&EPA
United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
EPA 600/2-79-055
July 1979
Research and Development
Chemical Primary
Sludge Thickening
and Dewatering
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution-sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-79-055
July 1979
CHEMICAL PRIMARY SLUDGE
THICKENING AND DEWATERING
by
David Di Gregorio
J. Brian Ainsworth
Keith J. Mounteer
Envirotech Corporation
Salt Lake City, Utah 84110
Contract No. 68-03-0404
Project Officer
Roland Villiers
Wastewater Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental
Research Laboratory, 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.
-------
FOREWORD
The Environmental Protection Agency was created because of in-
creasing public and government concern about the dangers of
pollution to the health and welfare of the American people.
Noxious air, foul water, and spoiled land are tragic testimony
to the deterioration of our natural environment. The complexity
of that environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem
solution and it involves defining the problem, measuring its
impact, and searching for solutions. The Municipal Environmental
Research Laboratory develops new and improved technology and
systems for the prevention, treatment, and management of waste-
water and solid and hazardous waste pollutant discharges from
municipal and community sources, for the preservation and treat-
ment of public drinking water supplies, and to minimize the
adverse economic, social, health, and aesthetic effects of
pollution. This publication is one of the products of that
research; a most vital communications link between the researcher
and the user community.
This report summarizes the results of a pilot plant program which
developed application and design criteria for thickening and de-
watering waste sludges produced from chemical clarification of
municipal wastewater. The development of such information pro-
vides valuable insight to engineers in their efforts to design
efficient, cost-effective wastewater treatment systems.
Francis T. Mayo
Director
Municipal Environmental
Research Laboratory
ill
-------
ABSTRACT
This report presents the results of a ten month study of the
thickening and dewatering characteristics of chemical-primary
sludges. Alum-primary and ferric-primary sludges were produced
in parallel trains of a pilot plant operated using a municipal
wastewater. Each chemical treatment unit was operated under
several coagulant doses during the four phases of this study
resulting in the production of several chemical-primary sludges
with distinct characteristics.
Gravity thickening and dissolved air flotation thickening results
for each chemical-primary sludge are presented. Gravity thick-
ening was evaluated using continuous, pilot scale gravity thick-
eners; dissolved air flotation thickening evaluations were per-
formed using batch, bench-scale equipment. Sludge dewatering
evaluations were performed for all chemical-primary sludges
using a pilot scale solid bowl centrifuge, vacuum belt filter
and filter press.
The report presents correlations developed relating performance
of each unit operation to specific characteristics identified
for each chemical-primary sludge. An economic analysis of cen-
trifugation and vacuum belt filtration of each chemical-primary
sludge is presented.
This report was submitted in fulfillment of EPA Contract Number
68-03-0404, by the Eimco Process Machinery Division of Envirotech
Sanitary Engineering Technology Department, under the (partial)
sponsorship of the Environmental Protection Agency. Work was
completed as of January 1976.
xv
-------
CONTENTS
Foreword iii
Abstract iv
List of Figures vii
List of Tables xv
Abbreviations and Symbols xix
Acknowledgements xxii
I Introduction 1
II Conclusions 3
III Recommendations 7
IV Experimental System 8
Pilot Plant Equipment 8
Bench-Scale Equipment 15
V Experimental Methods and Procedures 18
General 18
Pilot Plant Process Control Procedures 19
Sludge Thickening and Dewatering
Procedures 21
Sampling and Laboratory Procedures 29
VI Experimental Results and Discussion 31
Chemical Treatment 31
Dissolved Air Flotation Thickening 44
Gravity Thickening 74
Vacuum Filtration Dewatering 116
Centrifugal Dewatering 175
Pressure Filtration Dewatering 221
VII Dewatering Performance and Cost Comparison 229
References 245
Appendices
A. Dissolved Air Flotation Data Reduction
Procedures 246
B. Specific Resistance Determination
Procedures 250
-------
CONTENTS (continued)
C. Laboratory Analyses and Scheduled Tests 253
D. Sample Jar Test Data Sheet 256
E. Sludge Production Determinations 257
F. Sludge Characterization Determinations 260
G. Cost Analysis Assumptions 261
-------
FIGURES
N°_±. Page
1 Pilot Plant Flowsheet Schematic 9
2 Flocculating Clarifier Treatment Unit 11
3 Schematic of Solids Bowl Centrifuge 14
4 Dissolved Air Flotation Thickening Test
Apparatus 16
5 Bench-Scale Filter Leaf Test Apparatus 17
6 Sludge Thickening Test Apparatus 23
7 Sample Data Plot for Filter Press Cloth
Blinding Analysis 28
8 Effect of Lime Conditioning Dose on Sludge
Specific Resistance 30
9 Dissolved Air Flotation Thickening of Alum-
Primary Sludge from Phase 1 46
10 Dissolved Air Flotation Thickening of Alum-
Primary Sludge from Phase 2 48
11 Dissolved Air Flotation Thickening of Alum-
Primary Sludge from Phase 3 50
12 Dissolved Air Flotation Thickening of Ferric-
Primary Sludge from Phase 1 53
13 Dissolved Air Flotation Thickening of Ferric-
Primary Sludge from Phase 2 55
14 Dissolved Air Flotation Thickening of Ferric-
Primary Sludge from Phase 3-1 58
15 Dissolved Air Flotation Thickening of Ferric-
Primary Sludge from Phase 3-II 59
VI1
-------
FIGURES (continued)
No. Page
16 Dissolved Air Flotation Thickening of Primary
Sludge from Phase 4 62
17 Float Solids Concentration versus Fraction of
Chemical Solids in Alum-Primary Sludge 63
18 Solids Loading Rate versus Fraction of Chemical
Solids in Alum-Primary Sludge 64
19 Hydraulic Loading Rate versus Fraction of
Chemical Solids in Alum-Primary Sludge 65
20 Float Solids Concentration versus Fraction of
Chemical Solids in Ferric-Primary Sludge 66
21 Solids Loading Rate versus Fraction of Chemical
Solids in Ferric-Primary Sludge 67
22 Hydraulic Loading .Rate versus Fraction of
Chemical Solids in Ferric-Primary Sludge 68
23 Comparative Float Solids Concentration of
Alum-Primary and Ferric-Primary Sludges 70
24 Comparative Solids Loading Rates of Alum-
Primary and Ferric-Primary Sludges 71
25 Comparative Hydraulic Loading Rate of Alum-
Primary and Ferric-Primary Sludges 72
26 Alum-Primary Sludge Thickening Profiles -
Phase 1 76
27 Predicted TSL versus Thickener Feed for Alum-
Primary Sludge - Phase 1 77
28 Maximum Predicted Underflow Concentration
versus Thickener Feed Concentration -
Phase 1 78
29 Alum-Primary Sludge Thickener Profiles -
Phase 2 80
30 Predicted TSL versus Thickener Feed for
Alum-Primary Sludge - Phase 2 81
31 Maximum Predicted Underflow Concentration
versus Thickener Feed Concentration -
Phase 2 82
viii
-------
FIGURES (continued)
No. Page
32 Alum-Primary Sludge Thickener Profiles -
Phase 3 84
33 Predicted TSL versus Thickener Feed for
Alum-Primary Sludge - Phase 3 86
34 Maximum Predicted Underflow Concentration
versus Thickener Feed Concentration -
Phase 3 87
35 Alum-Primary Sludge Thickener Profiles -
Phase 4 90
36 Predicted TSL versus Thickener Feed for
Alum-Primary Sludge - Phase 4 91
37 Maximum Predicted Underflow Concentration
versus Thickener Feed Concentration -
Phase 4 92
38 Ferric-Primary Sludge Thickener Profiles -
Phase 1 95
39 Predicted TSL versus Thickener Feed for
Ferric-Primary Sludge - Phase 1 96
40 Maximum Predicted Underflow Concentration
versus Thickener Feed Concentration -
Phase 1 97
41 Ferric-Primary Sludge Thickener Profiles -
Phase 2 100
42 Predicted TSL versus Thickener Feed for
Ferric-Primary Sludge - Phase 2 101
43 Maximum Predicted Underflow Solids Con-
centration versus Thickener Feed Con-
centration - Phase 2
44 Ferric-Primary Sludge Thickener Profiles
Phase 3
45 Predicted TSL versus Thickener Feed for
Ferric-Primary Sludge - Phase 3
Phase 3 104
105
46 Maximum Predicted Underflow Solids Concen-
tration versus Thickener Feed Concentration -
Phase 3 106
IX
-------
FIGURES (continued)
No. Page
47 Primary Sludge Thickener Profiles - Phase 4 109
48 Predicted TSL versus Thickener Feed for
Primary Sludge - Phase 4 110
49 Maximum Predicted Thickener Underflow Con-
centration versus Thickener Feed Concentration
for Primary Sludge - Phase 4 111
50 Range of Thickener Operating Periods for
Alum-Primary Sludge 112
51 Range of Thickener Operating Periods for
Ferric-Primary Sludge 113
52 Form Filtration Rate versus Form Time
for Alum-Primary Sludge - Phase 1 117
53 Full-Scale Filtration Rate versus Cycle
Time for Alum-Primary Sludge - Phase 1 120
54 Filter Cake Solids Content verus Corre-
lating Factor for Alum-Primary Cludge -
Phase 1 121
55 Form Filtration Rate versus Form Time for
Alum-Primary Sludge - Phase 2 126
56 Full-Scale Filtration Rate versus Cycle
Time for Alum-Primary Sludge - Phase 2 128
57 Filter Cake Solids Content versus Corre-
lating Factor for Alum-Primary Sludge -
Phase 2 129
58 Form Filtration Rate versus Form Time
for Alum-Primary Sludge - Phase 3 134
59 Full-Scale Filtration Rate versus Cycle
Time for Alum-Primary Sludge - Phase 3 135
60 Filter Cake Solids Content versus Corre-
lating Factor for Alum-Primary Sludge -
• Phase 3 136
61 Form Filtration Rate versus Form Time
for Alum-Primary Sludge - Phase 4 140
-------
FIGURES (continued)
No. Page
62 Full-Scale Filtration Rate versus Cycle Time
for Alum-Primary Sludge - Phase 4 141
63 Filter Cake Solids Content versus Correlating
Factor for Alum-Primary Sludge - Phase 4 142
64 Form Filtration Rate versus Form Time for
Ferric-Primary Sludge - Phase 1 145
65 Full-Scale Filtration Rate versus Cycle Time
for Ferric-Primary Sludge - Phase 1 147
66 Filter Cake Solids Content versus Correlating
Factor for Ferric-Primary Sludge - Phase 1 148
67 Form Filtration Rate versus Form Time for
Ferric-Primary Sludge - Phase 2 153
68 Full-Scale Filtration Rate versus Cycle Time
for Ferric-Primary Sludge - Phase 2 154
69 Filter Cake Solids Content versus Correlating
Factor for Ferric-Primary Sludge - Phase 2 155
70 Form Filtration Rate versus Form Time for
Ferric-Primary Sludge - Phase 3 159
71 Full-Scale Filtration Rate versus Cycle Time
for Ferric-Primary Sludge - Phase 3 160
72 Filter Cake Solids Content versus Correlating
Factor for Ferric-Primary Sludge - Phase 3 161
73 Form Filtration Rate versus Form Time for
Primary Sludge - Phase 4 164
74 Full-Scale Filtration Rate versus Cycle Time
for Primary Sludge - Phase 4 166
75 Filter Cake Solids Content versus Correlating
Factor for Primary Sludge - Phase 4 167
76 Effect of Chemical Solids Content of Alum-
Primary Sludge on Form Filtration Rate 169
77 Filter Cake Solids Content versus Correlating
Factor for Alum-Primary Sludge Phases 170
78 Effect of Chemical Solids Content of Ferric-
Primary Sludge on Form Filtration Rate 172
xi
-------
FIGURES (continued)
No. Page
79 Filter Cake Solids Content versus Correlating
Factor for all Ferric-Primary Sludge Phases 173
80 Solids Capture from Centrifugal Dewatering of
Alum-Primary Sludge - Phase 1 177
81 Cake Solids Content from Centrifugal Dewatering
of Alum-Primary Sludge - Phase 1 178
82 Solids Capture from Centrifugal Dewatering of
Alum-Primary Sludge - Phase 2 181
83 Cake Solids Content from Centrifugal Dewatering
of Alum-Primary Sludge - Phase 2 182
84 Solids Capture from Centrifugal Dewatering
of Alum-Primary Sludge - Phase 3 186
85 Cake Solids Content from Centrifugal Dewatering
of Alum-Primary Sludge - Phase 3 187
86 Solids Capture from Centrifugal Dewatering
of Ferric-Primary Sludge - Phase 1 189
87 Cake Solids Content from Centrifugal De-
watering of Ferric-Primary Sludge - Phase 1 190
88 Solids Capture from Centrifugal Dewatering
of Ferric-Primary Sludge - Phase 2 193
89 Cake Solids Content from Centrifugal De-
watering of Ferric-Primary Sludge - Phase 2 194
90 Solids Capture from Centrifugal Dewatering
of Ferric-Primary Sludge - Phase 3 198
91 Cake Solids Content from Centrifugal De-
watering of Ferric-Primary Sludge - Phase 3 199
92 Solids Capture from Centrifugal Dewatering
of Primary Sludge 201
93 Cake Solids Content from Centrifugal De-
watering of Primary Sludge 202
94 Summary of Solids Capture from Centrifugal
Dewatering of Alum-Primary Sludges from
all Phases 203
XII
-------
FIGURES (continued)
No. Page
95 Summary of Cake Solids Content from Centrifugal
Dewatering of Alum-Primary Sludge from all
Phases 204
96 Effect of Chemical Solids Content of Alum-
Primary Sludge on Maximum Total Solids
Recovery Without Polymer 206
97 Effect of Chemical Solids Content of Alum-
Primary Sludge on Cake Solids Content 207
98 Summary of Solids Capture from Centrifugal
Dewatering of Ferric-Primary Sludge from 208
all Phases
99 Summary of Cake Solids Content From Centri-
fugal Dewatering of Ferric-Primary Sludge
from all Phases 209
100 Effect of Chemical Solids Content of Ferric-
Primary Sludge on Cake Solids Content 211
101 Volatile Solids Capture versus Total Solids
Capture for Centrifugal Dewatering of Alum-
Primary Sludge 212
102 Volatile Solids Capture versus Total Solids
Capture for Centrifugal Dewatering of Ferric-
Primary Sludge 213
103 Aluminum Capture versus Total Solids Capture
for Centrifugal Dewatering of Alum-Primary
Sludge 214
104 Iron Capture versus Total Solids Capture for
Centrifugal Dewatering of Ferric-Primary Sludge 215
105 Phosphorus Capture versus Total Solids Capture
for Centrifugal Dewatering of Alum-Primary Sludge 216
106 Phosphorus Capture versus Total Solids Capture
for Centrifugal Dewatering of Ferric-Primary
Sludge 217
107 Volatile Solids Capture versus Total Solids
Capture for Centrifugal Dewatering of
Primary Sludge 219
XI11
-------
FIGURES (continued)
No. Page
108 Phosphorus Capture versus Total Solids Capture
for Centrifugal Dewatering of Primary Sludge 220
109 Pressure Filter Form Time versus Cake
Solids Concentration for Primary Sludge 224
110 Summary of Pressure Filtration Results for
Alum-Primary Sludge - Phase 4 225
111 Summary of Pressure Filtration Results
for Alum-Primary Sludge - Phase 3 226
112 Summary of Pressure Filtration Results
for Primary Sludge 227
113 Comparative Costs for Centrifugal and Vacuum
Filtration Dewatering of Alum-Primary Sludge 238
114 Comparative Costs of Centrifugal and Vacuum
Filtration Dewatering of Ferric-Primary Sludge 239
115 Comparative Total Costs of Centrifugal and
Vacuum Filtration Dewatering plus Incineration
of Alum-Primary Sludge 240
116 Comparative Total Costs of Centrifugal and
Vacuum Filtration Dewatering plus Incineration
of Ferric-Primary Sludge 241
Al Sample Plot of Typical Dissolved Air
Flotation Thickening Data 249
Bl Sample Plot of Specific Resistance Data 252
xiv
-------
TABLES
— Page
1 Centrifuge Machine Parameters 13
2 Summary of Results of Filter Press Cloth
Blinding Analysis 27
3 Summary of Flocculating Clarifier Operating
Conditions 32
4 Chemical Treatment Unit Performance Summary
from Phase 1 Operations 33
5 Chemical Treatment Unit Performance Summary
from Phase 2 Operations 34
6 Chemical Treatment Unit Performance Summary
from Phase 3A Operations 35
7 Chemical Treatment Unit Performance Summary
from Phase 3F-I Operations 36
8 Chemical Treatment Unit Performance Summary
from Phase 3F-II Operations 37
9 Chemical Treatment Unit Performance Summary
from Phase 4 Operations 38
10 Post-Precipitation Effect in CT-F Effluent 39
11 Sludge Production Summary 41
12 Sludge Characterization Summary 43
13 Dissolved Air Flotation Thickening of Alum-
Primary Sludge from Phase 1 45
14 Dissolved Air Flotation Thickening of Alum-
Primary Sludge from Phase 2 47
15 Dissolved Air Flotation Thickening of Alum-
Primary Sludge from Phase 3 49
16 Dissolved Air Flotation Thickening of Ferric-
Primary Sludge from Phase 1 52
xv
-------
TABLES (continued)
No. Page
17 Dissolved Air Flotation Thickening of Ferric-
Primary Sludge from Phase 2 54
18 Dissolved Air Flotation Thickening of Ferric-
Primary Sludge from Phase 3-1 56
19 Dissolved Air Flotation Thickening of Ferric-
Primary Sludge from Phase 3-II 57
20 Dissolved Air Flotation Thickening of Primary
Sludge from Phase 4 61
21 Effect of Phosphorus Removal on Dissolved
Air Flotation Properties of Alum-Primary
and Ferric-Primary Sludge 73
22 Alum-Primary Sludge Thickening Average
Phase 1 Pilot Plant Results 75
23 Alum-Primary Sludge Thickening Average
Phase 2 Pilot Plant Results 75
24 Alum-Primary Sludge Thickening Average
Phase 3 Pilot Plant Results 83
25 Alum-Primary Sludge Thickening Average
Phase 4 Pilot Plant Results 88
26 Alum-Primary Sludge Thickening Average
Phase 4 Pilot Plant Results 89
27 Ferric-Primary Sludge Thickening Average
Phase 1 Pilot Plant Results 94
28 Ferric-Primary Sludge Thickening Average
Phase 2 Pilot Plant Results 98
29 Ferric-Primary Sludge Thickening Average
Phase 3 Pilot Plant Results 103
30 Primary Sludge Thickening Average Phase 4
Pilot Plant Results 108
31 Effect of Phosphorus Removal on Gravity
Thickening Properties of Alum-Primary and
Ferric-Primary Sludge 116
32 Data Summary of Vacuum Filtration of Alum-
Primary Sludge from Phase 1 118
xvi
-------
TABLES (continued)
No. Page
33 Data Summary of Vacuum Filtration of Alum-
Primary Sludge from Phase 2 123
34 Data Summary of Vacuum Filtration of Alum-
Primary Sludge from Phase 3 130
35 Data Summary of Vacuum Filtration of Alum-
Primary Sludge from Phase 4 138
36 Data Summary of Vacuum Filtration of Ferric-
Primary Sludge from Phase 1 143
37 Data Summary of Vacuum Filtration of Ferric-
Primary Sludge from Phase 2 149
38 Data Summary of Vacuum Filtration of Ferric-
Primary Sludge from Phase 3 157
39 Data Summary of Vacuum Filtration Leaf Tests
of Primary Sludge from Phase 4 162
40 Data Summary of Centrifugal Dewatering of
Alum-Primary Sludge from Phase 1 176
41 Data Summary of Centrifugal Dewatering of
Alum-Primary Sludge from Phase 2 179
42. Data Summary of Centrifugal Dewatering of
Alum-Primary Sludge from Phase 3 184
43 Data Summary of Centrifugal Dewatering of
Ferric-Primary Sludge from Phase 1 188
44 Data Summary of Centrifugal Dewatering of
Ferric-Primary Sludge from Phase 2 191
45 Data Summary of Centrifugal Dewatering of
Ferric-Primary Sludge from Phase 3 196
46 Data Summary of Centrifugal Dewatering of
Primary Sludge from Phase 4 200
47 Summary of Pressure Filtration Results 222
48 Summary of Sludge Characterization from
Chemical Treatment of Municipal Wastewater 230
49 Summary of Dewatering Performance for Chemical-
Primary Sludges 231
xvii
-------
TABLES (continued)
No. Page
50 Cost Summary of Vacuum Filtration of Alum-
Primary Sludge 233
51 Cost Summary of Vacuum Filtration of Ferric-
Primary Sludge 234
52 Cost Summary of Centrifugation of Alum-Primary
Sludge 235
53 Cost Summary of Centrifugation of Ferric-
Primary Sludge 236
XVlll
-------
ABBREVIATIONS AND SYMBOLS
Al aluminum available after its utilization in
, mg/£
aluminum sulfate
A/S air to solids ratio, weight percent
BOD5 5-day biochemical oxygen demand, mg/£
C conditioned feed concentration to filter press,
C degrees Celsius
Ci thickener feed concentration, g/&
Cu thickener underflow concentration, g/£
COD chemical oxygen demand, mg/X-
CT chemical treatment unit - an abbreviation for
flocculating clarifier
CT-A flocculating clarifier in which alum was the
coagulant
CT-F flocculating clarifier in which ferric chloride
was the coagulant
Di cake dry solids weight, g
DAF ' dissolved air flotation
DC direct current
DS dry suspended solids
E excellent discharge of vacuum filter cake
F fair discharge of vacuum filter cake
xix
-------
Fer
FFR
FSFR
Ferric
Chloride
HLR
JTU
kw
ND
O&M
P
Phase
PH
Q
AP
rpm
Arpm
Sc
SBOD5
ABBREVIATIONS AND SYMBOLS (continued)
degrees Fahrenheit
iron available after its utilization in FePCK , mg/£
form filtration rate, Kg/hr-sq m
full-scale filtration rate, Kg/hr-sq m
a source of Fe+++ in liquid form
centrifugal force when used in centrifuge
description
good discharge of vacuum filter cake when used in
a vacuum filter description
hydraulic loading rate, cu m/sq m-hr
Jackson Turbidity Units
kilowatt
non-dischargeable vacuum filter cake
operating and maintenance
poor discharge of vacuum filter cake
a time period under which flocculating clarifier
was operated at selected chemical and/or polymer
dosages
negative logarithm of the hydrogen ion concentration
liquid flowrate
difference in feed and filtrate densities, g/i
revolutions per minute
difference between centrifuge bowl rpm and conveyor
rpm
cake dry solids, weight percent
soluble 5-day biochemical oxygen demand, mg/£
xx
-------
ABBREVIATIONS AND SYMBOLS (continued)
SCOD soluble chemical oxygen demand, mg/&
SLR solids loading rate, Kg/sq m-hr
TDS ton of dry suspended solids
TS total solids, g/fc
TSL thickener solids loading, Kg/day-sq m
TSS total suspended solids, mg/J,
6 a measurement of time
Q
D vacuum filter dry time, minutes
n
D/W a correlating factor used in describing vacuum
filter performance, min-sq m/Kg
2
F vacuum filter form time, minutes
AV final filtrate volume minus filtrate volume at
cycle time, t
W cake dry weight per unit area, Kg/sq m
Wi wet cake weight at end of test, g
xxi
-------
ACKNOWLEDGEMENTS
The support and indulgence of the Salt Lake City Commissioner of
Water and Sewers, city engineers and municipal sewage pumping
plant personnel is acknowledged.
Special acknowledgement is due Messrs. Doug Hendry, Steve Nelson,
Jack Otto, and Charles Thompson who nursed the pilot plant
through start-up and kept it operating. Mr. Hendry was of great
assistance in preparing the figures.
The support, counsel and patience of the Project Officer, Mr.
Roland Villiers, and Dr. Joseph Farrell of the Water Research
Division, Municipal Environmental Research Laboratory, Environ-
mental Protection Agency, Cincinnati, Ohio, is acknowledged with
sincere thanks.
xx 11
-------
SECTION I
INTRODUCTION
The discharge of phosphorus-containing wastewater into the sur-
face water of the United States has contributed to their over-
fertilization and eutrophication. As a result, efforts are now
being made to remove phosphorus from wastewater.
Phosphorus removal from raw wastewater is normally achieved by
precipitating it from solution with a metal salt of iron, alum-
inum or calcium. The use of phosphorus removal metal not only
increases the amount of sludge produced but the character of
sludge is changed. This results in thickening and dewatering
properties different from raw-primary sludge. It is important
that means of thickening and dewatering these combined chemical-
primary sludges be investigated and determined so equipment
sizing parameters can be determined.
