EPA-600/2-76-252
September 1976
Environmental Protection Technology Series
PAPERMILL WASTEWATER TREATMENT
BY MICROSTRAINING
industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1 Environmental Health Effects Research
2 Environmental Protection Technology
3 Ecological Research
4 Environmental Monitoring
5, Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution Tins
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161
-------
EPA-600/2-76-252
September 1976
PAPERMILL WASTEWATER
TREATMENT BY MICROSTRAINING
by
Frederick R. Bliss
Strathmore Paper Company
Turners Falls, Massachusetts 01376
Project No. 12040 FDE
Project Officer
Edward Conley
Regional Office I
John F. Kennedy Federal Building
Boston, Mass. 02203
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental
Research Laboratory - Cincinnati, U.S. Environmental Protection
Agency, and approved for publication. Approval does not signify
that the contents necessarily reflect the views and policies
of the U.S. Environmental Protection Agency, nor does mention
of trade names or commercial products constitute endorsement or
recommendation for use.
ii
-------
FOREWORD
When energy and material resources are extracted, processed,
converted, and used, the related pollutional impacts on our en-
vironment and even on our health often require that new and in-
creasingly more efficient pollution control methods be used. The
Industrial Environmental Research Laboratory - Cincinnati (lERL-Ci)
assists in developing and demonstrating new and improved methodo-
logies that will meet these needs both efficiently and economical-
Papermill Wastewater Treatment by Microstraining concerns the
use of microstrainer facilities for removal of waste components
contained in wastewater discharges originating from the manufacture
of fine papers. The wastewaters from two small mills were combined,
flocculated and passed through the microstrainer. Significant re-
moval of suspended solids and BOD was achieved by this method of
treatment. Such a system can be installed in situations where on-
ly limited space is available for the development of equivalent
primary facilities. Effectiveness of this method of treatment
should be first determined by suitable laboratory examination to
forecast expected full scale efficiencies. Waste treatment by micro-
straining may be particularly applicable to wastewater discharges
from small tissue mills where fiber losses form the principle pri-
mary waste component discharged by the industry. For further in-
formation contact the Food and Wood Products Branch, Industrial
Pollution Control Division, lERL-Ci.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
iii
-------
ABSTRACT
An original treatment system was designed, constructed, and operated for
removal of suspended solids, turbidity, color, and BOD from the waste-
waters of two paper mills which produce technical and other fine papers.
The treatment process involves coagulation and flocculation followed by
microstraining. Space and cost considerations were of paramount impor-
tance in selecting this process.
Fiber recovery was investigated, but was found to be uneconomical be-
cause of the high percentage of fillers being employed and unaccept-
able levels of color and dirt. The sludge is being discharged to the
municipal sewerage system.
Plant operating efficiencies over the past year indicated substancial
removal of the suspended solids and 5-day BOD. Effluent turbidities
averaged less than 30 Jackson turbidity units (JTU).
The estimated construction cost of the treatment facility is $689,000.
First-year operating costs including wages, power, supplies, chemicals,
microfabric, and maintenance totaled $36,175, which is approximately
equivalent to $1.50 per ton of paper produced.
It is expected that the techniques used in this operation may have
broad applicability to industries under similar space limitations
and using similar manufacturing methods. Cooperative ventures will
be attractive to many small firms to meet the new criteria for indus-
trial wastewater discharges.
This report was submitted in fulfillment of Project 12040 FDE by
Esleeck Manufacturing Conpany and Strathmore Paper Company under the
sponsorship of the Environmental Protection Agency. Work was com-
pleted as of April 1974.
IV
-------
TABLE OF CONTENTS
Section Page
I CONCLUSIONS 1
II INTRODUCTION 3
III DESIGN OF PROTOTYPE TREATMENT PLANT 8
IV OPERATION 32
V DISCUSSION 56
VI PHASE IV PLANT OPERATION OPTIMIZATION 92
VII SOLIDS HANDLING 96
VIII REFERENCES 108
IX GLOSSARY OF TERMS AND ABBREVIATIONS 109
X APPENDICES 110
XI EXHIBITS: TYPICAL WASTEWATER CHARACTERISTICS 117
-------
LIST OF TABLES
Table No. Page No.
1 Design Criteria for Joint Industrial Wastewater 11
Treatment Facility
2 Unit Sizes of Components of Joint Industrial 11
Wastewater Treatment Facility
3 Treatment Plant Alarms 25
4 Construction Highlights — Strathmore-Esleeck 28
Joint Industrial Wastewater Treatment Facility
5 Final Construction Costs, Joint Industrial 31
Wastewater Treatment Facility
6 Microstrainer Seals Utilized 45
7 Phase III Microstrainer 78
8 Pilot Sludge Studies Data 97
9 Microstrainer Backwash Sludge Design Parameters 99
10 Sludge Handling Facility, Unit Design Parameters 103
11 Sludge Handling Operating Costs 105
12 First Year Operating Costs 106
vi
-------
LIST OF FIGURES
Fig. No. Page No.
1 Location Plan 4
2 Yard Piping Schematic 10
3 Junction Manhole Schematic 15
4 Automatic Bypass System 16
5 Process Piping Schematic 17
6 Instrumentation Control Cabinet 21
7 Chemical Feed Schematic 23
8 Plan, Chemical Feed 24
9 Effluent Pump Control Cabinet 26
10 Typical Microstrainer Arrangement 33
11 Daily Operating and Analysis Reports 38-39
12 Plant Effluent Pumped to Canal 40
13 Esleeck Wastewater Flows 41
14 Strathmore Wastewater Flows 43
15 Microstrainer Sealing Arrangements 48-49-50
16 Phase II Suspended Solids Removal 57
17 Phase II Frequency Distribution of Suspended Solids
Removal 59
18 Phase II Turbidity Removal 60
19 Phase II Frequency Distribution of Turbidity Removal ... 61
20 Phase II Color Removal 63
21 Phase II Frequency Distribution of Color Removal 64
22 Phase II BOD Removal 65
23 Phase II Frequency Distribution of BOD Removal 66
24 Phase III Suspended Solids Removal 68
25 Phase III Frequency Distribution of Suspended Solids
Removal 69
26 Phase III Turbidity Removal 71
27 Phase III Frequency Distribution of Turbidity Removal. . . 72
28 Phase III Color Removal 73
29 Phase III Frequency Distribution of Color Removal 74
30 Phase III BOD Removal 75
31 Phase III Frequency Distribution of BOD Removal 76
32 Effect of Seal Leakage on Suspended Solids Removal .... 80
vii
-------
LIST OF FIGURES (CONT.)
Fig. No. Page No,
33 Suspended Solids Removal -- Selected Data 81
34 Turbidity Removal -- Selected Data 82
35 Color Removal -- Selected Data 84
36 Filterability Index 86
37 Long-Term BOD (Samples of 5/9/72) 88
38 Long-Term BOD (Samples of 9/11/72) 89
39 Long-Term BOD (Samples of 9/19/72 and 10/11/72). ... 91
40 Daily Variations of Backwash Sludge 100
41 Schematic Flow Diagram 102
A-l Project Schedule Ill
B-l Ultrasonic Filtration Schematic 116
viii
-------
ACKNOWLEDGMENTS
The investigations and studies reported herein were carried out by
the Esleeck Manufacturing Company, the Strathmore Paper Company, and
Camp Dresser & McKee, consulting engineers. The authors gratefully
acknowledge the cooperation during construction of the treatment
facility and throughout the duration of the study of Mr. Albert Dempsey,
plant manager; Mr. Gustave Sell, plant engineer; Mr. Frederick Bliss,
technical director; and Mr. Francis Vivier and Mr. Thomas Rodencal,
mill chemists of Strathmore Paper Company, and Mr. Irving Esleeck,
president; Mr. Charles Winslow, plant engineer, and Mr. Paul Coughlin,
technical director of Esleeck Manufacturing Company.
Operation and analytical work was assisted by Mr. Leo Mutti and
Mr. Robert Leighton, plant operators. Mr. John S. Mudgett, project
director and chief engineer of Strathmore Paper Company, assisted in
the coordination and administration of all project activities.
Camp Dresser & McKee, Inc., Boston, Massachusetts, was responsible
for the design and general supervision of the treatment plant con-
struction and the first year of operation. Mr. Darrell A. Root,
project officer, Mr. John C. Thompson, project engineer and Mr. Norton
G. True, field engineer, assisted in the preparation of this report.
Mr. Alan E. Rimer, formerly of Camp Dresser & McKee, assisted in the
initial phases of the project.
Mr. Thomas McMahon, director, Mr. Han Bonne, supervising sanitary
engineer and Mr. Alfred Ferullo of the Massachusetts Water Resources
Commission, Division of Water Pollution Control, also gave generously
of their time.
The support of this project by the Environmental Protection Agency —
Mr. George R. Webster, project manager, Mr. James D. Gallup, plant
engineer, and Mr. Edward Conley, project officer -- is greatly
appreciated.
ix
-------
SECTION I
CONCLUSIONS
The following conclusions have baan reached, based on the experience
of one year's operation:
1. The microstraining process was effective in treating
the paper mill wastewaters, when mechanical failures
did not hinder the operation of the treatment plant.
2. Chemical treatment was required to maintain an accept-
able degree of suspended solids, turbidity, and color
removal, but close analytical control, using bench
scale jar studies, was needed to maintain the best
chemical dosages for optimum removal efficiencies
and acceptable filtration rates.
3. A secondary coagulant, or polymer, together with pH
control, was needed practically all the time as a
flocculant aid and floe strengthener. Polymers were
effective in dosages of 0.5 to 2.0 tng/1. The most
common dosage overall was 1.0 mg/1. Only the strong,
aniomic, high molecular weight polymers were effective.
Polymers increased the flow capacity of the micro-
strainers by keeping the solids mat open, thus allow-
ing better drainage.
4. Microstrainer No. 2 (35-micron microfabric) was capable
of as good a removal efficiency as microstrainer No. 1
(23-micron microfabric) during chemical feed periods.
The data at times was primarily a result of the seal
leakage of microstrainer No. 1 (from a damaged drum).
5. Under normal operating conditions, microstrainer No. 2
had a higher flow capacity than microstrainer No. 1.
6. The microfabric on both micros trainers was not always
clean and free, because of low backwash water pressure
and large amounts of very fine suspended solids which
clogged the microfabric openings. Serious hydraulic
deficiencies sometimes developed because of this clog-
ging of the screen, especially on microstrainer No. 1.
A thorough cleaning of the microfabric with 10% phos-
phoric acid cleaning is needed once each week, on the
average, to overcome this.
-------
7. Average incoming wastewater flows were somewhat higher
than expected, and peak incoming flows and flow surges
at times exceeded the design capacity of the treatment
plant. These factors upset the operation and efficiency
of the plant.
8. The most effective seals to date for the microstrainer
are the knife edge plastic and the dumbbell seals, as
shown in Fig. 15. Constant attention is needed to be
sure the seals are clean and tight. Investigation is
continuing to find more effective saal configurations
and materials.
9. It was impossible to evaluate properly the ultrasonic
microfiltration process because of a malfunction in
the automatic controls of the ultrafilter and the
lack of technical assistance from the manufacturer
to correct it. It was not felt that this process
would be practical in treating wastewater with sus-
pended solids concentrations as high as those experi-
enced at the treatment plant.
10. The backwash sludge was easily dewatered to a consist-
ency of 20 to 30 percent solids by flotation thickening
followed by either a solid bowl centrifuge or a vacuum
coil filter. This dewatered sludge could either be dis-
posed of in a sanitary landfill, sold as a byproduct,
or recycled in the paper process. The paper mills feel,
at this time, that it would be impractical to recycle
the paper sludge because of the large percentage of
filler material and the high cost of conditioning for
reuse.
-------
SECTION II
INTRODUCTION
BACKGROUND OF THE INVESTIGATION
The three major dischargers of industrial wastewaters in New England
are pulp and paper manufacturers, the textile industry, and the tan-
ning industry. Because of ths increasing public concern with the
effect of these wastewaters on the receiving streams, and more strin-
gent federal requirements, many of these industries are faced with
costly cleanup programs or closing their operations.
Currently, several processes are used by paper manufacturers to treat
their wastewaters to varying qualities. Many of these processes are
biological, and require relatively large areas for the treatment
facilities.
To find an effective and economical treatment process to remove a sub-
stantial portion of the suspended matter and BOD (biochemical oxygen
demand) from the wastewaters of two paper companies (Esleeck Manufac-
turing Company and Strathmore Paper Company, Turners Falls, Massachusetts),
Camp Dresser & McKee, Inc. conducted pilot plant studies at each plant.
The two plants are located in the western part of Turners Falls on an
island bound by the Connecticut River on the northwest and a power canal
owned by the Western Massachusetts Electric Company on the southeast.
Fig. 1 shows the location of the mills.
Both companies have endeavored for years to determine an economic means
of reducing their pollutional discharges to the Connecticut River. In
this connection, extensive independent and joint studies have been con-
ducted for both mills by the engineering firm of Camp Dresser & McKee,
Inc. (COM). The most recent work for Esleeck Manufacturing Company was
completed by CDM in September 1967: "Report on Process Wastewater Treat-
ment." In June 1969, CDM completed a "Report on Process Wastewater
Treatment" for the Strathmore Paper Company.
Both reports recommended the installation of a mechanical process uti-
lizing microstrainers to remove suspended solids and turbidity from
the wastewaters. Other ^nore conventional methods of treatment such as
sedimentation, vacuum filtration, and full treatment in the town (Montague)
sewerage system were studied, but were not considered feasible either be-
cause of the lack of the excessive capital and operating costs associated
with the particular alternative.
-------
Treatment
facility
Turners Falls
Scale: 1 in. = 200 ft
Fig. 1. Location Plan
-------
In the 1967 pilot plant study conducted at the Esleeck Manufacturing
Company, microstraining alone was used to remove the suspended solids
in the wastewater. In the 1969 Strathmore Paper Company investigation,
flocculation and coagulation, as well as microstraining, were required
because of the higher degree of suspended solids and turbidity in the
wastewaters. The experience gained in the Esleeck studies was a sub-
stantial aid in evaluating procedures for the subsequent Strathmore
investigations.
Since October 1971, a CDM-designed plant, the first of its kind, has
been treating an average of over 1.8 mgd of wastewater from the Strath-
more and Esleeck plants in Turners Falls. These wastewaters, basically
the excess Whitewaters from both mills, are subjected to coagulation
and flocculation and are then sent through two 10-ft-diameter by 10-ft-
long microstrainers prior to discharge.
The decision to employ microstrainers was based on previous pilot plant
studies conducted at both mills, from which it had been concluded that:
1. The mills could not discharge untreated wastewater to
an adjacent power canal.
2. It was not economical, nor was space available, to
construct clarifiers.
3. Neither the Montague municipal sewerage system nor the
primary wastewater plant had the capacity to treat the
wastewaters of the two mills.
The present report discusses the results of the operation of the treat-
ment facility and some of the problems encountered in establishing the
routine operation of the plant.
PROJECT OBJECTIVES
Because of the significance of this project to the paper industry, the
cost of design, construction, and the first year of operation of the
treatment plant was partially funded by a research and construction grant
from the Environmental Protection Agency (EPA) to the amount of $252,345.
Under the terms of this grant, our project objectives were:
-------
1. To investigate the applicability of microstraining mill
wastewaters from two paper manufacturers (whose products
consisted of technical and other fine papers made from
rag or chemical wood pulps) to determine the degree of
removal of BOD, suspended solids, color, and turbidity
2. To investigate the use of a coagulant or coagulant aid,
such as a polyelectrolyte, for more effective removal
of turbidity from the wastewaters
3. To determine the possibility of reclaiming fibers from
the microstrained sludge (appropriate treatment processes
such as centrifugation or sedimentation may enable mills
to recover lost fibers economically)
4. To determine the annual operating costs for such a treat-
ment plant based on one year of operation
5. To determine design factors and to estimate the cost of
construction and operation of such a facility (the cost
of operation will be correlated with the retail market
value of the product)
6. To conduct tests on an ultra filter manufactured by the
American Process Equipment Corporation and supplied free
of charge (including shipping and major maintenance) by
the FWPCA, now called EPA.
Grant funds were committed in May 1970; engineering design was completed
in July 1970; construction was started in January 1970 and substantially
completed in October 1971. Operation of the treatment plant began in
November 1971 and the project objectives were completed in October 1972.
PLAN OF OPERATION
To accomplish the above objectives, a plan of operation consisting of
three phases was established. A detailed log of operations and a bar
graph of the phases are given in Appendix A.
-------
Phase I was tc consist of the evaluation of the ultrafilter described
above (two to three months of testing). Construction and operationg
costs were to be developed for a full scale plant, if the unit proved
suitable for this application. A description of the ultrafilter studies
and their results is given in Appendix B.
Phase II, in actuality tne initial phase, consisted of the operation of
the treatment plant, which was composed of an equalization basin, floc-
culation basins, micros trainers, and effluent pumps. The wastewater
streams to be treated averaged 1.8 mgd (million gallons per day), and
contained approximately 4,500 Ibs. of suspended solids daily. The
purpose of this phase was to demonstrate the applicability of micro-
straining without chemicals for paper mill waste treatment. During its
four-month duration, suspended solids, turbidity, color and BOD removals
were measured and evaluated.
Phase III consisted of the investigation of the use of a coagulant or
coagulant aid prior to the microstraining for more effective removal
of suspended solids, turbidity, color, and BOD. During this phase,
possible coagulants, such as those utilized in the Strathmore pilot
plant investigations (Nalco 636 and sodium alumiriate), as well as alum
and other polymers were tested to determine the most effective and ef-
ficient combination. Evaluation of thfi coagulant or coagulant aid was
based on its performance and a cost effective analysis. It was antici-
pated that three to four months would be necessary to determine the
optimum combination and dosage of coagulants and another four to five
months to collect adequate data over a wide operating range to estimate
the overall efficiency and effectiveness of the process.
Phase IV consisted of the 18-month period following the conclusion of
Phase III. Investigations leading to the successful operation of the
plant were concluded in this period.
During the course of our study, alternative methods of sludge disposal
were evaluated. Initially, we considered that the waste sludge averaging
75 gpm (gallons per minute) would be discharged to the municipal sewer-
age system, and thus tests were conducted at the municipal plant to eval-
uate the effect of this sludge on the sewerage system and on the treat-
ment plant itself. Preliminary tests indicated that the sludge would
present no undue burden on the system or the treatment plant.
