to an extent that reclaim water quality or hydraulic supply is un-
satisfactory for the recycle demand, emergency fresh water must be
supplied to the manufacturing plant. Currently, the emergency method
available for use is an automatic/manual fresh water makeup system
located at the reclaim distribution tank. Thus, if the distribution
tank does not contain water, the entire manufacturing facility will
not have a water supply. The automatic fresh water makeup system
can supply 200 to )50 gpm of fresh water to the distribution tank
through an automatically contrailed line. Much of the time over the
test period either the reclaim demand has been so large or the reclaim
supply so small (Le., due to AWI "bottleneck" problems outlined in
Section VI.5.), that this fresh water supply was not large enough.
Thus, a manually operated 4-inch fresh water supply line was installed
to meet periodic reclaim supply deficiencies. Furthermore, recycle
system operators are under extreme psychological pressure to main-
tain an adequate supply level within the reclaim distribution tank
since it is the only cooling and process water source for the entire
manufacturing facility. Thus, the net result is that excess fresh
water is often manually added to the distribution tank to avoid
potential supply deficiencies and alleviate operator worry.
In the near future, this entire makeup system will be studied
to define improvements which alleviate these conditions. Such improve-
ments could include: conversion of the four-inch supply line to auto-
matic control; lowering the low level automatic set point to increase
surge capacity of the distribution tank, and installation of new
flow meters on the entire makeup system.
78
-------
VIII. REFERENCES
1.
Loven, A.W., and Pintenich, L.L., "Industrial Wastewater Recir-
culation: Preliminary Engineering", EPA-600/2-77-043, (1977).
2.
Angelbeck, D.I., Reed, W., and Thomas, S.H., '~losed Loop
Recycle System for Fiberglass Insulation Hanufactive", Proceed-
in~s of the 26th Purdue Industrial Waste Conference, Purdue Uni-
versity. (1971).
3.
Thomas, S.H., and Walch, D.R., "An Industrial Wastewater Recircu-
lation System for the Fibrous Glass Textile Industry", Textile
Industry Technology Conference, Williamsburg, Virginia, (1978).
4.
West, A.W., "Plant Performance at the Owens-Corning Fiberglas
Corporation Wastewater Treatment Facility, Anderson, South
Carolina", prepared by EPA National Field Investigation Center,
(1973) .
5.
Pharis, P., and Monaghan, G., "Biological Treatment of Textile
and Sanitary Wastes from a Fiberglas Plant", Journal Water Pollu-
tion Control Federation, (1965).
6.
"Standard Methods for the Examination of Water and Wastewater",
14th Edition, American Public Health Association, Washington, D.C.,
(1976) .
79
-------
APPENDIX A
EST~MATION OF TOTAL DISSOLVED SOLIDS
LOST IN PARTIAL DISCHARGE
Given Information:
Quantity
Description
86 gpm
Average corrected discharge
73 gpm
Planned fresh water input to existing
system
97 gpm
Unplanned fresh water input to existing
system
300 gpm
Average flow of once through system
1/76 through 6/76
52 mg/l TDS
Average TDS concentration of fresh water
280 mg/l TDS
Average TDS concentration of AWT effluent
in once through system 1/76 through 6/76
510 mg/l TDS
Average TDS concentration of reclaimed
wastewater 1/79 through 9/79
0.835 (W)
Average mass loss rate lb/day through
drift in total recycle system, if cooling
system operation was as planned. (W) is
concentration of 'parameter in reclaim
supply (mg/l)
Reference
or
Source
Section IV.
Figure 4.2
Figure 4.2
Table 5.1
(2)
(2)
Table 4.6
(2)
Rate of expected TDS input - (input through fresh) + (input through manu-
facture and net treatment effects)* D (73 + 97) (52) (1440) (8.34) +
106
(300 - 73) (280 - ~2) (1440) (8.34)
10
106 + 622 = 728 lb/day
--
Rate of loss through partial discharge* D 86 (510)(8.~4)(1440)
10
.Treatment system using FeC13,clay. and cationic polymer
80
- 527 Ib/day
-------
Necessary drift loss rate for equi1ibrium* = 728 1b/day - 527 1b/day
= 201 1b/day
Expected drift 1055 rate for equi1ibrium* = 0.835W = 0.835 (510)
= 425 1b/day
Necessary rate (201 lb/day) < Expected rate (425 lb/day)
Thus, system with partial discharge is loosing more than enough solids.