In March, 1974, EPA and the Eimco Process Machinery Division of
Envirotech Corporation initiated a research contract to operate
a pilot plant to generate various chemical-primary sludges, and
to obtain performance data describing the thickening and dewater-
ing characteristics of these sludges. The project, entitled
"Chemical-Primary Sludge Thickening and Dewatering," was con-
ducted in facilities originally constructed in 1969 where two
prior major studies on physical-chemical treatment of municipal
wastewater had been conducted.1'2 This latest study was conduct-
ed by generating chemical-primary sludges, produced from either
alum addition or ferric chloride addition to effect phosphorus
removal from municipal wastewater, thickening the resultant
underflow sludge in gravity thickeners, and then dewatering it
by vacuum filtration or centrifugation (and in a few instances
by pressure filtration).
The general objectives of the study, as originally established,
were as follows:
1. To determine the thickening and dewatering character-
istics of chemical-primary sludges generated from
chemical clarification of municipal wastewater using
alum or ferric chloride, each at two levels of phos-
phorus removal.
-------
2. To establish sludge conditioning chemical requirements
for dewatering alum and ferric-primary sludges by
vacuum filtration and centrifugation.
3. To perform a comparison of centrifugal and vacuum fil-
tration dewatering of chemical-primary sludges generat-
ed from each operating condition during this study.
An important product of the investigation was to be an evaluation
of the impact of dewatering cost on the phosphate removal by
Fe+++ or A1+++ salts.
-------
SECTION II
CONCLUSIONS
1. Sludge production in the chemical-primary flocculating
clarifier was predicted with reasonable accuracy from
analysis of physical and chemical constituents around the
chemical-primary unit and knowledge of chemical coagulant
dose.
2. No evidence was found to indicate that the addition of poly-
mer to the primary chemical treatment unit had any effect
on subsequent sludge thickening or dewatering operations.
3. Dissolved air flotation thickening produced satisfactory
sludge solids concentration and solids capture without the
need for chemical flotation aids.
4. Dissolved air flotation thickening performance was adversely
affected by the presence of aluminum or iron chemical solids
in the feed sludge to the flotator. Allowable loading rates
and float solids concentration generally decreased as the
amount of chemical solids in the feed sludge increased.
5. Dissolved air flotation thickening performance characteris-
tics of alum-primary sludge were superior to those of ferric-
primary sludge at equivalent levels of phosphorus removal
of 80 and 95 percent.
6. Dissolved air flotation thickening of alum-primary sludge,
ferric-primary sludge and primary sludge produced thickened
sludge concentrations of 2.8-5.5, 3.5-4.5 and 7.0-8.0
weight percent, respectively depending upon the amount of
chemical solids present in the flotator feed sludge, and
the air to solids ratio chosen.
7. Gravity thickener underflow solids concentration was ad-
versely affected by the presence of aluminum or iron
chemical solids in the feed sludge to the thickener. Under-
flow solids concentration generally decreased as the amount
of chemical solids in the feed sludge increased. The
decrease in thickener underflow solids concentration with
increasing chemical solids in the feed sludge was more
significant for alum-primary sludge than ferric-primary
sludge.
-------
8. Gravity thickener performance characteristics of ferric-
primary sludge were superior to those of alum-primary sludge
at equivalent levels of phosphorus removal of 80 and 95
percent.
9. Thickened alum-primary sludge solids concentrations from
flotation thickening were higher than those achieved with
gravity thickening while the reverse was true for ferric-
primary sludge.
10. Gravity thickening of alum-primary, ferric-primary and
primary sludge produced thickened sludge concentrations of
2.5-4.5, 4.5-6.0 and 7.0-10.0 weight percent, respectively,
depending upon the amount of chemical solids present in the
thickener feed sludge and the thickener operating conditions
chosen.
11. Vacuum filtration performance, relative to required chemical
conditioning dose, filtration rate and filter cake solids
content, was adversely affected by the presence of aluminum
chemical solids in the feed sludge to the filter. The re-
quired chemical conditioning dose increased and the filtra-
tion rate decreased as the quantity of aluminum chemical
solids in the feed sludge increased. Cake solids content
was insensitive to the amount of aluminum chemical solids
present in the filter feed sludge with the exception that
primary sludge dewatered to cake solids content levels
higher than alum-primary sludge.
12. Vacuum filtration performance, relative to required chemical
conditioning dose and filtration rate, was adversely affect-
ed by the presence of iron chemical solids in the feed
sludge to the filter. In general, required chemical condi-
tioning dose increased and filtration rate decreased as the
quantity of iron chemical solids in the feed sludge increas-
ed. Cake solids content was insensitive to the amount of
iron chemical solids present in the filter feed sludge with
the exception that ferric-primary sludge dewatered to cake
solids content levels higher than primary sludge.
13. Volatile solids, phosphorus and the primary coagulant metal
were efficiently captured during vacuum filtration of alum-
primary or ferric-primary sludge.
14. Vacuum filtration of alum-primary, ferric-primary and
primary sludge produced filter cakes of 25-27, 34-35 and
26-29 weight percent total solids, respectively, depending
upon the filter operating conditions chosen.
-------
15. Centrifugal dewatering performance, relative to polymer re-
quirements and cake solids concentrations, was adversely
affected as the quantity of aluminum chemical solids present
in the feed sludge increased. Machine capacity, to achieve
a given level of solids capture, was not significantly
affected by the presence of aluminum chemical solids in the
feed sludge.
16. Centrifugal dewatering performance, relative to polymer re-
quirements and machine capacity to achieve a given level of
solids capture, was not affected by the quantity of iron
chemical solids present in the feed sludge. Cake solids
concentrations increased as the amount of iron chemical
solids present in the feed sludge increased.
17. Centrifugal dewatering of ferric-primary sludge produced
cake solids concentrations significantly higher than those
achieved with alum-primary sludge.
18. Centrifugal dewatering of alum-primary, ferric-primary and
primary sludge produced cakes of 15-18, 22-25 and 20-21
weight percent total solids, respectively, at total solids
capture of approximately 95 percent.
19. Volatile solids, phosphorus and the primary coagulant metal
were efficiently captured during centrifugal dewatering of
alum-primary or ferric-primary sludge at 90-95 percent
total solids capture.
20. Pressure filtration cake solids concentrations were adversely
affected by the amount of aluminum chemical solids in the
feed sludge to the press.
21. Pressure filtration of alum-primary sludge produced cakes
ranging from 25-35 weight percent total solids and 30-41
weight percent total solids for primary sludge depending
upon cake thickness, cycle time, and the amount of aluminum
chemical solids present in the feed sludge to the press.
22. It was costlier by 10-20 percent to centrifugally dewater
and 15-25 percent to vacuum filter dewater alum-primary
or ferric-primary sludge produced from chemical treatment
aimed at 95 percent phosphorus removal as compared to 80
percent phosphorus removal. This was caused by the in-
creased quantities of sludge and a generally more difficult
sludge to dewater for the 95 percent phosphorus removal
case.
23. There was no significant difference between centrifugal
and vacuum filtration dewatering costs for either alum-
primary or ferric-primary sludge produced from chemical
treatment aimed at 95 percent phosphorus removal. The
-------
total costs (dewatering plus cake incineration) were 25-30
percent higher when centrifugation was used as compared to
vacuum filtration.
24. Centrifugal dewatering was 10-15 percent more expensive than
vacuum filtration of either alum-primary or ferric-primary
sludge produced from chemical treatment aimed at 80 percent
phosphorus removal. The total costs (dewatering plus cake
incineration) were 10-35 percent higher when centrifugation
was used as compared to vacuum filtration for the alum-
primary sludge case. The total costs (dewatering plus cake
incineration were 30-45 percent higher when centrifugation
was used as compared to vacuum filtration for the ferric-
primary sludge case.
25. The total costs for disposal (dewatering plus cake incinera-
tion) of alum-primary sludge produced from 95 percent phos-
phorus removal were approximately 25 percent higher when
centrifugation was used as compared to vacuum filtration.
The cost differential increased to approximately 30 percent
for the case of ferric-primary sludge produced from 95 per-
cent phosphorus removal.
26. The total costs for disposal (dewatering plus cake incinera-
tion) of alum-primary sludge produced from 80 percent phos-
phorus removal averaged approximately 20 percent higher
when centrifugation was used as compared to vacuum filtra-
tion. The cost differential increased to approximately
40 percent for the case of ferric-primary sludge produced
from 80 percent phosphorus removal.
-------
SECTION III
RECOMMENDATIONS
1. Operating control techniques should be developed for pre-
venting addition of excessive amounts of primary coagulant
(alum or ferric chloride) to raw wastewater for specific
levels of phosphorus removal. The need for these controls
was demonstrated during this study from documented deteriora-
tion in thickening and dewatering properties of chemical-
primary sludges as the dosage of primary coagulant increased.
2. Further studies would appear to be warranted in the area of
gravity thickening of chemical-primary sludges using organic
flocculants.
3. Further study of alternative dewatering devices should be
performed, particularly pressure filtration and horizontal
belt filtration.
-------
SECTION IV
EXPERIMENTAL SYSTEM
PILOT PLANT EQUIPMENT
The pilot facility consisted of two identical chemical treatment
and sludge thickening trains. Raw wastewater, following degritt-
ing in a hydrocyclone and screening through a 12.7 mm (.50 inch)
opening mesh screen, was contacted with alum in one train and
with ferric chloride in the other. The chemical primary sludges
produced were gravity thickened. The sludges were then dewater-
ed by vacuum filtration, centrifugation, or pressure filtration.
A schematic of the pilot plant flowsheet is shown on Figure 1.
An auxiliary line tapped into a discharge header of the Metro-
politan Salt Lake City pump station supplied raw wastewater to
the pilot plant. Approximately 1000 5,/minute (265 gpm) of raw
wastewater was pumped through a 15.2 cm (6 inch) diameter Hydro-
cyclone* for grit removal. The degritted wastewater was dis-
charged to a 1500 H (400 gallon) wet well and screened through
a 12.7 mm (0.50 inch) opening wire mesh screen. Since only 150-
300 A/minute (40-80 gpm) of degritted screened wastewater was
used for pilot plant operations, the majority of the flow enter-
ing the wet well overflowed to a drain and returned to the pump
station. Hydrocyclone operation was continuous, whereas routine,
manual cleaning of the wire mesh was required.
Degritted, screened wastewater was pumped from the wet well to
the pilot plant where the flow was split into two streams. Con-
trol of the flow to each chemical treatment unit was achieved
by adjustments of the feed pump speed and was fine-tuned by
throttling gate valves installed in each line.
Feed flow measurement for the ferric chemical treatment unit
(CT-F) was achieved using a magnetic flow meter. The magnetic
flow meter signal was directed to a recording flow rate indica-
tor and flow totalizer.
Because of continual delays in delivery of a second magnetic
flow meter, an orifice plate and mercury manometer were used to
Wemco Division, Envirotech Corporation
-------
FLASH MIX
CHEMICAL
CT-F
MAG.
FLOWMETER
FLASH MIX.
CHEMICAL
GRIT
REMOVAL
JCJ£_^JU
TO TK-F
TO TK-A
ORIFICE
PLATE
FEEDWELL
RAW SEWAGE PUMP
TK-A
TO SLUDGE DEWATERING
FEED TANK
S-F
FROM B D-F
TO DRAIN
LEGEND^
CT-CHEMICAL TREATMENT
B/D-BLOW DOWN
TK-THICKENER
S-STORAGE
0-OVERFLOW
SLUDGE DEWATERING
SLUDGE FEED-
FLOCCULATOR
VACUUM FILTER
POLYMER
CENTRIFUGE
SLUDGE TANK
CAKE
LIME
FEED
B
V
CAKE
'//////
CENTRATE TO DRAIN
SLUDGE FEED
FIGURE 1. PILOT PLANT FLOWSHEET SCHEMATIC.
-------
measure the rate of flow in the alum chemical treatment unit
(CT-A) feed line. Sixty degree V-notch weirs were used to verify
the flow rate to each chemical treatment unit. The weirs were
located in the effluent overflow box of each unit.
The two CT units used in this pilot plant study were flocculator-
clarifier type chemical treatment units. Each was 3.05 m (10 ft)
in diameter equipped with a variable-speed vertical paddle wheel
in the flocculation zone.
Chemical treatment of wastewater involved the following processes
(see Figure 2). Wastewater entered a 210 £. (55 gallon) flash
mix tank (fitted with a 175 rpm mixer) from the bottom as did
the chemical being used for coagulation; retention time in the
flash mix tank was 2 minutes for a clarifier loading of 0.204
£ps/sq m (0.20 gpm/sq ft). The rapid mix tank was fitted with an
overflow pipe through which chemically coagulated wastewater
flowed to the flocculating clarifier. When polymer was being
used, it was fed into this overflow pipe where coagulated waste-
water entered the flocculation zone of the flocculating clarifier.
Each flocculating clarifier consisted of a flocculation zone and
a clarification zone. Hydraulic retention time in the floccula-
tion zone was 1.8 hours for a loading rate of 0.204 £pm/sq m
(0.30 gpm/sq ft). For this same loading rate, the clarification
hydraulic retention time was 4.8 hours. Flocculation was effect-
ed by a "paddle wheel" mounted vertically inside the flocculation
zone. The paddle flocculator was connected to a variable speed
drive for control of rotational speed as required.
A raking mechanism (also controlled by a variable speed drive)
continually brought the thickened sludge to a center cone for
withdrawal. A plate was mounted directly on top of the raking
mechanism to minimize hydraylic-induced turbulence in the sludge
thickening and withdrawal zone.
Four sample taps were located in the sidewall of each floccula-
ting clarifier. These taps were located at 36 cm (14 inch),
79 cm (31 inch), 99 cm (39 inch) and 150 cm (59 inch) above the
floor of each flocculating clarifier and aided in its operation
by allowing plant operators to withdraw sludge samples from them.
A launder around the perimeter of the flocculating clarifier
carried the effluent to a collection box; from the collection
box the effluent gravity flowed to drain. Two sampling pipes
were located immediately ahead of the collection box. One pipe
supplied effluent to the composite sampler and the other supplied
effluent to the turbidimeter. On the discharge side of the coll-
ection box, a V-notch weir was located. Whenever weir measure-
ments were taken to check the flow rate, both of the sample pipes
were closed.
10
-------
COAGULANT
FEED
SEWAGE
3-<-FEED
FIGURE 2. FLOCCULATING CLARIFIER TREATMENT UNIT.
11
-------
Thickened sludge was removed from each CT unit and fed to a
gravity thickener on a timed basis. The thickener for ferric-
primary sludge was 107 cm (42 inch) diameter, and that for alum
primary sludge was 91 cm (36 inch) diameter. Each thickener had
a side wall depth of 152 cm (60 inch), a 45 cone on the bottom,
was equipped with pickets and rakes, and had an automatic de-
sludging valve.
The overflow from each gravity thickener flowed to a separate
storage tank. Each storage tank was 152 cm (60 inch) diameter
with a capacity of 3560 £ (940 gallon). As required, the con-
tents of these tanks were mixed, inventoried and manually
drained.
Thickened sludge from each gravity thickener was removed on a
timed basis and retained in storage tanks. Each tank was 152 cm
(60 inch) diameter with a capacity of 2910 I (770 gallon). As
required, the contents of these tanks were mixed, inventoried
and transferred for dewatering studies.
The vacuum filter* used in the dewatering studies was 91.4 cm
(36 inch) in diameter by 30.5 cm (12 inch) wide with a continuous
belt. The filter drum and vat were made of a thermoplastic
material. The vacuum filter, with flocculator, vacuum pump,
filtrate pump, and receiver was mounted on a common platform.
A 210 £ (55 gallon) tank equipped with a vertical rapid mixer
and positive displacement pump coupled to a variable speed drive
served as the conditioning lime slurry feed system.
Centrifugal dewatering of all sludges generated during this
study was evaluated using a 15.2 cm (6.0 inch) diameter, hori-
zontal scroll, solid bowl machine.** Pertinent centrifuge
machine parameters are listed in Table 1 and illustrated on
Figure 3.
The filter press used was a 30 cm (12 inch) polypropylene, center
>feed, corner discharge unit with ratchet closing. Maximum
recommended operating pressure was 10.5 Kg/sq cm (150 Ib/sq in).
The filter press was equipped with two chambers; cake thickness
was variable from 2.5 cm to 3.8 cm (1.0-1.5 inch) by using
various combinations of chamber spacers. Sewn-center cloths
were used, with the area of one side of a cloth being approxi-
mately 500 sq cm (0.54 sq ft). The skid-mounted filter press
system included an air compressor (maximum pressure = 14 Kg/sq
cm (200 Ib/sq in)) which functioned to pressurize the 115 £ (30
gallon) feed tank displacing feed sludge into the press chambers.
*Eimco-Belt Filter, Eimco PMD of Envirotech Corporation, Salt
Lake City, Utah
**P600 Sharpies Super Decanter, Sharpies Division of Pennwalt
Corporation, Warminster, Pennyslvania
12
-------
TABLE 1
CENTRIFUGE MACHINE PARAMETERS
Sharpies Model P600
Parameter
Bowl diameter (cm)
Pool length* (cm)
Pool depth* (cm)
Beach length* (cm)
Beach angle (degrees)
Pool volume* (&)
Conveyer pitch (cm)
Bowl rpm
Nominal G
Differential rpm
Values
15.2
31.4
1.27
2.54
10.0
1.56
5.08
5000
2100
10-25
Corresponds to No. 3 pool setting
13
-------
SOLIDS
CENTRATE
FIGURE 3. SCHEMATIC OF SOLID BOWL CENTRIFUGE.
14
-------
The feed tank was mounted on a "rocking" mechanism which was used
to keep the sludge solids in suspension. An air pressure regula-
tor was located in the line between the compressor and feed tank
to provide for a controlled pressure build-up as desired.
BENCH-SCALE EQUIPMENT
Dissolved air flotation (DAF) thickening was evaluated using a
batch, bench-scale dissolved air flotation device. DAF thicken-
ing evaluations were performed on all sludges generated during
this pilot plant study. A 61.5 sq cm (0.066 sq ft) bench-scale
batch DAF device, as shown on Figure 4, was used for the evalua-
tions. The flotation device consisted of a calibrated 1 I flo-
tation chamber, pressurization chamber and ancillary pressure
controls, piping and sampling tap.
Vacuum filtration of the primary sludge generated during Phase 4
was evaluated using a bench-scale, 93 sq cm (0.1 sq ft) filter
leaf. As schematically shown on Figure 5, the test apparatus
consisted of a vacuum source and receiver, filter leaf and
sludge slurry container.
15
-------
AIR BLEED
PRESSURE GAUGE
SUBNATANT
SAMPLE
TAP
FLOTATION
CHAMBER
RECYCLE
PRESSURIZATION
CHAMBER
FLOW
DISTRIBUTOR
K
DRAIN VALVE
I
RECYCLE VALVE
AIR SUPPLY
FIGURE 4. DISSOLVED AIR FLOTATION THICKENING
TEST APPARATUS.
16
-------
Flexible
Vacuum Hose
To Vacuum
Filter
Leaf
FIGURE 5. BENCH SCALE FILTER LEAF TEST APPARATUS.
-------
SECTION V
EXPERIMENTAL METHODS AND PROCEDURES
GENERAL
The study was divided into four phases. Each phase consisted of
chemically treating raw wastewater with alum in one train of the
pilot plant and with ferric chloride in the other. Variations
in the properties of the chemical-primary sludges produced were
achieved by varying the chemical dosage and/or with the use of
polymer during chemical treatment.
The approach used in pilot plant operation under each phase in-
volved establishment of stable chemical treatment operating con-
ditions followed by pilot plant equipment performance evalua-
tions. Chemical treatment operating conditions were established
in accordance with the following specific program objectives:
Phase 1 (P-l)
1. High alum dosage, no polymer
2. High ferric chloride dosage, no polymer
Phase 2 (P-2)
1. High alum dosage, with polymer
2. High ferric chloride dosage, with polymer
Phase 3 (P-3)
1. Low alum dosage, with polymer
2. Low ferric chloride dosage, with polymer
Phase 4 P-4)
1. High alum dosage, with polymer
2. No chemical addition, no polymer addition; primary
sludge
High chemical dosage was defined as the minimum chemical dosage
adequate to achieve approximately 95 percent phosphorus removal,
Low chemical dosage was defined as the minimum chemical dosage
adequate to achieve approximately 80 percent phosphorus removal,
However, experience has shown that a chemical dosage sufficient
18
-------
to effect 80 percent phosphorus removal could result in poor eff-
luent clarity. Therefore, either 80 percent phosphorus removal
or acceptable effluent clarity was established as the criterion
for determining the low chemical dosage. From previous studies2,
an effluent suspended solids concentration of 30 mg/& or less
was considered acceptable.
PILOT PLANT PROCESS CONTROL PROCEDURES
Chemical Treatment Unit Operation
Wastewater was metered to a flash mix tank ahead of each chemical
treatment unit where it was contacted with inorganic coagulant.
Either liquid alum or liquid ferric chloride was used. The
liquid alum (8.25 weight percent as AlaOs) was stored in a 2500 £
(660 gallon), rubber lined, steel storage tank, and fed to the
alum chemical treatment unit (CT-A) using a diaphragm pump*
driven by a variable speed DC motor. The liquid ferric chloride
(15.2 weight percent as Fe) was stored and fed in an identical
manner.
When polymer was used, it was added immediately before the con-
tents of the flash mix tank entered the flocculation zone of the
chemical treatment unit. Polymer was prepared by mixing approxi-
mately 200 g of anionic polymer with 200 £ (53 gallon) water in
a 210 H (55 gallon) container and vigorously agitating for one-
half hour. Following mixing, polymer solution was gravity drain-
ed to a 210 H (55 gallon) container where it was metered to the
chemical treatment unit using a pump identical to that used for
inorganic chemical feeding.
Feed was pumped from the wet well to each flocculating-clarifier
by a centrifugal pump controlled by a DC motor. A magnetic flow-
meter was installed in one feed line, and an orifice plate with
mercury manometer was installed in the other. Sixty degree V-
notch weirs were located in the discharge box of each floccula-
ting-clarifier .
By setting the proper flow through the feed line equipped with
the magnetic flowmeter, effluent flow over the V-notch weir was
noted and this measurement was used to set the same flow to the
unit without a magnetic flowmeter installed in the feed line.
When installation of a second magnetic flowmeter proved impracti-
cal, due to a long delivery time, an orifice plate and mercury
manometer were installed in the feed line. This system worked
well; sewage flow was checked several times per day by viewing
the manometer. The V-notch weirs were then used as a secondary
check.
*Wallace and Tiernan, Model 44-213
19
-------
Flocculation within the CT unit was developed by slow rotation
of the paddle assembly in the baffled flocculation chamber. The
flocculated wastewater flowed out from the bottom of the floccu-
lation chamber into a clarification chamber. A portion of the
overflow from each CT unit flowed through separate turbidimeters*
which continuously measured and recorded effluent turbidity.
Chemical treatment unit operation consisted mainly of routine
monitoring of feed, chemical and polymer flow rates. Attempts
were made to maintain a constant sludge blanket level in each CT
unit. Samples were routinely withdrawn from the sample taps
located in the sidewall of the clarification zone and settled in
2 a graduated cylinders. Sludge volume after a 10 minute settl-
ing period was recorded for each sample. The results of this
test were used by the operators to adjust the timer-controlled
sludge blowdown system. As the operators gained experience in
interpreting these settling tests, they were able to predict
(within one-half percent) the underflow solids concentration re-
moved from the flocculating clarifier. The basic objective in
attempting to maintain a constant sludge blanket level in each
CT unit was to retain the sludge in the thickening zone for as
short a period of time as practical, consistent with achieving
reasonably thick underflow sludge. In practice, maintaining a
sludge blanket depth of approximately one foot at the sidewall
resulted in a solids retention time of less than one day without
measureable diminution of underflow concentration due to the
minimal sludge depth.
Thickener Operation
Each chemical treatment unit was equipped with a timer-controlled,
pneumatic actuated de-sludge valve that allowed sludge to be
withdrawn and flow to a surge tank. From the surge tank, sludge
was fed directly to the thickener, again on a timed basis; the
contents of each surge tank were mixed at timed intervals coin-
cident with sludge drainoff from each surge tank.
Pilot plant operation during most of the phases was such that
more sludge was produced than could be thickened with the exist-
ing equipment. Therefore, when either surge tank became full,
its contents were inventoried (for subsequent sludge production
calculations) and wasted.
Sludge was pumped to each gravity thickener on an intermittent,
timed basis. The thickener feeding schedule was modified as re-
quired to maintain a predetermined solids loading rate to each
thickener by selecting the frequency and duration of the feed
schedule with electric timers. The operating procedure estab-
lished for gravity thickener operation involved routine sludge
*Hach, Model 1720 Low Range
20
-------
blanket level monitoring with thickener sludge blowdown timer
adjustments as necessary to maintain a constant thickener sludge
blanket level.
Both thickener underflow and supernatant flows were inventoried
in separate storage tanks to permit calculation of pertinent
thickener operating and performance parameters.
SLUDGE THICKENING AND DEWATERING PROCEDURES
Dissolved Air Flotation Thickening
Bench-scale dissolved air flotation (DAF) thickening evaluations
were performed on all sludges generated during this pilot plant
study. Chemical flotation aids were not used at any time during
the DAF study.