It was considered possible that appropriate mechanical processes such
as centrifugation or vacuum filtration could be used as a first step
to reclaim the paper fibers which could then bft treated for ultimate
reuse. These processes would be evaluated.
-------
SECTION III
DESIGN OF PROTOTYPE
TREATMENT PLANT
BRIEF DESCRIPTION OF MILLS
Strathmore Paper Company
The Strathmore Paper Company, a division of Hammermill Paper Company,
Inc., produces a wide variety of opaque and colored paper products.
Paper is manufactured from virgin chemical pulp, rag pulp, and reused
fibrous material on three 86-in. Fourdrinier paper machines, capable
of producing 54 tons per day. The plant employs about 115 persons and
operates three 8-hour shifts daily, six days per week.
The primary sources of wastewater discharges from the Strathmore mill
are the Whitewater, seal box, and wire pit overflows of the three paper
machines. Smaller intermittent flows from the pulper and various wash-
up operations also contribute to the discharge. Before and during con-
struction of the treatment plant, Strathmore began construction of a
Whitewater recirculation system designed to reduce their total indus-
trial wastewater discharge from 2.30 to 0.75 mgd. Under this system,
the relatively clear Whitewater flows would be segregated from the
other wastewaters for transmission to the treatment plant separately.
Wastewater from the Strathmore paper machines flows directly to the
treatment plant. Save-alls are not used at this mill to recover paper
fibers because of the numerous paper grade changes.
Esleeck Manufacturing Company
The Esleeck Manufacturing Company produces chiefly high-grade onionskin
and thin bond papers. The mill output is approximately 16 tons of paper
per day, manufactured primarily from rag pulp on a single 122-in. Four-
drinier paper machine. Other processes used to produce the paper in-
clude rag cooking, rag bleaching, and rag pulping operations. Sheet
finishing and sheet coating services are also provided. The plant em-
ploys approximately 120 persons and operates three 8-hour shifts daily,
five days per week.
Paper machine Whitewater from the Esleeck mill is pumped to the treat-
ment plant after a large portion of the suspended solids (SS) is ef-
fectively removed in a flotation-type save-all. The smallest percentage
-------
of the mill's wastewater (37 percent), which results from the rag
washing operations, flows by gravity to the plant via a separate line.
Because of the high BOD and sulfite concentration, the rag cooking
liquors from two rotary boilers are discharged directly to the Montague
sewerage system by a centrifugal pump. Fig. 2 shows the yard piping
layout from the two mills to the treatment plant.
Water for the various plant processes of both mills is obtained from
the Western Massachusetts Electric Company (WMECO) power canal. Be-
fore use, it undergoes coagulation and filtration in separate treat-
ment plants. Water for sanitary services is obtained from the Turners
Falls fire district. Total process water consumption presently averages
over 1 mgd for each mill. Domestic wastewater from both mills is dis-
charged to the Montague sanitary sewerage system through individual
ejector stations.
DESIGN CRITERIA OF TREATMENT PLANT
Based on the pilot plant studies described in the 1967 and 1969 COM
reports, the treatment plant was designed to provide adequate mixing,
coagulation, and microstraining of the combined wastewater from the
two mills (except the rotary boiler wastes from the Esleeck mill) to
maintain the Class B quality in the power canal and in the Connecticut
River. The primary intent of the design was to remove suspended solids,
color, and turbidity from the wastewaters of the two mills. Removal
of suspended solids (the fines, such as clays and other fill-j^s asso-
ciated with paper making operations) was enhanced by the coagulation
process, and this resulted in clearer effluents and greater efficiency
of BOD removal.
The design of the treatment facility was based on the flow and solids
loadings outlined in Table 1.* Table 2 describes the basic unit sizes
in the plant. The equalization basin provides time for damping of
surges in the solids loadings before the introduction of chemicals in-
to the wastewater in the rapid mix basins. Flucculation of the waste-
water occurs in the slow mix basin before the solids are removed by the
two micros trainers, which employ different, sizes of microfabrics. Screen
mesh sizes and drum sizes are directly proportional to those used in
the previous pilot studies, which were based on a thorough study of the
filterability index. Filtered effluent from the plant is pumped to the
WMECO power canal for dilution. Backwash sludge solids are pumped to
Montague sanitary sewers for transfer to the Montague sewage treatment
plant.
-------
Esleeck Mfg.
Co. mill
Junction manhole
P 4
• r S S S f *\
^Primary x J
/ coagulant y
^transfer ^
s
Primary coagulant'
storage room
Strathmore Paper Co.
xx / y s / / s s
/ s s
Rag boiler
effluent line
Strathmore Whitewater
Strathmore wirewater
Esleeck Whitewater
Esleeck washerwater
Plant bypass
LH)
m
Legend
Plant bypass
Plant influent
Plant influent
Plant overflow and drain
Backwash sludge to sewer
Effluent to canal
Plant bypass influent
to wet well
Gate valve No. 24
Fig. 2. Yard piping schematic
-------
TABLE 1
DESIGN CRITERIA FOR JOINT INDUSTRIAL
WASTEWATER TREATMENT FACILITY
Waste Source
Esleeck
White water
Washerwater
Strathmore
Whitewater
Wirewater
Flow,
Average
0.65
0.37
0.46
0.26
mgd
Maximum
1.08
0.57
0.72
0.43
Average SS,
Ib/day
1,200
500
3,200
300
Total
1.74
2.80
5,200
*Exhibit A lists the actual characteristics of the wastewater flow received
at the plant.
TABLE 2
UNIT SIZES OF COMPONENTS OF JOINT INDUSTRIAL
WASTEWATER TREATMENT FACILITY
Design flow, mgd
Average
Maximum
Equalization basin
Dimensions, ft.
Capacity, gal.
Detention time, min
Average
Minimum
Mixar size, hp
Rapid mix chamber (two)
Dimensions, ft.
Capacity, gal.
Detention time, min.
1.74
2.80
26 by 17 by 9 side water depth
30,000
25
15
10
5 by 8 by 9 side water depth
2,700
11
-------
TABLE 2. UNIT SIZES OF COMPONENTS OF JOINT INDUSTRIAL
WASTEWATER TREATMENT FACILITY (Cont.)
Average 2.3
Minimum 1.4
Mixer size, hp 3
Flocculation chamber
Dimensions, ft. 17 by 17 by 8.5 side water depth
Capacity, gal. 18,500
Detention time, min.
Average 15
Minimum 9.5
Mixer size, hp 1.5
Micros trainer (two, 10 by 10 drum)
Maximum screen area, sq. ft./min. 1,500
Maximum head loss, in. 6
Screen opening size, microns 23 and 35
Pu,p wet well
Dimensions, ft. 3.25-by 27 by 8 side water depth
Capacity, gal. 5,300
Detention time, min.
Average 4
Minimum 3
Effluent pumps
Variable speed pumps Three (one standby)
Capacity of each, gpm 1,200
Force main to canal, in. 12 (internal diameter)
Sludge disposal*
Fixed discharge pump
Number Two (one standby)
Capacity of each, gpm 120
*Discharge to 4-in. force main to Montague sewer system.
12
-------
UNIT OPERATIONS OF TREATMENT PLANT
In-Mi11 Changes
Prior to design and construction of the treatment plant, both mills made
major in-mill piping changes to collect all extraneous sources of waste-
water discharge and to reduce the total flow. Esleeck's changes con-
sisted of the construction of a concrete holding tank and new gravity
piping connections from th*3 paper machina and Whitewater save-all and
other smaller wastewater lines. Two 700-gpm pumps with automatic level
controls then deliver the Whitewater to the treatment plant through a
1C-in. line. The flows from the individual rag washers (two units) are
collected and the flow by gravity directly to the treatment plant through
an 8-in. cast iron pipe. The rotary boiler wastes (cooking liquor) are
discharged via a metering pump to the town sanitary sewer through the
4-in. cast iron line that joins the backwash sludge force main outside
the treatment plant.
Strathmore's changes consisted of the following for each paper machine:
1. Collecting the paper machine Whitewater flows and diverting
them to a Whitewater holding tank
2. Providing recirculation pumps to deliver Whitewater from
the holding tank to the hydropulper and to the headbox
of the machine for the makeup water
3. Constructing new seal box holding tanks and providing
recirculation pumps to the Whitewater tanks to reduce
the amount of overflow from the seal box
4. Collecting the wirewaters (water that washes the forming
wire) into a separate line for delivery to the treatment
plant or to the Whitewater tank as needed for makeup water.
For the entire mill (all three machines), Strathmore made the following
changes:
1. Constructing an in-mill drainage system that would collect
all wastewater overflows, including the Whitewater tanks,
seal boxes, floor drains, stock chest drains, and other
miscellaneous sources that contain wastewater
2. Conveying these flows, of which Whitewater is the largest
volume, to the wastewater treatment plant through a 12-in.
cast iron gravity sewer
13
-------
3. Discharging the wirewater flows through a separate 8-in.
cast iron gravity sewer to the treatment plant
4. Separating all starch wastes from internal and external
size operations and diverting these to the existing
domestic waste collection system.
Junction Manhole
Wastewaters from the two mills pass through a junction manhole in the
driveway beside the treatment plant (Fig. 2). The purpose of this man-
hole is to house the magnetic flow meter equipment and to combine the
four separate wastewater flows into the plant influent line. Fig. 3
is a schematic layout of the junction manhole. Whitewater from the
two mills is measured by individual magnetic flow meters before it flows
directly to the plant for treatment. The washerwater from the Esleeck
mill and the wirewater from the Strathmore mill, both usually clear,
are measured by separate magnetic flow meters and are then passed
through diverting plug valves. An automatic sampling and turbidity
monitoring system controls these valves.
Automatic Bypass System
The automatic bypass system is designed to divert the flow from the
Esleeck washerwater line, the Strathmore wirewater line, or both to
the power canal when these flows are clear, thus reducing the hydraulic
load on the plant and saving chemical costs. This system consists of
a hydraulically operated three-way plug valve controlled by an auto-
matic sampling and turbidity monitoring system (Hach flow-through tur-
bidimeter). If the turbidity monitoring system senses a high turbidity,
the flow is automatically diverted to the treatment, plant by means of
the plug valve. When the turbidity drops below a preset level, generally
50 JTU, the flow is diverted to the effluent wet well, mixed with the
effluent from the screens, and pumped to the power canal. Fig. 4 is a
schematic layout of the automatic bypass system.
Equalization Basin
The equalization basin is designed as a holding tank to provide th*:
necessary detention time for complete mixing of the separate waste
streams entering the plant, and as a surge tank to control peak flows
and solids loadings prior to entering the rapid mix basins. As shown
in Table 2, detention time at average plant flow is 25 minutes. Waste-
water enters the basin through a 14-in. cast iron influent line and ex-
its through one or more of three 30-by 24-in. slide gates controlling the
flow through one or both of the rapid mix basins or into microstrainer
No. 1. Fig. 5 is a process piping schematic of the treatment plant.
14
-------
Sample
pump
Sample
pump
PV
PV
-Q
Legend
Strathmore Whitewater
Strathmore wlrewater
Esleeck Whitewater
Esleeck washerwater
Three-way plug valve
Three-way plug valve
Magnetic flow meters
Plant bypass line
Plant bypass line
Plant influent line
Fig. 3. Junction manhole schematic
-------
Sample pump
From timer
in control panel
1/2 in. water for hydraulic
operators
alarm relays
or manual control
•"'in control panel
Limit switch
output to signal
switch in control
anel
Sample
pump
Hydraulically
Start-stop
on control
panel and remote
To river
or pumps
Turbidimeter
wall mounted
in laboratory
To
, converter in
i I' control panel
I
To open drain
115 v,"60 Hz
Fig. 4. Automatic bypass system
-------
P.V.
M.S.
-o
Legend
Direction of flow
Sluice gate, slide gate,
gate valve, or mud valve
Three-way plug valve
Micros trainer
Magnetic flow meter
Plant influent
Plant overflow and
drain line
Backwash sludge to pumps
Backwash sludge to sewer
Effluent to pumps
Effluent to Canal
Effluent recirculation
Equalization basin
suction to pumps
Plant bypass influent
to wet well and pumps
Rapid mix basins
Equalization basin
M.S.
No. 2
M.S.
No. 1
Effluent wet well
/ - -^feHD-
/No. 1 No. 2^
•—Backwash sludge pumps
Fig. 5. Process piping schematic
-------
To keep the wastewater in suspension in the equalization basin, a
Lightnin 10-hp axial flow turbine mixer is mounted in the center of
the tank. A 14-in. overflow pipe is provided for emergency overflow
if the plant operating water l^vel becomes too high. A 10-in. cast
iron suction line is also provided in this basin to facilitate de-
watering.
Rapid Mix Lasins
From the equalization basin, the wastewater flows into one or both
rapid mix basins where chemical flocculation normally begins. A
Lightnin 3-hp axial flow i'.win turbine mixer in eacn basin provides
rlash mixing of the wastewaters „ Detention time, at an average flow
of 1.8 mgd, is 2 rr.inutes if the basins are used in parallel or 4 min-
utes if they are used in series. Primary coagulants (alum and/or
sodium aluminate) are introduced into the wastewater through diffusers
in the slide gate channels entering the basin.
Slow Mix Basin
The slow mix basin provides the necessary mixing and detention time to
complete flocculation of the suspended material before the flow is
passed to the microstrainers through 48-by 30-in. slide gates. The
detention time in this basin is 1.5 minutes at average flow, and mixing
is supplied by a 1.5-hp Lightnin variable speed turbine flocculator.
Primary coagulant and/or a coagulant aid (polymer) can be fed through
diffusers in the inlet channels from the rapid mix basins.
Microstrainers and Microstrajner Chambers
After passing through the equalization, ri^'J, and slow mix basins, the
wastewater enters the mi:rostrainer chambers which house the 10-ft-
diameter by 10-ft-long microstrainers -- the basic treatment units of
this wastewater treatment plant. Flow can enter these units from the
equalization basin, the slow mix basin, or both. Flow enters the micro-
strainer through one end, called the upstream end, and exits through the
sides of the drum which are covered with the microfabric. The filtering
action takes place on this microfabric. Under normal operating conditions,
the drum is submerged in tha fleeing water to about two-thirds of its
depth.
As the waste water passes out through the micrcfabric, suspended material
is retained on the inside and forms a continuous blanket of sludge. As
the drum rotates, this blanket of sludge moves to the top and is washed
off by a backwash spray from two spray heads. The sludge then falls in-
to a hopper and is carried by gravity to the backwash sludge pumps. The
filtered wastawater, or filtrate, then flows over the effluent weir into
the effluent wet well.
18
-------
The two Glenfield & Kennedy micros trainer units were furnished by the
Ccchrane Division of the Crane Company, King of Prussia, Pennsylvania.
The two units are identical except for the mesh size of the microfabric
(unit No. 1: 23 micron; unit No. 2: 35 microns). These units have
the following common features:
Diameter and length of drum
Microfabric
Screen area
Gear drive and motor
Drum speed
Speed control
Backwash spray system
Backwash flow
Backwash supply
Effluent Wet Well
10 ft by 10 ft.
Stainless steel (monofilament wound)
314± sq ft (each unit)
10-hp hydraulic variable spaed
0 to 150 fpm (feet per minute) (0 to
4.78 rpm)
Pneumatic, automatic, and manual
32-jet header system (each unit)
49 gpm at 40± psi (each unit)
2-in mill water or 3-in. filtered
effluent
The effluent wet well maintains a constant flood suction for tha effluent
pumps; its normal operating depth is 5 ft of filtered effluent from the
microstrainer chambers. Connections to the wet well are a 14-in effluent
pump suction line and a 12-in. inlet connection from the plant bypass line,
Effluent Pumps
Since that reach of the Connecticut River in the area of the treatment
plant often has no flow, it was not possible to discharge the plant ef-
fluent directly into the river and still achieve an acceptable degree of
dilution. It is therefore necessary to pump the effluent from the wet
well through a 12-in. force main into an 8-in. diffuser in the bottom of
the power canal. Three 50-hp Warren model 6PH-15 centrifugal pumps,
rated at 1,200 gpm at 115 ft. total dynamic head (TDH), are installed in
the pump room.
19
-------
The pump speed and level control system is an automatic variable frequency
system furnished by Electric Machinery Company, Minneapolis, Minnesota.
This system operates any two of the three effluent pumps according to the
liquid level in the wet well. The third pump is kept on standby with an
on/off full speed control only. A manual pump selector switch permits
any combination of pumps to be run variable speed (ampli-cycle) and manual
speed.
Backwash Sludge Pumps
Two sludge pumps were provided to pump the backwash sludge generated by
the micros trainers to the Canal Streat sanitary sewer for transmission
to the Montague wastewater treatment plant. Each pump is rated at 120
gpm at 100 ft TDH of mixed flow backwash sludge. Pumps are on/off full
speed control: one is kept on standby.
The sludge pumps originally selected for the plant were Megator "snore"
pumps, which were recommended by Gle:afield & Kennedy-because they had
employed them previously for pumping microstrainer sludge. For reasons
discussed in another section of this report, the Megator.pumps were
later replaced by Warren Solids-Master end suction impeller pumps.
Flow Measurement and Instrumentation
Flow measuring devices in the treatment plant are all of the magnetic
flow meter type. Meters are located on the four incoming wastewater
lines in the junction manhole, as well as the plant effluent and back-
wash sludge force mains. Strip chart recorders and totalizers are lo-
cated in the central instrument control cabinet furnished by the Wallace
& Tiernan Division of the Pennwalt Corporation. This cabinet, situated
in the office/laboratory area adjacent to the motor control center, is
shown in Fig. 6. Instrumentation for most of the treatment plant is
also located in this control cabinet.
Chemical Feed System
The alum feed system for the treatment plant consists of:
1. Two 2,000-gal. fiber glass holding tanks for liquid alum storage
2. Transfer piping and a 50-gal. day service tank (located in the
treatment plant itself), .with liquid level probes for maintain-
ing a constant level
3. Chemical feed metering pumps and a rotameter broad for diverting
flow to one or more chemical diffuser locations.