*Treatment system using FeC13' clay. and cationic polymer
81
-------
APPENDIX B
EXCERPTS FROM PRIMARY AND SECONDARY
OPERATIONS MANUAL
A.
SURGE TANKS
1.
DESCRIPTION
The purpose of the surge tanks is to store chemical dumps
from the chemical plant and meter them on a controlled basis into
the waste treatment plant. There are three surge tanks (see
Figure lA). Two of these tanks hold 25,000 gallons while the
other holds 10 ,000 gallons. Each tank has an inlet and outlet
for the wastewater; an inlet for acid and an inlet for caustic
in case it is necessary to adj ust the pH in the surge tanks, an
inlet for mixing air. and a scale on the side of the tank showing
the liquid height inside the tank. There is a by-pass line through
which the wastewater can be put directly into the plant without
going through the tanks. Also, there is a pump placed on the inlet
side of the tanks that would allow a tank truck to hook up and pump
directly into or out of the tanks.
2.
NORMAL OPERATION
The wastewater (Figure 2A), as it comes from the chemical
plant is collected in one of the tanks. When the tank is full, the
wastewater is manually diverted into another tank. A high level
overflow between tanks prevents them from overflowing. A sample
is collected from the full tank and analyzed for total organic
carbon.
The small surge tank (capacity 10,000 gal.) is reserved
for an unusually concentrated waste which sometimes comes from
the chemical plant. The high TOC on this water dictates a very
slow release into the system. The chemical plant operator has
been instructed to inform the treatment plant operator of this
kind of dump. The treatment plant operator should then divert
the waste to the small tank. When he is certain all the waste
has been collected, the operator switches back to the previously
used tank and determines the release rate for the small tank.
Under normal conditions, the tanks fill slowly and there is
plenty of time to dump one tank before the other. There are
certain times when one tank is down or when heavy rains or wash-
down cause excessive flows. If there is not a place to put
additional water, the operator will have to drain the tanks. How-
ever, the TOC of the tanks should be logged and the cause of the
excessive flows should be identified. The tank with the lowest
TOC should be drained first.
82
-------
Surge
Tan ks
Chem Factory
ex>
"'"
Mat Line
AFQ
Causti.:
Bu1~ Stor~ge
"'-
pH ~~0n i tor
...----
:PSF
~---....
Opera t ions
Building
--.0.
\~aste From
A.B.D. & E Factories
Sludge Lagoon
Supernatant
r--------------- ---
,
,
,
I
.
,
Figure lA
WASTEWATER TREATMENT PLANT
SCHE}1ATIC
:: 3 AI:? rat ion
T;\nk
{)--------"'"
::3 Sec. C1arif.
::~ S~c.
::2 :\e rat ion
Tank
"
'i:
I "8cSF
::1 Aeration ~
'- - - - _~n_k - - - - - - - -Ilill
XSF Clarifier
S1udf;e Pl!i:1pS
To
Sand
:i It~r
---------
----- . .. "---
(..
)
! Duck and Fish
P on d
j
.
l__-
- - -.....-
---.,
r--- - .
,
. ~ ._-
,
Sludge
lagoo:1
-------
B n;:=: R I
APPt.Ic.AT~
BZ!.'-O:!?
1I~!;~:::
!
_.
ilESIN
M.\.""'!'~
FIGURE 2A
EXISTING WASTEWATER SOURCES
SOURCES OF WASTEWATER FROM THE ANDERSON PLANT
AIR
SCRCBBERS
SYS':!:H
O\IF.lIFI'..c'II
COOLING
WATER
REINFO~O
MAT
FABRICATION
SANITARY
r
INDUSTRIAL SANITARY
WAS'lE TREATMENT PLANT WASTE
TREATMENT PLAh"T
-------
3.
POTENTIAL PROBLEMS
During power failures, feed to the surge tanks will be shut
down. However, the tanks will continue to drain. If a power
failure of longer than 30 minutes occurs, the drain valve should
be closed.
Feed from the surge tanks should be completely shut off if
the main plant is in an upset condition or experiencing high TOC's
from some other area. Adjustments to the feed rate should be noted
in the plant operating log.
B.
EQUALIZATION BASINS
1.