Process supernatant was aerated at 4.6 Kg/sq cm (65 Ib/sq in)
gauge pressure for approximately 15 minutes within the pressuriza-
tion chamber in order to saturate the liquid with air. A contin-
uous air bleed from the pressurization chamber was maintained
such that the compressed air sparged into the liquid violently
mixed the liquid during the saturation period. After approxi-
mately 15 minutes of pressurization, the air bleed and compressed
air supply were simultaneously turned off such that the satura-
ted liquid gauge pressure was maintained at 4.6 Kg/sq cm (65 lb/
sq in).
The sample volume of sludge to be thickened was immediately
placed in the flotation chamber. A valve was then opened allow-
ing a predetermined volume of saturated liquid to quickly flow
from the pressurization chamber into the flotation chamber and
intimately mix with the sludge. After this operation was com-
plete, timed measurements of sludge interface height were re-
corded for a period of 15 minutes. During the initial stages
of flotation, interface height measurements were recorded every
10 seconds, whereas, measurements at one minute intervals were
sufficient near the end of the flotation period.
At the end of the flotation period, samples of float and subna-
tant were collected. Float was sampled from the float layer pro-
duced and subnatant was collected from the sample tap located
in the subnatant zone. Total suspended solids were then perform-
ed on the samples of feed, float and subnatant.
Data reduction for all flotation tests was in accordance with
the procedures outlined in Appendix A. In addition to solids
loading rate, the limiting rise rate as calculated from the
solids loading rate and bulk separation rate were also determin-
ed. All DAF data presented represents predicted, maximum full-
scale performance. The scale up factors used were developed from
Eimco-PMD experience and are indicated in Appendix A.
21
-------
Gravity Thickening
All sludges generated during this study were thickened in pilot
scale gravity thickeners. Sludge was discharged at 75-100 Ji/min
(20-26 gal/min) from each flocculating clarifier into separate
surge tanks on an automatically, timer-controlled basis. Sludge
discharge duration was adjusted routinely to maintain a reason-
ably constant sludge blanket level in each flocculating clarifier.
Sludge was pumped at 5.7-7.5 £/min (1.5-2.0 gal/min) for 1.0-2.5
minute durations at 15-30 minute intervals from the surge tanks
to each thickener. Specific pumping intervals and durations
were adjusted routinely by resetting timers in order to maintain
reasonably constant thickener solids loading rates.
Routine daily data collection pertinent to the gravity thickening
units involved determinations of feed solids concentrations, over-
flow and underflow volumes and thickener sludge depth. Underflow
and overflow suspended solids concentrations were determined
intermittently when the underflow and overflow storage tanks
were full.
Solids balances around the thickeners using operating data collec-
ted over extended periods of time (several days to several weeks)
permitted calculation of average thickener solids loading rate,
solids capture and underflow solids concentration.
The approach used to identify pilot thickener performance was
to monitor feed and underflow suspended solids concentration,
sludge depth and thickener solids loading rate. Thickener per-
formance was identified from periods of stable thickener opera-
tion in terms of solids loading rate and sludge depth.
Bench-scale thickening tests were performed routinely in order
to expand the thickener performance data base for a variety of
thickener operating conditions. Tests were performed using a
2 liter graduated cylinder equipped with a picket rake mechanism,
as shown on Figure 6. A plot of sludge interface height versus
elapsed settling time was made for each test. Data was then re-
duced using procedures presented elsewhere.2
Vacuum Filter
A typical vacuum filter run involved pumping thickened sludge
and lime slurry of known concentrations to the vacuum filter
flocculator. The conditioned sludge would flow by gravity from
the flocculator to the vacuum filter vat. After starting the
filtrate and vacuum pumps, the filter drum would be operated at
its lowest speed setting, thereby generally resulting in an
easily dischargeable cake. Samples of feed sludge (with and
without chemical conditioning), discharged filter cake and
filtrate were collected. Data recorded consisted of sludge
feed rate, lime slurry feed rate, drum speed, form and dry vacuum
22
-------
Ring Stand-
•« Picket Thickener
Mechanism
« Two-Liter
Graduated
Cylinder
FIGURE 6. SLUDGE THICKENING TEST APPARATUS.
23
-------
level, cake thickness and dischargeability, filtrate volumetric
flow rate, and conditioned sludge pH and temperature.
Prior to operation of the vacuum filter, a series of bench-scale
filter leaf tests were conducted for the purpose of selecting a
filter medium. Three filter media,* NY-317F, PO-801 and POPR-859,
were initially chosen and evaluated. The medias were tested
using alum-primary sludge conditioned with several different lime
dosages. The test results indicated that the NY-317F media was
superior in terms of cake formation rate, cake dryness, filtrate
clarity and filter cake discharge. A similar series of leaf
tests were performed on thickened ferric-primary sludge. The
results indicated the NY-317F media was again the most effective
from the standpoint of cake formation rate, cake dryness and
discharge.
The drum submergence was kept at a constant 31 percent by main-
taining a feed rate to the vat slightly in excess of that capable
of being dewatered; the remainder overflowed to drain.
A bridge block arrangement in the vacuum filter valve allowed
accurate control of the cake form and dry times. These were
arranged so that cake formation occurred during 17.9 percent
of one drum revolution with 34.7 percent for cake drying.
During each filter run, cake samples were taken and analyzed for
total solids content. To ensure uniform sampling, a round
"cookie cutter" with a face area of 77 sq cm (0.08 sq ft) was
used. Cake samples were taken after stable operation of the
vacuum filter was achieved under a given set of conditions.
Usually two or three cake samples were taken during each filter
run with the reported cake solids content being an average.
One important observation made during operation of the vacuum
filter was a visual judgement of filter cake discharge. The basis
for this judgement was a performance code established prior to
routine operation. This code established levels of performance
with consideration given to cake discharge conditions such as
dischargeability, cake cracking, and uniform cake thickness.
Filter cake discharge was judged as excellent, good, fair, poor
or not discharging.
An "excellent" (E) discharge meant:
A. The filter cake would readily discharge from the media.
B. Some or frequent cake cracking was observed during the
drying portion of the filter cycle.
*Media designations are those of the Eimco-PMD, Envirotech
Corporation
24
-------
C. Uniform cake dryness was observed between the middle of
the sector and the seam.
D. The cake would break into pieces as it passed over the
discharge roll.
A "good" (G) discharge meant:
A. The cake would separate and discharge from the media.
B. The cake did not crack during the drying cycle.
C. The observed moisture content was slightly higher in the
cake formed on top of the sector seams as compared to
the center of the sector.
D. The cake did not crack after going over the discharge
roll.
A "fair" (F) discharge meant:
A. The cake would discharge from the media except for
occasional timed intervals when a scraper was required
to keep the cake discharging.
B. Some filter cake would stick to the seams usually be-
cause of a higher moisture content.
C. The cake did not crack after going over the discharge
roll.
A "poor" (P) discharge meant:
A. The cake formed on top of the sector seams would not
discharge unless aided by a scraper.
B. Cake moisture was visually somewhat higher than the
filter cake judged as "fair".
A cake described as not discharging (ND) would have to be
scraped off the media.
Centrifuge
The approach used in collecting centrifugal dewatering perfor-
mance data involved operation at several operating conditions.
Feed sludge during any operating day was withdrawn from a single
sludge inventory. For each operating condition, samples of feed,
centrate, and cake were collected and flowrates of feed and poly-
mer were determined from volume displacements from calibrated
feed and polymer tanks. Measurements of bowl rpm and differen-
tial rpm were made for each centrifuge run. Solids recovery for
25
-------
each centrifuge run was calculated from a solids balance around
the centrifuge. Performance, as outlined above, would be char-
acterized over a range of feed rate, feed concentration, and
polymer dose.
An investigation of centrifuge machine variables was performed
during previous contract work at this pilot facility on sludges
similar to those generated during this study. The results of
this previous investigation7 were used in an effort to minimize
the number of machine variables requiring evaluation during this
study.
The capacity of larger centrifuges, for the purposes of the eco-
nomic analysis, was predicted from the performance results de-
veloped on the 15.2 cm (6.0 in) diameter machine used during
this study. The scale-up factors for solids bowl machines were
obtained from the manufacturer.3
Filter Press
Prior to each filter press run, sludge conditioning was effected
by adding lime on a predetermined dry weight percentage basis.
Lime was always added in a slurry form of a known concentration.
Following lime addition to the sludge in the feed tank, the exact
volume of conditioned sludge feed was measured. At the end of
the run, the final volume of the conditioned sludge was measured
which permitted a determination of the amount of sludge used.
This served as a cross-check of the techniques used for measure-
ment of filtrate volume generated and the sludge cake volume in
the press.
The mechanics of executing each filter press run involved the
following established procedures. The feed tank and press were
closed and the air compressor started. The pressure regulator
was then manually operated to provide a rate of chamber pressure
increase of 0.35 Kg/sq cm-min (5.0 Ib/sq in-min). After 20
minutes of operation, when a chamber pressure of 7.0 Kg/sq cm
(100 Ib/sq in) was reached, the pressure regulator was set to
maintain this pressure. Data collection during the initial 20
minutes of operation, as well as the remainder of the press run,
involved measurement of filtrate flowrate from each chamber.
When the filtrate flowrate decreased to 1/20 of the 5 minute
filtrate flowrate, the filter press run was terminated. Filter
cakes were subsequently removed and analyzed for cake thickness,
cake volume, cake wet weight, and cake weight percent total
solids. Filtrate, composited from each chamber during the filter
press run, was analyzed for suspended solids.
26
-------
The criteria used in all filter press runs, 0.35 Kg/sq cm-min
(5 Ib/sq in-min), pressure increase rate to terminal maximum
pressure, and filter press run termination at a filtrate rate of
1/20 of the 5 minute rate was arrived at from discussions with
various manufacturers of filter press equipment.5'6
The media employed in all filter press tests was SA-625* (a saran
material). This material was selected from experience with other
applications and because of its abrasion resistant properties.
A cloth blinding study was performed on the SA-625 media and in-
dicated it was an acceptable media and did not blind on alum-
primary sludge. The study was conducted by performing four con-
secutive filter press runs on the same sludge (alum-primary in
this case) under the same conditions without washing the media
between runs. During each run, a plot of form time/filtrate
volume versus filtrate volume was made (see Figure 7). The
straight line portion of the curve was extended to intercept the
time/filtrate volume axis. By using this procedure on each run,
the intercept values were compared to determine if cloth blind-
ing was taking place. If blinding was occurring, the intercept
would increase in value with progressive filter press runs. From
an examination of the intercept values from each run (see Table
2), for both chambers, it was determined that no cloth blinding
was occurring.
TABLE 2
SUMMARY OF RESULTS OF FILTER PRESS
CLOTH BLINDING ANALYSIS
-9-/V Intercept
Run_Number Front Chamber Back Chamber
1 .078 .052
2 .054 .042
3 .058 .046
4 .062 .040
X = .063 X = .045
*Media designations are those of the Eimco-PMD, Envirotech
Corporation
27
-------
0.4
0.3
O
UJ
V)
-------
Specific resistance tests were conducted prior to each filter
press test series to determine an acceptable level of lime con-
ditioning (it was learned very early that inadequate sludge con-
ditioning could lead to exorbitantly long form times). Appendix
B contains a description of the specific resistance test proce-
dure used.
Figure 8 illustrates the effect of lime dose on specific resis-
tance of alum-primary sludge. Note the rapid drop in specific
resistance as the lime dosage increased from 5 percent to 15 per-
cent. Following this rapid drop, a much more gradual decrease
occurred between 15 percent and 30 percent.
SAMPLING AND LABORATORY PROCEDURES
Grab samples of alum-primary and ferric-primary sludges from the
blowdown surge tanks were collected each weekday. One sample per
day was considered acceptable since the underflow blowdown from
each CT unit was effectively composited in the 3600 X, (950 gallon)
blowdown surge tanks.
Once stable operating conditions were established in each chem-
ical treatment unit, sampling of all pertinent wastewater streams
was initiated. Beginning each Monday morning and continuing for
48 consecutive hours, composite samples of degritted, screened
wastewater and overflow from CT-A and from CT-F were collected.
Compositing of the influent was accomplished using an automatic
sampler which collected approximately 1 £ (0.26 gallon) of sample
volume per hour, while the effluent sampler collected approxi-
mately 0.1 H (0.026 gallon) of sample volume per hour. Each
sample gravity flowed to its respective compositing container
which was located in a conventional, household refrigerator main-
tained at 4°C. After compositing, a representative sample from
each container was taken for laboratory analysis of BODs, SBODs/
COD, SCOD, turbidity, TSS, total and soluble phosphorus, pH, and
methyl orange alkalinity.
In addition to the analyses discussed above, continuous turbidity
measurements were automatically recorded for the overflow from
each CT unit. These recordings, which were visible in the lab-
oratory, were used by the pilot plant operators for remote
evaluation of the operating status of each CT unit. These re-
cordings proved very valuable since they made it possible for
pilot plant process performance to be documented at all times,
even when the pilot plant was unattended during late night hours.
An analytical laboratory was maintained at the pilot plant site
and was sufficiently equipped for the majority of the laboratory
analyses performed during this study. All analyses requiring
atomic absorption spectrophotometric capabilities were performed
in the Instruments Laboratory at Eimco PMD, Envirotech Corpora-
tion. Details of specific laboratory analyses performed during
this study are discussed in Appendix C.
29
-------
-2?
o
X
UJ
o
CO
CO
UI
oc
o
u.
o
UJ
0.
CO
20
18
16
14
10
8
ALUM-PRIMARY SLUDGE
FEEDTS = 48g/l
10 20 30 40 50
LIME DOSAGE — WEIGHT PER CENT
FIGURE 8. EFFECT OF LIME CONDITIONING DOSE ON SLUDGE
SPECIFIC RESISTANCE.
30
-------
SECTION VI
EXPERIMENTAL RESULTS AND DISCUSSION
CHEMICAL TREATMENT
LIQUID PROCESSING SUMMARY
As mentioned, two different chemical dosages were specified to
effect two different degrees of phosphorus removal. A high chem-
ical dosage was defined as that dosage required to achieve ap-
proximately 95 percent phosphorus removal. The low chemical
dosage was defined as the minimum dosage adequate to achieve
approximately 80 percent phosphorus removal, or acceptable clar-
ity.
Table 3 shows the average chemical dosage and average polymer
dosage for each phase conducted. Also shown is the degritted,
screened wastewater flowrate to the flocculating clarifiers and
the resulting clarification zone overflow rates. During any
phase, any difference in flowrate would arise from a period of
downtime in one unit and not the other. When polymer was used,
the dosage goal was 0.50 mg/£ (this dosage was determined from
the jar tests). Note that Phase 4A was essentially identical to
Phase 2A; this was the operating goal.
Percent removals were computed for phosphorus, TSS, BOD and COD
for each phase and are shown in Tables 4-9. All average figures
were calculated from the values determined from each 48 hour
composite sample.
From examination of the effluent suspended solids and phosphorus
results listed in Tables 4-5, it can be seen that 95 percent
phosphorus removal was generally achieved during the high chem-
ical dosage operation. Phase IF was an exception with a total
phosphorus recovery of only 88 percent.
Prior to initiating the low chemical dosage operation of Phase
3, jar tests were conducted to assist in determining chemical
dosages required to effect 80 percent phosphorus removal. When
these dosages were used in the flocculating clarifiers, however,
it was discovered that the ferric chloride dosage initially em-
ployed was too high and provided a phosphorus removal greater
than that desired. Accordingly, the ferric chloride dosage was
decreased after two weeks of operation. The new dosage selected
31
-------
TABLE 3
SUMMARY OF FLOCCULATING
CLARIFIER OPERATING CONDITIONS
Phase*
1A
IF
2A
2F
3A
3F-I
3F-II
4A
Average
Chemical
Dosage
(mg/Jl as
Fe/Al )
14.7
28.3
15.9
32.8
8.1
22.8
10.9
15.6
Average
Polymer
Dosage
(mg/A )
-
-
0.46
0.46
0.48
0.70
0.55
0.49
Average Raw
Sewage
Flowrate
(Aps)
1.40
1.39
2.56
2.56
2.47
2.44
2.47
2.49
Clarifier
Loading
(Jips/sq m)
0.228
0.227
0.417
0.417
0.403
0.398
0.403
0.407
1A = Phase 1 using alum;
IF = Phase 1 using ferric chloride, etc.
Clarifier loading based on 6.13 sq m clarification area
32
-------
CO
TABLE 4
CHEMICAL TREATMENT UNIT PERFORMANCE SUMMARY
FROM PHASE 1 OPERATIONS
Value
Measured
Total
BODs
rag/ 1
Total
COD
mg/-
TSS mg/,.
Total
Phos. as
P mg/£
Sol. Phos.
as P mg/?
Total
0-PO^ as
P mg/?.
Sol. 0-PO,,
as P mg/'1.
Average
Value
76
88
66
4.4
3.0
2.7
2.4
Range
70-81
74-103
37-85
4 . 0-4 .8
2.2-3.5
2. 5-3. 4
2.1-3.0
E E D
Number of
Observa-
tions
5
5
5
5
3
5
5
CT-A EFFLUENT
Number of
Average Observa-
Value Range tions
'17 12-22 5
31 27-37 5
11 6-15 5
.23 .17-. 38 5
.06 .008-. 19 5
.06 .03-. 08 ri
.012 .006-. 020 5
Percent
Removal
78
65
83
95
98
98
100
C
Average
Value
15
27
(10) 34
.55
.09
.30
. 007
T - F E F
Range
12-20
22-32
21-65
.27-. 76
.02-. 28
.11-. 52
.; .006-. 01
F L U E N
Number of
Observa-
tions
; 5
5
5
5
5
5
5
T
; Percent
Removal
80
69
(85)48
88
97
89
100
-------
TABLE 5
CHEMICAL TREATMENT UNIT PERFORMANCE SUMMARY
FROM PHASE 2 OPERATIONS
RAW FEED
CT-A EFFLUENT
CT-F EFFLUEN
Value
Measured
Total
BOD 5
Total
COD
TSS
Total
Phos . P
Sol.
Phos . P
Total
0-PO,, P
Average
Value
81
89
61
5.7
4. 1
3.4
Range
46-116
56-111
40-95
4.4-7. 3
3. 5-. 52
2.3-4.5
Number of
Observa-
tions
7
7
7
7
7
7
Average
Value
17
29
12
. 38
.22
.08
Range
10-26
21-40
7-24
.24-. 52
.05-. 50
.04-. 13
Number of
Observa-
tions
7
7
7
7
7
7
Percent
Removal
79
67
80
93
95
98
Average
Value
14
27
11
.35
.27
.06
Range
7-23
23-33
3-19
.15-. 57
.004-. 57
.03-. 11
Number of
Observa-
tions
7
7
7
7
7
7
,,°™ „ 3.1 2.2-3.8 7 .019 .006-. 048 7 99 .013 - . 006-.026 ' 7
Percent
Removal
83
70
82
94
93
98
i nn
UJ
K P
-------
TABLE 6
CHEMICAL TREATMENT UNIT PERFORMANCE SUMMARY
FROM PHASE 3A OPERATIONS
Value
Measured
Total
BOD s
Total
COD
TSS
Total
Phos . P
Sol.
Phos. P
Total
Sol.
Average
Value
138
133
108
5.43
3.52
3.46
3.41
Range
102-182
90-163
80-140
4.76-5.97
2.18-4.57
1.90-4.80
1.68-4.54
Number of
Observa-
tions
8
8
8
7
7
7
7
CT-A EFFLUENT
Average
Value
28
45
17
.86
.27
.52
.20
Range
18-37
36-56
4-35
.12-1.32
.04-. 42
.14-. 67
.02-. 32
Number of
Observa-
tions
7
7
7
6
6
7
7
Percent
Removal
80
66
84
84
92
85
94
Ui
-------
TABLE 7
CHEMICAL TREATMENT UNIT PERFORMANCE SUMMARY
FROM PHASE 3F-1 OPERATIONS
Value
Measured
Total
BOD,;
Total
COD
TSS
Total
Phos . P
Sol.
Phos . P
Total
0-P(K P
Col.
0-PCK P
RAW FEED
Average
Value
126
114
88
5.21
2.49
3.53
3.53
Range
105-146
90-137
80-95
4.76-5.66
2.18-2.80
2.53-4.53
2.53-4.53
Number of
Observa-
tions
2
2
2
2
2
2
2
CT-A EFFLUENT
Average
Value
21
35
10
.27
.07
.09
.02
Range
12-23
34-35
7-12
.21-. 33
.06-. 08
.07-. 10
.01-. 02
Number of
Observa-
tions
2
2
2
3
2
2
2
Percent
Removal
83
69
89
95
97
97
99
U)
-------
TABLE 8
CHEMICAL TREATMENT UNIT PERFORMANCE SUMMARY
FROM PHASE 3F-2 OPERATIONS
Value
Measured
Total
BOD 5
Total
COD
TSS
Total
Phos . P
Sol.
Phos. P
Total
0-P>» P
Sol.
0-P« P
====i
R A
Average
Value
142
139
115
5.52
3.94
3.44
3.36
— —
\ W FEE
Range
102-182
117-163
63-140
4.83-5.97
3.00-4.57
1.90-4.80
1.68-4.54
.
D
Number of
Observa-
tions
6
6
6
5
5
5
5
c •:
Average
Value
24
49
10
.86
.36
.50
.25
___________
? - F El
Range
21-26
40-65
6-19
.62-1.07
.11-. 71
.29-. 65
.03-. 45
1 F L U E N '
Number of
Observa-
tions
6
6
6
5
5
6
6 '
r
i
Percent
Removal
83
65
91
84
91
85
93
OJ
-------
TABLE 9
CHEMICAL TREATMENT UNIT PERFORMANCE SUMMARY
FROM PHASE 4 OPERATIONS
CJ
CO
Value
Measured
Total
BODb
Total
TSS
Total
Phos. P
Sol.
Phos. P
Total
0-PO,, P
So.
0-Pu P
Average
Value
142
135
120
5.79
4.0]
3.68
3.26
Number of
Observa- Average
Range tions Value
90-225
119-180
62-228
5.13-7.70
3. 53-4.57
2.80-5.06
1.56-4.40
7
7
7
7
7
7
7
24
43
8
0.28
.07
.09
.04
C T -
A E F F L U E
Number of
Observa-
Range tions
21-
36-
4-
.12-
.03-
.03-
.01-
30 7
51 7
17 6
.50 7
.20 7
.17 7
.18 7
N T
Percent
Removal
83
68
93
95
98
98
99
P R I
MARY EFFLUENT
Number of ,
Average Observa- : Percent
Value Range tions Removal
45
69
32
3.99
3.18
2.49
2.35
39
60
18
3. 38
2.66
0.84
0. 34
-55 7
-76 7
-65 7
-4.36 ; 7
-4.00 7
-2.90 7
-3.13 7
68
49
73
21
; 3i
i
1 32
28
-------
reduced the average phosphorus removal to 84 percent, just slight-
ly greater than the 80 percent removal objective. The alum dos-
age maintained during Phase 3 resulted in a total phosphorus
removal of 84 percent. It should be noted that the acceptable
effluent clarify criterion was controlling in determining the
chemical dosages during Phase 3 operations. A performance sum-
mary of Phase 3 chemical treatment operations is shown in Tables
6-8.
During Phase 4, only one of the chemical treatment units was
operated to achieve phosphorus removal. Alum was used as the
coagulant and the total phosphorus removal goal of 95 percent
was achieved. Table 9 summarizes the results of Phase 4 opera-
tions.
Discussion of the total suspended solids data shown in Tables
4-9 is warranted. In Table 4 the effluent TSS average concentra-
tion from CT-F is shown as 34 mg/£ (only a 48 percent removal).
As indicated previously, the effluent streams from both floccula-
ting clarifiers were continuously monitored by a recording tur-
bidimeter. Consistently throughout the study, the effluent tur-
bidity continuously recorded from the flocculating clarifier
where ferric chloride was being used was lower than that from
the alum side of the pilot plant (typically 2-3 versus 5-7 JTU).
A grab sample of the effluent from CT-F was collected and its
turbidity measured after several sample storage time periods.
During the course of this experiment, it could readily be seen
that a post-precipitation effect was taking place as the sample
was stored (the test sample was refrigerated to make conditions
identical to those of the composited effluent sample). The re-
sults of the experiment, shown in Table 10, substantiated the
visual observation. It can be deduced from Table 10 that the
turbidity of the sample at 48 hours would be approximately 30
JTU. ,This figure is approximately 10 JTU below the figure actu-
ally determined from the 48 composited sample. Since the con-
tinuous monitoring turbidimeter indicated a turbidity slightly
lower for CT-F effluent than that for the CT-A effluent, it was
reasoned that the TSS concentration in CT-F effluent was
TABLE 10
Post-Precipitation Effect in CT-F Effluent
Date: 6-12-74
Time Elapsed (hr) Sample Turbidity (JTU)
0
24
119
1.6
15.6
56
39
-------
equal to or less than that in the CT-A effluent. Since no post-
precipitation effects were noted in Phases 2 or 3 (due to opera-
tion which was aimed at eliminating conditions conducive to long
flocculating clarifier sludge retention times) and since the tur-
bidity and TSS concentrations of the CT-F effluent was consistent-
ly lower than that of the CT-A effluent (see Tables 5-8 for TSS
data), it was estimated that the TSS effluent concentration for
CT-F during Phase 1 was 10 mg/&, resulting in an 85 percent re-
moval. These figures are indicated parenthetically in Table 4.