20
-------
I PANEL NAMEPLATES
Fig. 6. Instrumentation control cabinet
21
-------
Originally, the alum feed system described above was used for sodium
aluminate. When it became apparent that alum was needed, and in far
greater quantities than sodium aluminate, the system was converted.
Because of the flexibility built into the original system, it was possi-
ble to divide the polymer feed system and create a completely separate
sodium aluminate system consisting of:
1. One 900-gal. holding tank for liquid sodium aluminate
2. Chemical feed metering pumps and rotameter board for diverting
flow to the inlet of one or both rapid mix basins.
The polymer feed system for the treatment plant consists of:
1. One 1,200-gal. makeup and holding tank equipped with both
dry and liquid polymer makeup facilities
2. Polymer metering and diverting facilities similar to the alum
and sodium aluminate systems.
Fig. 7 is a schematic diagram of the complete chemical feed system.
Fig. 8 shows the outlet locations of the chemical diffusers in the
sub-structure of the treatment plant.
Alarms and Alarm Systems
A fully automatic alarm system has been installed to prevent damage to
the equipment or interruption in treatment efficiency due to failure of
any operation. The alarms and their pertinent systems are listed in
Table 3.
Alarm lights are mounted on the effluent pump panel (see Fig. 9). An
audible alarm buzzer and alarm silencer and reset button are supplied.
The alarm light will not go out until the condition has been corrected.
A four-way selector switch is mounted adjacent to the alarm lights for
use in sounding alarms in one or all of the following places:
Position No. Location
L All stations
2 Treatment plant control area
3 Strathmore Paper Company boiler room
4 Esleeck Manufacturing Company boiler room
An alarm lamp will light and an audible alarm will sound in either or both
mill boiler rooms when any one of the 12 alarms trips in the control panel
at the treatment plant. It is anticipated that this complex system of a-
larms will insure constant and efficient operation of the treatment plant,
even in the absence of the treatment plant operator during night shifts
and weekends.
22
-------
NJ
u>
Note: PC 2 connects to PC 1, 3, 4, 5
P 7 connects to P6, 6A., 8,
8A, 9, 10
/Throttling valve
reducer
to
IX
Pressure^
regulator
Rotametei
(typ)
Solution tank
(1,000-gal. polymer
makeup tank)
1/2 in. primary
coagulant
service !!
primary coagulant
tanks
relief valve
polymer day service
in. water
•Dry polymer
disperser
Level probes
Tor continuation, see Fig. 8
50-gal. primary coagulant
PC Primary Coagulant
f Polymer
• Ball valve
Fig. 7. Chemical feed schematic
-------
tsJ
Microstrainer No. 1
Microstrainer No. 2
For continuation, see Fig.
PC Primary coagulant
P Polymer
Note:
PC 2 connects to PC 1,
3, 4, 5.
P7 connects to P6, 6A,
8, 8A, 9, 10.
Equalization basin
PC
Fig. 8. Plan, chemical feed
-------
TABLE 3
TREATMENT PLANT ALARMS
Alarm
High condensate
High wet well level
Low wet well level
High effluent turbidity
Microstrainer low air
Micros trainer No. 1, high
differential •
Microstrainer No. 2, high
differential
Microstrainer bearing water
low flow, each drum
Microstrainer drive overload,
each drum
Equalization basin, high
level
Backwash sludge, no flow
Power failure
System
Pumps for condensate return for steam
heating system
Wet well for plant effluent pumps
Wet well for plant effluent pumps
Plant effluent sampling and turbidity
monitoring system
Air supply for microstrainer automatic
control system
Hydraulic head loss through the screens
exceeds 10 in. of water
Hydraulic head loss through the screens
10 in. of water
Little or no flushing and lubricating
water being supplied to the bearings
Hydraulic oil pressure in variable
speed drum drive; indicates high head
loss or drum bearing problems
High operating level in treatment plant;
indicates condition of approaching emer-
gency overflow
Backwash sludge pumping system; indicates
pump failure
Indicates no electrical power to treat-
ment plant
25
-------
Fig. 9. Effluent pump control cabinet
-------
ACTUAL CONSTRUCTION
The Strathmore Paper Company and the Esleeck Manufacturing Company shared
equally the construction costs of the joint industrial wastewater treat-
ment facility and, on a prorated basis, the operating costs. Bids for
construction were let in the fall of 1970; Nalews Construction Company
of Laconia, New Hampshire, was the low bidder. After some negotiations,
a contract for the construction was signed on December 12, 1970.
Work on the project began on January 18, 1971, when the contractor moved
equipment to the site; however, excavation for the building did not be-
gin until January 25, 1971. Highlights of the construction activity,
some of which are described below, are given in Table 4.
After one month of construction during a very cold January (average
temperature was less than 15°F., with a recorded low of -33°F., the base
slab and piers were ready to be formed and poured. Because of unusually
heavy rains and other problems encountered in the excavation (such as
old mill debris), the final pours for the base slab were not completed
until March 24. The contractor utilized a critical path scheduling
technique to insure the timely finish of the project, and it was during
this two-month period that the technique illustrated its greatest flexi-
bility. Other tasks such as yard piping, some manhole construction, and
work not affected by weather were completed. Thus jobs which could not
be finished on schedule were allotted time at the end of the project, so
that the actual date of completion need not be changed.
On April 13, the excavation and wall forms, half of which had been erected,
were inundated. Recent work on the Northfield pumped storage project,
one of the nation's largest, required modification of several dams on
the Connecticut River. The Montague dam, just upstream of the treatment
plant site, was not quite complete and a tainter gate on the dam was
raised but could not be lowered. Water in the Connecticut River rose
nearly level with the proposed elevation of the main floor slab of the
plant.
No permanent damage was done either to the excavation or the form work and
slab, which were in place, but heavy deposits of silt were left in the
construction area. Although the high water receded within about eight hours,
the general level of runoff in the Connecticut River basin was such that
the operating authority at the dam could not insure that flooding would
not recur. It was therefore decided to halt construction on the building
until a more certain prediction of flood conditions could be made.
27
-------
TABLE 4
CONSTRUCTION HIGHLIGHTS -- STRATHMORE-ESLEECK
JOINT INDUSTRIAL WASTEWATER TREATMENT FACILITY
1972 Date
January 18
January 25
February 26
March 24
April 13
May 10
May 20
May 27
June 15
June 23
June 28
July 13
July 20
July 22
July 26
July 28
August 24
September 7
October 5
October 14
Contractor began work on site by constructing work
sheds, etc.
Excavation for main building begun
Piers for base slab foundation poured
Base slab, first major concrete pour
Flood waters inundate site
Excavation for 12-in. effluent line begun
Flood cleanup commenced
First wall pour for below-grade structure
Final wall of below-grade structure poured
First slab poured, excavation for junction manhole begun
Cement block wall construction begun
Plant operator hired by Strathmore (on their time)
to be present during final stages of construction
Microstrainers arrived (one damaged), precast T-beams
arrived (four damaged)
First microstrainer placed in plant
Second microstrainer placed in plant
Precast T-beams placed on roof
Roof and all other exterior portions of building completed
Microstrainer fabric secured to drum
Pump tests begun for plant startup
Plant startup, shakedown and punch list begun
_
-------
Work on the project did not stop however. Construction continued on
the interior piping, rag boiler line, sludge line, and 12-in. effluent
line, which was started on May 10. On May 20, permission was given to
clean up the substructure, forms, and everything affected by the flood.
By May 27, the forms had been cleaned and the first wall pour for the
below-grade structure was made. Work in this area continued smoothly,
and by June 15 all below-grade concrete work was completed. At this
date all construction was behind by nearly 1.5 months, but the September
completion date still appeared realistic if overtime were authorized.
On June 23 the first floor slab was poured and excavation for the junction
manhole was begun. The construction of the cement block walls began one
week later, and was completed by early July.
Strathmore, which was responsible for the operation of the plant, hired
an operator before the completion of the project. On July 13 the oper-
ator was assigned to the plant full time so that he could become familiar
with equipment and construction before the plant itself was placed in
operation.
On July 20, 1971, the microstrainers arrived at the site by truck after
shipment from England. The drums were unloaded and swung to a temporary
location by a 60-ton truck crane. It was not until July 22, after an
inspection of both drums, in which one was found slightly damaged, that
one of them was finally placed in its well in the plant. By July 26,
the second unit was placed and provisions were being made to pour the
concrete end wall around each microstrainer.
On July 20, the precast T-beams also arrived (their arrival was scheduled
to utilize the truck crane to lower these beams to the site as the micro-
strainers had been). When the beams arrived, it was noted that four of
them were cracked, some badly, at or near the neutral axis. The cracks
extended upward through the beam to the flanges. It was determined that
the beams should be unloaded, but they were not accepted pending an ex-
tensive period of testing and study by the manufacturer, engineer, client,
and a special consultant. It was not until after a sustained loading test
several days later that the beams were declared suitable, and on July 28
they were placed on the roof.
By August 24 the roof and all other exterior work were completed, and the
major effort of the contractor was concentrated in the building where the
interior piping was being installed. By September 7 most of the interior
29
-------
painting was finished, and it was possible to affix the microfabric
to the microstrainers. Prior to this time, it was felt that the fabric
would be clogged with paint, dust, etc. The work of attaching the
screens took over two weeks with at least two people working full time.
Pump tests were begun on October 5 when the wet wells, etc., were filled
with clear water. Finally, on October 14, the plant was filled with waste-
water and filtration began. Although the startup was about one month late,
most of this delay was attributable to problems beyond the contractor's
control. The plant shakedown and punch list items were then completed,
as would be expected in a project of this size.
TREATMENT PLANT COSTS
Table 5 lists the construction costs of this facility. Based on a total
cost of $689,000, and assuming an average life of the equipment and build-
ing of 25 years, the annual capital cost is $27,600. The two mills pro-
duce approximately 70 tons of paper per day, and operate a minimum of five
days per week, 50 weeks per year. Thus the capital cost of the treatment
plant is approximately $1.50 per ton. An equal amount should be allocated
for interest charges on financing the project at 8% interest.
Operating costs are discussed in Section IV.
30
-------
TABLE 5
FINAL CONSTRUCTION COSTS, JOINT INDUSTRIAL
WASTEWATER TREATMENT FACILITY
Exterior piping $ 54,500
Excavation, site work, building, and other
miscellaneous expenses 279,700
Equipment
Pumps, piping, valves 61,300
Microstrainers (two units) 120,000
Mixers 14,700
Chemical feeding and plant control equipment 37,000
Gates, hoists, and miscellaneous steel 19,000
Electrical 43,700
Subtotal $629,000
Engineering and contingencies (engineering
includes design, resident engineer, and services
during construction) 60,000
Total $689,900
31
-------
SECTION IV
OPERATION
TREATMENT PROCESS DESCRIPTION
Microstraining is a relatively simple method of solids-liquid separation,
which utilizes direct filtration of the wastewater through a very fine
screen (called a microfabric) mounted on the outer periphery of a cylin-
drical drum. Wastewater is introduced into the inside of the drum through
the open end, called the upstream end, and passes through the screen be-
cause of the hydraulic head differential across the screen. Suspended
material is trapped by the microfabric and deposited on the inside of '.:n-;
drum. To insure a continuous flow, the drum rotates constantly. Wh^n
deposited solids are brought to the top of the drum, two rows of backwash
spray heads spray process or city water through the microfabric and wash.
the solids mat into a V-shaped hopper mounted on the center axis of the
drum. Thus the microfabric, as it returns into the wastewater flow, is
clean and ready to filter more solids.
Two 10-ft-diameter by 10-ft-long Crane-Glenfield microstrainers are used
at the wastewater treatment plant. Microstrainer No. 1 has a 23-microu
(Mark 0) opening, 316 stainless steel microfabric, and No. 2 has a 35-
micron (Mark 1) microfabric. Both micros trainers are identical in all
other aspects of design and operation. Each microstrainer is sized to
filter 1.75 mgd of combined industrial wastewater from the two mills.
The filtered water, or effluent, fills the outside chamber of the drum
and flows over a weir to the effluent wet well. The weir maintains an
operating water level in and around the drum of about two-thirds the drum
depth. The solids washed off into the hopper (backwash sludge) flow by
gravity to the backwash sludge pumps. Fig. 10 is a cutaway view of a
typical microstrainer installation.
The most important criteria to be met. by the microstrainer are its hydraulL;
capacity and its solids removal efficiency. Factors affecting the hydrauij.,
capacity of a microstrainer are: (1) size of screen opening (in microns),
(2) head loss through the screen, (3) filtering area, and (4) character-
istics of the wastewater. To accommodate varying hydraulic conditions,
the plant microstrainers have a pneumatically operated automatic control
system that maintains a relatively constant head loss across the screens
(approximately 6 in.) by increasing or decreasing the filtering area per
unit time (i.e., the drum speed) according to the increase or decrease in
flow.
Filtering efficiency is affected primarily by the size of the microfabric.
and characteristics of the wastewater. However, other conditions such as
head loss and drum speed seem to be more significant in the treatment plant:
These are discussed later in this report.
32
-------
Drive
/— Washwater
sprays
Sealing bands
Microfabric
Fig. 10. Typical microstrainer
arrangement
-------
To optimize the removal of suspended solids, color, and turbidity,
chemical feeding and flocculation facilities were supplied as part of
the overall treatment system. Flocculation of the fine suspended matter
and filler material as well as the fibrous material creates large particles
that filter more easily, and creates a fibrous mat on the microfabric that
also becomes a filter aid. Chemical feeds at several locations in the
rapid and slow mix basin (as described in Section II) allow additional
flexibility of operation.
Operation of the microstrainers was designed for both parallel and series
flow sequences. Piping was installed to recirculate filtered effluent
to microstrainer No. 1 for further screening through the 23-micron screen,
or to treat the plant wastes completely when the unit is dewatered. Manual
override speed control was also provided, for flexibility of operation.
All features of this system were designed for maximum flexibility of
operation to determine the most effective and efficient means of util-
izing the process of microstraining in treating this type of wastewater.
TESTING METHODS AND PROCEDURES
Data collection from the treatment plant began at the onset of operation
in late November and early December 1971, but was sporadic because of
repeated mechanical problems. Sampling was stopped until the treatment
plant became operational. Data collection was resumed May 1, 1972, and
continued through October 17, 1972 (the completion of the studies by
COM). The same sampling and testing procedures were used throughout the
duration of the straight screening and chemical feed portions of the proj-
ect, except for periods of special studies and sampling programs.
Sampling
Composite samples (16-hour) were collected by hand hourly and composited
on an equal volume basis. Automatic samplers were tried but abandoned
because high concentrations of suspended solids clogged the sample lines.
Routine day shift sampling was accomplished by the treatment plant oper-
ator under the supervision of Camp Dresser & McKee personnel, who con-
ducted second-shift sampling. Manual 24-hour composites were not possible,
because of the unavailability of personnel during the night hours.
Samples were not composited on the basis of flow for two reasons: (1)
the flows from the two paper mills were so sporadic that compositing on
the basis of flow would not necessarily be representative; and (2) the
concentrations of solids, color, turbidity, and organic matter fluctuated
so greatly, and not necessarily in proportion to flow, that it would have
been impossible to catch every fluctuation in wastewater concentration.
34
-------
Samples were taken at the following points within the treatment plant:
1. Combined influent from the equalization basin
2. Filtered effluent from microstrainer No. 1
3. Filtered effluent from microstrainer No. 2
4. Combined effluent from both microstrainers plus any
bypassed flow taken at the effluent force main
5. Backwash sludge from the backwash sludge force main.
Combined influent and combined effluent samples were taken at sample
cocks at the laboratory sink. Effluent samples from each microstrainer
were taken with a ssrr.ple dipper at the effluent weirs. After collection,
samples were refrigerated at 40°F. until analysis was started.
Analytical Methods
All analyses of the composite samples were in accordance with Standard
Methods for the Examination of Water arid Wastewater, thirteenth edition,
1971, except for color analysis. Color, at best, is defined by arbitrary
standards and is difficult to monitor, particularly when there are so
many varied colors resulting from the papermaking processes. For the
purposes of this study, therefore, we based our color determinations of
both the influent and effluent on a dilution test recommended by the
Massachusetts Division of Water Pollution Control. The tests consisted
of determining the quantity of sample which had to be added to 1 liter
of tap water before color was noticeable.
Daily composite samples were analyzed and the results recorded on daily
operating forms. The following analyses were conducted on each sample:
?H
Turbidity
Settlesble solids
Total solids and total volatile solids
BOD (5 day, 20° C)
COD
Color
Aluminum
All analyses were conducted at the fully-equipped treatment plant laboratory
by Camp Dresser & McKee personnel assisted by the treatment plant operator.
35
-------
Jar Studies
Jar studies were conducted daily on grab samples of the influent to
determine the most effective and most efficient combination of chemi-
cals for the particular wastewater. Because the wastewater character-
istics changed rapidly, and because some flows were more difficult to
treat than others, on many days more than one jar test was required to
determine the chemical dosages necessary to meet the change. During
the straight screening phase (Phase II) of the project, many jar tests
were run to evaluate several different combinations of primary coagulant
and polymer.
Jar tests were run in series, six jars at a time, using a Phipps & Bird
six-unit paddle stirrer and 2-liter battery jars. Chemicals were added
into 1,500-ml samples in increasing dosages. Primary coagulants such
as alum and/or sodium aluminate were added prior to a 2-minute flash
mixing at 100 rpm, followed by a 10- to 20-minute slow mix at 15 to 20
rpm. Before the slow mix, increasing dosages of a polymer were added.
After completing the flocculation process, a portion of the flocculated
wastewater from each jar was filtered through a small hand microstrainer
unit supplied by the microstrainer manufacturer. This hand unit con-
sisted of a 3-in. diameter plastic column 6 in. high with a section of
Mark 1 (35-micron) microfabric enclosing the bottom end. A sample was
poured into this column and the volume of filtrate passed in 9 seconds
was noted. This method was used instead of settling the flocculated
sample, because it more closely duplicated the actual conditions ex-
pected in the full scale plant.
Evaluation of the most effective chemicals and dosages were based on
several factors: (1) size and characteristics of the floe, (2) filter-
ability through the screen in comparison with the raw wastewater filter-
ability, and (3) clarity of the filtered effluent. The chemical require-
ments determined by the jar tests were then utilized in the full scale
wastewater flow during the chemical feed portion (Phase III) of the study.