DESCRIPTION
There are two equalization basins. Basin No.1 is 34
ft. in diameter and 15 ft. deep while Basin No.2 is 50 ft.
in diameter and 17 ft. deep. There are two direct coupled,
centrifugal pumps having capacity of 950 gallons per minute
each that pump water from Basin No.2 to Basin No.1. Both
basins have diffused air which provides the mixing action.
The two basins are connected by a distribution box.
2.
NORMAL OPERATION
The two equalization basins serve to throughly mix the
waste so that any shock to the plant will be somewhat diluted.
Basin No.1 is fed by a line that comes from the matline, the
off-spec pond, the sludge lagoon. Basin No.2 is fed by
wastewater from Factories A, D and E. Drain lines from the
chemical plant surge tanks also enter this basin.
The wastewater from Basin No.2 is constantly being pumped
into Basin No.1 by one of the two pumps described earlier.
The overflow from Basin No.1 goes through a valve to the
distribution box. From the distribution box, part of the flow
goes into the flash mixer while the rest goes back into Basin No.2.
The amount of flow going either way is controlled by a sluice
gate in the distribution box. Also in this distribution box
there are two pipes for chemical addition. These chemicals act
as coagulants so that most of the suspended and colloidal
solids can be removed in the primary clarifiers.
3.
POTENTIAL PROBL~IS
The equalization basins require very little operator
attention. Some of the problems experienced in the past
include:
85
-------
b.
COAGULANTS. The Anderson plant uses ferric chloride
FeCl~ ' Bentonite Clay and an anionic polymer. 1 The
requ~red concentration of each chemical is best deter-
mined by a jar test.
The required concentration can then be converted to the proper
pump flow rate.
h POTENT IAL PROBLEHS
In addition to mechanical or electrical failure of the
flash mix or flocculator drive motor, the major problem involves
the coagulation process itself. If the primary clarifiers become
cloudy, the coagulant aids have become ineffective and a jar
test should be run to determine new concentrations.
The flash mixer shaft can occasionally become entangled
with glass or rags. When this happens, the mixer should be
stopped and the shaft cleaned. If the shaft is badly entangled,
the flash mix tank can be drained by shutting off forward flow
and closing the sluice gate between the flash mix and equali-
zation basins.
D.
PRIMARY CLARIFIERS
1.
DESCRIPTION
From the flocculation tanks waste flows through five tele-
scopic valves to five primary clarifiers, when the suspended
and coagulated solids are removed from the wastewater by
gravity. The clarifiers are of the rimflow design where waste-
water enters the tank tangentially, flows under a skirt and
the supernatant overflows a weir in the center. Each clari-
fier is 14 feet in diameter. Sludge is withdrawn through a
telescopic valve into a sludge well.
2.
NORMAL OPERATION
The clarifier mechanism will rotate continuously through the
bottom of each clarifier, pushing solids to the centerwell and
preventing excessive compaction.
3.
POTENTIAL PROBLID1S
There are three potential problems with the clarifier
operation. First, an obstruction in the clarifier can block the
rotation of the mechanism. This will cause a shear pin to
break, which in turn disconnects the gears of the dirve. The
motor will continue to operate, but the mechanism will not
rotate. Before the shear pin is replaced, the clarifier should
1
May be replaced with total polyelectrolyte system.
86
-------
be pumped down to find the cause of failure.
The second problem involves sludge withdrawal itself.
It is possible that a "hole" can be pulled in the sludge blanket
and clear water will be drained into the sludge well. When
this happens, it is necessary to break up the sludge blanket
with an air line.
The third potential problem is the most common. If the
coagulation system fails, carryover from the clarifier will be
very turbid. This problem is usually associated with the
chemical feed system upstream. However. if there are discrete
particles in the clarifier overflow, it indicate the sludge
blanket is too high and solids are being swept from the blanket
surface.
E.
ACTIVITATED SLUDGE SYST~t
1.
DESCRIPTION
The heart of the Anderson treatment plant is the activated
sludge system. Physical components of the process include
three 140,000 gallon aeration basins, three 10 x 50 rectan-
gular clarifiers, two return sludge pumps. one waste sludge
pump, one forward flow pump, ferric chloride and polyelectrolyte
chemical feed systems, and positive displacement blowers.