SLUDGE PRODUCTION AND CHARACTERIZATION SUMMARY
Sludge produced in each flocculating clarifier was carefully in-
ventoried so total sludge production could be determined. With
the exception of Phase 1, total sludge production for the entire
phase was readily computed because all sludge wasted or used
for dewatering studies was inventoried.
In Phase 1, it was not possible to use this procedure because
inventories and discharge of tanks were not recorded during the
entire phase. Therefore, in Phase 1, a representative period
of sludge production was taken when it was apparent that steady
state operation, in terms of flocculating clarifier sludge
blanket level, had been achieved (this representative period for
both sludges was plotted and provided a linear slope of sludge
production versus elapsed time of operation).
By keeping an accurate inventory of all sludge removed from each
flocculating clarifier over the time period of each phase, sludge
removal in Kg DS/day could be determined. By knowing the con-
centration of the effluent suspended solids from each floccula-
ting clarifier (from the weekly 48 hours composited), it was also
possible to compute an estimate of the solids lost over the
effluent weir. Adding this to the sludge removal calculated
from tank inventories gave an estimate for total sludge produc-
tion from each unit.
An attempt was made to predict total sludge production by de-
vising a model based on influent suspended solids, total phos-
phorusy , and coagulant fed to the flocculating clarifier. The
following formula was used to predict total sludge production
(including solids lost over the effluent weir) when using alum.
K9 DS = 0.0864 Q {TSS -I- (P) (3.94 mg Al PO^/mg P) + Al
day (2.89 mg Al(OH)3/mg Al} r (1)
where:
DS = dry suspended solids
Q = raw feed flowrate, I/sec
TSS = total suspended solids in influent,
P = total phosphorus in influent, mg/fc
40
-------
Al = aluminum available after its utilization in AlPOi,/
r mg/fc
The following formula was used to predict total sludge production
(including solids lost over the weir) when using ferric chloride:
K? ^S = 0.0864 Q {TSS + (P)(4.87 mg FePO^/mg P) + Fe
y (1.89 mg Fe(OH)3/mg Fe} r (2)
whe re:
Fer = iron available after its utilization in FePOu , mg/2,
Assuming that the coagulant fed to the flocculating clarifier
combined with the phosphorus as FePOij or AlPOi», and that excess
metal was always fed to the unit, it was assumed, for calculation
purposes, that excess metal would exhibit itself as Fe(OH)3 or
Al(OH) 3.
Table 11 shows the sludge production determined from tank inven-
tories and the predicted sludge production from the above formu-
las. A summary of the calculations is contained in Appendix E.
TABLE 11
SLUDGE PRODUCTION SUMMARY
Predicted by Formula Determined from
Phase (Kg DS/day) Inventories (Kg PS/day)
1A
IF.
2A
2F
3A
3F-I
3F-II
4A
13.9
15.0
25.4
29.2
29.7
29.1
30.7
37.4
17.8
11.5
33.5
30.2
26.9
19.2
26.3
31.1
Since the indicated formulas take influent suspended solids con-
centrations, influent phosphorus concentrations, and coagulant
dosages fed to the flocculating clarifier into account, it was
reasoned that the formula (mass balance) would provide an accur-
ate estimate of the total sludge production. Sludge production
as determined from inventories, as listed in Table 11, quantify
total sludge production (sludge that was inventoried plus an
estimate of solids lost with the effluent). Those solids lost
with the effluent, which were based on one 48 hour composite
sample per week, represented only 7-8 percent of the total
sludge production.
41
-------
With the exception of Phase 3F-I, reasonable agreement in sludge
production determined by formula and inventory was achieved. The
maximum deviation in sludge production from the two methods was
approximately 30 percent and the average deviation was zero (ex-
cluding Phase 3F-I results).
SLUDGE CHARACTERIZATION
Total sludge production was predicted for each type of sludge
using mass-balance calculations for each flocculating clarifier
(Equations 1 and 2). The sludge produced was further character-
ized to reflect the amount of chemical solids contained in the
chemical-primary sludge. Table 12 lists the sludge characteriza-
tion for each phase. Percent chemical, as indicated in Table 12
represents the chemical constituents (i.e., FePCK plus Fe(OH)3)
as a percentage of the total sludge dry solids by weight.
Inspection of Table 12 indicates that the volatile fraction of
the sludge increased when the chemical dosage decreased. Vola-
tiles for Phases 1 and 2 remained relatively constant. Inerts
in Phases 3 and 4 were slightly higher than those in Phases 1
and 2 because the total suspended solids in the raw influent
were much higher in the latter two phases. A typical calculation
of sludge characterization is given in Appendix F.
42
-------
TABLE 12
SLUDGE CHARACTERIZATION SUMMARY
Phase
1A
IF
2A
2F
3A
3F-I
3F-II
4A
4P
Chemical
Dose
(mg/£)
14.7
28.3
15.9
32.8
8.1
22.8
10.9
15.6
-
TSSi
(mg/A)
66
66
61
61
108
88
115
120
120
Pi
(mg/A)
4.4
4.4
5.7
5.7
5.4
5.2
5.5
5.8
5.8
Total
Sludge
Production
(mg/£)
115
125
115
132
139
138
144
174
120
Calculated
Balance
% Volatiles
47
43
43
38
64
52
65
56
82
Calculated
Balance
% Chemical*
43
47
48
54
22
36
20
31
0
Calculated
Balance
% Inert
10
10
9
8
14
12
15
13
18
OJ
Expressed as FePO^ + Fe(OH)3 or Al PO,, + Al (OH) 3 as a percentage of the total DS
-------
DISSOLVED AIR FLOTATION THICKENIITC
Alum-Primary Sludge
Phase 1 (High Alum; no Polymer)
A summary of alum-primary sludge dissolved air flotation (DAF)
thickening results obtained during Phase 1 operations are shown
in Table 13 and on Figure 9.
A minimum air to solids weight ratio of approximately 1.4 percent
was required for effective flotation thickening to occur. Maxi-
mum predicted loading rates were observed to increase with in-
creasing air to solids ratios, whereas, float total solids con-
centration was insensitive to air to solids ratio above the mini-
mum air to solids ratio necessary to achieve flotation. Total
solids capture of 95 percent or higher was achieved at air to
solids ratios above the minimum necessary to achieve flotation.
In addition to the successful DAF tests conducted during this
phase, the tabulated data summary also includes tests which were
classified as marginal or unsuccessful. Tests were rated mar-
ginal if flotation occurred but with significant masses of sludge
breaking away from the sludge blanket and settling out. Unsuc-
cessful tests were those with operating conditions which resulted
in a failure to float the solids.
Phase 2 (High Alum; with Polymer)
A summary of Phase 2 alum-primary sludge DAF thickening results
is shown in Table 14 and on Figure 10.
Maximum predicted hydraulic and solids loading rates were found
to increase with increasing air to solids ratio while float TSS
concentration was insensitive to air to solids ratio. Total
solids capture of at least 98 percent was achieved.
Phase 3 (Low Alum; with Polymer)
Table 15 and Figure 11 summarize the DAF thickening results
obtained on alum-primary sludge produced during Phase 3.
44
-------
Ul
TABLE 13
DISSOLVED AIR FLOTATION THICKENING OF
ALUM-PRIMARY SLUDGE FROM PHASE 1
(High Alum; no Polymer)
Test
No.
1*
2*
3*
4*
5*
6*
7*
8
9
*Implies
Notes
Feed
TSS
20.3
29.1
29.1
20.3
29.1
11.4
20.3
11.4
11.4
marginal
Float
TSS Subnatant A/S Solids Loading
(%) TSS (mg/£) (%) (Kg/sq m-hr)
0.53
0.56
0.74
0.80
0.94
0.95
1.1
3.1 216 1.4 7.51
3.2 246 1.9 8.05
or failure to successfully float thicken
Hydraulic Loading
(cu m/sq m-hr)
_
-
-
-
-
-
-
1.33
1.80
Kg/sq m-hr x 0.204 = Ib/hr-sq ft
cu m/sq m-hr x 0.407 = gal/min-sq ft
-------
a£=
-i cc 5
§2
0)
^s
og
—I ft
u. "•
6
4
2
0
1
0
0.4
0.8
1.2
1.6
2.0
AIR/SOLIDS—WEIGHT PERCENT
FIGURE 9. DISSOLVED AIR FLOTATION THICKENING
OF ALUM-PRIMARY SLUDGE FROM PHASE 1
(High Alum; no Polymer)
46
-------
TABLE 14
DISSOLVED AIR FLOTATION THICKENING OF
ALUM-PRIMARY SLUDGE FROM PHASE 2
(High Alum; with Polymer)
Test
No.
1
2*
3
4
5*
6
7
8
9
10
*Implies
Notes
Feed Float
TSS TSS Subnatant A/S
(q/H) (%) TSS (mg/A) (%)
14.
8.
8.
14.
21.
21.
13.
8.
13.
8.
5
00
00
5
5
5
8
00
8
00
marginal
2.
-
2.
2.
-
2.
2.
2.
2.
2.
or
3
3
3
8
9
7
7
7
failure
104
-
101
128
-
60
84
116
70
132
to
0.
0.
0.
0.
0.
0.
1.
1.
1.
1.
36
64
64
71
73
97
1
3
5
9
successfully
Solids Loading
(Kg/sq m-hr)
7
4
4
6
4
7
5
6
.55
-
.36
.16
-
.62
.61
.89
.54
.72
Hydraulic Loading
(cu m/sq m-hr)
0
0
0
0
0
1
1
2
.418
-
.786
.319
-
.909
.835
.97
.18
.09
float thicken
Kg/sq m-hr x 0.204 = Ib/hr-sq ft
cu m/sq m-hr x 0.407 = gal/min-sq ft
-------
UJ
OC Z (/)
5
c
O «
9 *
O
8
6
4
2
0
3
2
1
0
0.4
0.8
1.2
1.4
AIR/SOLIDS—WEIGHT PERCENT
FIGURE 10* DISSOLVED AIR FLOTATION THICKENING OF
ALUM-PRIMARY SLUDGE FROM PHASE 2.
(High Alum; with Polymer)
2.0
48
-------
TABLE 15
DISSOLVED AIR FLOTATION THICKENING OF
ALUM-PRIMARY SLUDGE FROM PHASE 3
(Low Alum; with Polymer)
Test
No.
1*
2
3
4
5
6
7
8
* Implies
Notes
Feed
TSS
(g/*)
27.8
17.5
27.8
27.8
17.5
27.8
17.5
17.5
marginal
Float
TSS Subnatant A/S
(%) TSS (mg/£) (%)
-
4.2
8.4
5.1
5.0
10.0
4.9
5.2
or failure
-
306
424
410
314
480
250
96
to
0.19
0.30
0.38
0.58
0.61
0.77
0.91
1.2
successfully
Solids Loading
(Kg/sq m-hr)
-
6.76
7.45
9.07
10.2
16.7
14.4
24.9
float thicken
Hydraulic Loading
(cu m/sq m-hr)
-
0.580
0.400
0.816
1.16
1.80
2.04
4.32
Kg/sq m-hr x 0.204 = Ib/hr-sq fr
cu m/sq m-hr x 0.407 = gal/min-sq ft
-------
UJ
UJ
5
QC
11
Q S
£0
O
O
(0
rH Q.
0 0.4 0.8 1.2 1.6
AIR/SOLIDS—WEIGHT PERCENT
FIGURE 11. DISSOLVED AIR FLOTATION THICKENING OF
ALUM-PRIMARY SLUDGE FROM PHASE 3.
(Low Alum; with Polymer)
2.0
50
-------
The effects of air to solids ratio on loading rates and float
concentration observed during the first two phases were consis-
tent with those observed during Phase 3. Loading rates and
float concentration, however, were higher for the Phase 3 alum-
primary sludge. This was reasoned to be due to a lower fraction
of chemical solids contained in Phase 3 alum-primary sludge due
to the low coagulant dose used during this phase.
FERRIC-PRIMARY SLUDGE
Phase 1 (High Ferric; no Polymer)
Table 16 and Figure 12 summarize the DAF thickening results ob-
tained on ferric-primary sludge produced during Phase 1. Con-
sistent with the trends noted for alum-primary sludge, maximum
predicted loading rates increased with increasing air to solids
ratio while float concentration was insensitive to air to solids
ratio.
Phase 2 (High Ferric; with Polymer)
A summary of the ferric-primary sludge DAF thickening results
obtained during Phase 2 is shown in Table 17 and on Figure 13.
Although relatively more scatter was observed in the results
from this phase, trends were consistent with previous observa-
tions.
Phase 3 (Low Ferric; with Polymer)
Tables 18 and 19 and Figures 14 and 15 summarize the DAF thick-
ening results obtained on ferric-primary sludges produced
during Phase 3.
Two distinct ferric-primary sludges were produced during Phase 3
Sludge produced during the initial part of the operating period
was generated from operation of the flocculating clarifier with
a significantly higher ferric chloride coagulant dose than that
used during the latter part. The fraction of chemical solids
in the ferric-primary sludge was significantly different, as
documented in Section VI.
A review of the DAF thickening results shown on Figures 14 and
15 reveal that the maximum predicted loading rates and float
concentration were higher for the Phase 3 ferric-primary sludge
containing the lower fraction of chemical solids.
51
-------
01
ro
TABLE 16
DISSOLVED AIR FLOTATION THICKENING OF
FERRIC-PRIMARY SLUDGE FROM PHASE 1
(High Ferric; no Polymer)
Test
No.
1*
2*
3*
4
5
6
7
* Implies
Notes
Feed Float
TSS TSS Subnatant A/S
(g/fc) (%) TSS (rag/ A) (%)
16.
29.
41.
29.
16.
29.
16.
5
3
9
3
5
3
5
marginal
4.0
-
-
4.8
4.8
4.8
4.8
or failure
246 0
0
0
0
153 0
0
172 0
.33
.37
.52
.55
.66
.73
.99
to successfully
Solids Loading
(Kg/sq m-hr)
8
9
8
10
10
.04
-
-
.41
.24
.39
.74
Hydraulic Loading
(cu m/sq m-hr)
0.
0.
1.
1.
1.
737
_
_
629
01
06
62
float thicken
Kg/sq m-hr x 0.204 = Ib/hr-sq ft
cu m/sq m-hr x 0.407 = gal/min-sq ft
-------
UJff
OC Z (0
UJ
5
ff
<0
O 0)
o
O
0)
0)
a)
o
oc
ui
Q.
1
0
10
8
6
4
2
0
5
4
3
0.4
0.8
1.2
1.6
2.0
AIR/SOLIDS—WEIGHT PERCENT
FIGURE 12. DISSOLVED AIR FLOTATION THICKENING OF
FERRIC-PRIMARY SLUDGE FROM PHASE 1.
(High Ferric; no Polymer)
53
-------
en
TABLE 17
DISSOLVED AIR FLOTATION THICKENING OF
FERRIC-PRIMARY SLUDGE FROM PHASE 2
(High Ferric; with Polymer)
Test
No.
1*
2*
3*
4*
5*
6
7
8
9
10
11
*Implies
Notes
TSS
(gA)
11
20
29
9
29
20
11
20
11
11
11
.6
.4
.1
.33
.1
.4
.6
.4
.6
.6
.6
marginal
TSS Subnatant A/S
(%) TSS (mg/£) (%)
3.3
-
-
-
-
4.0
2.8
5.1
3.4
3.5
3.7
or failure
122 0.
-
-
-
-
158
102
94
120
140
185
to
0.
0.
0.
0.
0.
0.
1.
1.
1.
1.
46
50
54
56
72
77
93
0
3
8
8
successfully
Solids Loading
(Kg/sq m-hr)
5
7
4
13
7
7
8
.83
-
-
-
-
.79
.65
.0
.40
.35
.58
Hydraulic Loading
(cu m/sq m-hr)
0
0
0
1
0
1
2
.344
_
_
_
_
.762
.688
.94
.762
.06
.24
float thicken
Kg/sq m-hr x 0.204 = Ib/hr-sq ft
cu m/sq m-hr x 0.407 = gal/min-sq ft
-------
oc z
-------
VI
en
TABLE 18
DISSOLVED AIR FLOTATION THICKENING OF
FERRIC-PRIMARY SLUDGE FROM PHASE 3-1
(Low Ferric; with Polymer)
Test
No.
1*
2*
3
4
5
6
7
8
9
10
11
* Imp lies
Notes
Feed
TSS
(q/A)
23.
33.
13.
23.
33.
33.
23.
13.
23.
13.
13.
5
0
5
5
0
0
5
5
5
5
5
marginal
Float
TSS Subnatant A/S
(%) TSS (mg/£) (%)
3.
3.
4.
4.
4.
3.
4.
4.
3.
or
4
9
7
3
6
4
5
2
9
failure
-
-
128
164
158
124
154
142
112
128
150
to
0.
0.
0.
0.
0.
0.
0.
0.
0.
1.
1.
23
32
40
46
48
63
68
79
91
2
6
successfully
Solids Loading
(Kg/sg m-hr)
_
3.
3.
5.
5.
5.
5.
9.
7.
7.
97
92
10
54
05
78
22
30
94
Hydraulic Loading
(cu m/sq m-hr)
0
0
0
0
0
0
1
1
1
__
.442
.253
.334
.516
.541
.860
.13
.35
.77
float thicken
Kg/sq m-hr x 0.204 = Ib/hr-sq ft
cu m/sq m-hr x 0.407 = gal/min-sq ft
-------
01
TABLE 19
DISSOLVED AIR FLOTATION THICKENING OF
FERRIC-PRIMARY SLUDGE FROM PHASE 3-II
(Low Ferric; with Polymer)
Test
No.
1
2*
3
4
5
6
7
8
9
10
11
* Imp lies
Notes
Feed Float
TSS TSS Subnatant A/S
(g/£) (%) TSS (mg/£) (%)
31.
22.
31.
22.
31.
31.
22.
13.
22.
13.
13.
7
8
7
8
7
7
8
6
8
6
6
marginal
4.
4.
4.
4.
5.
5.
5.
4.
5.
3.
4.
or
6
3
9
0
0
3
2
0
1
9
9
failure
•» * /•»
697
289
372
201
443
463
198
123
216
138
182
to
*— .
0.
0.
0.
0.
0.
0.
0.
0.
0.
1.
1.
18
24
35
50
53
70
76
79
94
2
6
successfully
Solids Loading
(Kg/sq m-hr)
6.
5.
6.
7.
13.
45.
9.
5.
11.
8.
16.
37
46
76
01
4
7
8
59
8
97
8
Hydraulic Loading
(cu m/sq m-hr)
0.
0.
0.
0.
1.
2.
1.
0.
1.
1.
3.
278
344
428
617
06
90
08
835
57
65
51
float thicken
J\\J/ Ov4 ill 11J- .o. v • «- v -» j-*-f/ ***. ** T, — *-
cu m/sq m-hr x 0.407 = gal/min-sq ft
-------
-------
o
UJ
5
DC
<0
O (/)
o
(O
o
q uj
rr1 Q-
4
3
2
1
0
16
12
8
4
0
5
4
3
0.4
0.8
1.2
1.6
2.0
AIR/SOLIDS—WEIGHT PERCENT
FIGURE 15. DISSOLVED AIR FLOTATION THICKENING OF
FERRIC-PRIMARY SLUDGE FROM PHASE 3-II.
(Low Ferric; with Polymer)
59
-------
PRIMARY SLUDGE
One of the pilot plant flocculating clarifiers was operated as a
primary clarifier during Phase 4. The objective was to generate
primary sludge (zero percent chemical solids) whose thickening
and dewatering characteristics could be related to the chemical-
primary sludges produced during previous phases.
Table 20 and Figure 16 summarize the DAF thickening results ob-
tained on the primary sludge produced during Phase 4. Although
a substantial amount of scatter in the results occurred, pre-
dicted maximum loading rates and float concentration were sig-
nificantly higher for the primary sludge compared to the chemi-
cal-primary sludges. The relationship between air to solids
ratio and flotation properties was similar to that observed
for alum-primary and ferric-primary sludge. Float total solids
content, however, was observed to be more sensitive to air to
solids ratio than either the alum-primary or ferric-primary
sludge.
DISCUSSION
The weight percent of chemical solids present in alum-primary
sludge was correlated with float concentration, maximum predicted
solids loading rate and maximum predicted hydraulic loading
rate. Chemical solids were defined as the sum of the metal phos-
phate and metal hydroxide (see Appendix F). Figures 17-19
summarize these correlations. Included in the correlation was
the DAF performance results generated on primary sludge (zero
percent chemical solids).
Figures 17-19, which include all alum-primary sludge DAF results
collected and the primary sludge DAF results collected during
Phase 4, were plotted for four levels of air to solids ratio
ranging from 0.75 to 2.0 weight percent. The trends shown in-
dicate a worsening in flotation properties with increasing
weight fractions of chemical solids in the sludge.
As shown on Figures 17-19, increasing air to solids ratios
above approximately 1.5 weight percent had little effect on
float concentration or solids loading rate, whereas, maximum
allowable hydraulic loading rate increased with further increases
in air to solids up to 2.0 weight percent.
Figures 20-22 correlate the weight percent of chemical solids in
the ferric-primary sludge versus DAF properties for four ratios
of air to solids. The DAF results collected on primary sludge
from Phase 4 were included in the summary on Figures in order to
extend the analysis to zero percent chemical solids. '
60
-------
TABLE 20
DISSOLVED AIR FLOTATION THICKENING OF
PRIMARY SLUDGE FROM PHASE 4
Test
No.
1*
2
3
4
5
6
7
8
*Implies
Notes
Feed Float
TSS TSS Subnatant A/S
(g/M (%) TSS (mg/fc) (%)
27.
16.
27.
27.
16.
27.
17.
17.
1
5
1
1
5
1
4
4
marginal
-
5.
6.
6.
6.
7.
7.
7.
or
3
2
6
7
0
6
5
failure
-
1800
604
640
1400
508
1400
1200
to
0
0
0
0
0
0
0
1
.21
.33
.43
.64
.65
.85
.93
.2
successfully
Solids Loading
(Kg/sq m-hr)
-
55.
59.
40.
57.
48.
53.
67.
9
8
2
4
0
4
6
Hydraulic Loading
(cu m/sq m-hr)
-
5.
4.
3.
7.
5.
7.
11.
06
42
69
03
41
86
8
float thicken
Kg/sq m-hr x 0.204 = Ib/hr-sq ft
cu m/sq m-hr x 0.407 = gal/min-sq ft
-------
I
00
-I (0
OS
O
OC
O
z
UJ
I- OC
§25
9<5S
III
o*
40
20
0
7
6
5
0.4
0.8
1.2
1.6
2.0
AIR/SOLIDS—WEIGHT PERCENT
FIGURE 16. DISSOLVED AIR FLOTATION THICKENING OF PRIMARY
SLUDGE FROM PHASE 4.
62
-------
UJ
O
Jl
a
8
AIR/SOLIDS — WT PCT
Ml!
0.75
1.0
1.5
2.0
10 20 30 40
CHEMICAL SOLIDS IN SLUDGE—WEIGHT PERCENT
50
FIGURE 17 FLOAT SOLIDS CONCENTRATION VERSUS FRACTION
OF CHEMICAL SOLIDS IN ALUM-PRIMARY SLUDGE.
63
-------
cc
x
O
(0
o
s
cc
O
<
O
o
(A
60
50
40
30
20
10
AIR/SOLIDS—WT PCT
10
20
30
40
50
CHEMICAL SOLIDS IN SLUDGE—WEIGHT PERCENT
FIGURE 18. SOLIDS LOADING RATE VERSUS FRACTION OF
CHEMICAL SOLIDS IN ALUM-PRIMARY SLUDGE.
64
-------
o
(ft
o,
T
UJ
s
oc
O
i
o
o
_l
<
DC
o
AIR/SOLIDS—WT PCT
0 20 40 60
CHEMICAL SOLIDS IN SLUDGE—WEIGHT PERCENT
FIGURE 19. HYDRAULIC LOADING RATE VERSUS FRACTION OF
CHEMICAL SOLIDS IN ALUM-PRIMARY SLUDGE.
65
-------
UJ
o
cc
8!
J,
o
o
§
8
6
10
20
30
40
50
CHEMICAL SOLIDS IN SLUDGE—WEIGHT PERCENT
FIGURE 20. FLOAT SOLIDS CONCENTRATION VERSUS FRACTION OF
CHEMICAL SOLIDS IN FERRIC-PRIMARY SLUDGE.