Records
Complete and accurate records were collected at the treatment plant using
the standard forms shown in Fig. 11. These forms were prepared for the
operating manual and were designed to be used not only for laboratory
analysis reporting but also for recording complete backup information
pertaining to the wastewater streams entering the plant each day (i.e.,
individual hydraulic flows, turbidity, and grades of paper made at each
mill). In addition, complete daily records of the chemicals and dosages
fed were kept for future reference and also for cost analysis purposes.
36
-------
Besides the forms shown in Fig. 11, individual daily work sheets were
provided for recording analyses of solids, BOD, and COD. A monthly
form for reporting analyses and operating conditions at the treatment
plant was provided (after being accepted by the state) for the required
monthly reports to the State of Massachusetts Division of Water Pollu-
tion Control.
A bound operating log was maintained to record all operating conditions
and related problems during the first year of operation.
OPERATING PROBLEMS
The Strathmore-Esleeck wastewater treatment plant is an original facility
for treatment of paper mill Whitewaters, and is the first plant in the
United States that uses microstraining with coagulants. Therefore, one
of the objectives of the post-construction studies was to report on the
performance of the related equipment and any unusual operating problems.
The following paragraphs describe some of the problems encountered dur-
ing the first year of operation.
Minor problems centered on the microstrainer control system and pump
plugging. There were, however, major problems concerning the overall
plant flows, the microstrainer seals and capacity, the sludge pumps,
the plant effluent pump controls, and the automatic bypass system. Some
of these are not yet fully resolved.
Plant Flows
Incoming wastewater flows have been consistently higher than anticipated,
as shown in Fig. 12, which illustrates the total plant flows pumped to
the canal. This figure indicates that the design average flow was 1.74
mgd (1,200 gpm), but the actual pumped flow to the canal has been closer
to 2.0 mgd (1,400 gpm).
That portion of the flow from the Esleeck mill has been approximately as
expected -- the average Whitewater and washerwater flows have been about
1.00 mgd. However, during the early stages of operation, the Whitewater
flows were sporadic. Whitewater is pumped from Esleeck (the only pumped
flow to the treatment plant) by a duplex pump station with two pumps
rated at 750 gpm each. Initially, these pumps worked from a level con-
trol in the wet well. When this control system was used, the flow to
the treatment facility fluctuated drastically, upsetting operations. To
correct this condition, a flow throttling valve was installed on the dis-
charge line, which resulted in a relatively stable pumping rate and al-
leviated much of the flow surging. Fig. 13 illustrates the actual Esleeck
Whitewater and washerwater flows versus their design values.
37
-------
DATE
OPERATOR
STRATH MORE WHITEWATER
STRATHMORE WIREWATER
ESLEECK WHITEWATER
ESLEECK WASHERWATER
TOTAL FLOW
BACKWASH SLUDGE
FLOWS (MGD)
MIN.
AVG.
MAX.
TREATED
DIVERTED
PUMPED
00
STRATHMORE
1 GRADE
COLOR
PAPER GRADE
MACH.*7
MACH. *8
MACH. #9
ESLEECK
GRADE
COLOR
MACH.*I
PRIMARY COAGULANT
( )
POLYMER
( ]
CHEMICALS LISTED
FEED RATE
igph)
DOSAGE
(mfl/l )
(Ibs/doy)
STRATHMORE WIREWATER
ESLEECK WASHERWATER
COMBINED INFLUENT
EFFLUENT
TURBIDITY UTU)
MIN.
AVG.
MAX.
UNIT *l
UNIT *2
MICROSTRAINERS
DRUM SPO
AVG.(fpm)
DIFF. HD.
AVG.lin)
EFFLUENT SOLIDS
SETTL.(ml/IJ
SUSP(mg/l)
Fig. 11. Daily operating and analysis report
(daily operating report and data sheet)
-------
DATE
OPERATOR.
ANALYSIS (1) (2)
TEMPERATURE fF)
PH
ACIDITY
TURBIDITY (JTU)
Q
Ii
0
tO
Settleoble(mlVL)
Totol
Total Volatile
Suspended
Volatile
Suspended
BOD 5DAY-20°C
COD
COLOR
ALUMINUM
TITANIUM
COMBINED
INFLUENT
MICROSTRAINER
* 1 EFF.
*2 EFF.
PLANT
EFFLUENT
%
REMOVAL
BACKWASH
SLUDGE
(I) See Table _6_ for frequency of testing
(2) All results in mg/L unless otherwise noted
Fig. 11. Daily operating and analysis report
(laboratory analysis report)
39
-------
3*00-,
£ M«H
oo
.g tOOO-
0)
o.
I»OO
cd
4J IZOO
•00
01
u
§
C
4)
Design
average flow h
t-Esleeck shutdown
Low points
indicate weekend flow
Strathmore shutdown
4/2t 4/30 S/O* 5/l»
S/t4 t/OI 6/0» C/IT »/28 7/03 7/0* 7/1* 7/Z» »AJt */!•
Date of Observation, 1972
B/ZI »X>7
9/1 T
n/D7 10/17
Fig. 12. Plant effluent pumped to canal
-------
1000-I
800-
I
no
. 600-
o
f*
-------
Prior to design, estimates were made of the total wastewater flows
expected (after providing for recirculation of certain wastewaters).
These flows were confirmed by Strathmore. However, during actual oper-
ation it was found that those values were exceeded at the plant.
Flows from the Strathmore mill fluctuated even more widely than those
from Esleeck during the early stages of the project. There were several
reasons for this: (1) flows were discharged by gravity to the treatment
plant, and were thus more erratic; (2) three paper machines were contrib-
uting flows, and corresponding sequences of grade changes and washups of
two or more machines resulted in peak surges in flow; and (3) at times,
only limited amounts of Whitewater were recirculated at one or more
machines.
Flows from the wirewater pits were below the expected average by some 70
percent, or about 60 gpm. This was due in part to the incorrect imple-
mentation of the wirewater collection system, which resulted in the wire
pit flow going to the Whitewater storage tanks instead of directly to
the treatment plant. The mill is now taking the necessary action to
correct this problem.
Whitewater flows were nearly 700 gpm, which was well above the predicted
average. This was partially the result of inefficient use of the white-
water recirculation system -- primarily that portion relating to white-
water reuse in the hydropulper operation. The high Whitewater flows
varied in duration. Short-term peak flows caused severe hydraulic loads
on the plant and micros trainers, thus reducing plant efficiency. These
peak flows were generally caused by chest dumps and washups of the paper
machines. Proper scheduling of machine washups and more consideration
of the effect of quick dumping of chests has generally relieved the con-
dition. However, the mill is constantly monitoring this problem.
Whitewater flows from the mill still exceed the estimates used for design.
Originally it was anticipated that Whitewater flows could be reduced to
almost 300 gpm. The mill is in the process of further reusing Whitewater
to reduce the flow to more nearly conform to the design parameters. Fig. 14
shows the Whitewater and wirewater flows for the study period.
Micro strainers
No significant problems of installation of the micros trainers were en-
countered, although one microstrainer was slightly damaged in transit.
The only serious problems encountered with the microstrainers were foul-
ing of the drum bearings with paper stock and drum seal leakage.
42
-------
1600-
to
Whitewater
o Wirewater
Whitewater
design average flowy
Not correcl
flow meter
in error
5/20
6/9
6/29 7/19 8/8 8/28
Date of Observation, 1972
9/1 r
10/7
10/27
Fig. 14. Strathmore wastewater flows
-------
Bearings -- Three water-lubricated bearings were provided on each drum.
The bearings on each end of the drum were supplied directly from a source
of fresh water and in turn supplied the middle bearing. When the bearings
became packed with paper stock, causing an eventual stoppage of drum ro-
tation, it was assumed that this was caused by the high solids content
of the waste and low pressure in the bearing water feed line.
To alleviate the problem eacft of the three bearings on each microstrainer
was connected to its own separate water supply and the supply lines en-
larged. The modification immediately relieved the problem. A rotameter
was installed on each flow line, so that a drop in the preselected flow
rate at the bearing water line would automatically warn the operator to
take appropriate action.
Prior to removing the paper stock from the bearings and installing ad-
ditional water supply lines to them, it was necessary to raise the micro-
strainer assembly with jacks. As a result of an improperly placed jack,
one of the microstrainers was severely damaged. One end of the drum was
bent so that seal leakage, discussed below, increased for an extended
period of time until the damage was repaired.
Sea_ls_ -- Leakage of the bands sealing the fixed and rotating portions of
the microstrainer (Fig. 15) was a problem far more difficult to resolve.
Failure of a seal allowed leakage of wastewater to bypass the microfabric
and pass untreated to the plant effluent. Seal failure was a result of
the accumulation of fibrous suspended material (paper stock) under the
unattached edge of the seal. Small particles of fibrous material became
attached to the seal and, from the rotation of the drum, these rolled in-
to small balls which continued to grow in size under the seal. Therefore,
a seal had to be designed to insure that fibers would not be passed under
it.
Seven different seals were tried on one or both drums, including the orig-
inal equipment seals furnished with the units. In all, nine different
combinations and/or seal arrangements were tried. Table 6 lists the type
of seal employed, the length of service, and average suspended solids re-
moval. The table is keyed to Fig. 15, which illustrates the various seals.
As indicated in Table 6, only varying degrees of success were experienced
with most of the seals tried.
The flat seals tried (such as arrangements a and b, Fig. 15) failed almost
immediately. The most effective seals to date were the cove molding seal
(arrangement d), the knife edge seal (arrangement f). The cove molding
seals worked well except when there was a large percentage of cotton fiber
in the wastewater. Cotton rolls much faster and into larger balls than
wood fiber. Once these seals failed, the treatment plant had to be shut
44
-------
TABLE 6
MICROSTRAINER SEALS UTILIZED
Date of Service
Design
Material
11/8-12/13/71
(drums Nos. 1 and 2)
12/31/71-2/18/72
(drum No. 1)
12/14/71-2/11/72
(drum No. 2)
2/18/-3/2/72
(drum No. 2)
3/4-4/28/72
(drum No. 1)
4/29-5/15/72
(drum No. 1)
5/16-7/3/72
(drum No. 1)
7/5-9/12/72
(both drums)
9/13-10/20/72
(drum No. 1)
9/13/72 to date
(drum No. 2)
10/21/72-11/15/72
Standard Glenfield & Kennedy Nylon outside, felt
seals and brass inside
Flat plastic
Knife edge
Cove molding curved edge
Double overlapping
Reinstalled cover molding
seals
Dumbbell type cylinder mold
Plastic Orkot
Plastic
Rubber
60-durometer rubber
Rubber
PVC
Drum shutdown -- bent drum
caused excessive seal leakage
Reinstalled knife edge seals Plastic
Flat plastic feather edge
Combination
Combination with flushing water
Plastic - brass
45
-------
TABLE 6 (cont.)
MICROSTRAINER SEALS UTILIZED
Key
S.S Removal,
percentage
Comments
15a
15b
15c
15d
15e
15d
15f
15c
15g
15h
151
40-45
36
No data
70-75
63
68
70-75
(with chemical
feed)
85-90
(with chemical
feed)
40-50
Seals plugged and leaked immediately
As above
Seals failed immediately because of
out-of-round drum, downstream end
(see text)
Seals worked well, except for down-
stream end drum No. 1
Failed because of high friction be-
tween two rubber seals
Worked well except for periods of
high cotton fiber in waste (see text)
Worked well except for periods of
rapidly changing head loss (down-
stream end No. 1 drum leaked ex-
cessively)
Worked well
Failed because of solids buildup under
seal, causing seal to pull away from
drums
Not tried, see above
46
-------
down to clean them. The dumbbell seal worked well except for periods
of rapidly changing head loss, when the seals tended to leak at the
newly established water level. Although this seal appeared to be sat-
isfactory, continuing research was being carried out on the knife edge
seal similar to arrangement c. When this seal was originally installed,
as shown in Table 6, it was put on the downstream end of microstrainer
No, 1. At that time, it was not known that this microstrainer (which
was the one damaged in shipment) was so out-of-round that it caused the
almost immediate failure of all seals installed. Because this seal was
not given a reasonable trial, it was placed on microstrainer No. 2 in
the middle of September. This seal held tight, and did not fail immed-
iately or stop the drum.
The plastic feather edge and brass combination seal tried on microstrainer
No. 1 failed due to solids buildup under the seal. This buildup pulled
the plastic seal from the drum, and stretched it so that it did not re-
turn to the drum surface after cleaning. For this reason, use of the
plastic seal in combination with the outside dumbbell seal (Fig. 1.5, h
and i) was abandoned.
Controls -- As previously described, the speed of the microstrainer
drums is controlled by a pneumatic automatic and/or manual system. This
system worked well, except for a few minor problems. A constantly re-
curring problem was the plugging with paper st ;. .k of the screen on the
air bubbler tubes, used to record the head differential. This created an
incorrect head differential reading and consequently affected the oper-
ation of the automatic control system. By raising the elevation of these
tubes and providing a blowdown system to periodically clear these lines,
the problem was eliminated.
Another problem was the overall adjustment of the automatic control system.
To function properly, this system had to maintain a fine adjustment at
three locations: (1) the hydraulic drive, (2) the differential air to cur-
rent converter, and (3) the differential controller. Incorrect adjustment
at any one of these locations caused inefficient operation of the micro-
strainer control system. For the first few months of operation, the
microstrainer controls were always out of adjustment, even with constant
maintenance by the manufacturer's service representative and Strathnsore
instrumentation personnel. Eventually this system became less trouble-
some, but to date minor adjustments are still needed.
Drives -- The Carter "F" type hydraulic variable speed drives that were
supplied with the micros trainers in general performed well with proper
maintenance. On two separate occasions, once on each drum, the hydraulic
drives failed. Failure was directly related to torque overloads on the
47
-------
Outside of drum
Nylon band
Rotating
end
Attaching bolt
Inside of drum
Flow
Outside of drum
(a) Original Equipment Design
No outside seal
Rotating
end
Attaching bolt
Inside of drum
Plastic
Copper band
(b) Inside Band Only
Flow
Outside of drum
No outside seal
Rotating end
Attaching bolt
Inside of drum
Plastic knife edge seal
Flow
(c) Knife Edge Seal
Fig. 15. Microstrainer sealing arrangements
48
-------
Outside of drum
No outside seal
Rotating end
Attaching bolts
Cove molding
(as used in kitchens)
Flow
Inside of drum
Outside of drum
(d) Cove Molding
No outside seal
Rotating end
60-durometer rubber
Attaching bolts-
Inside of drum
(e) Double Overlapping
Flow
Outside of drum
Fixed
End
Dumbbell tightened by
wire rope through core
of each dumbbell
Rotating
end
No inside seal
Inside of drum
Flow
(f) Dumbbell Seal
Fig. 15. Microstrainer sealing arrangements (Cont.)
49
-------
Outside of drum
No outside seal
Fixed
end
Attaching bolt
Inside of drum
Outside of drum
Rotating end
-— Plastic
Brass backing strip
(g) Plastic-Brass
Dumbbell tightened by
wire rope through core
of each dumbbell
Rotating
end
Flow
Attaching bolt
Inside of drum
Outside of drum
Plastic
Brass backing strip
Flow
(h) Combination
Dumbbell tightened by
wire rope through core
of each dumbbell
Rotating
end
Attaching bolt
Flushing water
Inside of drum
Brass backing strip
Flow
(i) Combination Plus Flushing
Fig. 15. Microstrainer sealing arrangements (Cont.)
50
-------
drive when the microstrainer drum bearings became jammed with paper stock,
as mentioned above. This resulted in a braking action on the drum. Once
the drum bearing problems were corrected and the hydraulic drives were
rebuilt, the condition did no; recur. To insure that the drives would
not become overloaded to the point of damage again, indicating hydraulic
pressure gages with alarm devices and a preset pressure adjustment were
installed. The alarm signal was added to the plant alarm system located
in the operating control center.
Microfabric Cleaning -- One of the most important factors in the efficient
operation of the micros trainers and their hydraulic capacity is the con-
dition of the microfabric. Because the openings in the microfabric are
minute, and because the microfabric is continuously filtering wastewater
containing fine suspended material, it is imperative that it be kept
clean and unplugged. This is accomplished by the backwash sprays that
spray through the microfabric continuously as it reaches the high point
in the drum.
One of the most troublesome problems with the micros trainers, aside from
seal leakage, was the continual plugging of the microfabric. This was
due in part to the very fine particles of titanium dioxide and filler
material and in part to the backwash pressure available. (It was ini-
tially proposed by Strathmore that sufficient water pressure would be
available.) The washwater sprays did not clean the microfabric properly
when the drum speed was greater than 100 feet per minute (drum speeds
range from 25 to 150 fpm), because of a deflection of the washwater sprays
at the higher drum speeds.
As a result of the microfabric plugging problem, frequent washups with a
strong cleaning solution had to be employed. The manufacturer recommended
that the microfabric be cleaned by spraying phosphoric acid solution on
the screen as it rotated slowly, and that the microstrainer chamber be
completely drained during the process. This soon became prohibitive be-
cause of the complete disruption of the treatment process. Frequency of
cleanings during the first year of operation averaged one per week and
sometimes more often on microstrainer No. 1, which has the finer mesh
microfabric. A highly alkaline foam cleaner was substituted for the acid
and was applied without dewatering the chamber. This allowed the screens
to be cleaned without interrupting plant flow longer than 30 minutes.
Two alternative cleaning methods were tried during the latter part of the
first year's operation. These were: (1) daily injection of a liquid
cleaner into the washwater spray head intermittently, and (2) intermittent
cleaning of the microfabric with high pressure steam nozzles after the
backwash sprays. It is believed thul either procedure will allow the micro-
fabric to be cleaned more regularly, perhaps daily. This will increase
the efficiency of the straining process, including the hydraulic capacity
of the microstrainers, lengthen the life of the microfabric, and reduce
the cost of cleaning.
51
-------
Sludge Pumps
The sludge pumps originally selected for the plant were Mega tor snore
stainless steel fitted pumps. The pumps were recommended by Glenfield
& Kennedy, who had employed them previously for pumping micros trainer
sludge. The manufacturer stated that these pumps would pump soft sus-
pended solids and grit.