The combined purpose of these units is to biologically remove
soluble and colloidal organics from the wastewater. The
operator should be familiar with basic terminology and concepts
of activated sludge.
Figure lA shows the basic process schematic of the acti-
vated sludge system. From the aeration basins, waste flows
into a splitter box and then into three final clarifiers. A
coagulant aid (ferric chloride or polyelectrolyte) is added to
the splitter box. Sludge is drawn from the clarifiers through
three telescopic valves to a sump pump where the recycle sludge
pumps return sludge to the front of the aeration basins. When
required, liquid fertilizer is added to the recycle sludge
pump pit. Effluent from the clarifiers flows by gravity
through a parshall flume to the sand filters.
2.
PROCESS
To produce an acceptable effluent, a biologically active
sludge system must meet the following requirements:
a.
There must be an adequate number of microorganisms
exposed to the food source for sufficient time.
b.
The aeration basin environment must be acceptable
87
-------
3.
to the organisms.
c.
The sludge solids must readily separate from the
treated wastewater in the clarifier.
NORMAL OPERATION
Control of an activated sludge system that is treating a
complex industrial waste is difficult under the best circum-
stances. It is the operator's responsibility to maintain a
proper environment for the system and to collect accurate data
for subsequent process modifications.
The major monitoring is done once/day during what is called
the "Operational Control Tests". The results of these tests
are swmnarized on the chart shown in Figure 3A. The following
summarizes the purpose of each test.
a.
Flows.
The operator first records the daily flows.
AFI ~erator !,low E1
RSF
!!.,ecycle .2,ludge Flow
WSF
!iaste !ludge Flow
b.
Settleometer. The sludge settling rate is monitored in a
Mallory Settleometer. This calibrated two liter beaker is
filled with mixed liquor from the splitter box. The mixed
liquor will form a blanket as it settles and the blanket
is read at 5. 15 and 30 minutes.
c.
Centrifuge. The solids level and the compaction tendency
are monitored with a spin test. Samples from each aeration
basin and the recycle sludge line are poured into an API
tube and centrifuged. The percent solids are recorded
on the sheet. This test is used primarily for balancing
solids in each aeration basin and historical reference.
'.
d.
Sludge Blanket Depth. A sludge blanket
for monitoring the blanket level in the
The blanket should never be higher than
from the water surface.
finder is used
final clarifier.
five (5) feet
e.
Turbidity. The efficiency of the solids removal in the
final clarifier is gauged by ~he turbidity test. This
test is quickly performed using a laboratory turbidimeter.
£.
Aeration Tank DO. Dissolved oxygen is essential for the
proper function of aerobic organisms. However t too much
88
-------
FIGURE 3A OPERATIONAL CONTROL TESTS
Day
Date
Flows AFI API API
GPH RSF RSF RSF
WSF WSIo' WSF
SST SST SSC SSV SSO SSV SSO
~ 5
~ 15
30
j 45
I-<
I;; 60
tJ) Blne Time Hrs. Hrs. Hrs.
ATC 01 ATC 01 A TC ill
ATC 02 ATC '2 ATC 12
ATC 113 ATC 113 ATC 13
~
t:) ATC Comb. ATC Comb. ATC Comb.
;:>
~
....
~
z RSC III RSC III RSC III
~
u KSC 1/2 RSC 112 RSC 12
RSC 113 RS(; /J 3 RSC (/3
KSC Comb. RSe. Comb. RSC Comb.
Sludge
Blanket ooB ooB DOB
Depth
Turbidi ty Tn! to In!t. In! t.
Me te r I Ur. 1 H.:. 1 Hr.
- ---u----- -
Temperature
Pond DO
Flow Adjustments
89
-------
DO wastes energy and causes excessive shear on the sludge
mass. In a balanced, properly operated system, the DO
should range from 1.0 to 3.0 mg/1. If the DO is above this
level, it indicates the amount of food is low or the
organisms have been shocked and available food is not being
metabolized. If the DO is below 1.0 mg/1, the system is
experiencing a high growth rate and filamentioUB organisms
and poor settling may result.
g.
~. The activated sludge process will best perform at a
pH range of 7.0 to 8.0. Acceptable performance can be
achieved in a range of 6.0 to 9.0. The operator should be
aware that carbon dioxide is a biological by-product of
metabolism and forms carbonic acid in water. The incoming
waste should be at least 7.0 to insure the aeration basin
effluent is well above 6.0.
h.