66
-------
80
O
O
(£
O
5
o
V)
60
40
20
AIR/SOLIDS—WT PCT
_i_-
]
iHit: 0.75
-fzbH.O
:t±m1-5
2.0
10
20
30
40
50
CHEMICAL SOLIDS IN SLUDGE— WEIGHT PERCENT
FIGURE 21, SOLIDS LOADING RATE VERSUS FRACTION OF
CHEMICAL SOLIDS IN FERRIC-PRIMARY SLUDGE.
67
-------
O
V)
o
T
UJ
cc
o
Q
O
O
QC
Q
X
8
AIR/SOLIDS—WT PCT
»4^-i4-4
0.75
20 40 60
CHEMICAL SOLIDS IN SLUDGE—WEIGHT PERCENT
FIGURE 22. HYDRAULIC LOADING RATE VERSUS FRACTION OF
CHEMICAL SOLIDS IN FERRIC-PRIMARY SLUDGE
68
-------
Increased fractions of chemical solids in the sludge generally
resulted in poorer ferric-primary sludge DAF properties. Float
total solids content was maximized at an air to solids ratio at
0.75 weight percent, and maximum allowable flotator solids load-
ing rate was maximized at an air to solids ratio of about 1.5
weight percent.
From the correlations presented on Figures 17-22, a comparison
in DAF properties may be made between the alum-primary and
ferric-primary sludges generated during this study. The curves
on Figures 23-25 were reproduced in order to facilitate this
comparison.
Although an air to solids ratio of 1.5 weight percent resulted
in maximization of flotator performance in terms of float con-
centration and solids loading rate, a ratio of 1.0 weight per-
cent was selected as the basis for the comparison. This was
done since a broader and better defined data base was available
to an air to solids ratio of 1.0 weight percent as compared to
1.5 weight percent. Inspection of Figures 17-22 revealed that
the general outcome of the comparison at a 1.0 weight percent
air to solids ratio was similar to that which could be developed
at an air to solids ratio of 1.5 weight percent.
Table 21 shows a comparison of DAF performance parameters for
alum-primary and ferric-primary sludge at various levels of
phosphorus removal. To review DAF performance for these sludges,
a raw wastewater containing 100 mg/H TSS and 5.0 mg/£ total P
was assumed. These characteristics correspond to average Salt
Lake City wastewater concentrations observed during this study.
The molar ratios of metal to phosphorus required to achieve
each level of total phosphorus removal were determined from
this study and from previous studies.1'2
The results in Table 21 indicates that, for similar levels of
phosphorus removal, the DAF performance parameters for alum-
primary sludge were superior to those for ferric-primary sludge.
The results also indicate a decrease in DAF performance para-
meters as additional inorganic coagulant is added to achieve
higher total phosphorus removal efficiency.
69
-------
Ul
o
DC
0
Ji
O
(0
§
AIR/SOLIDS = 1.0 WT PCT
0 20 40
CHEMICAL SOLIDS IN SLUDGE—WEIGHT PERCENT
FIGURE 23. COMPARATIVE FLOAT SOLIDS CONCENTRATION
OF ALUM-PRIMARY AND FERRIC-PRIMARY SLUDGES.
60
70
-------
80
oc
X
o
(A
O
I
UJ
oc
o
z
Q
(A
O
60
40
20
AIR/SOLIDS = 1.0WTPCT
0 20 40
CHEMICAL SOLIDS IN SLUDGE—WEIGHT PERCENT
FIGURE 24. COMPARATIVE SOLIDS LOADING RATES OF
ALUM-PRIMARY AND FERRIC-PRIMARY SLUDGES.
60
71
-------
ec
I
8
O
s
oc
O
z
5
OC
e
8
6
AIR/SOLIDS = 1.0 WT PCT
0 20 40 60
CHEMICAL SOLIDS IN SLUDGE-WEIGHT PERCENT
FIGURE 25. COMPARATIVE HYDRAULIC LOADING RATE OF
ALUM-PRIMARY AND FERRIC-PRIMARY SLUDGES.
72
-------
TABLE 21
EFFECT OF PHOSPHORUS REMOVAL ON DISSOLVED AIR
FLOTATION PROPERTIES OF ALUM-PRIMARY AND FERRIC-PRIMARY SLUDGE
Total Phosphorus
Removal-Percent
80
90
95
Alum-Primary Sludge
Chemical
Sludge
Weight
Percent
18
23
32
Float
Percent
TS
5.6
5.1
4.2
SLR
Kg/sq-
m hr
24
18
11
HLR
cu m/
sq m-hr
3.0
2.3
1.4
Ferric-Primary Sludge
Chemical
Sludge
Weight
Percent
22
28
38
Float
Percent
TS
4.4
4.2
3.9
SLR
Kg/sq
m hr
9.0
7.5
6.8
HLR
cu m/
sq m-hr
1.6
1.3
1.0
BASIS: Raw wastewater with 100 mg/£ TSS and total phosphorus of 5 mg/H as P.
DAF operation at 1% air to solids ratio.
-------
GRAVITY THICKENING
ALUM-PRIMARY SLUDGE
Phase 1 (High Alum; no Polymer)
A summary of gravity thickener performance for alum-primary
sludge during Phase 1 is shown in Table 22. As shown in Table
22, three separate stable operating periods were obtained.
Feed concentration ranged from 21-26 g/£, with corresponding
underflow concentrations ranging from 23-29 g/&. Clearly, the
alum-primary sludge generated during Phase 1 did not gravity
thicken to any appreciable extent. Underflow concentration was
not affected by solids loading rate over the range studied (at a
sludge age of 1.7-2.2 days and sludge depth of 0.95-1.2 m {3.1-
3.9 ft}). The only trend discernable was that as the feed con-
centration increased, the underflow concentration also increased.
The effect of sludge depth within the thickener was studied by
sampling the sludge blanket at intervals of 15.2 cm (6 inches)
and analyzing these samples for total solids. The results of
three typical thickener profiles are shown on Figure 26. These
data indicate that a thickener sludge blanket depth of approxi-
mately 40 cm (1.3 ft) was required to achieve maximum underflow
concentrat ion.
Figures 27 and 28 summarize the results of bench-scale thicken-
ing tests performed on alum-primary sludge generated during
Phase 1. Plotted on Figure 27 are the pilot plant thickener
performance results presented in Table 22. None of the three
pilot plant operating periods produced an underflow concentra-
tion equal to that predicted by the laboratory tests. Highest
underflow concentration from the pilot plant thickener was 2.9%,
on Figure 27, this operating point appears in an area where the
bench-scale tests predicted an underflow concentration in excess
of 5.0%. Figure 28 shows the positive relation between thicken-
er feed concentration and maximum predicted underflow concentra-
tion; this trend is substantiated by the pilot plant thickener
results.
Phase 2 (High Alum; with Polymer)
A gravity thickenej: performance summary for alum-primary sludge
generated during Phase 2 is shown in Table 23. As shown in
Table 23, three stable operating periods were obtained.
74
-------
TABLE 22
ALUM-PRIMARY SLUDGE THICKENING
AVERAGE PHASE 1 PILOT PLANT RESULTS
(High Alum; no Polymer)
Parameter
Values
Evaluation period, days 5
Thickener feed, g/H 21
Thickener loading, £/min-sq m .53
Thickener loading, Kg/day-sq m 15.8
Thickener sludge age, days 1.7
Thickener sludge depth, m 1.2
Solids capture, percent 98
Thickener underflow sludge, g/2, 23
12
22
.32
10.1
2.2
.95
99
24
6
26
.45
16.8
1.9
1.1
99
29
TABLE 23
ALUM-PRIMARY SLUDGE THICKENING
AVERAGE PHASE 2 PILOT PLANT RESULTS
(High Alum; with Polymer)
Parameter
Values
Evaluation period, days
Thickener feed, g/&
Thickener loading, £/min-sq m
Thickener loading, Kg/day-sq m
Thickener sludge age, days
Thickener sludge depth, m
Solids capture, percent
Thickener underflow sludge, g/&
9
21
.53
15.9
1.9
1.2
99
27
6
25
.51
18.1
1.5
1.0
99
30
9
O 1
£L
.68
O f\ f\
20.9
1f\
. 0
O Q
.00
Q Q
y y
o c.
2 3
75
-------
o
tu
o
u.
tr
UJ
O
O
_i
UJ
CD
Ul
o
UJ
o
o
20
40
60
80
100
120
140
FEEDTS=21-24g/l
-t-r
01234
TOTAL SOLIDS—WEIGHT PERCENT
FIGURE 26. ALUM-PRIMARY SLUDGE THICKENER PROFILES
PHASE I.
(High Alum; no Polymer)
76
-------
O
«P
<
Q
I
o
z
Q
g
CO
o
o
CO
tr
z
UJ
o
o
UJ
o
O
111
oc
Q.
200
175
150
125
100
75
50
25
UNDERFLOW CONCENTRATION
Q Cu = 2.0%
A Cu = 3.0%
Q Cu = 4.0%
Cu = 5.0%
PILOT PLANT
FIGURE 27.
10 20
THICKENER FEED — g/l
PREDICTED TSL VS THICKENER FEED
FOR ALUM-PRIMARY SLUDGE — PHASE I
(High Alum; no Polymer)
30
77
-------
;ENER UNDERFLOW CONCENTRATIONS/I
W * Ul O>
o o o o
-f-
!
! '
.-_
,
>i
-
0
1
1
3ENCH TESTS
PILOT PLANT THICKENER
,
-•
•-• - - -
O
O
—
—
k_j
• — i
O
0
o
-
RESULTS
i
—
—
! ' '
, —
i —
-
-
• i i ' ' l 1
O
"]••"
— o
o
--J
— L-,
•
1
_J
1
- ' -
0
-44—
__^j_;_
~j-r-^- •
o;
.
D
Q
UJ
O
UJ
CC
0.
X
20
10
10 20
THICKENER FEED — g/l
30
FIGURE 28. MAXIMUM PREDICTED UNDERFLOW CONCENTRATION
VS THICKENER FEED CONCENTRATION — PHASE I.
(High Alum; no Polymer)
78
-------
Feed solids concentration ranged from 21-25 g/£, and the result-
ing underflow solids ranged from 25-30 g/&. Solids loading rate,
in the range studied, did not affect the underflow solids (at a
sludge age of 1.0-1.9 days and sludge depth of 0.88-1.2 m (2.9-
3.9 ft). Solids capture was excellent, with all operating per-
iods exhibiting 99% capture. Data given in Table 23 tend to show
that a higher feed solids concentration resulted in a thicker
underflow concentration.
By conducting thickener profile tests, the effect of sludge depth
within the thickener on underflow solids concentration was stud-
ied. The sludge blanket was sampled at 15.2 cm (6 inch) depths
and these samples analyzed for total solids concentration. Re-
sults of three typical tests are shown on Figure 29. All pro-
files indicate that a sludge depth of only about 20 cm (0.66 ft)
was required to achieve maximum underflow concentration. Sludge
depths greater than 20 cm (0.66 ft) were of little value in in-
creasing the underflow solids concentration.
The results of bench-scale thickening tests conducted on the alum-
primary sludge generated during Phase 2 are summarized on Figures
30 and 31. Also plotted on Figure 30 are the thickener solids
loading from the pilot plant thickener presented in Table 23.
None of the pilot plant operating periods produced an underflow
solids concentration equal to that predicted by the laboratory
tests. Highest underflow solids from the pilot plant thickener
was 3.0%; on Figure 30, this operating point appears in an area
where the bench-scale tests predicted an underflow concentration
in excess of 5.0%.
Figure 31 shows the relation between thickener feed solids con-
centration and the maximum predicted underflow concentration; the
trend is substantiated by the pilot thickener results.
Phase 3 (Low Alum; with Polymer)
Results of gravity thickener performance during Phase 3 are sum-
marized in Table 24. As shown in Table 24, two stable operating
periods were obtained. Feed solids concentration of 32-37 g/£
provided resultant underflow solids concentrations of 44-48 g/&.
Solids loading rates of 13.5-21.2 Kg/day-sq m (2.8-4.3 Ib/day-
sq ft) did not affect the underflow solids of the alum-primary
sludge. Solids capture ranged from 95% to 99%. The data tend
to show that higher thickener feed solids concentration increases
underflow solids concentration.
Figure 32 shows the results of three separate thickener profiles
conducted during Phase 3. Inspection of this figure shows a
trend not apparent during the first two phases. In this case,
it is apparent that the full depth of the thickener was being
utilized in thickening of this sludge, and that as the sludge
depth increased, the underflow solids increased.
79
-------
U
I
DC
UJ
O
CC
111
O
I
UJ
CD
UJ
O
UJ
O
Q
(A
20
40
60
80
100
120
140
____U—.4-
FEEDTS = 19-26g/l
D
TOTAL SOLIDS—WEIGHT PERCENT
FIGURE 29. ALUM-PRIMARY SLUDGE THICKENER PROFILES
PHASE 2.
(Low Alum; with Polymer)
80
-------
o
0)
i
I
o
z
o
CO
o
_i
o
CO
cc
Ul
UJ
*
o
£
o
UJ
s
Q
UJ
DC
Q.
225
200
175
150
125
100
75
50
25
UNDERFLOW CONCENTRATION
Cu = 2%
Cu = 3%
Q Cu = 4%
Cu = 5%
PILOT PLANT
THICKENER
FIGURE 30.
10 20
THICKENER FEED — g/l
PREDICTED TSL VS THICKENER FEED FOR
ALUM-PRIMARY SLUDGE — PHASE 2.
(High Alum; with Polymer)
81
-------
UJ
O
O
O
QC
UJ
O
CC
UJ
UJ
o
UJ
O
O
Ul
cc
0.
D
X
70
60
50
40
30
20
10
O BENCH TEST RESULTS
1 PILOT PLANT THICKENER RESULTS
-I ^-^.Lj.-
^B±
<$>
10 20
THICKENER FEED — g/l
30
FIGURE 31- MAXIMUM PREDICTED UNDERFLOW CONCENTRATION
VS THICKENER FEED CONCENTRATIOIi-PHASE 2,
(High Alum; with Polymer)
82
-------
TABLE 24
ALUM-PRIMARY SLUDGE THICKENING
AVERAGE PHASE 3 PILOT PLANT RESULTS
(Low Alum; with Polymer)
00
u>
Parameter
Evaluation period, days
Thickener feed, g/&
Thickener loading, £/min-sq m
Thickener loading, Kg/day-sq m
Thickener sludge age, days
Solids capture, percent
Thickener underflow sludge, g/£
Values
5
32
.29
13.5
1.6
.52
95
44
11
37
.40
21.2
1.0
.45
99
48
-------
I
LU
O
u.
cc
UJ
O
O
UJ
00
i
UJ
O
UJ
O
O
20
40
60
80
100
120
140
FEED TS = 29-39 g/l
t +-
12345
TOTAL SOLIDS—WEIGHT PERCENT
FIGURE 32. ALUM-PRIMARY SLUDGE THICKENER PROFILES
PHASE 3.
(Low Alum; with Polymer)
84
-------
Figures 33 and 34 summarize the results of bench-scale thicken-
ing tests conducted on the alum-primary sludge generated during
Phase 3. Plotted on Figure 33 are the results of the two stable
pilot thickener operating conditions shown in Table 24. Neither
of the operating periods produced an underflow comparable to that
indicated from the bench-scale tests shown on Figure 33. Figure
34 shows a direct relationship between thickener feed solids con-
centration and maximum underflow solids concentration. This is
substantiated by the results from the pilot plant thickener.
Phase 4 (High Alum; with Polymer)
Results of gravity thickener performance from Phase 4 are summar-
ized in Tables 25 and 26. During Phase 4, thickener performance
was closely monitored to determine the effect of feed solids con-
centration and polymer conditioning of the thickener feed. From
Table 25, the effect of diluting the thickener feed is apparent;
no improvement in thickener performance was observed. Hydraulic
loading was, of course, high when feeding a dilute feed, but
solids loadings were generally in line with solids loading ex-
perienced when using a more concentrated feed. Solids captures
of 98-99% were experienced, even with the highest hydraulic load-
ing of 5.78 fcpm/sq m (0.14 gal/min-sq ft). Again, the trend of
higher underflow concentrations resulting from higher feed solids
concentrations was present.
The results of the thickening operations conducted when polymer
was added to the thickener feed are summarized in Table 26. The
results show that adding polymer to a dilute thickener feed im-
proved thickener performance. Dilution of the feed plus polymer
conditioning resulted in an underflow solids concentration of 56
g/i which was the highest achieved for alum-primary type sludge.
Future studies of feed dilution and polymer addition should ex-
amine the effects of sludge age, sludge depth, thickener feed
concentration, polymer type and dosage.
Figure 35 shows the same pattern of gravity thickening that was
exhibited during Phases 1 and 2. All the profiles on Figure 35
illustrated the effect of sludge depth on underflow concentra-
tions. Only 30-40 cm (1.0-1.3 ft) sludge depth was required to
achieve maximum underflow concentrations; sludge depths greater
than 40 cm (1.3 ft) were of little use in increasing the solids
concentration.
The results of bench-scale thickening tests conducted on alum-
primary sludge generated during Phase 4 are given on Figures 36
and 37. Also plotted on these two figures are the results of
pilot plant thickening performance from six stable operating
periods. With the exception of the period representing polymer
feed to the thickener, none of the pilot plant operating periods
85
-------
,1
<
O
o
o
<
o
(/)
o
_j
o
cc
HI
UJ
Q
UJ
O
Q
UJ
CC
Q.
450
400
350
300
250
UNDERFLOW CONCENTRATION
Cu = 2%
Cu = 3%
• Cu = 5%
A PILOT PLANT
50
FIGURE 33.
10 20
THICKENER FEED — g/l
PREDICTED TSL VS THICKENER FEED FOR
ALUM-PRIMARY SLUDGE—PHASE 3.
(Low Alum; with Polymer)
86
-------
MAXIMUM PREDICTED THICKENER UNDERFLOW CONCENTRATION- g/l
-*roc*>-ptuio>->ioo
-------
00
00
TABLE 25
ALUM-PRIMARY SLUDGE THICKENING
AVERAGE PHASE 4 PILOT PLANT RESULTS
(High Alum; with Polymer)
Parameter
Evaluation period, days
Thickener feed, g/Jl
Thickener loading, Jl/min-sq m
Thickener loading, Kg/day-sq m
Thickener sludge age, days
Thickener sludge depth, m
Solids capture, percent
Thickener underflow sludge, g/£
Values
5
36
.22
11.4
3.42
1.02
99
39
5
22
.59
18.7
.07
.05
99
29
4
8*
1.95
22.1
.68
.65
98
30
2
4*
3.51
22.4
.55
.58
98
27
3
3*
5.78
21.2
.27
.28
99
26
Dilution of thickener feed with flocculating clarifier effluent
-------
TABLE 26
oo
vo
ALUM-PRIMARY SLUDGE THICKENING
AVERAGE PHASE 4 PILOT PLANT RESULTS
(High Alum; with Polymer)
Parameter
Values
(Polymer Fed to Thickener)
Evaluation period, days
Thickener feed, g/£
Thickener loading, Z/min-sq m
Thickener loading, Kg/day-sq m
Thickener sludge age, days
Thickener sludge depth, m
Polymer dosage, % by weight
Solids capture, percent
Thickener underflow sludge, g/Ji
3
2*
6.37
21.9
0.56
0.29
0.07+
97
56
Dilution of thickener feed with flocculating
clarifier effluent.
Dow AP-30.
-------
o
I
cc
UJ
o
cc
Ul
o
o
UJ
CO
Q.
UJ
Q
UJ
o
Q
20
40
60
80
100
120
140
FEED TS - 24-29 g/l
NO POLYMER FED
TO THICKENER
FIGURE 35.
11234
TOTAL SOLIDS—WEIGHT PERCENT
ALUM-PRIMARY SLUDGE THICKENER PROFILES
PHASE 4,
(High Alum; with Polymer)
90
-------
o
V)
o
o
*
I
o
z
Q
o
3
o
(/>
OC
UJ
Z
UJ
*
o
z
Q
6
o
ui
oc
Q.
225
200
175
150
125
100
75
50
25
UNDERFLOW CONCENTRATION
D
O Cu = 1%
&• Cu = 2%
a Cu = 3%
_* Cu = 4%
A PILOT PLANT
THICKENER RESULTS
• PILOT PLANT
THICKENER RESULTS
USING POLYMER
10 15 20
THICKENER FEED —g/l
25
30
FIGURE 36 PREDICTED TSL VS THICKENER FEED FOR ALUM
PRIMARY SLUDGE — PHASE 4
(High Alum; with Polymer)
91
-------
ENER UNDERFLOW CONCENTRATION- g
o
£
O
O
UJ
DC
Q.
5
D
BENCH TEST RESULTS
PILOT PLANT THICKENER RESULTS
PILOT PLANT THICKENER RESULTS
USING POLYMER
0 0
O
o
o o
o I
0 •
FIGURE 37.
5 10 15 20 25
THICKENER FEED — g/l
MAXIMUM PREDICTED THICKENER UNDERFLOW
CONCENTRATION VS THICKENER FEED
CONCENTRATION — PHASE 4.
(High Alum; with Polymer)
30
92
-------
produced an underflow solids concentration equal to that predict-
ed by the laboratory tests, although the results of the three
periods representing a dilute feed to the thickener approach
the laboratory results. Figure 37 shows the relation between
thickener feed solids concentration and the maximum predicted
underflow solids concentration; the trend is generally sub-
stantiated by the pilot plant results.
FERRIC-PRIMARY SLUDGE
Phase 1 (High Ferric; no Polymer)
In Table 27, two stable operating periods from Phase 1 summarize
the gravity thickener performance during this phase. The long
sludge ages and relatively low solids loading rates for this
period were due to operational inexperience.
The first operating period (five days) had an exorbitantly long
sludge age; thickener feed solids of 38 g/£ resulted in an
underflow solids concentration of 48 g/&. For a shorter sludge
age and slightly higher thickener solids loading, a feed solids
concentration of 18 g/i resulted in an underflow of 64 g/£.
However, in this case solids capture was only 87%.
Figure 38 shows the effect of sludge depth in the thickener on
underflow solids concentration. Both curves on Figure 38 show
the same pattern; the first 40-50 cm (1.3-1.6 ft) of sludge
depth produced the majority of the thickening.
Figures 39 and 40 present the results of bench-scale thickening
tests performed on ferric-primary sludge generated during Phase
1. Also plotted on Figure 39 are the two pilot plant operating
points. Results from one of the thickener operating periods
appears consistent with the bench-scale test results, whereas,
the pilot performance measured at a long thickener sludge age
was significantly poorer than bench-scale results.
Figure 40 illustrates a definite trend in the relationship be-
tween thickener feed solids and thickener underflow solids.
Pilot plant thickener operation did not substantiate this, but
this was probably due to an excessively long thickener sludge
age during one of the operating periods.
Phase 2 (High Ferric; with Polymer)
Table 28 summarizes the results of the pilot plant thickening
studies conducted on ferric-primary sludge during Phase 2. Four
separate operating periods are shown, each of relatively long
duration. Thickener feed solids ranged from 17-22 g/£ with
underflows ranging from 45-57 g/&. Inspection of the three
operating periods in which the thickener feed solids were
identical (22 g/£) shows an interesting phenomenon. For the
93
-------
TABLE 27
FERRIC-PRIMARY SLUDGE THICKENING
AVERAGE PHASE 1 PILOT PLANT RESULTS
(Hiqh Ferric; no Polymer)
Parameter Values
Evaluation period, days
Thickener feed, g/i
Thickener loading, A/min-sq m
Thickener loading, Kg/day-sq m
Thickener sludge age, days
Thickener sludge depth, m
Solids capture, percent
Thickener underflow sludge, g/&
5
38
0.13
7.1
6.5
1.02
99
48
8
18
0.36
9.2
3.4
0.62
87
64
94
-------
o
I
cc
i
1
u.
DC
Ul
g
1
UJ
m
u
Q
UJ
O
o
D
_J
(A
20
40
60
80
100
120
140
FEED TS = 23-25 g/l
8 9 10
TOTAL SOLIDS-WEIGHT PERCENT
FIGURE 38 FERRIC-PRIMARY SLUDGE THICKENER PROFILES
PHASE 1.
(High Ferric; no Polymer)
95
-------
250
UNDERFLOW CONCENTRATION
Cu = 3%
Cu = 4%
Cu = 5%
PILOT PLANT THICKENER
FIGURE 39.
15 20 25
THICKENER FEED — g/l
PREDICTED JSL VS THICKENER FEED FOR FERRIC
PRIMARY SLUDGE — PHASE 1.
(High Ferric; no Polymer)
96
-------
-90
O)
I
p80
^
oc
z
Ul
O
O
O
60
u.