Each pump, rated at approximately 120 gpm at 100 ft total dynamic head
(TDK), was capable of discharging all the microstrainer sludge. The
design of these pu.mps is such that they have three shoes mounted on an
eccentric cam shaft and riding on a port plate. Excessive wear occurred
on the shaft, shoes, and plate. Finally, after 1,500 to 1,700 hours
of service, both pumps failed. The manufacturer claimed that failure
was caused by excessive solids and low pH (4 to 5) which attacked the
stainless steel.
After replacing the shaft and cam assembly, shoes, and port plate, one
pump was again placed in service. After approximately 150 hours of
operation, this pump again failed from excessive wear.
The Megator pumps were then replaced with Warren Solids -- Master pumps,
which are open end impeller pumps. These are working satisfactorily,
with no sign of wear.
Effluent Pumps
The three effluent pumps installed in the treatment plant (Warren paper
stock pumps) are each rated at 1,200 gpm at 112 TDK. Thus any two pumps
are capable of pumping the peak plant flow, with the third pump as stand-
by. To equalize pumping and minimize the size of the required wet well,
variable speed drives were chosen for the pumps. Magnetic drive units
and ampli-cycle drive units were considered. The latter were chosen as
the best compromise, considering initial cost, needed floor space, and
cooling requirements. Although the ampli-cycle unit has a slightly
higher initial capital cost than the magnetic drive, it requires less
floor space, since It does not fit in the drive train between the motor
and pump.
The ampli-cycle unit was purchased from Electric Machinery Company, and
was one of the first of its kind used in such a facility. The unit is
unusual; only one drive is required to serve all three pumps. As one
pump reaches maximum speed, it is automatically transferred to the bus
and the second pump is then activated to pump at a variable speed. Fail-
ure of the ampli-cycle unit does not put all pumps out of operation, be-
cause an alternative system of automatic on/off control, operated by a
bubbler system which indicates a wet well level, can be used.
52
-------
Problems were encountered with the ampli-cycle system. The problems
could be attributed essentially to the initial adjustment of the unit
and failure of electrical components. Since this was an original unit,
field adjustments were required several times during startup. If the
ampli-cycle system does fail, the system can be switched to the auto-
matic bubbler operation which only operates one pump. Thus, if a
second pump is required, an operator may be needed 24 hours a day to
operate the second pump to accommodate the random high flows.
The continued smooth operation of this system was marred by the in-
stability of certain components in the unit; in particular, some re-
sistors were particularly sensitive to heat. These resistors failed
at critical times, and it was not until after several failures that
the problem was pinpointed.
An apparent problem also developed in meeting the rated capacity of
the three effluent pumps. During several pump tests, design conditions
could not be met. Repeated high flows made it sometimes impossible to
handle the flow with two pumps. This condition was finally traced to
a partially plugged wastewater distributor located in the power canal.
After cleaning this distributor during plant shutdown in July, the
problem was corrected. A final pump test indicated that the pumps
were capable of meeting their requirements. Steps have been taken
to monitor the distributor head losses continuously to alleviate
future pumping difficulties.
Sampling and Turbidity System
The sampling systems for the automatic turbidity monitoring systems
(Strathmore wirewater-Esleeck washerwater and plant influent-effluent)
were designed to operate continuously, on an alternating basis, with
two waste streams tributary to one wall-mounted Hach model 1800 tur-
bidimeter. Automatic solenoid diverting valves controlled by a timer
for each sample line divert one waste stream into the turbidimeter
and one to waste. Individual sample pumps continuously deliver waste-
water from three of the four sources to the diverting valve system,
while the plant effluent is delivered under pressure from the effluent
force main.
A problem that was experienced with these sample systems was plugging
of the pumps, sample lines, and solenoid valves, as well as the tur-
bidimeter measuring chamber, with suspended solids (paper stock) in
the wastewater.
Replacing the sample pumps with pumps of a different design and in-
stalling basket strainers in the lines solved the problem of frequent
plugging. However, frequent cleaning of the microstrainers is required,
53
-------
Excessive wear of the impellers of the new pumps appears to be due
to running the pumps dry during periods when the basket strainers are
plugged.
Replacing the solenoid valves with specially designed three-way valves
and enlarging the drains from the turbidimeter corrected the failure
of the throttling system at the turbidimeters. Solids continued to
accumulate in the overflow chamber of the turbidimeter, but a regular
program of cleaning and flushing the chambers solved the problem and
permitted uninterrupted service.
Alarm System
As previously described, a comprehensive alarm system was provided in
the operating area, with branch alarms to each mill, to detect any mal-
function of equipment or change in operation that would cause damage
or interrupt the treatment process. During the first months of oper-
ation, it became evident that further alarms were needed to insure ef-
ficient and regular operation. Initially, only the first seven alarms
listed in Table 3 were installed; the remaining five were added in late
October. Each one of the last five alarms was added as a result of
some breakdown or malfunction experienced with that system, or repeated
conditions causing interruption in treatment. Fig. 9 showed the alarm
panel with the selector switch.
Since the installation of this last set of alarms, the treatment plant
has become more flexible in operation, and closer monitoring of the
major mechanical equipment involved is now possible.
Chemical Feed System
A major factor in efficient operation of the treatment plant is the
ability to add the proper chemicals at various points in the plant
basins to flocculate the wastewater effectively. The basis of this
system design was to feed a primary coagulant and/or coagulant aid
(polymer) at the various points, as shown in Figs. 7 and 8. In gen-
eral, this system proved to be very effective in its flexibility of
operation and accuracy of control.
One problem which developed with the primary coagulant feed system
(only sodium aluminate at the beginning) was the frequent plugging
which occurred in the small plastic distribution lines leading from
the rotameter board to and including the diffusers located in the
sludge gate openings. In addition, the 1-in. PVC transfer line and
transfer pump (from the holding tanks in the Strathmore mill to the
50-gal. day service tank) were continually plugged because of the high
strength and viscosity of the 43 percent sodium aluminate used. These
54
-------
plugging problems were solved by: (1) adding flushing connections at
the feed end of the 1-in. transfer line and on the suction side of the
metering pump, (2) enlarging the diffuser holes to 0.25± in., and (3)
replacing the transfer pump by a motorized ball valve controlled by
level probes in the day service tank, after it was determined that
gravity flow from the holding tanks would be sufficient to supply the
feed requirements.
The polymer feed system, as designed and built, worked well with no
major problems. The system provided facilities for making up polymer
solutions from either dry or liquid polymer bases; and had a holding
tank and feeding capacity, via a two-pump system, for a wide range of
feed concentrations.
One major change in the chemical feed system was made after repeated
jar test data indicated that alum in place of or in conjunction with
sodium aluminate was needed. It was determined that the primary coag-
ulant would be alum, and sodium aluminate would be used only when waste-
water was highly colored and turbid. Therefore, it became necessary to
provide holding and feeding facilities for liquid alum. Because the
design of the existing facilities was flexible, it was possible to de-
velop a third system for feeding alum without any major additional con-
struction. This was accomplished by:
1. Converting the two 2,000-gal. tanks from sodium
aluminate to alum storage
2. Converting the transfer piping day service tank,
metering pump, splitting and diffuser system for
primary use of alum
3. Converting the 1,000-gal. polymer day service
holding tank to a sodium aluminate storage tank
4. Converting one of the two polymer feed metering
pumps to a sodium aluminate metering pump
5. Rearranging the necessary suction and discharge
piping from the polymer makeup tank and the new
sodium aluminate holding tank to the respective
metering pumps, and from the pumps to the rota-
meter splitting board
6. Utilizing the polymer feed lines connected to the
rapid mix basins to convey sodium aluminate. This
would eliminate the possible application of poly-
mer to the rc'id mix basins.
55
-------
SECTION V
DISCUSSION
As stated in the introduction, the overall objectives of this study, as
far as the treatment process is concerned, were:
1. To compare the efficiency of the two microstrainers in
screening wastewater without prior chemical coagulation
2. To compare the efficiency of the two microstrainers in
screening wastewater when primary coagulant aids were
added prior to screening
3. To compare the overall efficiency of the micros training
process with and without the above-mentioned coagulants
4. To investigate the capacity of each microstrainer and
compare the two.
This section describes and compares the Phase II and Phase III studies.
Special studies and alternatives are discussed according to their rel-
evance to the microstrainers or microstraining process. Also included
are the results of the backwash sludge dewatering studies and a de-
scription of full scale units that would be needed to handle the sludge
on site.
PHASE II STUDIES
Phase II comprised approximately three months of straight screening
studies (parallel screening of the wastewater through the microstrainers).
During this period, however, microstrainer performance was hampered by
the leaking sealing bands. The data presented will therefore reflect
days when efficiency was poor because of seal leakage. Unfortunately,
the seals leaked sporadically throughout the study, so there are no
distinct periods of good or bad seal performance. Only 20 percent of
the data on microstrainer No. 1 and 40 percent of the data on micro-
strainer No. 2 are representative of no leakage.
Suspended Solids Removal
Suspended solids removals for this period generally ranged from 50 to 85
percent for microstrainer No. 1 and 60 to 87 percent for microstrainer
No. 2. Fig. 16 shows the suspended solids removals for Phase II. Major
seal leakage on microstrainer No. 1 is indicated by the removals near or
less than 50 percent. An average of selected data on days when there was
little or no seal leakage shows removals of 84 percent and 82 percent for
microstrainers No. 1 and No. 2, respectively.
56
-------
1000
ttf
0)
60
cfl
CJ
l-i
<1>
eu
4/SO 9/10 9/2O 8/30 C/O» «/!• 6/29 7/0» 7/1*
100.0
4/30
9/10
9/20 9/30
«A>»
•/!•
7/O9 7/H
7/Z»
Date of Observation, 1972
Fig. 16. Phase II suspended solids removal
57
-------
Fig. 17 shows the frequency distribution of suspended solids removal for
this phase. These curves indicate that 50 percent of the time micro-
strainers No. 1 and No. 2 had 73 percent or better suspended solids re-
moval; 75 percent of the time the removal for microstrainer No. 1 was 60
percent or better, and for microstrainer No. 2 was 62 percent or better.
Turbidity Removal
Turbidity removals during Phase II fluctuated greatly, partly because of
the wide range of influent turbidity (generally between 100 and 3,000
JTU) of the wastewaters. Most of the time, however, low turbidity re-
movals were caused by the passing of very fine suspended titanium dioxide
(10-to 20-micron particle) through the screens and also through the seal
leaks.*
Microstrainer No. 1 had turbidity removals ranging from 10 percent to 90
percent, with an average of about 40 percent. Microstrainer No. 2 had a
similar range, with an average of 35 percent. During days or no seal
leakage, microstrainer No. 1 had an average turbidity removal of 51 per-
cent and microstrainer No. 2 had a removal of 50 percent. Fig. 18 illus-
trates turbidity removals for this period.
The frequency distribution of turbidity removal (Fig. 19) indicates that
microstrainers No. 1 and No. 2 averaged 35 percent and 38 percent or
better removal 50 percent of the time, respectively, and 47 percent and
52 percent or better 25 percent of the time, respectively.
Color Removal
Color removal during Phase II also depended on the characteristics of the
incoming wastewater. Color changes in the wastewater varied in frequency
from once every two hours to once every several days, and at least 50 dif-
ferent color combinations were experienced. The intensity and nature of
the colors varied from color that was attached to the suspended solids to
completely soluble colors that could not be removed with the suspended
solids.
Individual microstrainer effluent color determinations did not begin until
halfway through Phase II; thereafter, ranges of removal varied from 20 to
75 percent for microstrainer No. 1 and 10 to 65 percent for microstrainer
No. 2. Average removals were 45 and 30 percent for microstrainers No. 1
and No. 2, respectively.
* Titanium dioxide is used as a filler material in the paper making oper-
ation to increase opacity and brightness.
58
-------
100.0
U1
o Microstrainer No. 1
Microstrainer No. 2
30 40 SO 60 70
Suspended Solids, Percentage Removed
Fig. 17. Phase II Frequency distribution of suspended solids removal
-------
100.0
OS
0)
60
a)
4J
0)
u
OjO
04/jo oSTTo OMOos/so otvb*
100.0
0.00, „ . . . , . .
O4/5O 09/10 OVtO 08/30 O*A» 06/1* Ot/29 07/0» O7/I9 OT/M
Date of Observation, 1972
Fig. 18. Phase II Turbidity removal
60
-------
100.0
oMlcrostrainer No. 1
+ Microstrainer No. 2
10.0
20.0
so o
•o.o
30.0 40.0 50.0 ftO.O 70.0
Turbidity, Percentage Removed
Fig. 19. Phase II frequency distribution of turbidity removal
100.0
-------
Color removal was not expected to be as high during the Phase II studies
as during the Phase III studies, when preliminary coagulants would aid
in removal of colloidal color particles. Fig. 20 illustrates Phase II
color removal, and Fig. 21 gives the frequency distribution of color
removal. Fig. 21 shows that tnicrostrainer No. 1 had 41 percent or
greater removal 50 percent of the time and 57 percent or greater re-
moval 25 percent of the time. Microstrainer No. 2 had 32 and 46 per-
cent removals 50 and 25 percent of the time, respectively.
BOD Removal
BOD removals during this study were not expected to be as high as the
suspended solids and turbidity removals, because much of the BOD was
caused by the starch used in the sizing process in the paper mills.
However, the amount of this soluble BOD was considered to be so low
that an acceptable reduction could be attained with the removal of the
suspended solids. It was expected that BOD removal would be proportional
to the suspended solids removal.
BOD removals during Phase II were generally low, as shown by Fig. 22,
partly because of seal leakage (especially on microstrainer No. 1).
However, a general trend toward lower removal efficiencies can be seen
as this phase progressed, and this is not due to lower suspended solids
removal efficiencies. Tests on the influent showed a high percentage
of soluble BOD. The studies conducted to locate the source of this
soluble BOD are discussed in more detail later in this section.
BOD removals averaged 50 percent for microstrainer No. 1 and 55 to 60
percent for microstrainer No. 2. The frequency distribution of BOD re-
movals is shown on Fig. 23. Both microstrainers showed 42 percent or
better removal 50 percent of the time and only 52 percent or better 25
percent of the time.
Microstrainer Comparisons
We expected that the smaller screen (23 microns) would be superior in all
areas of removal to the larger screen (35 microns). However, a comparison
of the frequency distribution curves (Figs. 17, 19, 21, and 23) reveals
that microstrainer No. 1 demonstrated only slightly higher removal of
turbidity and color than microstrainer No. 2, and that suspended solids
and BOD removal efficiencies are nearly the same for each. This was
caused partly by the considerable seal leakage on microstrainer No. 1
during this phase. As discussed previously, average suspended solids
removals on days of little or no seal leakage were 84 percent for micro-
strainer No. 1 and 82 percent for microstrainer No. 2. This is reason-
able, if the nature of the suspended solids to be removed is considered.
62
-------
78.0
L| 600
I
CO
2
U
cd
S,
2
§
§
&
4/SO 5/10 5/tO 3/JO t/0* I/I* t/t» 7/0* 7/lt
r/tt
•7.5
74.0
t-l
It S
.
Q
M 800
1-1
cd
M S7.»
4J
09
O
O tlO
*ri
33
12 S
0.0
4/
SO B/
10 B/
to »v
'SO «/
-A
»
L
Ir
'0» I/
^
V
1* 1
\
^t* 7)
/
/
j
w
p 1
'0* 7
^
\\
X
fit 7/
Date of Observation, 1972
Fig. 20. Phase II color removal
63
-------
100.0
43
Wl
87. S
n
oo
3
a
T3.0
62.3
30.0
97.3
at
DO
«0
g
o
M
23.0
12.5
0.0
I I \ L
J L
Legend
0 Microstrainer No. 1
+Microstrainer No. 2
0.0 8.0 16.0
240 32.0 4O.O 4«.0 56.0
Color, Percentage Removed
Fig. 21. Phase II frequency distribution of color removal
64-0 72.0 80-0
-------
'OO.O
CM
£
H
3
2
*J
CO
2
u
1
OJ
g
U
M
80.0
ST.5
4/JO 9/10 S/tO e/SO CAM I/I* «/2* T/O* 7/19 7/2*
75.G
0)
2
4J
CD
s
U
SO.O
ST.*
ts.o
0.0
4/30 9/10 8/20 a/50 I/O* «/!» a/2* T/O* 7/1* 7/2*
Date of Observation, 1972
Fig. 22. Phase II BOD removal
65
-------
100.0
87.5
2 7SO
IB
62.5
I
50.0
O 37.S
Q>
60
flj
4J
8 25.0
S
PM
12.5
0.0
10
Legend
o Microstrainer No. 1
+ Microstrainer No. 2
20
30
40
SO
60
80
90
IOO
BOD, Percentage Removed
Fig. 23. Phase II frequency distribution of BOD removal
-------
Straight micros training of the wastewater removed only the large and
medium sized suspended solids — the fines and filler materials passed
through the screen. Because of the size of these fines and filler nr.at-
erials, the 23-micron screen did not retain a significantly higher amount
than the 35-micron screen. Phase III, the chemical feed period of this
study, illustrates that coagulation assists the removal of more of these
fine suspended solids.
PHASE III STUDIES
Phase III comprised the chemical feed studies. Coagulants and coagulant
aids were added to the influent to increase removals of suspended solids,
turbidity, BOD, and color. The primary coagulant used, aluminum sulfate
and/or sodium aluminate, increased the particle size and density for more
efficient removal by microstraining. Cationic polyeletrolytes, the sec-
ondary coagulants, were used primarily to strengthen the coagulated solids
produced by the primary coagulant.
The percentage removal efficiency data are highly variable, because of
continuing seal leakage. In the following discussion, suspended solids
removals less than 50 percent indicate very high leakage around the drum
seals.
Suspended Solids Removal
The suspended solids removal for microstrainer No. 1 ranged from 89 per-
cent to 10 percent with an average of 68 percent. For microstrainer
No. 2, the data show two distinct averages. Initially, an outside dumb-
bell drum seal was used, and during this period the suspended solids re-
moval averaged 68 percent with a range of 91 to 24 percent. Following
this, an inside knife-edge plastic drum seal was used. During this
period the average removal was 82 percent, with a maximum and minimum
values of 95 percent and 67 percent, respectively. These data are shown
in Fig. 24.