Temperature. All biological processes are temperature
dependent. A decrease of 100C will cut the biological
activity approximately in half. It is important that
aeration basin temperature be monitored in the fall and
winter.
In addition to the operational control tests, operators
should be vigilant for conditions which could affect the
activated sludge environment. Conditions which signal
potential upset include:
a.
Increased Flow. Flows in excess of 600 gpm (.864
mgd) can cause problems in the clarifiers.
b.
Increased TOC. TOC's in excess of certain limits
can upset the activated sludge system. These limits
will vary with time, season and the condition of the
sludge mass.
c.
Low or High DO. A sharp change in the dissolved
oxygen level can indicate either an organic overload
(low DO) or a shock load condition (high DO).
d.
~. The pH in the aeration basin should never get
below 6.0 or above 9.0.
e.
!2!!. Excessive foaming can indicate a binder dump
or an organic overload.
4.
POTENTIAL PROBLEMS
The major problems with activated sludge systems concerns
separating the biological solids in the secondary clarifier.
90
-------
5.
There are three adjustments that can be made when this problem
occurs.
a.
Chemical Dosage. The most common adj ustment is to change
the concentration of coagulant aids. This can best be
determined with ajar test.
b.
Sludge Recycle Rate. The sludge recycle rate can be adjusted
by valving the discharge.
c.
Slud~e Age. The basic control parameter on the secondary
system is the sludge age. This procedure is based on the
length of time the average mic~oorganism remains in the system.
The age directly relates to the amount of excess sludge
produced and to the settling characteristics of the cell
mass. Sludge age can be calculated on the Anderson
system as follows:
Sludge Age -
Vol. of Sludge in Basin
Vol. of Sludge Wasted/day
This calculation assumes sludge is wasted from the aerator.
If sludge is wasted from the returned sludge line, the
following calculation should be used:
Sludge Age ...
MLSS (mg/l) x .42
RAS (mg/l x Vol Wasted/Day
REFERENCES
Additional information on the activated sludge process can be
found in the following references:
a.
Operational Control Procedures for the Activated Sludge Process,
U.S. EPA, Office of Enforcement and General Counsel, 1973.
Part 1
Observations
Part 2
Control Tests
Appendix
b.
Operation of Wastewater Treatment Plant, Manual of Practice
No. II, Published by the Water Pollution Control Federation,
1976.
c.
Operator's Pocket Guide to Activated Sludge.
Part 1
The Basics
91
-------
Part 2
Process Control and Trouble Shooting, Prepared
by Stevens, Thomas & Runyan.
d.
Wastewater Engineering, Ed. by Metcalfe and Eddy, Inc.,
McGraw Hill, 1979.
F.
SLUDGE SYSTEM
1.
DESCRIPTION
The sludge system consists of a 300,000 gallon aerobic
digester and a two acre sludge lagoon. The digester was originally
an anaerobic digester from ~hich the floating cover was removed.
The digester is 45 feet in diameter, 20 foot sidewall and a 5 foot
deep conical bottom. The digester can be filled to within 2 feet
of the top and has a usable capacity of 234,000 gallons. Super-
natant from the sludge lagoon is pumped back to the equalization
basin.
2.
NORMAL OPERATION
The digester is drained by gravity about once a week on the
average. The daily input averages 24,000 gallons. The digester
removes organic matter similar to the activated sludge system.
Solids that carry through the digester will collect in the
sludge pond. The sludge pond is a "facultative" process. This
means that biological activity in the upper pond layers is
aerobic, while the activity in the bottom is anaerobic.
Maintenance of an aerobic layer is very important since it
oxidizes hydrogen sulfide and minimizes the characteristic
septic small of anaerobic processes.
3.
POTENTIAL PROBLEMS
The digester is a biological system, similar to activated
sludge and must produce an acceptable environment for the
organisms to properly function. The major item that the
operator can control is the oxygen input. He should insure
that at least 1.0 mgl1 02 is maintained in the tank.
Biological activity in the pond is temperature dependent
and bio reactions will be much more rapid in water weather.
The solids level in the pond will build over the winter, and
as spring warms the pond up, the accumulated sludge layer
will undergo rapid anaerobic degradation. Vigorous bubbling
can be seen and sludge masses will often float. The greater
risk of sludge odors will occur at this time. The operator
can minimize odors by:
92
-------
a.