OC
Ul
O
z
3 50
OC
UJ
Z
UJ
9 40
X
Q
UJ
O 30
0
UJ
OC
a
20
X
<
10
BENCH SCALE TESTS
PILOT PLANT THICKENER RESULTS
O
10
20
30
40
FIGURE 40
THICKENER FEED — g/l
MAXIMUM PREDICTED THICKENER UNDERFLOW
CONCENTRATION VS THICKENER FEED
CONCENTRATION — PHASE 1
(High Ferric; no Polymer)
97
-------
TABLE 28
vo
00
FERRIC-PRIMARY SLUDGE THICKENING
AVERAGE PHASE 2 PILOT PLANT RESULTS
(High Ferric; with Polymer)
Parameter
Evaluation period, days
Thickener feed, g/£
Thickener loading, £/min-sq m
Thickener loading, Kg/day-sq m
Thickener sludge age, days
Thickener sludge depth, m
Solids capture, percent
Thickener underflow sludge, g/£
7
17
.52
12.9
3. 3
1.14
99
46
5
22
.53
16.5
2.5
.93
99
53
Values
8
22
.44
13.9
2.8
.83
99
57
8
Rfi
17.5
2 7
1 22
97
45
-------
same sludge age (2.5-2.8 days) and the same feed solids, a
higher thickener solids loading caused a reduction in the thick-
ener solids underflow concentration. From the data, it can be
seen that exceeding a loading of 16.5 Kg/day-sq m (3.4 Ib/day
sq ft) caused a deterioration in the underflow solids (53 to 45
g/£). For all loadings except the highest one, solids capture
was 99%; for the highest loading, solids capture dropped slightly
to 97%.
Figure 41 shows three typical thickener solids profiles from the
ferric-primary sludge of Phase 2. Clearly, it can be seen that
the full depth of the thickener blanket was being utilized in
thickening this type of sludge; as the sludge blanket depth
increased, the underflow total solids concentration increased.
Figures 42 and 43 summarize the results of the bench-scale
thickening tests of ferric-primary sludge from Phase 2. Also
shown on Figure 42 are the results of the four pilot plant
thickening periods presented in Table 28. Generally, pilot plant
results were consistent with predicted performance from bench-
scale tests. Figure 43 shows a relationship between the thick-
ener feed solids and predicted underflow solids. This is
basically substantiated by the pilot plant thickener results
presented in Table 28.
Phase 3 (Low Ferric; with Polymer)
Average pilot plant thickening results for three operating
periods from Phase 3 are presented in Table 29. Thickener feed
solids ranged from 33-52 g/A, with underflow solids ranging from
43-62 g/£. Higher thickener solids loadings were achieved
during this phase, and underflows produced were generally
slightly higher than those produced in the prior two phases.
High solids captures were also experienced (98-99%).
Figure 44 shows three typical thickener solids profiles. Near
maximum underflow concentration was achieved in the first 40 cm
(1.3 ft) of blanket depth with only nominal increases in under-
flow concentration occurring with further increases in blanket
depth.
Bench-scale thickening results are summarized on Figures 45 and
46. Shown on Figure 45 are the pilot plant thickener results.
Pilot thickener results are generally inconsistent with pre-
dicted performance from bench-scale results. Figure 46 shows
a definite relationship between the thickener feed and pre-
dicted thickener underflow concentration. This trend was sub-
stantiated by the results of the pilot plant thickener.
99
-------
o
I
IE
iu
o
u.
DC
UJ
o
o
UJ
CD
Q.
LU
O
UJ
C3
O
D
20
40
60
80
100
120
140
TOTAL SOLIDS—WEIGHT PERCENT
FIGURE 41. FERRIC-PRIMARY SLUDGE THICKENER PROFILES
PHASE 2.
(High Ferric; with Polymer)
100
-------
UNDERFLOW CONCENTRATION
o
(0
I
>
<
o
o
^
I
cs
z
Q
o
(A
DC
HI
Z
UJ
*
o
Q
UJ
Q
Ul
cc
Q.
225
200
175
150
125
100
75
50
25
O Cu = 2.0%
Q Cu - 3.0%
=4.0%
Cu = 5.0%
Cu = 6.0%
A PILOT PLANT
~ THICKENER
25
30
35
0 5 10 15 20
THICKENER FEED — g/l
FIGURE 42. PREDICTED TSL VS THICKENER FEED FOR FERRIC —
PRIMARY SLUDGE — PHASE 2.
(High Ferric; with Polymer)
101
-------
o> 80
I
O
F
cc
1- 70
Z
UJ
o
0
o
0 60
u_
GC
UJ
Q
Z
3
QC 50
QJ 3U
Z
UJ
o
I~"
O 40
HI
5
UJ
cc
Q.
5 30
X
S
•: :-;•:•:-;;
- — -
! i , i i ' !
: ! ! l \
r • ! : ' I i
1 ' ' i ' i j '
1 ' , • ' i ! ! !
• ''••;! I '
' ; ; '
1 !
J j_
^1 1 M ' ' •1~|-tT-~r
J 1
T-- -)— -T 1— L-
i J T~
!
i 1
1
1 i
6 BENCH-SCALE TESTS
• PILOT PLANT THICKENER RESULTS
I
i
- T
O
O 0
• • • -----
. ... . ,-: ......
' ' ' i
! ! II1' ' '
._„._,
- --r—
' • : ! ; , '
1 : , ' ' " '
..--4-^.-,.-T-_-..--^.-_.^T,....
-T-- rt 7 • f---'--f — r --+• — : T
1 ; i
'
4
i
O
o
0 n
A; Q
- 9- ' '
i !~
1 I I 1 1 '* "
J , | , i . ,,
1 — pr- n-r
rt i . t~r
! ' ' * ' "
!_.|-_T__;_._..J_Li..
ry-j— r-~'--i— f :~ -
i i
' ' ' ~T" u : ;
r -t- r- —.-..-.
— - -
o
• 0 |
... . . , ,
~*
A A
O
O °
• - - • -
10 15 20
THICKENER FEED-g/|
25
30
FIGURE 43.
MAXIMUM PREDICTED THICKENER UNDERFLOW
CONCENTRATION VS THICKENER FEED
CONCENTRATION — PHASE 2.
(High Ferric; with Polymer)
102
-------
TABLE 29
FERRIC-PRIMARY SLUDGE THICKENING
AVERAGE PHASE 3 PILOT PLANT RESULTS
(Low Ferric; with Polymer)
o
CO
Parameter
Values
Evaluation period, days
Thickener feed, g/£
Thickener loading, l/min-sq m
Thickener loading, Kg/day-sq m
Thickener sludge age, days
Thickener sludge depth, m
Solids capture, percent
Thickener underflow sludge, g/£
4
33
.65
30.8
.35
.27
99
43
4
52
.35
26.1
1.6
.71
99
62
8
38
.40
22.0
1.5
.63
98
60
-------
o
I
cc
UJ
o
EC
1U
O
O
_l
Ul
CD
U
O
UJ
o
o
20
40
60
80
100
120
140
D
FEEDTS = 21-53g/l
0123456 789 10
TOTAL SOLIDS—WEIGHT PERCENT
FIGURE 44. FERRIC-PRIMARY SLUDGE THICKENER PROFILES
PHASE 3.
(Low Ferric; with Polymer)
104
-------
I
>•
o
a
I
o
z
o
(0
DC
UJ
Z
III
500
450
400
350
300
250
200
y 150
Q
UJ
9 100
a
UJ
oc
a.
50
T-t
l.J . . I-
UNDERFLOW CONCENTRATION
O Cu = 3%
2T Cu = 4%
a Cu = 5%
• Cu = 6%
A Cu = 7%
• PILOT PLANT THICKENER
10 20 30
THICKENER FEED — g/l
FIGURE 45. PREDICTED TSL VS THICKENER FEED FOR FERRIC
PRIMARY SLUDGE — PHASE 3.
(Low Ferric; with Polymer)
105
-------
100
g
z
o
5
CC
I-
UJ
o
z
o
o
o
u.
DC
UJ
Q
Z
3
or
UJ
z
UJ
*
o
z
o
£
o
Q
UJ
flC
0.
5
5
X
<
O BENCH SCALE TESTS
• PILOT PLANT THICKENER RESULTS
L
o
10 20 30 40 50
THICKENER FEED — g/l
60
70
FIGURE 46.
MAXIMUM PREDICTED THICKENER UNDERFLOW
CONCENTRATION VS THICKENER FEED
CONCENTRATION — PHASE 3.
(Low Ferric; with Polymer)
106
-------
PRIMARY SLUDGE
Phase 4
During Phase 4, ferric-primary sludge was no longer produced.
Primary sludge instead was produced to attempt to expand the
thickening data base.
Table 30 summarizes the results of the three operating periods
from Phase 4. As expected, and as indicated by the first two
operating periods shown in Table 30, once chemical coagulant
addition to the flocculating clarifier was terminated, a sludge
with greatly enhanced thickening properties was produced. The
first operating period, for example, resulted in an average
thickener underflow concentration of 83
Interconnecting piping and tank elevations were not designed to
permit primary sludge, at such concentrations, to gravity flow
from the thickener to the underflow sludge storage tank. There-
fore, during the second and third operating periods, thickener
solids loading was increased and blanket depth was decreased in
an attempt to reduce the thickener underflow concentration.
Figure 47 shows the results of one thickener solids profile
conducted on primary sludge. This profile was taken when the
sludge level was very low, however, considerable thickening
still occurred.
Figures 48 and 49 summarize bench-scale thickening results for
primary sludge generated during Phase 4. The results of the
three pilot plant operating periods are also plotted on
Figure 48. Ignoring the operating period with the underflow
of 18 g/fc, due to the operating conditions it was run under,
pilot plant thickener performance correlated well with pre-
dicted bench-scale performance.
Figure 49 shows the results of the thickener feed versus thick-
ener underflow solids concentration. No pattern is discernable.
THICKENING COMPARISONS : ALUM-PRIMARY AND FERRIC-PRIMARY SLUDGES
Figures 50 and 51 show the results of all stable pilot plant
thickening periods experienced during the study. Maximum thick-
ener underflow solids are the maximum solids obtained in the
thickener underflow for any stable thickener operating period.
Figure 50 shows the impact of percent chemical solids present
in the thickener feed sludge on the maximum attainable thickener
underflow solids. Because of thickener operational problems,
discussed previously, the underflow concentration for 0%
chemical solids (i.e., primary sludge) is considered to be
lower than could have been achieved.
107
-------
TABLE 30
PRIMARY SLUDGE THICKENING
AVERAGE PHASE 4 PILOT PLANT RESULTS
Parameter Values
~~*.*-**o v-a^uuic, t'ement H4 95
Thickener underflow sludge, g/£ 83 66
97
o Evaluation period, days 17 6 14
Thickener feed, g/Jl 10 26 in
Thickener loading, Vmin-sq m .52 67 9
Thickener loading, Kg/day-sq m 7.3 25*3 13*9
Thickener sludge age, days 2.0 10 *02
Thickener sludge depth, m .24 [49 "02
Solids capture, percent 84 96 87*
18
-------
20
fcw
— 40
ig ~w
O
1
cc
? 60
O
LJ_
CC
UJ
° 80
O
UJ
ffi
x
Si 100
0
UJ
o
o
:t4:.j...J44±.t
.L , .--i-j-j- ..[-
... .; H-H— - •
D; "i ~ ! '""!"
I ! : j ' ~"t
i _ j- _|_ . _|.._
i ' ~" " *~ *~ "*"
. . '...,-, . '. !
. . : - :.. L. .
p: ; : : : i :
o —
o
. .
4- -14 -i!J
- j-- • !-!-(• : -t--
-\- — !-[-•:- - !
- • - - ~H i h-
-\--\-}- -]-! ;
-.J-4.-U- -4-| --j
-;- -1 -;-ft ;
•- - !- -[- '.- -- l
[ ! ' ! ' •
...j ... _U 1 ..;._,. ,
-.1 . _.. 1 . ,. . i
. L- ...... ,_;!;;
.
;- ; • i i ;•;-(-•{-
. ..] :._: 4 |...| .(.. !
- -L ::.;., i :
. -j. l. . : [ t ; (
1 ' i
, . . . . . I j .
,- l , ; ....
:,.,., (. i
.
* ! i ' I 1 i
1 i '1
-j4 ' i-l i 4.
"t"i'h4'i: it
i ; , i . -j j
...... j ,
.-,.]., i j .
-.;.... , . .
' !
FEED TS = 26 g/l
i
'
14,4;;-
! ! ' *
._i. .; M~ .
-h j.. - -1- -T
f. l-f. -.j ...
!-(---,.. 1 {__
1 - , 1 1— J , \ ..
', i ! - i . : ', , ''.
-.; .; i'1
i •
1 : •
•
1 !
O ;
o
120
140
8 9 10
TOTAL SOLIDS—WEIGHT PERCENT
FIGURE 47. PRIMARY SLUDGE THICKENER PROFILES — PHASE 4.
109
-------
o
I
>•
<
Q
O
X
I
O
z
0
(0
9
_i
o
(A
DC
lit
Z
UJ
*
o
Q
UJ
O
UJ
£C
Q.
1000
900
800
700
UNDERFLOW
I, , L i ! , -. ,
tttiJrCONCENTRATION
ffirS
• I / i i >. '•.
a cu = 4%
Cu = 5%
A PILOT PLANT
THICKENER RESULTS
200
100
10 15 20 25
THICKENER FEED — g/l
FIGURE 48- PREDICTED TSL VS THICKENER FEED FOR PRIMARY
SLUDGE — PHASE 4.
no
-------
g
I 70
UJ
o
z 60
O
o
I
d 50
DC
Ul
Q
Z
D
EC *u
UJ
UJ
Z
§ 2o
UJ
oc
0.
I 10
pmTT
.t-f-M-j-i--^.
tbutt'
BENCH—SCALE TESTS
I-F
=:rff
-r-* .....
10
15
20
25
30
35
THICKENER FEED — g/l
FIGURE 49. MAXIMUM PREDICTED THICKENER UNDERFLOW
CONCENTRATION VS THICKENER FEED
CONCENTRATION FOR PRIMARY SLUDGE — PHASE 4
ill
-------
o>
I
0)
o
u.
oc
UJ
o
z
D
OC
UJ
UJ
*
u
X
<
90
80
70
60
50
40
30
20
10
0 10 20 30 40 50 60 70
CHEMICAL SOLIDS IN SLUDGE - WEIGHT PERCENT
FIGURE 50 RANGE OF THICKENER OPERATING PERIODS FOR
ALUM-PRIMARY SLUDGE.
112
-------
100
o>
0)
Q
O
(A
O
a:
UJ
Q
DC
UJ
z
111
*
O
X
<
10
10 20 30 40 50 60
CHEMICAL SOLIDS IN SLUDGE - WEIGHT PERCENT
70
FIGURE 51. RANGE OF THICKENER OPERATING PERIODS FOR
FERRIC-PRIMARY SLUDGE.
113
-------
Figure 51 shows the relationship of percent chemical solids in
ferric-primary sludge on the maximum underflow solids. The
marked reduction in maximum attainable underflow solids that was
observed for alum-primary sludge was not present here. Also,
there is essentially no difference in maximum underflow solids
during any of the three phases conducted. As previously noted,
the range of attainable underflow solids for primary sludge is
considered to be somewhat lower than could have been achieved.
Table 31 shows a comparison of gravity thickener performance
parameters for alum-primary and ferric-primary sludge at various
levels of phosphorus removal. To review performance for these
sludges, a raw wastewater containing 100 mg/& TSS and 5.0 mg/£
total P was assumed. These characteristics correspond to
average Salt Lake City wastewater concentrations observed during
this study. The molar ratios of metal to phosphorus required
to achieve each level of phosphorus removal were deter-
mined from this study and from previous studies1'2.
The results in Table 31 indicate that, for similar levels of
phosphorus removal, the gravity thickening performance parameters
for ferric-primary sludge were superior to those for alum-primary
sludge. The results also indicate a decrease in thickener
underflow solids concentration as additional inorganic coagulant
is added to achieve greater phosphorus removal.
114
-------
TABLE 31
EFFECT OF PHOSPHORUS REMOVAL ON GRAVITY THICKENING
PROPERTIES OF ALUM-PRIMARY AND FERRIC-PRIMARY SLUDGE
Total Phosphorus
Removal-Percent
80
90
95
ALUM-PRIMARY SLUDGE
Chemical Sludge
Weight Percent
18
23
32
Underflow
Percent
TS
4.7
4.1
3.3
FERRIC -PRIMARY SLUDGE
Chemical Sludge
Weight Percent
22
28
38
Underflow
Percent
TS
5.5
5.4
5.3
Ul
BASIS: Raw wastewater with 100 ing/2, TSS and total phosphorus of 5 mg/fc P.
Solids loading rate -20 Kg/day-sq m
-------
VACUUM FILTRATION DEWATERING
Alum-Primary Sludge
Phase 1 (High Alum; no Polymer)
Alum-primary sludge vacuum filtration data obtained during Phase
1 is given in Table 32.
Form filtration rates are shown on Figure 52. These rates were
calculated exclusive of chemical conditioner in the filter cake
in order to simplify an evaluation of the effects of conditioning
dose. Solid lines through data points were extended over a range
of form time which resulted in filter cake dischargeability with
a rating of at least "fair". Dotted lines through data points
represent the range of form times which resulted in a cake dis-
charge rating of poor to non-dischargeable.
From Figure 52, it can be observed that filtration performance
was poor when the alum-primary feed sludge concentration was
approximately two weight percent. A lime conditioning dose of
37 weight percent resulted in poor discharge and low filtration
rates. One test conducted using a lime dose of 55 weight percent
did demonstrate dischargeable cakes; however, filtration rate
was also low.
When operating with an alum-primary sludge of approximately
three weight percent, dischargeable cakes were produced at a
lime dose of 37 weight percent. When the lime dose was reduced
to 29 weight percent, a "poor" cake discharge resulted, in
addition to lower filtration rates.
Figures 53 and 54 represent full-scale vacuum filter operating
curves predicted from the data in Table 32 and on Figure 52. The
filtration rates indicated on Figure 53 are pertinent to an alum-
primary sludge similar to that produced during Phase 1 thickened
to approximately three weight percent and dosed with 37 weight
percent lime. Note that a filter submergence of 30 percent was
assumed in constructing the full-scale filtration curve and that
conditioning chemical was included in the quoted rates.
Use of Figures 53 and 54 for design will be illustrated with the
following example. Assume that it would be desired to operate
a full-scale vacuum filter to produce a filter cake of 29 weight
percent total solids. From Figure 54, a correlating factor
116
-------
GO
<
O
I 20
til
I
O
O
O
X
I 10
O
CO
cc
O
UJ
cc
g
<
cc
O
O
SLOPE = -0.50
29.4 g/l Feed @ 37% Ca (OH)2
D 29.4 g/l Feed @ 29% Ca (OH)2
A 21.7 g/l Feed
© 21.7 g/l Feed
55% Ca (OH)2
37% Ca (OH)2
0.5 1.0 1.5
FORM TIME—MINUTES
2.0
3.0
FIGURE 52.
FORM FILTRATION RATE VS. FORM TIME FOR
ALUM-PRIMARY SLUDGE — PHASE 1.
(High Alum; no Polymer)
117
-------
00
TABLE 32
DATA SUMMARY OF VACUUM FILTRATION OF
ALUM-PRIMARY SLUDGE FROM PHASE 1
(High Alum; no Polymer)
== • . — —
Feed total solids, g/£
Conditioned feed total solids, g/fc
Lime dose, weight percent
Filter form time (9f ) , min.
Filter dry time (9 ,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
6d/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
Percent
— —
1
28.4
31.0
24
2.0
3.8
640
560
P
0.80
0.24
16
7.2
28.7
99
2
28.4
31.7
31
2.0
3.8
640
530
F-P
0.80
0.34
11
10.2
26.2
99
RUN
3
21.7
24.9
37
2.0
3.8
580
530
ND
1.2
0.24
16
7.2
28.0
98
NUMB
4
21,7
24.5
37
1.6
3.1
580
560
F
1.6
0.20
16
7.5
25.8
99
E R
5
21.7
37.3
55
2.0
3.8
610
560
P-F
1.6
0.39
10
11.7
25.9
99
— in
6
29.4
27.5
29
2.0
3.8
510
460
P
1.2
0.40
10
12.0
27.6
97
(continued)
-------
\D
TABLE 32. (continued 1
Feed total solids, g/i
Conditioned feed total solids, g/H
Lime dose, weight percent
Filter form time (9f), min.
Filter dry time (9^)/ min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Ca'ke thickness, mm
Cake dry weight (W) , Kg/sq m
0 ,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
7
29.4
37.7
29
1.3
2.6
510
460
P
1.2
0.34
7.6
15.7
27.8
99
RUN
8
29.4
37.4
37
1.3
2.6
510
460
F
1.2
0.41
6.3
18.9
27.8
97
N U M B E
9
29.4
38.0
37
2.0
3.8
510
460
F
1.6
0.56
6.8
16.8
29.2
99
R
10
29.4
37.8
37
1.0
1.9
510
460
P-F
1.2
0.38
5.0
22.8
28.7
99
11
29.4
37.8
37
0.8
1.5
510
510
P
0.80
0.39
3.9
28.5
28.3
97
Percent
-------
CO
<
o
5
UJ
u
o
5
D
_l
O
o
cp
cc
X
o
I
UJ
DC
z
o
cc
UJ
o
CO
10
9
8
7
6
5
29.4 g/l Feed @ 37% Ca(OH)2
30% SUBMERGENCE
6 7 8 9 10
15
CYCLE TIME—MINUTES
FIGURE 53.
FULL-SCALE FILTRATION RATE VS. CYCLE TIME FOR
ALUM-PRIMARY SLUDGE — PHASE 1.
(High Alum; no Polymer)
120
-------
ill
o
oc
Ul
Q.
I-
g
UJ
LLI
z
o
o
UJ
O
QC
UJ
40
35
30 -
25
20
29.4 g/l Feed @ 37% Ca (OH)2
6
8
CORRELATING FACTOR -0d/W (KG-MIN/SQ M)
FIGURE 54.
FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR ALUM-PRIMARY SLUDGE — PHASE 1
(High Alum; no Polymer)
121
-------
(6,/W) of 6.0 would be required
Since, W = (FSF?> (?^e Time) (3)
60 min/hr l *'
6d = 0.5 (Cycle Time)* (4)
where:
W = cake dry weight, Kg/sq m
FSFR = full-scale filtration rate, Kg/hr-sq m
Q
d = filter dry time, min
Then, ^d = 30.0
W FSPR
In the example, a full-scale filtration rate (FSFR) of 5.0 Kg/
hr-sq m (1.0 Ib/hr-sq ft) would be calculated with a 9d/W corre-
lating factor of 6.0. Referring to Figure 53, the filter would
have to be operated at a cycle time of 6.5 minutes, which is
within the acceptable range for cake discharge.
It should be noted that the FSFR of 5.0 Kg/hr-sq m (1.0 Ib/hr-
sq ft) includes chemical conditioning. Subtracting out the con-
ditioning lime gives a filter yield rate based on dry sludge
feed solids of 3.7 Kg/hr-sq m (0.73 Ib/hr-sq ft).
With the exception of volatile suspended solids, capture of feed
sludge constituents across the vacuum filter were not determined
during Phase 1. Volatile suspended solids capture were evaluated
and ranged from 97-99 weight percent with an average of 98
weight percent.
Phase 2 (High Alum; with Polymer)
Table 33 summarizes the results of the vacuum filter runs per-
formed during Phase 2 on alum-primary sludge.
A review of this data, shown on Figure 55, revealed that filtra-
tion rate was maximized at lime conditioning dose of approximate-
ly 35 weight percent as Ca(OH)2« Tests conducted at a lime dose
of 38 weight percent resulted in filtration rates equal to or
less than that observed at the 35 weight percent lime dose.
Similarly, lower filtration rates were observed at lime doses of
less than 35 weight percent.
*Most belt vacuum filters are constructed such that approximate-
ly 50 percent of the cycle time functions as cake dry time.
Other designs may be considered by modifying the constant, 0.5,
in equation (4).
122
-------
ro
TABLE 33
DATA SUMMARY OF VACUUM FILTRATION OF
ALUM-PRIMARY SLUDGE FROM PHASE 2
(High Alum; with Polymer)
Feed total solids, g/i
Conditioned feed total solids, g/£
Lime dose, weight percent
Filter form time (9f ) , min.
Filter dry time, (9,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
9 ,/W, Kg-min/sq m
Form filtration rate, (Kg.hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
=^==
1
28.0
32.7
15
2.0
3.8
510
480
E
1.2
0.37
10
11.1
28.0
99
2
28.0
32.7
15
1.3
2.6
510
480
G-E
1.2
0.41
6.3
18.9
27.6
98
RUN
3
28.0
32.7
15
1.0
1.9
530
510
G
0.80
0.32
5.9
19.2
27.0
98
N U
4
28.0
32.7
15
0.75
1.5
530
510
F
0.80
0.24
6.3
19.2
26.0
94
M B E R
5
28.0
34.4
30
2.0
3.8
510
480
E
2.5
0.56
6.8
16.8
27.8
99
6
28.0
34.4
30
1.3
2.6
510
480
G-E
1.6
0.48
5.4
22.2
27.7
96
7
28.0
34.4
30
0.75
1.5
530
510
G-E
1.2
0.38
3.9
30.4
28.0
99
(continued)
-------
10
TABLE 33 . (continued)
Feed total solids, g/£
Conditioned feed total solid, q/l
Lime dose, weight percent
Filter form time (0_ ) , min.