The relative performance of each drum in removing suspended solids is
shown in Fig. 25. Those curves, based on all the data collected, could
lead to the false conclusion that increased suspended solids removal is
proportional to increased screen opening. Actually, there was serious
seal leakage in microstrainer No. 1, and this effectively shifts the
curve to the left. In addition, during much of the period leakage was
so bad that operation of microstrainer No. 1 was impossible. It is ex-
pected that the curve for microstrainer No. 1 in Fig. 25 would lie to
the right of No. 2; however, no valid comparison can be made from these
data.
67
-------
0)
M
3
£
-------
100.0,
O.O
II.0
19 O
zr.o
83-0
91.0
35.0 43.0 51.0 99.0 67.0 75.0
Suspended Solids, Percentage Removed
Fig. 25. Phase III frequency distribution of suspended solids removal
-------
From Fig. 25 it can be seen that the suspended solids removal of micro-
strainer No. 2 was 66 percent or better 75 percent of the time. When
data for selected dates in the chemical feed period when seal leakage
was minor or nonexistent were compared, microstrainer No. 1 had an aver-
age suspended solids removal of 75 percent and microstrainer No. 2 had
an average removal of 83 percent.
Turbidity Removal
The turbidity removals of Phase III are shown in Fig. 26. The average
removal for microstrainer No. 1 was 46 percent, with maximum and mini-
mum values of 75 and 5 percent, respectively. Microstrainer No. 2 had
about the same average removal, and maximum and minimum efficiencies of
88 and 10 percent, respectively. Fig. 27 shows the frequency distribu-
tion of Phase III turbidity removals.
Average turbidity removals for periods of no seal leakage were 46 and
55 percent for microstrainers No. 1 and No. 2, respectively. The accum-
ulated data for microstrainer No. 1 are of doubtful value. Operation
of this microstrainer was discontinued from September 16 to the end of
Phase III because of the seal leakage.
Color Removal
Phase III color removals are shown in Fig. 28. The average removal for
microstrainer No. 2 was 55 percent, with maximum and minimum values of
80 and 4 percent, respectively. The percentile ranking curves in Fig.
29 indicate that 75 percent of the time color removal of microstrainer
No. 1 was 32 percent or better. Microstrainer No. 2 showed a removal
of 39 percent or better 75 percent of the time.
BOD Removal
The BOD removal efficiencies were highly variable (see Fig. 30). Con-
sidering all the data, microstrainer No. 1 had an average BOD removal
of 45 percent. The maximum and minimum removals were 80 and 2 percent,
respectively. Microstrainer No. 2, which had little difficulty with
seal leakage, showed less variable data. The average removal was 42
percent, with maximum and minimum values of 75 and 11 percent, respect-
ively. The percentile ranking curves in Fig. 31 indicate that 50 per-
cent of the time BOD removal for microstrainer No. 1 was 42 percent or
better. Microstrainer No. 2 was slightly better, with a BOD removal of
49 percent or higher.
70
-------
100.0,
s
O
T/2» •/• */!•
»/7 »/IT t/ZT 10/7 10/17 IOW
o.o I
•/• •/!• •/«• »/T t/17 9/2T 10/7 10/IT I0/t7
Date of Observation, 1972
Fig. 26. Phase III turbidity removal
71
-------
IOO.O
NJ
8.0 16.0
24.0
3Z.O 40.0
48.0
56.0 64.0
72-0
80.0
Fig. 27.
Turbidity, Percentage Removed
Phase III frequency distribution of turbidity removal
-------
100.0
7/11
4)
to
«d
I
u
O
S3
4J
to
O
M
U
I/I* t/li */7 1/17 »/ir 10/T 10/17 10/tT
7/M I/I l/ll l/t* 1/7 1/17 t/27 10/7 10/17 10/S7
Date of Observation, 1972
Fig. 28. Phase III color removal
73
-------
IOO.O
87.5 -
to
CO
-------
0)
»
4J
g
U
M
01
IOO.O.
•7.5
100.0
T/t* •/• •/!• VI* »/T »/IT VtT W/T IO/IT IO/IT
Date of Observation, 1972
Fig. 30. Phase III BOD removal
75
-------
100.0
ON
J3
a
87.5
7S.O
n
a
a
M 62.3
-------
Polymer Feed
Throughout Phase III, several strongly cationic polymers were evaluated,
either singly or in addition to the primary coagulants. Polymers were
evaluated on the bases of their efficiency in floe strengthening and
their effect on drainage characteristics. Table 7 lists some of the
more promising polymers and dosages tried and the primary coagulant
dosages for selected days when no seal leakage occurred. For compari-
son purposes, percentage removals of suspended solids and turbidity and
chemical costs are also shown. Total chemical cost ranged from $8.00
to $156.80 per day or $7.40 to $149.00 per million gallon treated. It
should be noted that the cost of polymers is not a major portion of the
total chemical cost.
COMPARISON OF PHASE II AND PHASE III STUDIES
The selected data used in the preceding paragraphs are for those operating
days when there was no leakage around the seals. This leakage, of course,
reduced removal efficiencies, because less material was removed by the
microstrainers. The effect of seal leakage, using suspended solids data,
is illustrated in Fig. 32. The steep slope of the selected data is a
relative indication of process reliability, since it shows that for a
large percentage of the time the removal efficiency was nearly constant.
The inclusion of all data implies highly variable removal efficiencies
over relatively short periods.
The following discussion uses only no-leakage data for comparisons of
and/or conclusions on microstrainer performance during Phases II and III.
It should be noted that there is a limited amount of selected data for
microstrainer No. 1. The performance data are shown in Figs. 33, 34 and
*5 J •
Fig. 33 indicates that straight screening (i.e., Phase II) results were
almost equivalent for each screen size. The suspended solids removal
was approximately 80 percent or better 70 percent of the time. During
Phase III, the performance of microstrainer No. 2 was about the same as
in Phase II, with a removal of 78 percent or less 70 percent of the time.
For the same period, microstrainer No. 1 had a removal of only 55 percent
or better 70 percent of the time.
Similar comparisons may be made from the turbidity removal efficiencies,
Fig. 34. Here again there was no major difference in performance between
the micros trainers in Phase II. During Phase II, the turbidity removal
for both microstrainers was about 35 percent or better 70 percent of the
time. In Phase HI, however, better turbidity removal was exhibited by
both microstrainers. These removals were 40 and 45 percent or better
for microstrainers No. 1 and No. 2, respectively, 70 percent of the time.
77
-------
TABLE 7
PHASE III MICROSTRAINER PERFORMANCE AND CHEMICAL COSTS
Date,
1972
8/4
8/7
8/9
8/10
8/11
8/12
8/24
9/11
9/12
9/16
9/19
9/20
9/21
9/24
9/25
9/26
9/27
9/28
9/29
9/30
10/3
10/6
10/7
10/10
10/12
10/13
10/14
Flow,
mgd
1.03
1.74
1.77
1.81
1.90
--
2.14
2.11
2.34
1.19
2.20
2.13
2.13
0.96
2.20
2.23
1.95
2.01
1.04
1.13
1.81
1.05
1.08
0.94
1.08
2.00
1.71
Alum,
mg/1
0
50
25
50
25
25
50
50
25
25
75
75
75
75
0
25
0
0
50
0
100
150
0
0
50
50
50
Primary Coagulant
Polymer
Sodium
Aluminate, Dose,
mg/1 Name mg/1
0 Nalco 636 HD 1.0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0 1
^
2.0
1.0
1.0
2.0
r
0 Natron 88
0
0
0
30
100
0
/
1.0
2.0
1.0
1.0
3.0
1.0
0 Nalco 71D09 2.0
0 1 1.0
0 1 1.0
0 Natron 88 2.0
78
-------
TABLE 7 (Cont.)
PHASE III MICROSTRAINER PERFORMANCE AND CHEMICAL COSTS
PERFORMANCE AND CHEMICAL COSTS
Suspended Solids Removal,
percent
Turbidity Removal,
percent
Chemical Costs
Microstrainer
No. 1
70.4
61.5
85.1
81.3
74.7
Microstrainer
No. 2
88.5
73.6
77.1
77.6
68.0
80.2
81.3
82.3
89.1
85.9
91.1
95.6
80.6
93.0
80.5
83.2
84.1
86.7
71.2
92.6
74.3
81.6
86.1
93.6
82.7
88.2
77.8
Microstrainer
No. 1
26.0
40.3
37.5
50.0
45.4
55.5
55.5
-
58.0
Microstrainer
No. 2
50.0
32.0
44.4
33.3
42.0
86.9
51.1
58.3
66.6
58.0
70.5
70.4
45.5
58.3
66.0
64.6
58.7
75.3
37.5
14.7
35.5
46.1
50.0
55.7
91.3
54.7
33.6
Per Mil Gal.
$ 12.91
29.59
21.40
29.50
21.21
-
29.57
42.51
21.23
21.21
50.68
50.84
50.84
50.93
25.81
34.17
14.82
7.41
31.53
7.43
63.00
149.00
7.40
13.50
26.08
47.89
31.60
Per Day
$ 13.30
51.50
36.60
53.40
40.30
-
63.30
89.70
49.70
25.25
107.50
108.30
108.30
48.90
56.80
76.20
28.90
14.90
32.80
8.40
114.60
156.80
8.00
12.70
28.20
95.70
54.10
79
-------
100
o»
O
o>
O
OT
§ 40
£
n
oo
1)
|J 20
10
va
0)
«
Phase II
micro strainer No. 1
Total data
(includes seal leakage),
Selected data '
(no seal leakage)'
20
40
60
80
100
Percentage of Time
w * o>
-------
100
80
60
CO
co
a
01
u
60
40
20
Phase III
Microstrainer No. 1 ,
Microstrainer No. 2
NX
0 20 40 60 80 TOO
Suspended Solids, Percentage Removed
Fig. 33. Suspended solids removal-selected data
81
-------
Microstrainer No. 1
Microstrainer No. 2
l i
100
20 40 60 80
Turbidity, Percentage Removed
100
Fig. 34. Turbidity removal—selected data
82
-------
The effect of chemical addition on color removal is shown in Fig. 35.
Microstrainer No. 2 showed significantly better removal in Phase III,
compared to Phase II when straight screening was practiced. Due to
lack of usable data during Phase III, no correlation can be made for
microstrainer No. 1.
MICROSTRAINER CAPACITY AND FILTERABILITY INDEX
The hydraulic capacity of the microstrainers is a function of: (1)
the characteristics of the wastewater being filtered, (2) the opening
size of the microfabric, (3) the freeness of the solids mat on the micro-
fabric, (4) head loss through the microfabric, and (5) the speed of the
drum, (i.e., filtering area per minute).
These parameters are related according to the following formula:
N1Q/S
MQCfe
H =
L A
Where HL = head loss, in.
Q = flow rate, gpm
Cf = hydraulic resistance of the screen
(23-microns = 1.8 ft. water, 35 microns =
1.0 ft.)
A = effective submerged area, sq. ft.
1 = filterability index
S = speed of microstrainer, sq.ft/min.
M = 0.0267, N = 0.1337
e = base of the natural system of logarithms
(2.71828...)
All of these values can be determined by field measurements, except the
filterability index (1). This is an index of the rate of clogging of
the microfabric, and is a function of the characteristics of the waste-
water, ttje solids mat, and the condition of the microfabric (i.e., cleanli-
ness), olving the equation for 1 yields:
ln AH
83
-------
100
CO
e
09
CO
-------
For the purpose of sizing the micros trainers, a value of 1 = 24 was
given as an estimated average filterability index of the waste, based
on previous studies and the manufacturer's past experience.
The capacity of each microstrainer can be determined in two ways. First,
the actual flow through the microstrainer and other related conditions
can be measured directly, the equation can be solved for the filterability
index, and this can be compared with the given index of 24. Secondly,
using the given index of 24, the manufacturer's rated capacity (Q) of the
microstrainer can be compared with the actual measured flow (Q) passed.
Neither method is accurate for a determination of the capacity of the
microstrainers because of the following factors:
1. The filterability index varies with the rapidly changing
characteristics of the wastewater, and may not be equal
to 24.
2. The flow capacity of the screen is computed assuming
that the microfabric is clean and completely open --
most of the time it was partially plugged. The intensity
of plugging, however, depended on the backwash pressure,
speed of the drum, and fine suspended material in the
washwater.
3. The values of 1 reported are based on average values of
head loss, flow, and drum speed.
Daily readings of head loss, flow, and drum speed were averaged and the
daily filterability indexes computed assuming a clean screen. Fig. 36
illustrates the fluctuations in these indexes for micros trainers No. 1
and No. 2 during the straight screening and chemical feed periods, and
also shows their relation to the given average design index of 24. Gen-
erally, points located below the 24 index line indicate that the micro-
strainer has a capacity greater than design capacity; points located
above the line indicate a lower capacity than design capacity.
The indexes for microstrainer No. 1 show that most of the time it did
not meet design capacity in Phase II, but did during Phase III. Micro-
strainer No. 2 met its design capacity most of the time during Phase H
and Phase III.
It is impossible to evaluate properly the capacity of microstrainer No. 1
until new methods, now being tested, of cleaning the screen are perfected
or the backwash pressure is increased to meet the manufacturer's recom-
mendations. Microstrainer No. 2 has the ability to meet its design capacity
under normal operating conditions.
85
-------
ao
X
01
•H
,0
A>T |0/t7
Date of Observation, 1972
Fig. 36. Filterability index
-------
SPECIAL STUDIES
Long-Term BOD's
Throughout Phase III, several long-term BOD tests were run to study the
extended oxygen demand of organic matter in the wastewater. These tests
also compared the extended BOD to the standard 5-day BOD. It is generally
considered that the 5-day BOD values of domestic sewage are approximately
75 percent of the ultimate BOD. However, not all organic solids demon-
strate the same rate of biodegradation, especially those found in industrial
wastes. For this reason, long-term, or ultimate, BOD tests are important
to determine the rate and amount of biochemical oxygen demand over a given
time.
Tests for long-term BOD's were conducted on composite samples of the treat-
ment plant influent, effluent, and backwash sludge. BOD analyses were con-
ducted on samples at 5-, 10-, 20-, and 30-day intervals. Fig. 37 shows the
development curves for a 10-day period on samples taken 5/9/72. The influ-
ent and sludge curves indicate that the 5-day BOD is 67 percent of the 10-
day BOD. From the effluent curve, the ratio of 5-to 10-day BOD is 80 per-
cent (because of the lack of solids in the effluent). This can also be
shown by the higher values of 10-day BOD for the influent sample and sludge
samples; 210 mg/1 and 3,000 mg/1, respectively. These values correspond
to the increased solids concentrations of the two samples. It will be ob-
served that an apparent plateau is reached at the 10-day interval. This
is not, however, an indication that the samples have reached their ultimate
level of BOD, as shown by further studies.
An example of a 30-day BOD is given in Fig. 38. This family of curves shows
a plateau or decreasing rate of oxygen depletion for the influent and sludge
samples between the 5- and 10-day periods, and then a rapid increase through
the 30-day period. In addition, the curves are still climbing at 30 days,
which indicates an even longer time needed for complete degradation (ultimate
or total BOD). The effluent curve shows a slow rate of oxygen depletion,
as denoted by its flat slope. The ratios of 5- to 30-day BOD values for
the sludge and influent samples are 8.7 and 9.7 percent, respectively;
the same ratio for the effluent samples is 23 percent.
BOD removal efficiencies are greatly affected by the longer term BOD tests.
The ratio of influent to effluent BOD values, shown in Fig. 38, is only
46 percent for the 5-day period, but 79 percent for the 20-day period and
81 percent for the 30-day period.
87
-------
50OO
500
4000
3000
oo
6
00
00
2000
1000
OJ
00
-o
rH
CO
400
—— Sludge
Time, days
Fig. 37. Long-term BOD (samples of 5/9/72)
-------
7,000
•,000
8,000
i/»oo
9
S
4,000
1,000
1,000
1,000
V
60
Influent
Effluent
Sludge
18 tO
Time, days
Fig. 38. Long-term BOD (samples of 9/11/72)
89
-------
Fig. 39 shows families of 20- to 30-day curves for samples taken 3/19/72
and 10/11/72, respectively. Effluent BOD compared to influent BOD shows
5- and 20-day removals of 68 and 80 percent, respectively, for samples
of 9/19/72.
A slower BOD development is shown in the 10/11/72 samples of this figure.
The 5- and 30-day BOD removal efficiencies for the influent and effluent
samples of 10/11/72 are 59 and 65 percent, respectively.
To sum up: the BOD's of the influent and backwash sludge samples for
10-, 20-, and 30-day periods are much higher than the 5-day standard BOD,
but the effluent BOD's for 10, 20, and 30 days are more nearly equal to
the 5-day values. The actual long-term BOD's for influent and sludge
samples fluctuated from day to day according to the fluctuations in the
characteristics of the wastewater. The effluent long-term BOD values,
however, were relatively constant. This is an indication of the relative
consistency of effluent quality, even though removal rates varied.
Soluble BOD
The previously mentioned problems with high percentages of soluble BOD
in the wastewater required further study to locate their source. Since
most of the soluble BOD is attributed to the starch in the internal and
surface sizing operations of the paper mills, it was assumed that one or
several sources of waste from the sizing operations were finding their
way into the treatment plant.
Three separate special sampling programs were conducted to find the source
of the soluble BOD. The first two sample programs indicated that the
Strathmore Whitewater had a high percentage of soluble BOD. This was
attributed in part to the resolubilized surface size on the reused fibrous
material and the internal size. The third study showed a decrease in the
amount of soluble BOD in the Whitewaters at both the Whitewater tank over-
flows and the paper machines (tray water). This in soluble BOD found in the
latter program was from the relocation of certain size tarik drains entering
the treatment plant to discharge to the sanitary sewers. Relocation of all
wash drains has been*completed.
90
-------
3,000 r 600
2,000
• 400
8
1,000
- ZOO
o Lrt o
4-1 .U 0
0) CO
oo
-------
SECTION VI
PHASE IV
PLANT OPERATION OPTIMIZATION
Problems
1) Inability to handle surges or high flows -- even within design
range. Often, normal flows could not be filtered.
2) Leaking seals.
3) Failure to treat effluent adequately so regulatory agency re-
quirements could be met consistently or even a high percent
of the time.
4) The effluent pump drive was unreliable suffering frequent
failure and causing use of backup and even manual controls.