Keeping the pond surface clear by removing floating
sludge
b.
Increasing digester retention time by daily drawoff of
small quantities of sludge.
In extreme cases, odors can be reduced by spreading
sludge accumulations to other parts of the pond with a
dragline or using chemical disinfectants such as
chlorine to hold down biological activity.
93
-------
APPEND IX C
r
SPECIAL MICROBIAL PROCEDURES
In order to evaluate the public health aspects of the new water reuse
and recycling system at the Anderson, South Carolina, Owens-Corning
Fiberglas Plant, JTC Environmental Consultants, Inc., was asked to provide
bacterial and viral analyses on water circulating through the plant. The
virus sample was collected on April 17, 1978, and the bacterial sample was
taken on April 26, 1978.
A.
VIRUS CONCENTRATING PROCEDURE
The basic virus monitoring apparatus and procedure is similar to that
set forth in Standard Methods for the Analysis of Water and Wastewater (6).
A summary of the procedure followed by JTC and the specific filters used for
the concentration steps are described here.
The virus concentrating apparatus without the filters in place was
sterilized with chlorine solution containing 25 mg per liter of free chlorine.
The solution was forced into the unit and maintained there for a 30 minute
contact period. This solution was then rinsed from the virus concentrator
by allowing the water to be tested to flow through the system until all
traces of chlorine were dissipated. The system was then flushed of all water
by using air pressure and allowing adequate time for the water to drain.
The filter cartridge holders were then dismantled and the fiberglas
filters installed using aseptic technique. Fiberglas filters for adsorbing
virus consisted of honeycomb filters, Model A-27, obtained from Commercial
Filters Division, Carborundum Company and 0.45 m filters from Filterite
Corporation. Prefiltering was not utilized because of the low level sus-
pended solids in the water and the possibility of virus loss during pre-
filtering.
Virus adsorption was enchanced by pH adjustment to 3.5 and addition
of aluminum chloride. Hydrochloric acid and aluminum chloride solution
were added by means of a proportional dosing pump set to provide a pH of
3.5 during the test runs and a final aluminum chloride concentration of
0.0015 M. The virus concentrator was operating a flow rate between one
and three gallons per minute and periodically checked for pH to assure a
value of 3.5. When necessary the proportioning pump was adj usted to main-
tain pH.
A standard solution of thiosulfate was pumped into the system through
the use of a second proportioning pump to provide neutralization of chlorine.
The thiosulfate solution was previously prepared to neutralize 10 ppm of
total chloride. Upon testing of the water discharge from the virus concen-
trator, a chlorine residual of approximately 0.1 ppm was found in the processed
water. Higher feed rates of thiosulfate did not remove the chlorine resi-
dual as determined by the orthotolidine method. Laboratory investigations
94
-------
indicated that the positive chlorine residual was a result of interferences
from components in the wastewater when using the orthotolidine method and
not chlorine. However, to insure chlorine removal the thiosulfate solution
was prepared to neutralize 20 ppm of total chlorine.
A total water sample of 93.1 gallons was processed. After processing
this volume, all water that had passed through the concentrator was drained
from the unit and the elutant vessel containing glycine buffer at pH 11.5
was connected to the filter. Air pressure was used to force the elutant
through the filters. It was collected and immediately adjusted to pH 7.0.
The eluate was dosed with penicillin and fungizone and frozen in dry ice
for shipping to the laboratory.
Field efficiency for virus detection was tested by processing 78 gallons
of wastewater under conditions identical to those described above for the
wastewater except for the addition of a stock solution of polio virus I,
CHAT strain. A total of 253 ml of stock virus solution was added to a total
of 4,780 ml of thiosulfate solution. This thiosulfate solution was added
to the process water by a proportioning pump during the field efficiency run.
After completion of the efficiency run the thiosulfate-virus solution was
measured at 2,520 ml resulting in a net use of 126.7 ml of stock virus
solution. A sample of stock virus solution was taken in the field, packed
in dry ice and returned to the laboratory for assay to establish total number
of virus units added to the process water during the efficiency run.
The elution procedure during the field efficiency run was identical to
that described above for the test run.
1.