Filter dry time (6,), min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
0 ,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
8
28.0
34.4
30
0.59
1.2
530
510
G
1.2
0.32
3.8
32.5
27.9
98
9
29.0
37.3
35
2.0
3.8
510
480
E
2.4
0.77
4.9
23.1
28.8
95
R U
10
29.0
37.3
35
1.0
1.9
510
480
G
1.6
0.55
3.5
33.0
28.0
97
N N U
11
29.0
37.3
35
.59
1.2
530
510
F-G
1.2
0.40
3.0
40.7
25.7
99
M B E R
12
29.0
37.3
35
0.45
0.87
530
510
F-P
0.8
0.34
2.6
45.3
25.7
98
13
29.0
37.3
35
1.3
2.6
510
480
G-E
1.6
0.56
4.6
25.8
72.0
99
^=8=-
14
29.1
34.9
18
2.0
3.8
510
460
F-G
1.6
0.46
8.3
13.8
27.0
99
(continued)
-------
TABLE 33. (continued)
to
tn
Feed total solids, g/i
Conditioned feed total solids, g/jl
Lime dose, weight percent
Filter form time (6f ) , min.
Filter dry time (9 ,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
8 j/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
15
29.1
34.9
18
1.3
2.6
510
480
F-G
0.8
0.36
7.2
16.6
28.0
98
16
29.1
37.3
25
1.3
2.6
510
480
F-G
1.6
0.53
4.9
24.5
27.0
96
R U
17
29.1
37.7
25
1.0
1.9
510
480
F
1.6
0.47
4.0
28.2
26.4
96
N N U
18
29.1
36.6
25
0.75
1.5
510
480
F-P
0.8
0.37
4.1
29.6
26.3
93
M B E R
19
29.1
36.9
25
2.0
3.8
510
480
G
1.6
0.56
6.8
16.8
27.4
98
20
29.1
37.1
38
2.0
3.8
510
480
G-E
3.2
0.71
5.4
21.3
27.7
98
=ssss
21
29.1
37.1
38
1.0
1.9
510
480
F-G
1.6
0.49
3.9
29.4
27.4
96
percent
-------
W
g
UJ
o
o
o
X
UJ
o
00
flC
X
o
oc
z
o
OC
I-
oc
£
40
20
10
SLOPE=-0.50
28.0-29.1 g/l Feed
15-18% Ca(OH)2
25-30% Ca(OH)2
35% Ca(OH)2
. „_.____
0.5
1.0
2.0
3.0
FORM TIME—MINUTES
FIGURE 55-
FORM FILTRATION RATE VS. FORM TIME FOR
ALUM-PRIMARY SLUDGE — PHASE 2.
(High Alum; with Polymer)
126
-------
Figures 56 and 57 summarize the full-scale operating curves for
two levels of lime conditioning. Use of these curves is identi-
cal to the description contained in the previous section. The
data shown on Figure 57 indicate that cake dry solids concentra-
tion was independent of lime conditioning dose. This is con-
trary to what was expected. It was hypothesized that the cake
was at near maximum solids concentration due to the thin cakes
0.8-3.2 mm (0.3-0.13 inch) produced and, therefore, the effect
of the lime component was minimal.
Capture of the volatile suspended solids and aluminum constitu-
ents of the filter feed sludge were determined. Volatile solids
capture ranged from 93-99 weight percent, with an average of 98
weight percent. Aluminum capture averaged 88 weight percent.
Phase 3 (Low Alum; with Polymer)
Table 34 summarizes the vacuum filter data developed from opera-
tion on the alum-primary sludge generated during Phase 3.
Figure 58 shows the vacuum filter results for the five operating
conditions evaluated. From Figure 58, it can be seen that form
filtration rate increased as the lime conditioning dose increas-
ed from 15 weight percent to 26 weight percent over a feed solids
concentration range of 47.2 to 61.4 g/£ TS. The data also
dempnstrated that increases in feed TS concentration resulted
in increases in form filtration rate at constant lime con-
ditioning dose.
From Table 34, the results of runs 12-15 indicate that increas-
ing the lime dose to 35 weight percent resulted in lower filtra-
tion rates than those shown on Figure 58 for 25 weight percent
lime addition at feed concentrations ranging from 47.2 to 61.4
g/£ TS. With lower feed concentrations, 40.6 q/i TS, increasing
the lime dose to 37 weight percent may be deduced to have
resulted in an increase in filtration rate.
Figures 59 and 60 show the predicted full-scale filtration per-
formance for the four operating conditions judged to be most
typical for the alum-primary sludge generated in Phase 3. As
observed in Phase 2, filter cake solids content was found to
be independent of feed sludge concentration and lime condition-
ing dose.
Phosphorus, volatile suspended solids and aluminum constituent
captures were determined during Phase 3. Volatile suspended
solids capture ranged from 96 to 99 weight percent with an
average of 98 weight percent. Both phosphorus and aluminum
capture averaged 97 weight percent.
127
-------
2
<
O
i
UJ
I
u
o
1
_i
o
o
0)
i
I
o
X
,1
5
CC
z
cc
UJ
g
0)
20
15
10
9
8
7
6
1
TKtfHi-
±Hi
28.0-29.1 g/l Feed
30% SUBMERGENCE
1-.-.+-- 4—I--, (--, j
ri-r-rrrrr
6 7 8 9 10
CYCLE TIME—MINUTES
FIGURE 56.
FULL-SCALE FILTRATION RATE VS. CYCLE TIME
FOR ALUM-PRIMARY SLUDGE — PHASE 2.
(High Alum; with Polymer)
128
-------
UJ
o
cc
UJ
Q.
g
UJ
o
o
0)
g
_i
o
UJ
*
<
o
QC
UJ
35
30
25
20
FIGURE 57
28.0-29.1 g/l Feed
25-30% Ca(OH)2
35% Ca(OH)2
6
8
CORRELATING FACTOR-Gd/W (KG-MIN/SQ M)
FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR ALUM-PRIMARY SLUDGE — PHASE 2
(High Alum; with Polymer)
129
-------
OJ
o
TABLE 34
DATA SUMMARY OF VACUUM FILTRATION OF
ALUM-PRIMARY SLUDGE FROM PHASE 3
(Low Alum; with Polymer)
Feed total solids, g/Z
Conditioned feed total solids, g/i
Lime dose, weight percent
Filter form time (9f ) , min.
Filter dry time (0 ,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weihgt (W) , Kg/sq m
6^/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
— — ™^^^^^^^^™«
1
40.6
53.2
37
2.0
3.8
510
460
E
4.8
1.3
2.9
39.0
26.3
99
=
2
40.6
53.2
37
1.0
1.9
510
460
E
3.2
0.96
2.0
57.6
25.8
99
RUN
3
40.6
53.2
37
0.59
1.2
510
480
G
2.4
0.73
1.6
74.2
26.2
99
N U M
4
40.6
53.2
37
0.41
0.80
510
480
F
1.6
0.59
1.4
86.3
26.2
98
B E R
5
47.2
48.8
15
2.0
3.8
510
460
G
2.4
0.64
5.9
19.2
24.8
99
6
47.2
48.8
15
1.0
1.9
510
430
F-G
1.6
0.47
4.0
28.2
25.2
99
(continued)
-------
TABLE 34. (continued)
Feed total solids, g/£
Conditioned feed total solids, g/£
Lime dose, weight percent
Filter form time (9f ) , min.
Filter dry time (63) / min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
0,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
7
47.2
48.8
15
0.75
1.5
510
460
F-P
1.2
0.38
3.9
30.4
25.2
98
8
47.2
51.6
25
1.0
1.9
530
480
G
2.4
0.63
3.0
37.8
26.0
98
R U
9
47.2
51.6
25
0.75
1.5
530
480
G
2.0
0.54
2.8
43.2
25.8
96
N N U M
10
47.2
51.6
25
0.59
1.2
530
460
F
1.6
0.52
2.3
52.9
24.0
98
B E R
11
47.2
51.6
25
2.0
2.0
530
480
E
3.2
0.89
4.3
26.7
25.6
97
12
57.7
68.8
35
2.0
2.0
510
450
G
1.6
0.79
4.8
23.7
35.4
99
(continued)
-------
TABLE 34. (continued)
— — — — — __ — —
Feed total solids, g/S,
Conditioned feed total solids, g/£
Lime dose, weight percent
Filter form time (6f ) , min.
Filter dry time (0,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
6 ,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
13
57.7
68.8
35
1.0
1.9
510
450
F
1.2
0.50
3.8
30.0
35.6
99
14
57.7
68.8
35
0.75
1.5
510
460
ND
0.8
0.43
3.5
34.4
34.1
RUN
15
57.7
68.8
35
1.3
2.6
510
430
F-G
1.2
0.56
4.6
25.8
35.1
99
NUMB
16
61.4
69.1
16
2.0
3.8
510
460
E
3.2
1.1
3.5
33.0
26.3
99
E R
17
61.4
69.1
16
1.0
1.9
510
460
F-G
2.4
0.76
2.5
45.6
27.6
99
18
61.4
69.1
16
0.59
1.2
510
460
P-F
1.6
0.59
2.0
60.0
26.7
99
(continued)
-------
TABLE 34. (continued)
Ul
to
Feed total solids, g/H
Conditioned feed total solids, g/£
Lime dose, weight percent
Filter form time (Qf ) , min.
Filter dry time (9 ,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, nun
Cake dry weight (W) , Kg/sq m
6,/W, Kg-min/sq m
.Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
19
61.4
69.1
16
1.3
2.6
510
460
G
2.8
0.80
3.3
36.9
26.8
98
20
61.4
70.0
26
2.0
3.8
510
480
E
4.8
1.4
2.7
42.0
26.0
99
RUN
21
61.4
70.0
26
1.0
1.9
510
460
E
4.0
1.1
1.7
66.0
24.6
99
NUMB
22
61.4
70.0
26
0.59
1.2
510
460
G
3.2
0.80
1.5
81.4
25.0
99
E R
23
61.4
70.0
26
0.41
0.80
510
460
F
2.4
0.67
1.2
98.0
25.0
98
24
61.4
70.0
26
0.32
0.62
510
480
P
1.6
0.59
1.1
111
25.0
98
percent
-------
s
5
u
O
o
5
D
O
X
UJ
100
90
80
70
60
50
40
30
o
UJ
cc
z
g
5
QC
CC
O
U.
20
10
SLOPE=-0.50
O 47.2 g/l Feed @ 15% Ca(OH)2
8 47.2 g/l Feed @ 25% Ca(OH)2
~TJ 61.4 g/l Feed @ 16% Ca(OH)2
—&• 61.4 g/l Feed @ 26% Ca(OH)2
_A. 40.6 g/l Feed @ 37% Ca(OH)2
!
—i
0.3
0.5 1.0
FORM TIME—MINUTES
2.0
FIGURE 58. FORM FILTRATION RATE VS. FORM TIME FOR
ALUM-PRIMARY SLUDGE — PHASE 3.
(Low Alum; with Polymer)
134
-------
(A ^« i=-±
O
i
UJ
O
O
S
D
O
00
oc
I
UJ
<
cc
O
Ul
_l
<
O
1 2 3 456789 10
CYCLE TIME—MINUTES
FIGURE 59. FULL-SCALE FILTRATION RATE VS. CYCLE TIME
FOR ALUM-PRIMARY SLUDGE — PHASE 3.
(Low Alum; with Polymer)
15
135
-------
FILTER CAKE SOLIDS CONTENT — WEIGHT PERCENT
-* i\j i\j
01 O 01 0 01
• j ,
1
1
i
i i
-
' . ' • i :
• ' < !
; '!|'.
-
-
/
__
^
r
L
—
^
TT
J
If
J*
n
H
X
iJ
+
i
i
i
i
; ;
i
i ! 1
3
V
<
[-:
j:
1
- •$
— •—
£
.—
•^M.
—
-
^•^^
i
--
C,
!
--T^
-----
—
±
; :
..._.
- .I—. — L_,
: • — -
1
,
g/l Feed @ 15% Ca(OH)a
g/l Feed @ 25% Ca(OH)2
g/l Feed @ 16% Ca(OH)2
g/l Feed @ 26% Ca(OH)2
g/l Feed @ 37% Ca(OH)2
£
-
>
—
-
-
- —
...
-
i !
,
1
1
i
i
—^-i-4-^-
—
— — i — i — : — :_:._._:_
— -
___..__
---
— -
-- - — : -
-- ---
—
, •_ _ .._ . .:_..
-— - --
8
CORRELATING FACTOR — Od/W (KG-MIN./SQ M)
10
FIGURE 60. FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR ALUM-PRIMARY SLUDGE — PHASE 3.
(Low Alum; with Polymer)
136
-------
Phase 4 (High Alum; with Polymer)
A summary of the filter runs conducted on the alum-primary sludge
generated during Phase 4 is shown in Table 35.
A review of this data, shown on Figures 61, indicated that form
filtration rates were maximized at a lime conditioning dose of
approximately 26 weight percent as Ca(OH)2.
The full-scale vacuum filter performance curves are plotted on
Figures 62 and 63. As was observed in all previous runs, filter
cake solids content was unaffected by lime conditioning dose
within the range studied.
Phosphorus, volatile suspended solids and aluminum captures were
determined during Phase 4. Volatile suspended solids capture
ranged from 94 to 99 weight percent. Phosphorus and aluminum
captures determined from results of one vacuum filter test
series indicated average captures of 80 and 89 weight percent,
respectively.
FERRIC-PRIMARY SLUDGE
Phase 1 (High Ferric; no Polymer)
Table 36 represents the operating data developed for vacuum
filtration of ferric-primary sludge generated during Phase 1.
Figure 64 shows a graphical summary of the three operating con-
ditions studied during this Phase. As indicated, the -0.55
slope of the logarithmic plot of form filtration rate versus
form time was only slightly in excess of the theoretical -0.5
slope.
The data on Figure 64 indicate an increase in form filtration
rate due to an increase in lime conditioning dose from 18 per-
cent to 30 percent. A further rate increase was observed with
a lime conditioning dose increase to 47 percent/ however, the
full extent of the rate increase was confused due to vacuum
fluctuations which were experienced at this lime dose.
Severe filter cake cracking during the cake drying cycle was
the cause of the vacuum fluctuations experienced at the 47
percent lime conditioning dose. The diminution in vacuum level
also resulted in wetter filter cakes. A review of the data in
Table 36 indicated that the vacuum diminution resulted in two to
eight percentage points decrease in filter cake dry solids con-
tent as compared to filter operation at 30 percent lime. The
conclusion was, therefore, that filter operation at a lime
conditioning dose of 47 percent would not be advisable.
137
-------
00
TABLE 35
DATA SUMMARY OF VACUUM FILTRATION OF
ALUM-PRIMARY SLUDGE FROM PHASE 4
(High Alum; with Polymer)
Feed total solids, g/£
Conditioned feed total solids, g/&
Lime dose, weight percent
Filter form time (9f ) , min.
Filter dry time (6j)/ min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
0 ,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
1
55.2
66.4
35
2.0
3.8
510
460
G-E
3.2
0.93
4.1
27.9
28.4
98
RUN
2
55.2
66.4
35
0.75
1.5
510
460
F-G
2.0
0.60
2.5
48.0
27.0
97
N U
3
55.2
66.4
35
0.41
0.80
510
460
P-F
1.2
0.42
1.9
61.5
26.7
89
M B E R
4
55.2
66.4
35
1.0
1.9
510
460
G
2.4
0.67
2.8
40.2
26.7
92
5
57.4
64.9
26
2.0
3.8
530
510
G-E
3.2
0.80
4.8
24.0
27.2
99
-------
TABLE 35. (continued)
CJ
Feed total solids, g/H
Conditioned feed total solids, g/H
Lime dose, weight percent
Filter form time (6 ) , min.
Filter dry time (0 ,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
9 ,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
RUN
6
57.4
64.9
26
1.0
1.9
510
480
F-G
1.6
0.58
3.3
34.8
26.8
99
NUMB
7
57.4
64.9
26
0.59
1.1
510
460
F-P
1.2
0.45
2.4
45.8
25.8
90
E R
8
57.4
64.9
26
0.75
1.5
530
480
F-G
1.6
0.47
3.2
37.6
25.3
90
percent
-------
GO
<
o
s
LU
O
O
o
X
UJ
50
40
30
o
GO
OC
I
O 20
I
UJ
s
oc
z
o
2 10
oc
o
SLOPE=-0.50
U ~ 55.2 g/l Feed @ 35% Ca(OH)2
O~ 57.4 g/l Feed @ 26% Ca(OH)2
I r
0.4 0.5 0.6 0.7
1.0
1.5
2.0
FORM TIME—MINUTES
FIGURE 61. FORM FILTRATION RATE VS. FORM TIME FOR
ALUM-PRIMARY SLUDGE — PHASE 4
(High Alum; with Polymer)
140
-------
<
o
UJ
o
o
S
o
o
(0
o
*
I
UJ
5
cc
z
cc
I-
UJ
<
o
en
10
30% SUBMERGENCE
6 7 8 9 10
15
FIGURE 62.
CYCLE TIME—MINUTES
FULL-SCALE FILTRATION RATE VS. CYCLE TIME
FOR ALUM-PRIMARY SLUDGE — PHASE 4.
(High Alum; with Polymer)
141
-------
UJ
o
DC
UJ
OL
UJ
.1
O
O
UJ
DC
UJ
35
30
25
20
15
55.2 g/l Feed @ 35% Ca(OH)2
157.4 g/l Feed @ 25% Ca(OH)2 i H
_U!_UU- L U
..1_LJ—I—' -!-•--.
"'
-frKii-rr
| I i ! ! |
8
CORRELATING FACTOR—9d/W (KG-MIN/SQ M)
FIGURE 63.
FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR ALUM-PRIMARY SLUDGE — PHASE 4.
(High Alum; with Polymer)
142
-------
u>
TABLE 36
DATA SUMMARY OF VACUUM FILTRATION OF
FERRIC-PRIMARY SLUDGE FROM PHASE 1
(High Ferric; no Polymer)
Feed total solids, g/£
Conditioned feed total solids, g/i
Lime dose, weight percent
Filter form time (6f ) , min.
Filter dry time (6,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
0 ,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
1
41.2
48.1
47
2.0
3.8
250
230
E
5.6
1.4
2.7
42.0
26.1
99
2
41.2
49.6
47
1.0
1.9
330
250
E
3.6
1.0
1.9
60.0
28.2
99
R U
3
41.2
50.0
47
0.59
1.2
410
360
G
2.4
0.83
1.5
84.4
29.6
99
N N U
4
41.2
50.5
47
0.41
0.80
470
420
F-G
1.6
0.68
1.2
99.5
31.7
98
M B E
5
46.5
47.2
18
2.0
3.8
580
510
G
1.6
0.58
6.6
17.4
27.8
97
R
6
46.5
47.2
18
1.3
2.6
580
510
F
1.2
0.44
5.9
20.3
29.5
98
7
46.5
47.2
18
1.6
3.1
580
510
F
1.2
0.49
6.3
18.4
30.5
98
(continued)
-------
TABLE 36.
Feed total solids, g/i
Conditioned feed total solids, g/l
Lime dose, weight percent
Filter form time (0f ) , min.
Filter dry time (6,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
6,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
3
46.5
47.2
18
1.1
2.2
580
510
P
0.80
0.39
5.6
21.3
31.7
99
(continued)
9
46.5
48.2
30
2.0
3.8
580
510
G
2.4
.68
5.6
20.4
33.7
-
P. U
10
46.5
48.2
30
1.6
3.1
580
510
G
2.0
0.67
4.6
25.1
33.6
99
N N U
11
46.5
48.2
30
1.1
2.2
580
510
F
2.0
0.63
3.5
34.4
33.7
99
M B E R
12
46.5
48.2
30
2.0
3.8
580
510
G-E
2.4
0.78
4.9
23.4
31.4
99
13
46.5
48.2
30
1.0
1.9
580
510
G
1.6
0.58
3.3
34.8
34.1
99
14
46.5
48.2
30
1.6
3.1
580
510
G
2.4
0.59
5.3
22.1
34.0
99
-------
3 70
I 60
ill
g 50
O
1 40
o
X
UJ
o
00
cc
0
I
UJ
cc
O
DC
O
30
20
10
_j. SLOPE=-0.55
A 41.2 g/l Feed @ 47% Ca(OH)2
^46.5 g/l Feed @ 18% Ca(OH)2
D 46.5 g/l Feed @ 30% Ca(OH)2
0.5 1.0
2.0
FORM TIME—MINUTES
FIGURE 64.
FORM FILTRATION RATE VS FORM TIME FOR
FERRIC-PRIMARY SLUDGE — PHASE 1.
(High Ferric; no Polymer)
145
-------
Figures 65 and 66 show the full-scale vacuum filter operating
curves for the ferric-primary sludge produced in Phase 1. Un-
like the observations from alum-primary sludge dewatering, filter
cakes were generally drier at a 30 percent lime dose than at an
18 percent dose.
With the exception of volatile suspended solids, capture of feed
sludge constituents across the vacuum filter was not determined
during Phase 1. Volatile suspended solids capture ranged from
97 to 99 weight percent and averaged 98 weight percent.
Phase 2 (High Ferric; with Polymer)
A tabulation of the vacuum filtration operating results for
ferric-primary sludge dewatering during Phase 2 is shown in
Table 37.
Filter cake cracking also occurred during this phase at lime con-
ditioning doses above approximately 20 percent. This resulted
in vacuum level diminution and a decrease in cake solids content
of zero to three percentage points as compared to the cakes pro-
duced at either a 10 to 20 percent lime dose. No effect on
cake solids content was observed at lime doses from 10 to 20
percent.
Figure 67 summarizes the cake formation rate versus form time
data developed during Phase 2. At a feed sludge concentration
of 43.6 to 44.6 g/H TS, benefits in form filtration rate were
not observed with an increase in lime dose from 10 to 20 per-
cent. An increase in form filtration rate did occur, however,
when the lime dose was increased further to 24 percent. At a
feed sludge concentration of 53.3 g/i, an increase in form fil-
tration rate was observed with an increase in lime conditioning
dose from 10 percent to 19 percent. A further increase in lime
dose to 26 percent (at a 60.2 g/£ feed TS) resulted in erratic
performance due to vacuum fluctuations which were caused by
cake cracking.
Figure 68 summarizes operating results from the more important
operating conditions of Phase 2 on ferric-primary sludge. These
curves, in addition to the information shown on Figure 69, rep-
resent full-scale vacuum filter performance and may be used for
predictive purposes as explained above.
Capture of the volatile suspended solids, phosphorus and iron
constituents of the filter feed sludge were determined. Vola-
tile suspended solids capture ranged from 95 to 99 weight per-
cent with an average of 98 weight percent. Both phosphorus and
iron capture averaged 99 weight percent.
146
-------
CO
2
2
Ul
O
O
z
Q
O
O
CO
OC
O
I
Ul
oc
z
O
QC
uT
UJ
O
GO
10
9
8
7
6
5
30% SUBMERGENCE
5 6 7 8 9 10
15
CYCLE TIME—MINUTES
FIGURE 65.
FULL-SCALE FILTRATION RATE VS. CYCLE TIME FOR
FERRIC-PRIMARY SLUDGE — PHASE 1.
(High Ferric; no Polymer)
147
-------
FILTER CAKE SOLIDS CONTENT — WEIGHT PERCENT
ro 10 co u -u •c*
o en o in o en
;:::::.;:::
• :
-
- -
:•::.. :O
: : : : • :Q
. , . _ . _
0-1—
-
.
-
-
1 --
1 —
-
.. . —
--
--
-
i i
w
- -----
4J
-n-
IP
_ .
- -;—
l-
-
—
r
JL
].
: O
..
— -
-
—
; i ill
j j i
ill '
I !
i i
i
i ' ;
-
-
i
;
| :
1 i
1
! i
! '
— 1 —
~
-
-
I I
: ! I
. i i i
i : '
^
1
X
H-t--^-
-— )—
i
^i '
, —
I
; •)
1
46.5 g/l Feed (ft 18% Ca(6H)2
46.5 g/l Feed @ 30% Ca(OH)2
• ; --
. . . ' ; J i
; i ! - i
i
;
-
1
T
I
6
8
FIGURE 66.
CORRELATING FACTOR—Gd/W (KG-MIN/SQ M)
FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR FERRIC-PRIMARY SLUDGE —
PHASE 1.
(High Ferric; no Polymer)
148
-------
vo
TABLE 37
DATA SUMMARY OF VACUUM FILTRATION OF
FERRIC-PRIMARY SLUDGE FROM PHASE 2
(High Ferric; with Polymer)
Feed total solids, g/H
Conditioned feed total solids, g/£
Lime dose, weight percent
Filter form time (0^) , min.