5) There appeared to be no hope of removing soluble BOD even
with good plant operation.
6) Chemical costs were higher than anticipated because of high
polymer and alum dosages required for drainage.
7) Federal regulations became more stringent than those for
which the plant was designed and approved while the plant
was being built.
8) Because of effluent bypassed during failures of equipment
during this shakedown period, the canal discharge manifold
plugged several times.
9) Hydraulic microstrainer drives failed because of high loads
imposed on them by excess drag from seals. The seals were
overtightened in an attempt to stop the leakage.
10) Chemical upsets of an unknown origin in the mills caused loss
of floe and drainage.
11) Much maintenance, technical and engineering time was required
to operate the plant and find solutions to the problems.
12) There were numerous less important problems such as the auto-
matic turbidity system not working and magnetic flowmeters
being kept in calibration.
92
-------
Flow surges
Esleeck originally had a collection sump which was pumped out rapidly when
it filled. They converted to a continuous pump system which eliminated
that source.
Strathmore has all gravity flows. The solution here was to control dumps
particularly from white water chests. The white water is released over a
relatively long period rather than all at once. This procedure does not
affect machine changeover time.
Another source of difficulty was several machines dumping at the same time.
Coordination of these activities by plant supervision has minimized this
type of surge.
Leaking Seals
As has been discussed earlier, many types of seals were tried but all
either leaked or put too much drag on the drums or wore out quickly.
Later discussion will show that the seals would not have been anywhere
near the problem they seemed to be had we discovered how to achieve a
free-draining floe and how to keep the screens clean. The seals finally
chosen were the dumbbell (Skelley)seals on the outside with no inside
seals. See (f) Fig. 15, page 49.
Composite samples have no measurable settleable solids normally with these
seals.
The dumbbell seals are inexpensive and can be used four times before dis-
carding by turning them over, turning end-for-end, and turning over again.
Only the edge against the moving chime wears, and there are four such
edges on each seal.
In addition, the dumbbell seals are simple to replace -- taking about 30
minutes. By contrast, inside seals require about 8 hours for each drum
with the plant down. The level need not be dropped to change a dumbbell
seal.
The reason that the seals leaked earlier was because a) the head differ-
entials across the screens were so great (8-10") and b) the drum speeds
were usually at their maximum to handle or attempt to handle the flow.
The head differentials resulted from a) dirty screens and b) poor floe.
93
-------
Soluble BOD
It was demonstrated in the lab and in the treatment plant, when it was
operating properly, that BOD removal would not be sufficient to meet either
the old 85% removal or the new 8 Ibs. per ton requirements. The problem
was almost entirely soluble BOD which could not be filtered or flocced.
Some small removals were obtained by flocculation. This was probably
finely colloidal starch.
It was shown by a series of experiments in the Strathmore mill that starch
accounted for more than 957, of the BOD from that mill.
A campaign was mounted to
1) Reduce starch usage
2) Increase starch retention
3) Decrease starch storage time
4) Avoid over-cooking of starch
5) Maximize material retention of all kinds
6) Police water use practices.
The results of this effort were dramatic. Solids losses from Strathmore
were cut in half and BOD was cut 65%.
It was then discovered that a high molecular weight, anionic polymer de-
veloped a large, tough, free-draining floe which greatly improved plant
performance. In spite of this development the plant still could not handle
high flows within the design range.
Screen Cleaning
COM evaluated many cleaning agents including phosphoric acid initially
recommended by Crane. The chemical cleaner chosen and used for a long
period of time was Oakite LSD, a highly alkaline detergent.
It was reasoned that the flow handling capability of the plant could be
improved by 30% by changing the No. 1 screen from 23 micron fabric to 35
micron fabric as was on No. 2 screen. No. 2 screen seemed to pass twice
the flow that No. 1 passed. No. 1 screen was changed during our 1973
summer shutdown.
Flow handling improved markedly. But over a few weeks' time, we again
could not handle the flows. It was noted that sludge flow to the town
sewers had gone up when the new screen was installed and gradually fell
off corresponding to the loss of flow handling capability. It seemed that
the knockoff showers were not penetrating the fabric.
94
-------
A piece of the removed, dirty, No. 1 fabric was cleaned with various
chemicals in the laboratory. Phosphoric acid appeared to do the best
job.
Both screens were cleaned by applying 10% phosphoric acid with a sprink-
ling can to the exposed portion of the drums. A 10-minute soak time was
allowed before rotating the drum to do the next section.
Following the cleaning, the plant could handle all flows. On one occasion
just one screen passed a 2.5 MGD flow rate. In conjunction with the ani-
onic polymer effluent quality was excellent. Sludge flow to the town went
from 60,000 gpd to over 200,000 gpd -- off the magmeter scale. Dirty
screens had obviously been the main problem all along.
November 1973 marked the beginning of 24-hour anionic polymer usage. (Stein
Hall's Polyhall 980.) It was discovered that pH had to be between 6.5 and
8.0 for adequate floe characteristics particularly as regards drainage. pH
7.5 was best.
Because of this requirement and the NEPDES Permit requirement for pH of
discharge, automatic pH control using NaOH was installed.
We were plagued by shortages of chemicals during early 1974. Caustic and
phosphoric acid were unavailable from time to time and finally we were
notified that the best polymer was not available at all.
When operating at neutral pH, a gradual deposition of aluminum hydroxide
occurs that eventually blinds the screen such that normal plant flows can-
not be handled. The deposition appears as a uniform, "plated-on" coating
in an SEM picture. X-ray energy dispersion analysis shows aluminum to be
the predominant substance present in the coating. At this writing, a sat-
isfactory solution to the blinding has not been found and the plant is by-
passed several times per day.
The total by-passed per day is in the range of 30-150,000 gallons; the
balance of the 2 MGD is treated and meets permit requirements.
A recent development is that with summer river water temperature the entire
effluent can be adequately treated with just one microstrainer. It seems
that the viscosity and surface tension of water from 55° F. (winter) to
90° F. (summer) have a marked effect on drainage and chemical reaction
time of floccing chemicals.
95
-------
SECTION VII
SOLIDS HANDLING
PILOT FACILITIES
To determine the dewaterability of the backwash sludge, small scale
thickening and dewatering units were set up and operated intermittently
during two months of the last part of the Phase III studies at the treat-
ment plant. These studies were a preliminary step to ascertain the possi-
bility of reclaiming paper fibers in the backwash sludge and reusing them
in the mill or producing a marketable byproduct for sale. After determining
whether the sludge could be dewatered efficiently and economically, the
paper mills would study the feasibility of reusing the reclaimed fiber.
The methods of sludge dewatering studied were centrifugation, vacuum
filtration, flotation, and gravity thickening. Pilot units included a
6-in. solid bowl centrifuge manufactured by Bird Machine Company, a 3-ft-
diameter by 1-ft-wide coil filter, and a 1-sq-ft thickening utilized a
55-gal. barrel fitted with side draw-offs and a side window.
Before commencing the studies, it was assumed that it would be possible
to dewater the backwash sludge (average suspended solids concentration
0.25 percent) directly in the centrifuge and on the coil filter units
operating in parallel. After some preliminary studies, it became apparent
that the backwash sludge was too low in solids for efficient operation of
the centrifuge and coil filter units. In fact, the coil filter unit would
not form a solids mat unless the suspended solids concentration in the
sludge was greater than 0.50 percent.
For this reason, gravity thickening was begun using the 55-gal. barrel
mentioned previously. In this procedure, the barrel was filled with back-
wash sludge and allowed to settle. About 90 percent of the time the solids
floated to the top because of entrainment of air caused by the backwash
water passing through the microfabric. Therefore, either the supernatant
or underflow was drawn off to leave only the thickened sludge. Thus, by
repeating this operation several times (three to five), the solids concen-
tration was doubled (i.e., 0.5 percent or greater solids concentration was
obtained).
It was anticipated that gravity thickening might not be feasible for full
scale operation, because of the need for large settling tanks and the in-
ability of this method to raise the solids concentration above 1 percent.
For this reason, a plant flotation thickener was set up and operated
during the last two weeks of the study. Operating results of the flota-
tion thickener as well as the coil filter and centrifuge are given in
Table 8. The flotation thickener was very effective in thickening the
96
-------
TABLE 8
PILOT SLUDGE STUDIES DATA
Bird 6-in.
Solid Bowl Centrifuge
Bowl Data
Date,
1972
8/30
9/7
9/8
9/11
9/12
9/18
9/19
10/2
10/3
10/5
10/11
10/12
10/13
10/14
10/15
10/16
10/17
Flow,
gpm
10*
2-3
2-34
2*
1-1. 5}
1.0}
1.5f
1.0-1.5
1.5-2. Of
2t
1.5f
2!
Speed,
rpm
5,000
5,000
5,000
5,000
5,000
5,000
5,000
4,000
4,000
4,000
5,000
5,000
Bowl
Depth
Medium
Medium
Maximum
Maximum
Maximum
Minimum
Medium
Medium
Medium
Maximum
Medium
Maximum
Sludge
Raw
0.32
0.27
0.394
0.394
0.524
0.5755
0.698
0.456
0.279
0.685
0.632
1.43
Solids, percent
Filtrate
0.12
0.08
0.103
0.109
0.099
0.084
0.076
0.058
0.094
0.202
0.142
0.31
Cake
13.4
25.8
30.5
76
32.6
21
29.5
34.8
37.1
44.8
45.6
43.5
Per-
centage
Recovery
62
70
74
71
81
85
89
87
62.3
70
78
78
* Unit plugged.
•} Gravity thickened sludge.
Flotation thickened sludge.
97
-------
TABLE 8 (Cont.)
PILOT SLUDGE STUDIES DATA
Komline-Sanderson 10-sq-ft Coil Filter
Drum Data
Sludge Solids,
percent
Submer-
Flow, Drum gence
gpm Speed Percent Raw
15-18 Slow
15-18 Slow
15-18 Med.
37 0.44
37 0.44
37 0.51
Per-
cent
Fil- Re-
trate Cake covej
0.14
0.19
0.053
21.0
22.2
37.4
68
57
90
Komline-Sande.rson 1-sq-ft
Flotation Thickener
Sludge Solids,
percent
Flow,
pm
Fil-
Raw trate Cake
6.5f Slow 37 0.58 0.31 20.5 47
5+ Slow 37 0.60 0.21 21.3 65
3f Slow 37 0.49 0.10 20.2 80
5f Slow 37 0.53 0.08 22.5 85
Slow 37 2.32 0.045 21 98
1.5-2.0
2
2.5
3
0.099
0.103
0.178
0.156
0.201
0.010
0.021
0.034
0.038
0.029
2.34
4.9
2.9
3.54
4.18
90
80
81
75
85
98
-------
raw backwash sludge from 0.25 to 2 to 4 percent solids, with good recovery
efficiencies. Further studies were conducted with the centrifuge and coil
filter using the flotation thickened sludge as a feed. These results are
also shown in Table 8.
FULL SCALE FACILITIES
Because the wastewater treatment plant is attended only 8 hours per day,
the solids handling facilities should be designed for similar operation.
Considering the time needed to start operation and to clean up, 6 hours
actual operating time were allowed.
Fig. 40 shows the daily variations of the backwash sludge flow, percentage
solids, and total solids coming from the microstrainer from April 30 through
October 20, 1972. From this figure and the results of the pilot plant studies
listed in Table 8, the backwash sludge design parameters for a full scale
solids handling facility were determined. These are given in Table 9.
TABLE 9
MICROSTRAINER BACKWASH SLUDGE DESIGN PARAMETERS
Average daily flow 0.15 mgd
Maximum daily flow 0.20 mgd
Percentage solids in sludge 0.25
Maximum total solids per day 4,200 Ib.
Fig. 41 is the schematic flow diagram for the microstrainer backwash water
sludge consolidation treatment plant showing the two alternatives (vacuum
filtration or centrifuging). An equalizing tank has been provided between
the microstrainer and the first treatment unit, the flotation thickener,
to hold the sludge during the 18-hour period that the treatment plant is
not operating. The tank has been sized for an 18-hour flow under maximum
flow conditions, i.e., 20,000 cu. ft. The tank's contents would be thor-
oughly mixed to insure that the sludge withdrawn for treatment was of uni-
form consistency.
The sludge would be pumped from the equalizing tank at a steady rate to
the air flotation thickener. Table 10 lists the design parameters used to
size this unit and other units* making up the sludge dewatering facility.
The conditioned sludge would flow from the flotation unit into a holding
99
-------
g
o
H
00
•o
-a
0)
T3
g
(X
(0
13
2
to
O
O
5000-
2500-
0
0.2
O.I
1/3
e
i-H
0>
4/30/72
6/I/T2
7/1/72
7/31/72
Fig. 40. Daily variations of backwash sludge
-------
8/1/72
9/1/72 10/1/72
Fig. 40. Daily variations of backwash sludge (Cont.)
10/20/72
-------
o
N>
Sludge
\ 7 cake
'—' hopper
Paper mill
influent
I Alternative No. 2
Fig. 41. Schematic flow diagram
-------
TABLE 10
SLUDGE HANDLING FACILITY, UNIT DESIGN PARAMETERS
Equalizing Tank
Number One
Volume, cu. ft. 20,000
Raw sludge pumps
Number Three (including one standby)
Capacity each, gpm 300
Flotation thickener
Surface area, sq. ft. 400
Solids loading, Ib/sq. ft./hr. 1.75
Volume of raw sludge, gpm 555
Volume of consolidated sludge, gpm 41.2
Percentage solids in consolidated sludge 3.0
Volume of underflow, gpm 513.8
Percentage solids recovery 89.0
Conditioned sludge holding tank
Volume, cu. ft. 2,850
Percentage of maximum day 100
Conditioned sludge pump
Number Two (including one standby)
Capacity each, gpm 45
Vacuum filter
Number One
Surface area, sq. ft. 100
Loading, Ib/sq. ft./hr. 6.2
Volume of sludge cake, cu. ft/day 292
Percentage solids in cake 20
Volume of filtrate, gpm 35
Percentage solids in filtrate 0.05
Percentage solids recovery 98.5
Sludge cake hopper, cu. ft. 300*
Centrifuge
Number One
Diameter, in. 24
Loading, Ib/hr. 620
Volume of sludge cake, cu. ft./day 120
Percentage solids in cake 45
Volume of centrate, gpm 38.7
Percentage solids in centrate 0.3
Percentage solids recovery 90.5
Sludge cake hopper, cu. ft. 125*
*Allows for bulking and assumes emptying twice per day.
103
-------
tank to allow continuous operation of the vacuum filter (or, alternatively,
the centrifuge) with sludge of uniform consistency. The holding tank would
be capable of holding one day's conditioned sludge production for the flot-
ation unit at maximum daily flow. The tank would be fitted with diffused
air mixing facilities.
The use of either a vacuum filter or centrifuge as the final unit in the
sludge dewatering process has been considered. These are shown as Al-
ternatives Nos. 1 and 2 in Fig. 41. Sludge pumps would transfer the
sludge directly from the holding tank to the final process unit.
The vacuum filter would have a surface area of 100 sq. ft. and would be
loaded at a rate of 6.2 Ib./sq. ft./hr. at maximum conditions. We es-
timate from the results of the pilot plant work that the percentage sol-
ids recovery would be 98.5 and the sludge cake would have a solids con-
tent of 20 percent and a maximum volume of 292 cu. ft. per day.
The sludge cake from either of the alternative final units would be
transferred by conveyor to an elevated sludge hopper for storage prior
to hauling by truck for reuse in the paper mill or for off-site disposal.
The flotation thickener underflow, the vacuum filter filtrate, or the
centrate would be returned by gravity to the wastewater treatment plant
influent. The treatment plant and sludge processing facility would thus
be a closed system, since the total sludge removed by the microstrainers
will be recovered. The microstrainer backwash sludge solids content
will increase somewhat as the underflow and filtrate or centrate are
returned to the microstrainer. No provision has been made for dupli-
cation of these principal units for standby purposes, since it would
probably be more economic to discharge the sludge into the Montague
sewerage system at times of emergency shutdown.
OPERATING COSTS
Table 11 lists the estimated costs of operating the microstrainer sludge
handling facility. We estimate that two men per day could operate the
treatment plant and the sludge handling facility, allowing 1.5 man-days
per day for the running of the sludge facility. A nominal amount only
has been allowed for chemical conditioning, because the wastewater nor-
mally contains suitable chemicals from the paper manufacturing process
and the wastewater treatment process.
104
-------
TABLE 11
SLUDGE HANDLING OPERATING COSTS*
Item
Chemicals
Labor
Power
Cake disposal**
Total
Cost per Ton
Dry Solids-j-
$ 3.19
44.00
7.47
28.68
$ 83.34
Cost per
Dayt
$ 5.00
69.00
11.72
45.00
$130.72
* Based on average daily flow.
f 1.38 tons/day.
J Based on 7-day week, 8-hour day.
** Assuming cake hauled to municipal disposal site 5 miles from mill.
The costs are based on an average daily flow of 0.15 mgd of backwash water.
The solids recovered equal the total solids entering -- 1.57 tons dry
weight per day. For the purpose of these estimates, it has been assumed
that the sludge cake is hauled to the local municipal landfill for final
disposal.
The estimated operating cost per day is $130.72, or $83.34 per ton of dry
sludge.
First Year Operating Costs
Table 12 lists the operating costs according to general categories for the
first year of operation of the wastewater treatment plant. These costs
were taken from a summary sheet prepared by the Strathmore Paper Company
showing their detailed operating expenses for the period October 1971 to
4 December 1972. For the purpose of this study, however, only those costs
from 1 November 1971 to 1 November 1972 were used (start of operation of
the treatment plant until the end of the post-construction studies). An
explanation of each of the categories follows.
105
-------
TABLE 12
FIRST YEAR OPERATING COSTS*
1971 1972 Total
Chemicals $2,930} $ 1,810 $ 4,740
Electricity 2,705 10,750 13,455
Supervisor 490 1,535 2,025
Operator 1,200 9,890 11,090
Maintenance 109 4,445 4,554
Laboratory supplies 1,110$ 20° 1,310
Miscellaneous 13 565 578
Major repair and labor -- 1,305 1,305
Microstrainers repair and labor -- 2,395 2,395
Heating and ventilation 424 1,690 2,115
Process water 400 1,590 1,990
Total $9,382 $36,175 $45,567
* Compiled from 1 November 1971 to 1 November 1972.
f Includes initial purchase of bulk chemicals.
| Includes initial purchase of basic supplies.