Reconcentration Steps
The frozen eluates samples were thawed upon receipt at the labora-
tory and reconcentrated by precipitation with ferric chloride. The
pH of each eluate was adjusted to 3.5. Ferric chloride was added to
give a concentration of 10-3M. The solutions were then flocculated
for 20 minutes, allowed to settle for 1/2 hour and the supernanant
discarded. After centrifugation the precipitate was suspended
in heat-inactivated calf serum and adjusted to pH 7.5. This
solution was mixed on a magnetic stirrer to provide virus release.
Bacteria and suspended mater were then removed by centrifugation
at 7,000 to 8,000 rpm for 15 minutes. The supernatant was with-
drawn, dosed with antibiotics of penicillin and fungizone and
frozen prior to assay on primary rhesus monkey kidney cells and
HeLa cells. Both the wastewater test run as well as the field
efficiency run were handled identical during this reconcen-
tration procedure. The reconcentration provided a final volume
of 36 ml for the test sample and 63 ml for the field efficiency
run.
95
-------
B.
VIRUS ASSAY PROCEDURES
The waste sample and virus seed concentrates were tested using the
plaque assay method of two cell cultures. Petri dishes of primary rhesus
monkey kidney (PRMK) tissue culture and HeLa cell culture were rinsed with
37°C minimal essential media (~mM), inocul~ted with 0.4 ml of the waste
sample or 0.2 ml of dilutions of the virus preparations, and rocked
continuously at 370C for 90 minutes to allow even distribution of the
virus. The cultures were then overlaid with an agar medium and incubated
at 37oC. Cultures were checked daily for plaque formation and overlaid
a second time with an agar media containing neutral red when plaques
were noted. Plawues ere counted after 24 hours additional incubation.
C.
BACTERIAL TESTING PROCEDURES
1.
Fecal Coliforms
Duplicate, samples of 250 ml, 100 ml, and 10 ml were
filtered through Hillipore filters. The filter funnel was
washed down after each sample with 10 - 20 ml of sterile
distilled water. In addition to the 6 samples, dupli-
cate control samples of 10 ml of sterile distilled water
were filtered. The filters were placed on M-FC broth and
incubated at 450C for 48 hours.
2.
Fecal Streptococcus
Duplicate samples were filtered in a manner identi-
cal to that used for the fecal coliforms (described above).
The filters were placed on KF streptococcus agar and
incubated at 350C for 48 hours.
3.
Total Count
Duplicate samples of 100, 10, and 1 ml were filtered on
the Millipore fi-lter apparatus. Between each sample the
filter funnel was washed down with 10 - 20 ml sterile dis-
tilled water. In addition, duplicate control samples of
10 ml of sterile distilled water were filtered. The filters
were placed on M-Standard Methods broth and incubated at 350C
for 48 hours. Counts of colonies were taken at 24 hours
and again at 48 hours.
4.
Staphylococcus
Duplicate samples were filtered in a manner identical
to that used for fecal coli forms (see above). Duplicate
control samples were also filtered as for fecal coliform
determination. The filters were placed on Chapman-Stone
Agar and incubated at 350C for 48 hours.
96
-------
5.
Total Facultative Organisms
In order to determine the number of facultative
organisms, approximately 40 ml of sample as bubbled with
pure oxygen through a sterile, cotton-plugged pipet for
two hours at room temperature. It was felt that the
exposure to pure oxygen would kill any obligate anae-
robes persent in this sample. Following aeration, four
samples each of five ml and one ml were inoculated by a
pour-plate method in nutrient agar. Two plates of each
sample plus two uninoculated control plates were incubated
aerobically at 350C for 48 hours. The remaining two
plates were incubated in an anaerobic condition at 350C
for 48 hours. Colonies appearing on the anaerobic
plates were considered to be facultative anaerobes.
6.
Obligate Anaerobic Bacteria
Samples of 5 ml and 1 ml were set up in triplicate
or quadruplicate in nutrient agar by the pour-plate
technique. One or two plates of each sample volume were
incubated aerobically at 350C for 48 hours. The
remaining plates of each sample volume were incubated
anaerobically at 350C for 48 hours. In addition, one
uninoculated control plate was incubated aerobically and
one anaerobically in an identical manner to the inoculated
plates.
97
-------
D.
EXAMPLE CALCULATIONS
1.