Chemical costs refer to the purchase of aluminum sulfate and sodium alum-
inate (used as primary coagulant) and polymers (used as coagulant aids).
These costs are not necessarily distributed throughout 1971 and 1972,
because some chemicals were purchased in bulk at the treatment plant
startup.
Electrical costs are the direct costs of the electrical power used. The
rate was approximately 2.5 cents per kilowatt-hour.
Supervisor costs refer to the time spent by Strathmore supervisory per-
sonnel in overseeing the operation of the treatment plant. These costs
are based on 10 percent of the payroll costs, including fringe benefits,
of operation and maintenance personnel.
Operator costs include the payroll costs of the operating personnel plus
30 percent for fringe benefits.
Maintenance costs include the payroll costs plus 30 percent for fringe
benefits of the Strathmore maintenance personnel needed for the day-to-
day preventive maintenance and repair needed at the plant.
106
-------
Laboratory supplies costs comprise all expenses of the operation of the
laboratory and wastewater testing. They include some small initial pur-
chases, but for the most part are related to the normal replacement pur-
chases of chemicals and apparatus needed. Costs of additional glassware
and chemicals needed to conduct the post-construction studies are not
included, but amounted to approximately $500.00.
Major repair and labor costs include all costs of the repair of mechanical
equipment and piping, excluding the microstrainers, such as repair and
replacement of the sludge pumps, fixing and cleaning valves, piping,
wastewater distribution, etc. Also included are costs for alterations
or modifications made to the treatment plant for more efficient operation.
Miscellaneous costs include small items such as cleaning materials and
tools needed for the operation and maintenance of the treatment plant.
Microstrainer repair and labor costs include costs for major maintenance
of the microstrainer, and major repair of items such as the main drum
bearings, bearing water flush system, drive controls, and microfabric
repair. Also included in, and a major part of, this cost were the re-
placement and maintenance of the several different sealing operations.
Heating and ventilating costs include maintenance and repair of the
heating, ventilating, and air conditioning equipment.
Process water costs include the mill process water supplied by Strathmore
for use in the backwash spray systems for each microstrainer, as cooling
water for the effluent and backwash sludge pumps, and for microstrainer
washups. This cost is based on 6 cents per 1,000 gallons of water supplied.
All of the costs in Table 12 are not considered normal yearly operating
costs, but are the actual first year costs. A portion of the maintenance,
major repair, and microstrainer repair costs is due to unexpected mechanical
failures encountered in the first year. It is difficult, however, to es-
timate a yearly cost that could be used for these categories, and no attempt
is made to do so.
Based on the total operating costs of $45,567 and an average daily flow
of 1.8 million gallons (6 days per week, 50 weeks per year), the first
year's operating costs per million gallons is $84.30. Based on an approx-
imate paper production of 70 tons per day from the two mills, the first
year's operating cost per ton of paper is $2.20.
107
-------
SECTION VIII
REFERENCES
American Process Equipment Corporation, "Ultrasonic Filtration of
Combined Sever Overflovs" for Environmental Protection Agency Water
Quality Office, (June 1970).
Camp Dresser & McKee, "Strathmore Paper Company and Esleeck Mfg.
Company—Joint Industrial Wastevater Treatment Plant Operating
Manual" (December 1972).
Bliss, Frederick H., "Strathmore Paper Company—Industrial Wastevater
Report" (September 1968).
Bliss, Frederick R. , "Strathmore Paper Company—Process Waste-water
Treatment" (September 1967).
Bliss, Frederick R., "Strathmore Paper Company—Process Wastewater
Treatment" (June 1969).
Esleeck Mfg. Company, "Industrial Wastevater Disposal" (May 1965),
Rimer, Alan E., "Microstraining Paper Mill Wastevater," Journal
WPCF (July 1971). Vol. U3, pp. 1528-15UO.
Bliss, Frederick R., "Papermill Wastevaters: A Prototype Study of
Microstraining," paper presented at the U5th Annual Conference of
the WPCF (October 1972). Atlanta Georgia.
108
-------
SECTION IX
GLOSSARY OF TERMS AND ABBREVIATIONS
BOD
5-Day BOD
Coagulant aid
COD
Filterability index
gpm
mgd
Microfabric
Micros trainer
Microstraining
Primary coagulant
SS
TDK
Ultrafilter
Ultrasonic microfiltration
Biochemical oxygen demand
Biochemical oxygen demand, 5-day, 20°C.
Chemical used to assist coagulation
process (e.g., polymer)
Chemical oxygen demand
Index of rate of clogging of microfabric
(function of wastewater characteristics,
solids mat, and condition of microfabric)
Gallons per day
Gallons per minute
Million gallons per day
Stainless steel, monofilament wound cloth
mounted on the rotary drum of a micro-
strainer; opening size of fabric is
measured in microns
Rotary drum covered with microfabric,
used for filtration of wastewater
Process of solids/liquid separation
utilizing microstrainer
Chemical, such as aluminum sulfate (alum)
or sodium aluminate, used in the rapid
mix basins
Suspended solids
Total dynamic head
Unit employing ultrasonic microfiltration
Solids/liquid separation using microporous
stainless steel canisters for ultrasonic
cleaning
109
-------
APPENDIX A
LOG OF OPERATION
A schedule of operation was set up at the beginning of the post-construction
studies in November 1971, when treatment plant operation was initiated. Al-
though sufficient time had been allowed to accomplish the contract objective,
repeated startup problems and mechanical equipment malfunctions (the micro-
strainers in particular) prevented effective operation and data collection
on any of the scheduled phases of study. For this reason, scheduled studies
were interrupted in February 1972, although the plant operations continued.
The studies were begun again in April 1972, and data collection on the Phase
II studies was initiated on May 1, 1972. The project schedule, Fig. A-l, was
revised in accordance with the new dates. This appendix presents the re-
vised log of operation.
MAY 1 TO JULY 22, 1972
During this period, straight screening studies, utilizing microstrainers
only to treat the industrial wastewaters, commenced. The purpose of these
studies was to evaluate the efficiency of the two microfabrics (23-tnicron
opening on microstrainer No. 1 and 35-micron opening on microstrainer No. 2)
to evaluate the efficiency of microstraining alone in removing suspended
solids, turbidity, color and BOD.
Drum speed and head loss were varied in each microstrainer, and series and
parallel screening sequences were attempted by rescreening some of the
plant effluent through microstrainer No. 1. The latter process proved in-
effective, and had to be abandoned due to high plant flows and leaking seals.
Throughout this period of operation, two major problems persisted that greatly
affected the data collected. The first problem, excessively high plant flows,
hydraulically overloaded the microstrainers and caused problems in evaluating
efficiencies at various drum speeds and head losses. The second and most
important problem was the seal leakage experienced during almost the entire
period. The results of this leakage varied in degree from almost completely
cancelling the effectiveness of the microstraining process to consistently
lowering the efficiency somewhat. This proved especially critical in de-
termining the removal efficiencies of the two microstrainers. To solve this
problem, several different types and/or designs of sealing bands were in-
stalled and evaluated.
110
-------
Scheduled phases (see plan of operation)
(1) Phase 1 (2) Phase II (3) Phase III
Plant startup
Operator training
Operating manual
Jar studies
Straight screening
Screening with coagulants
Sludge tests
Ultrafilter tests
Special tests
(2)
(2)
(3)
(3)
(1)
Fig. A-l. Project schedule
-------
Numerous jar studies were conducted in the laboratory during this period
in an attempt to gather sufficient data on the use of various combinations
of coagulants and coagulant aids. These data were utilized in the full
scale studies of Phase III.
From July 10 to 26, ultrafilter studies were conducted. The results of
these studies are given in Appendix B.
JULY 28 TO OCTOBER 16, 1972
Chemicals were added full scale to the treatment plant under the Phase III
portion of the schedule. Various combinations of primary coagulants (alum
and/or sodium aluminate) and coagulant aids (polymers) were tried and their
effectiveness in flocculating the fine suspended solids was evaluated. The
purpose of this chemical addition was to increase solids removal by the
micros trainers. Chemical feed requirements were determined by jar studies
of daily wastewaters. The most effective dosages were then used full scale
and compared with the bench scale results. The efficiencies of the differ-
ent microfabrics were also evaluated during this period, as well as the
filterability index of each microstrainer (see Section IV).
Again, high flows and leaking seals affected the data collected. Micro-
strainer No. 1 had excessive seal leakage at times because of an improperly
seated sealing band. From September 22 to October 16, microstrainer No. 1
was shut down because of this leakage. At this time, a new seal was in-
stalled on microstrainer No. 2, and it proved effective. The operating data
obtained during this period were among the best, as far as treatment effi-
ciency was concerned.
Several special studies were conducted on the wastewater entering the treat-
ment plant and also on other sources affected by the treatment plant efflu-
ents. Several sets of 30-day BOD's were run on the influent and effluent
wastewaters. Weekly samples of the canal water, upstream and downstream
of the wastewater distributor, were taken to study the effect of the plant
effluent on the canal. Sampling programs were conducted at the Montague
sewage treatment plant to determine the effect of the backwash sludge dis-
charged to this system.
AUGUST 16 TO OCTOBER 16, 1972
During this period, equipment was assembled and operating data were obtained
on several pilot sludge dewatering units to evaluate their effectiveness in
dewatering the backwash sludge for alternative methods of sludge disposal
or recovery. Methods of sludge dewatering tried included: gravity thick-
ening, air flotation thickening, vacuum coil filtration, and centrifugation.
112
-------
APPENDIX B
ULTRAFILTER STUDIES
An ultrasonic microfiltration system designed and built by the American
Process Equipment Corporation, Hawthorne, California, was supplied by EPA
for pilot testing on the paper mill wastewaters. Previous studies with
this unit have shown that it is not successful in filtering wastewaters
with suspended solids concentrations in excess of 100 mg/1. The proposed
testing, therefore, was strictly for research purposes, since the suspended
solids concentrations of the Strathmore-Esleeck wastewaters far exceed 100
mg/1. Because the raw wastewater would be pretreated before it was intro-
duced to the ultrasonic unit, no valid comparison could be made between
ultrafiltration and microstraining.
The ultrafilter system, assembled on a flat bed trailer unit, consisted of:
Twenty stainless steel canisters, each fitted with an
influent port, an effluent port, a backwash port, a
drain, and an air vent in five banks of four canisters
each with each bank connected to the system manifold
through automatic and manual valves. Each canister
contained a cluster of microporous polyethylene filter
tubes which were sealed at the bottom and fitted at the
top with connections to the effluent and backwash mani-
folds (influent flow is into the canister body, through
the filter element, and out the top fitting to the ef-
fluent manifold). The bottom plate of each canister
holds an electrostrictive ultrasonic transducer driven
by an ultrasonic generator located in the electrical
control console. The transducer is activated during
frequent, short, backwash cycles (spritz) for one second
out of each 10 or 20 seconds, and during long-clean back-
wash for three minutes out of every 15 minutes.
Dial pressure gages were provided on the inlet side of
one canister in each bank, and one on each system mani-
fold line for each bank.
An air compressor and air storage tank supplied control
air to solenoid valves for operation of the automatic
flow control valves. Wastewater was supplied to the
filters by means of a 200-gpm capacity pump. A separate
pump provided backwash water.
113
-------
An electronic control panel contained two 700-w, 22-kHz
ultrasonic generators, stepping switches, control cir-
cuits, timers, manual overrides, and a lighted control
panel array showing the operational status of each bank
of filters at any moment.
Fig. B-l is a schematic diagram of a typical bank of filter canisters and
basic canister design.
The ultrafilter trailer was positioned and set up in the parking area near
the treatment plant, adjacent to the Strathmore mill for convenient elec-
trical power hookup. Influent pump suction piping connections were made
to the plant Influent wastewater line in the junction manhole and to a
250-gallon tank for dilution of the wastewater with city water. Filtrate
and drain lines were connected to 2,5-in. fire hoses which conducted these
to the river.
The intended plan of operation for the ultrafilter studies was to: (1)
dilute the paper mill wastewaters with city water to lower the suspended
solids below 100 mg/1; (2) depending on these results, gradually decrease
the dilutions until raw wastewater was introduced into the ultrafilter;
and (3) pretreat the raw wastewater by settling and then pumping the
supernatant into the ultrafilter.
After repeated mechanical failures, the ultrafilter was placed in oper-
ation on July 25, 1972. Raw influent wastewater was diluted with city
water approximately fourfold and the mixture pumped into the system.
After 10± minutes, the filter plugged. In addition, the drain control
panel indicated that these systems were operating. Repeated attempts
to locate the problem were fruitless, and the unit was shut down.
Results of samples taken during this limited run are as follows (flow,
20± gpm; cycle time, 20 sec.):
Test
Influent
Filtrate
7/25/72
PH
Turbidity, JTU
SS, mg/1
BOD, mg/1
COD, mg/1
1*
270
72
183
19
8
12
3
53.5 53.5
After 100± gal. filtered.
After 200± gal. filtered,
114
-------
On July 26, another attempt was made to operate the ultrafilter. This
time, influent wastewater was diluted approximately two times before
feeding. There was no plugging, and the filter operated at a much higher
flow; however, very little filtrate was produced. It appeared that the
influent was being discharged into the drain line. Again, the control
panel indicated that all sequences were operating properly, but the res-
pective valves were not responding as indicated. It became evident that
the control panel did not indicate the actual conditions at each bank of
canisters. Upon isolating single banks of canisters, it appeared that
the valve sequence was wrong for continuous filtering and backwashing
operations due to a faulty signal from the control panel. Again the
unit was shut down. Samples of that filtrate gave the following results
(flow, 45± gpm; cycle time, 20 sec):
Test
(7/26/72)
PH
Turbidity, JTU
SS, mg/1
BOD, mg/1
COD, mg/1
Influent
7.4
405
144
38.3
275
Filtrate
1*
7.4
65
7
15
22.9
2t
7.5
54
10
16.2
30.6
* After 3 minutes.
} After 5 minutes (approximately 200 gallons filtered).
Many more attempts to operate the ultrafilter system ended in failure.
The malfunction appeared to be in the electronic control panel which con-
trolled the sequence of operation. After several discussions with of-
ficials of the EPA concerning these problems, and after their repeated
attempts to obtain assistance from the manufacturer, they advised us to
discontinue our studies.
115
-------
Backwash
Vent
Spritz backwash —W—
Filtrate
D Pressure
gage
Hi
Supersonic
generator
Filter tubes
(35-micron porous
stainless steel)
— ^ Raw
wastewater
Ff=
Basic Canister Design
^T T X
i'
Vent
.Influent from pump
Backwash from pump
•Filtered influent
• Drains (spritz)
Typical Four-Unit Canister Bank
Legend
0 i Filter canister
o
Air-operated automatic
"red valve"
Manual shutoff valve
Dial pressure gage
Flow meter
Fig. B-l. Ultrasonic filtration schematic
116
-------
EXHIBIT A
TYPICAL WASTEWATER CHARACTERISTICS (THROUGH PHASE III)
Range Average Total*
Flow, average mgd
PH
Turbidity, JTU
Color, mg/1
Suspended solids, rag/1
BOD, mg/1
COD, mg/1
0.58-2.90
3.7-10.10
32-3,000
15-200
40-600
75-765
150-1,150
1.8
5.5
600
--
300
180
500
--
--
--
--
4,500
2,700
7,500
* Pounds per day, based on average flow and concentration.
EXHIBIT B
TYPICAL WASTEWATER CHARACTERISTICS (AFTER PHASE IV)
Range Average Total*
pH
Suspended solids, mg/1
Turbidity, JTU
BOD, mg/1
COD, mg/1
6.4-7.8
2-48
3-27
12-61
61-202
6.9
20.4
11.6
31.5
120
—
338
--
528
2,123
* Pounds per day, based on average flow and concentration.
117
-------
TECHNICAL REPORT DATA
(Please read iHslruciions on the reverse before completing)
i. m com NO.
EPA-600/2-76-252
4. TITLL ANUSUUmLE
PAPERMILL WASTEWATER TREATMENT BY MICROSTRAINING
5. REPORT DATE
September 1976(Issuing date)
6. PERFORMING ORGANIZATION CODE
RECIPIENT'S ACCESSION-NO.
7. AUTHOR(S)
Frederick R. Bliss
8. PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORG '\NIZATION NAME AND ADDRESS
Strathmore Paper Company
Turners Falls, Massachusetts 01376
10. PROGRAM ELEMENT NO.
1BB037
11. CONTRACT/GRANT NO.
12040 FDE
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Laboratory - Cin., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/12
16. SUPPLEMENTARY NOTES
is. ABSTRACT
origina"| treatment system was designed, constructed, and operated for re-
moval of suspended solids, turbidity, color, and BOD from the wastewaters of two paper
mills which produce technical and other fine papers. The treatment process involves
coagulation and flocculation followed by microstraining. Space and cost considera-
tions were of paramount importance in selecting this process.
Fiber recovery was investigated, but was found to be uneconomical because of the
high percentage of fillers being employed and unacceptable levels of color and dirt.
The sludge is being discharged to the municipal sewerage system.
Plant operating efficiencies over the past year indicated substantial removal of
the suspended solids and 5-day BOD. Effluent turbidities averaged less than 30
Jackson turbidity units (JTU).
The estimated construction cost of the treatment facility is $689,000. First-
year operating costs including wages, power, supplies, chemicals, microfabric, and
maintenance totaled $36,175, which is approximately equivalent to $1.50 per ton of
paper produced.
It is expected that the techniques used in this operation may have broad applica-
bility to industries under similar space limitations and using similar manufacturing
methods, and that cooperative ventures will make it possible for many small firms to
survive when faced with meeting the new criteria for industrial wastewater discharges.
17.
a.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Coagulation, Flocculation, Pollution
Abatement, Waste Water Treatment
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Microstraining, Paper
Wastes Treatment,
Suspended Solids Separa-
tion, Paper Mill Fiber
Recovery
13/b
18. DISTRIBUTION STATEMENT
Unlimited Distribution
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
128
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA form 2220-1 (9-73) »«S.60»BIIIII£llin»IITIII60rFK£:H77-757-056/5500
118
------- |