Field Efficiency Calculations
Virus Stock Pumped through the Concentrator:
253 m1 2520 = 126.7 m1
5033
Virus Concentration of Stock Virus Solutions as measured by:
5
a) PMK: 2.5 x 10 PFU/m1 7
times 126.7 m1 = 3.168 x 10 PFU
b) HeLa: 3.03 x 102 PFU/m1 4
times 126.7 m1 = 3.839 x 10 PFU
Virus Recovered in Field Efficiency Run as measured by:
a) PMK: 4.5 x 104 PFU/m1
times 63 m1 of final sample = 2.835 x 106 PFU
b) HeLa: 3.26 x 102 PFU/m1
times 63 ml-of final sample = 2.054 x 104 PFU
Recovery Efficiency:
a) PMK: 2 835 106
3: 168 X 101
b) HeLa: 2.054 X 104
3.839 104
x 100 = 9%
x 100 = 53%
98
-------
TECHNICAL REPORT DATA
(Please read IRs.1/Uct;ons on the reverse before completing)
1. REPORT NO. 12. 3. RECIPIENrs ACCESSION NO.
E P A - 600/2 - 80 - 040
4. TITLE AND SUBTITLE 5. REPORT DATE
Demonstration of a Closed Loop Reuse System in a J anuarv 1980
Fiberglas Textile Plant 6. PERFORMING ORGANIZATION CODE
7. AUTHOR IS) 8. PERFORMING ORGANIZATION REPORT NO.
S. H. Thomas "and D. R. Walch
9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PRC,GRAM ELEMENT NO.
Owens-Corning Fiberglas Corporation lBB610
Fiberglas Tower 11. CONTRACT/GRANT NO.
Toledo, Ohio 43601 S80ll73
12. SPONSORING AGENCY NAME AND ADDRESS 13. TYPE OF REPORT AND PERIOD COVERED
EPA, Office of Research and Development Fin::!l' 5/73 - 12/7~
Industrial Environmental Research Laboratory 14. SPONSORING AGENCY CODE
Research Triangle Park, NC 27711 EPA/600/13
15. SUPPLEMENTARY NOTES IERL-RTP project officer is Max Samfield, Mail Drop 62,9191
541-2547. 16. A8STRACT The report describes work done toward providing a totally recycled water
system for Owens-Corning's textile fiber manufacturing plant at Anderson, SC.
(The work was based on pre-1968 pilot plant work by Owens-Corning that resulted in
development of totally recycled industrial wastewater systems for all of their in-
sulation manufacturing plants. ) Water quality requirements for the Anderson plant
were considerably more stringent than for insulation manufacturing. Test and engi-
neering design work started in 1973. Design work was completed in March 1977
and actual field work was started. All sanitary wastes from the plant were segrega-
ted for separate treatment. Much modification to the existing treatment was required
to improve the quality of primary and secondary effluent for tertiary treatment. In
the final process, biologically treated effluent is sand-filtered, followed by activated
carbon adsorption and disinfection with chlorine. Major items of equipment added
were a commercial upflow pressure sand filter and three upflow activated carbon
columns with backwash capabilities. The quality of treated effluent is completely
satisfactory for all plant operating requirements. Total recycle of treated effluent
was realized in mid-1978 and has continued since then. Complete closed-loop oper-
tion had not yet been achieved. Discharge to Betsy Creek has been greatly reduced.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Pollution Ins ulation Pollution Control l3B l4B
Textile Indus try Sand Filters Stationary Sources lIE l3K
Textile Processes Activated Carbon l3H llG
Glass Fibers Chlorine lIB 07B
Waste Water
Water Treatment
18. DISTRIBUTION STATEMENT 19. SECURITY CLASS (This Reporr) 21. NO. OF PAGES
Unclass ified 107
Release to Public 20. SECURITY CLASS (This page) 22. PRICE
Unclassified
EPA Form 2220.1 (9-73)
99
-------
US ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Center for Environmental Research Information
Cincinnati. Ohio 45268
OFFICIAL BUSINESS
PENALTV FOR PR IV ATE USE S3OO
AN EQUAL OPPORTUNITY EMPLOYER
If you' address is incorrect, please change on the above label
tear oft, and return to the above address.
If you do not desire to continue receiving these technical
reports, CHECK HERE\H. tear off label, and return it to the
above address.
Publication No. EPA-600/2-80-040
------- |