Filter dry time (6d) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
•6 ,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
1
43.6
49.3
24
2.0
3.8
250
200
E
4.8
1.4
2.7
42.0
34.8
99
R U
2
43.6
49.3
24
0.41
0.80
460
360
F-P
1.6
0.63
1.3
92.2
33.9
95
N N
3
43.6
49.3
24
0.59
1.2
410
330
F-G
2.0
0.75
1.6
76.3
32.0
98
U M B E
4
43.6
49.3
24
1.0
1.9
330
280
G-E
3.2
0.95
2.0
57.0
32.3
96
R
5
44.6
45.5
10
2.0
3.8
510
480
E
3.2
0.87
4.4
26.1
34.4
99
6
44.6
45.5
10
1.0
1.9
530
510
G-E
1.6
0.60
3.2
36.0
34.2
98
(continued)
-------
Oi
O
TABLE 37.
Feed total solids, g/i
Conditioned feed total solids, g/i
Lime dose, weight percent
Filter form time (6f ) , min.
Filter dry time (0,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
9d/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
7
44.6
45.5
10
0.75
1.5
530
510
G
1.2
0.51
2.9
40.8
38.8
97
(continued)
8
44.6
45.5
10
1.3
2.6
530
510
G-E
1.6
0.57
4.6
26.3
35.0
99
RUN
9
44.6
47.4
20
2.0
3.8
510
480
G-E
1.6
0.65
5.8
19.5
35.7
99
NUMB
10
44.6
47.4
20
1.0
1.9
530
480
G
1.6
0.54
3.5
32.4
34.4
98
E R
11
44.6
47.4
20
0.75
1.5
510
480
F
1.2
0.51
2.9
40.8
34.5
95
==:
12
44.6
47.4
20
1.3
2.6
510
510
G-E
1.6
0.77
3.4
35.5
36.5
98
(continued)
-------
TABLE 37,
Feed total solids, g/i
Conditioned feed total solids, g/t.
Lime dose, weight percent
Filter form time (6 f) , min.
Filter dry time (0d) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
01 Cake thickness, mm
U-l
r^
Cake dry weight (W) , Kg/sq m
0,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
13
55.3
61.2
19
1.1
2.2
430
360
E
4.0
1.4
3.1
76.4
33.0
99
(continued)
14
55.3
61.2
19
0.41
0.80
510
460
E
2.0
0.88
0.91
129
35.1
99
RUN
15
55.3
61.2
19
0.59
1.2
510
460
E
3.2
1.0
1.2
102
34.8
99
N U M B E
16
55.3
61.2
19
0.36
0.69
510
480
G-F
2.4
0.77
0.90
128
34.2
98
R
17
55.3
58.7
10
1.1
2.2
510
480
G
2.4
0.95
2.3
51.8
34.2
99
18
55.3
58.7
10
0.75
1.5
510
460
G-F
1.6
0.78
1.9
62.4
33.8
99
(continued)
-------
TABLE 37. (continued)
to
Feed total solids, g/fc
Conditioned feed total solids, g/i
Lime dose, weight percent
Filter form time (8-) , min.
Filter dry time (9j)/ min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake Thickness, mm
Cake dry weight (W) , Kg/sq m
6,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
19
55.3
58.7
10
2.0
3.8
510
430
E
2.4
1.1
3.5
33.0
36.4
99
RUN
20
60.2
70.2
26
2.0
3.8
330
280
E
4.8
1.5
2.5
45.0
30.7
99
N U M
21
60.2
70.2
26
0.59
1.2
380
330
E
3.2
1.2
1.0
102
31.8
96
B E R
22
60.2
70.2
26
0.41
0.80
380
360
G-E
2.4
1.1
0.73
161
32.4
96
23
60.2
70.2
26
0.32
0.62
460
380
G-F
2.4
0.93
.67
174
32.8
96
percent
-------
300
3
<
O
5
LU
O
O
O
X
UJ
O
CO
cc
x
O
UJ
s
DC
O
oc
O
200
150
100
90
80
70
60
50
40
30
20
15
SLOPE=-0.50
A 60.2 g/l Feed @ 26% Ca(OH)2
--&• 43.6 g/l Feed @ 24% Ca(OH)2
O 44.6 g/l Feed @ 10% Ca(OH)2
© 53.3 g/l Feed @ 10% Ca(OH)2
0 44.6 g/l Feed @ 20% Ca(OH)2
® 53.3 g/l Feed @ 19% Ca(OH)2
0.3
0.5
1.0
1.5
2.0
3.0
FORM TIME—MINUTES
FIGURE 67
FORM FILTRATION RATE VS. FORM TIME FOR
FERRIC-PRIMARY SLUDGE — PHASE 2-
(High Ferric; with Polymer)
153
-------
00
o
5
iij
i
O
o
z
o
o
o
(/)
cc
i
o
*
I
LU
z
o
oc
LU
<
o
00
_J
30% SUBMERGENCE
3 4 56789 10
CYCLE TIME—MINUTES
15
FIGURE 68.
FULL-SCALE FILTRATION RATE VS. CYCLE TIME FOR
FERRIC-PRIMARY SLUDGE — PHASE 2.
(High Ferric; with Polymer)
154
-------
45
LU
O
CC
UJ
Q.
h-
g
UJ
UJ
o
o
CO
g
_i
o
CO
UJ
o
CC
UJ
40
35
30
25
20
O 44.6 g/l Feed @ 10 % Ca(OH)2
0 53.3 g/l Feed @ 10 % Ca(OH)2
D 44.6 g/l Feed @ 20 % Ca(OH)2
a 53.3 g/l Feed @ 19 % Ca(OH)2
8
10
FIGURE 69
CORRELATING FACTOR—9d/W (KG-MIN/SQ M)
FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR FERRIC-PRIMARY SLUDGE —
PHASE 2.
(High Ferric; with Polymer)
155
-------
Phase 3 (Low Ferric; with Polymer)
Table 38 summarizes the results of the vacuum filter runs per-
formed during Phase 3 on ferric-primary sludge. These results
are presented graphically on Figure 70.
As shown, feed solids concentration from 38.1 g/£ to 53.3
and lime conditioning doses from 16 percent to 56 percent did
not significantly affect form filtration rate. This observation
was inconsistent with much of the experience gained throughout
this study on both alum-primary and ferric-primary sludges.
A review of the operating results shown on Figure 70 indicated
that form filtration rate was maximized at a lime conditioning
dose of 31 percent for a feed TS of 38.1 g/£. Maximization
of form filtration rate was also achieved for a 53.3 g/i TS
feed conditioning with lime at a dose of 16 percent. It is not
known if the lime dose could have been further reduced for
either of these feed conditions without any reduction of form
filtration rate.
Figure 71 summarizes the two operating conditions discussed
above. The information provided on Figure 71 and on Figure 72
may be used for full-scale predictive purposes as described in
previous sections. Note that the results plotted on Figure 72
indicate that drier filter cakes were generally produced at
the higher lime conditioning doses used during these filter
runs.
Capture of vacuum filter feed sludge constituents evaluated
during Phase 3 included volatile suspended solids, phosphorus
and iron. Volatile suspended solids captures exceeded 99
weight percent whenever measured. Phosphorus capture averaged
92 weight percent and iron capture averaged 98 weight percent.
PRIMARY SLUDGE
Phase' 4
Since only a relatively small quantity of primary sludge was
produced each day, only bench-scale filter leaf tests were run
on primary sludge. Table 39 summarizes the results of these
tests.
Figure 73 shows a graphical summary of the bench-scale leaf
test results for the primary sludge. As indicated, the slope
of the form filtration rate - form time line was observed to
be significantly greater than the theoretical -0.5. There
are generally two reasons for a slope greater than -0.5.
156
-------
TABLE 38
DATA SUMMARY OF VACUUM FILTRATION OF
FERRIC-PRIMARY SLUDGE FROM PHASE 3
(Low Ferric; with Polymer)
Feed total solids, g/H
Conditioned feed total solids, g/fc
Lime dose, weight percent
Filter form time (B f) , min.
Filter dry time (6j)/ min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
6 /W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
1
38.1
47.8
31
2.0
3.8
510
460
F
1.2
0.54
7.0
16.2
34.0
99
2
38.1
47.8
31
1.3
2.6
510
460
P-F
0.8
.43
6.0
19.8
33.3
99
RUN
3
38.1
47.8
31
1.0
1.9
510
460
P-ND
0.8
0.35
5.4
21.0
34.3
99
N U
4
38.1
50.3
45
2.0
3.8
510
480
F
1.2
0.57
6.7
17.1
37.4
«.
M B E R
5
38.1
50.3
45
1.3
2.6
530
480
P-F
0.8
0.50
5.2
23.1
36.3
99
6
38.1
50.3
45
1.0
1.9
510
480
P-ND
0.8
0.41
4.6
24.6
36.2
99
7
38.1
51.4
56
2.0
3.8
510
480
F
1.2
0.61
6.2
18.3
37.3
99
(continued)
-------
TABLE 38. (continued)
Ul
00
Feed total solids, g/i
Conditioned feed total solids, g/£
Lime dose, weight percent
Filter form time (9f), min.
Filter dry time (9d) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
9,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
8
38.1
51.4
56
1.0
1.9
510
480
P
1.2
0.47
4.0
28.2
36.4
99
9
53.3
62.8
16
2.0
3.8
510
460
F-G
1.6
0.54
7.0
16.2
34.8
99
RUN
10
53.3
62.8
16
1.3
2.6
510
460
P-F
C.8
0.41
6.3
18.9
33.6
99
NUMB
11
53.3
62.8
16
1.6
3.1
510
460
F
0.8
0.46
6.7
17.3
33.4
99
E R
12
53.3
65.4
24
2.0
3.8
510
460
F-G
1.6
0.55
6.9
16.5
34.8
99
13
53.3
65.4
24
1.3
2.6
510
480
P
0.8
0.36
3.6
16.6
37.9
99
-------
30
o
LU
O
O
z
Q
3
_J
O
X
UJ
cc
X
o
HI
DC
Z
20
10
OC
O
O
D
SLOPE=-0.50
38.1 g/l Feed @ 31 % Ca(OH)2
38.1 g/l Feed @ 45 % Ca(OH)2
A 38.1 g/l Feed @ 56 % Ca(OH)2
e 53.3 g/l Feed @ 16 % Ca(OH)2
ffl 53.3 g/l Feed @ 24 % Ca(OH)2
0.5
1.0
1.5
2.0
3.0
FORM TIME—MINUTES
FIGURE 70.
FORM FILTRATION RATE VS. FORM TIME FOR
FERRIC-PRIMARY SLUDGE — PHASE 3.
(Low Ferric; with Polymer)
159
-------
~ 10
_
* 9
2 8
7
6
5
UJ
I
o
o
o
o
2 4
5
O
i 3
X
o
;
<
QC
Z
o
DC
I-
LU
O
0)
30% SUBMERGENCE
6 7 8 9 10
15
CYCLE TIME—MINUTES
FIGURE 71.
FULL-SCALE FILTRATION RATE VS. CYCLE TIME
FOR FERRIC- PRIMARY SLUDGE — PHASE 3.
(Low Ferric; with Polymer)
160
-------
UJ
o
cc
111
a.
x
g
UJ
z
UJ
z
o
o
9
o
UJ
o
oc
UJ
45
40
35
30
25
20
38.1 g/l Feed @ 31 % Ca(OH)2
38.1 g/l Feed @ 45 % Ca(OH)2
38.1 g/l Feed @ 56 % Ca(OH)2
53.3 g/l Feed @ 16 % Ca(OH)2
E 53.3 g/l Feed @ 24 % Ca(OH)2
-H44-H
O
D
A
8
FIGURE 72.
CORRELATING FACTOR—0d/W (KG-MIN/SQ M)
FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR FERRIC-PRIMARY SLUDGE — PHASE 3.
(Low Ferric; with Polymer)
161
-------
en
to
TABLE 39
DATA SUMMARY OF VACUUM FILTRATION LEAF
TESTS OF PRIMARY SLUDGE FROM PHASE 4
Feed total solids, g/fc
Conditioned feed total solids, g/£
Lime dose, weight percent
Ferric chloride dose, weight percent
Filter form time (6 _) , min.
Filter dry time (6,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
8,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, weight percent
Total solids recovery, weight
percent
1
35.2
39.2
12
3
1.0
1.0
510
510
G
5.6
1.2
0.83
12.0
17.5
95
2
35.2
39.2
12
3
0.5
2.0
510
510
E
4.8
0.86
2.3
103
24.0
95
R U
3
35.2
39.2
12
3
2.0
1.0
510
510
F
7.9
1.3
0.77
39.0
12.8
95
N N
4
35.2
39.2
12
3
4.0
2.0
510
510
F
12.7
1.7
1.2
25.5
11.4
95
U M B E
5
35.2
39.2
12
3
2.0
4.0
510
510
E
7.1
1.3
3.1
39.0
20.8
95
R
6
40.6
47.0
16
4
1.0
1.0
510
510
G
3.6
0.87
1.2
52.2
16.7
92
7
40.6
47.0
16
4
0.5
2.0
510
510
F
2.4
0.52
3.8
62.4
14.0
92
(.continued)
-------
to
TABLE
Feed total solids, g/fc
Conditioned feed total solids, g/H
Lime dose, weight percent
Ferric chloride dose, wt. %
Filter form time (6f ) , min.
Filter dry time (6,) , min.
Form vacuum, mm Hg
Dry vacuum, mm Hg
Discharge code
Cake thickness, mm
Cake dry weight (W) , Kg/sq m
9,/W, Kg-min/sq m
Form filtration rate (Kg/hr-sq m
(including chemicals)
Cake solids content, wt. %
Total solids recovery, wt. %
8
40.
47.
16
4
2.
1.
510
510
P
3.
0.
1.
15.
9.
92
39
6
0
0
0
2
52
9
6
5
(continued)
9
40.
47.
16
4
4.
2.
510
510
P
3.
0.
3.
8.
9.
92
6
0
0
0
2
53
8
0
8
10
40.6
47.0
16
4
2.0
4.0
510
510
P-F
2.4
0.54
7.4
16.2
13.0
92
RUN
11
41.3
49.6
20
5
1.0
1.0
510
510
E
7.1
1.7
.59
102
23.1
98
N U
12
41.
49.
20
5
0.
2.
510
510
E
5.
1.
1.
156
29.
98
M B
3
6
5
0
6
3
5
0
E R
13
41.3
49.6
20
5
2.0
1.0
510
510
E
7.9
1.8
0.56
54.0
22.7
98
14
41.3
49.6
20
5
4.0
2.0
510
510
E
6.4
1.7
1.2
25.5
25.6
98
15
41.3
49.6
20
5
2.0
4.0
510
510
E
4.8
1.2
3.3
36.0
29.9
98
-------
150
<
O
i
UJ
o
o
O
X
UJ
cc
x
o
I
UJ
oc
o
<
oc
oc
o
u.
35.2 g/l Feed @ 12 % Ca(OH)2 + 3% FeCb
(40.6 g/l Feed @ 16 % Ca(OH)2 + 4% FeCb
! 41.3 g/l Feed @ 20 % Ca(OH)2 + 5% FeCb
10
6
0.5
1.0 2.0
FORM TIME-MINUTES
3.0
4.0
FIGURE 73.— FORM FILTRATION RATE VS. FORM TIME FOR PRIMARY
SLUDGE—PHASE 4.
164
-------
One suggests that sludge constituents of a significantly smaller
particle size (fines) than the majority of the feed sludge tend
to migrate into the pores of the filter cake during filtration.
The net result is a less porous filter cake with relatively
poorer filtration characteristics than if no particle migration
occurred. The degree of particle migration and the magnitude
of its effect on cake porosity and cake formation characteris-
tics is determined by the specific particle size distribution
of the feed sludge dry solids. With fine particle migration and
longer cake form times, less filter cake is formed due to the
less porous cake. The result would be, as observed, lower form
filtration rates than expected at the longer cake form times.
The second explanation for a slope greater than -0.5 involves
cake sloughing during formed cake removal from the feed slurry
vat. The net effect of this, as observed, is lower form filtra-
tion rates than expected at the longer cake form times.
Observations made during the filter leaf tests indicate that
cake sloughing was responsible for the slopes which were cal-
culated. The filter cake appeared to form very quickly with
very little additional cake formed at longer form times. This
is best illustrated by the results plotted on Figure 73 which
indicated a slope of -0.98. The maximum possible negative
slope is -1.0. Observation of a -1.0 slope indicates that no
additional cake is formed over that formed at the lowest form
time evaluated. Evidently, this extreme situation was occurring
during the leaf tests which resulted in the -0.98 slope shown
on Figure 73.
A review of Figure 73 also indicates anomalous results for the
data with a -0.98 slope. The form filtration rates observed at
a lime dose of 16 percent were lower than those observed for
the tests conducted at lime doses of 12 or 16 percent. Because
of the observed slope of this data and because of its inconsis-
tency with the remainder of the leaf test data, the results of
the tests conducted at a 16 percent lime dose are considered
suspect, and were eliminated from the predicted full-scale
performance curves shown on Figures 74 and 75.
The scale-up of 0.8 from leaf test results to full-scale
filter performance is based on Eimco-PMD experience. Note on
Figure 75 that significantly drier cakes were produced with a
lime conditioning dose of 20 percent as compared to 12 percent.
DISCUSSION
The method chosen to rationalize vacuum filter performance
measured during the various phases of this study was based on
the chemical constituent content of the chemical-primary sludges
produced.
165
-------
o
5
UJ
o
o
5
D
_l
O
O
°?
DC
X
o
*
J,
O
<
QC
UJ
<
O
50
40
30
20
15
10
•; :r:.::
30% SUBMERGENCE -
0.8 SCALE-UP FACTOR
_i_ L
2 3 4 56789 10
CYCLE TIME—MINUTES
FIGURE 74. FULL-SCALE FILTRATION RATE VS. CYCLE TIME
FOR PRIMARY SLUDGE — PHASE 4.
166
-------
30
UJ
o
QC
UJ
Q.
g
Ul
.1
O
o
CO
9
_i
o
V)
UJ
o
oc
UJ
25
20
15
10
0 35.2 g/l Feed @ 12 % Ca(OH)2 + 3% FeCb
& 41.3 g/l Feed @ 20 % Ca(OH)2 + 5% FeCb
8
10
CORRELATING FACTOR-Od/W (KG-MIN/SQ M)
FIGURE 75.— FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR PRIMARY SLUDGE — PHASE 4.
167
-------
Figure 76 shows the relationship between maximum form filtration
rate since the performance results were selected for the maximum
feed solids concentrations which could be obtained by gravity
thickening during each phase. Differences in thickener under-
flow solids concentration were considered to be due to the rela-
tive fractions of chemical solids comprising the sludges pro-
duced .
For illustrative purposes, the results shown on Figure 76 are
pertinent to a 1.5 minute form time, and to a lime conditioning
dose of 25-30 weight percent as Ca (OH)2•
A review of the results shown on Figure 76 indicates that in-
creases in the fraction of alum associated chemical solids in
the sludge fed to the vacuum filter resulted in lower form fil-
tration rates. This observation means that the addition of
filter alum to raw sewage for the purpose of improving suspended
solids and phosphorus removal results in an alum-primary type
sludge which dewaters at lower filtration rates than the primary
sludge produced without the addition of alum.
As shown on Figure 76 the deterioration in form filtration rate
with increases in the percent alum associated chemical solids
in the sludge is extreme. The curve was not drawn through the
zero percent chemical solids data point (primary sludge from
Phase 4) since filtration data were collected only at a feed
sludge concentration of 41.3 g/&. A higher form filtration rate
would have been observed if the feed sludge concentration had
been closer to the gravity thickener underflow (70 to 100 g/A
TS). Nevertheless, the results on Figure 76 show a dramatic
effect of alum associated chemical solids in the sludge on form
filtration rate.
Figure 77 shows a summary of the relationship between filter
cake solids content and correlating factor for alum-primary
sludge dewatering for all phases. Little difference in filter
cake dry solids content was observed between the results for
Phase 1 through 4. The dotted line drawn on Figure 77 was trans-
ferred from Figure 74 for primary sludge. As shown primary
sludge dewatered to a higher cake dry solids content than the
alum-primary from any phase.
Based on the results shown on Figure 76, the observation of
little difference in filter cake dry solids between the phases
was unexpected. Also unexpected was the fact that no signifi-
cant increase in filter cake dry solids was observed with an
increase in lime dose from 15 to 37 weight percent. However,
it can be noted that the addition of filter alum to raw sewage
resulted in filter cakes with higher moisture contents.
168
-------
60
so
o
CO
DC
UJ
„«
40
30
QC
QC 20
O
x
< 10
FORMTIME1.5 MIN.
• PRIMARY SLUDGE @ 20% Ca(OH)2+ 5% FeCI3
a ALUM- PRIMARY SLUDGE @ 25-30% Ca(OH)2
20
40
60
80
100
CHEMICAL SOLIDS IN SLUDGE - WEIGHT PERCENT
FIGURE 76. EFFECT OF CHEMICAL SOLIDS CONTENT OF
ALUM-PRIMARY SLUDGE ON FORM FILTRATION
RATE.
169
-------
UJ
o
cc
UJ
o
UJ
I
o
V)
oc
Q
30
oc
25
20
O
OC
s
C 15
ALUM PRIMARY SLUDGE
PHASE 1
O
P PHASE 2
a PHASE 3
$ PHASE 4
ra/nm
15'30% Ca2
PRIMARY SLUDGE
_ 20%Ca(OH)2
5% FeCta
6
8
10
CORRELATING FACTOR-Od/W(KG-MIN/SQ M)
FIGURE 77 FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR ALL ALUM-PRIMARY SLUDGE
PHASES.
170
-------
A similar analysis was developed for ferric-primary sludge,
vacuum filtration dewatering. Figures 78 and 79 show the results
of this analysis.
An obvious relationship between maximum form filtration rate and
the fraction of chemical solids in the sludge was not documented.
Figure 78 shows that extremely erratic results were obtained.
The results from Phases 1 and 3 indicated poor filtration per-
formance while excellent filtration was achieved during Phase 2.
Difficulty in vacuum filtration dewatering of ferric-primary
sludge was experienced during a previous study.2 In this study,
raw wastewater was contacted with ferric chloride in a Reactor-
Clarifier operated in a solids contact mode. The ferric-primary
sludge proved to be nearly impossible to filter on a three foot
diameter by three foot face vacuum filter. Acceptable cake
discharge was not achieved even at lime conditioning doses of
up to 50 weight percent as Ca(OH)2.
The filtration problem experienced during this former study
was tentatively traced to the presence of ferrous sulfide in
the sludge. It was hypothesized that anaerobic conditions with-
in the sludge blanket of the Reactor-Clarifier caused the re-
duction of the ferric to ferrous iron with the formation of
ferrous sulfide.
The approach followed during the course of this work was to
maintain as low a sludge solids retention time as possible
within the flocculating clarifier. This was done in order to
maintain as "fresh" a sludge as possible with the goal of pre-
venting ferric reduction.
A review of the results shown on Figure 78 indicates that the
ferric-primary sludge produced during Phase 2 exhibited fil-
tration characteristics greatly superior to that produced in
Phases 1 or 3. This observation was inconsistent with the
hypothesis that increasing fractions of chemical solids result
in poorer filtration characteristics. No definitive conclusion
may be drawn from the results shown on Figure 78 as to the
effect of chemical solids in the ferric-primary sludge on fil-
tration characteristics.
It seems apparent, therefore, that the ferric-primary sludges
produced during this study were exposed to undocumented con-
ditions which drastically affected their thickening and de-
watering characteristics. It was hypothesized that the ferric-
primary sludges produced during Phases 1 and 3 may have been
retained in the chemical treatment unit for too long a period
of time, and that this may have resulted in ferric iron reduc-
tion.
171
-------
70
60
O
00
DC
X
O
UJ
50
oc
z
40
g 30
20
10
FORM TIME 1.0 MIN.
• PRIMARY SLUDGE @ 20% Ca(OH)2
----- + 5% FeCb
O FERRIC-PRIMARY SLUDGE
@ 26-30% Ca(OH)2
20
40
60
80
CHEMICAL SOLIDS IN SLUDGE - WEIGHT PERCENT
100
FIGURE 78.
EFFECT OF CHEMICAL SOLIDS CONTENT OF
FERRIC-PRIMARY SLUDGE ON FORM FILTRATION
RATE.
172
-------
UJ
O
DC
111
Q.
O
0)
cc
O
UJ
O
oc
UJ
40
35
25
20
15
— K ! 1 ; — -
—
er-i
n
/' U::::C
.
! ;
~^ ' ' ' ' jf
_;:::: :.:./r:':
/
/
/
/
:: r ;.rj.: :.
i •
— • !-- • • —
— - --^— - --
r^L'-lES
i i : , , !
i i", " n
±tl-~
1 1
" "" .
:
__ __^
HO
n «-^
: Q .:::
.*• **
, . .**...--. . .
^ ...
X"_ ....
_.:...__..
.._. . . „ —;-.-•;-
1
. . . . .
— - - -)-— Lf-r -;-
FE
..::;...:.:.tr.Pf
. . ; :^i- pj
•::::ir£-.W
i , [ i i
..L_L .. _... .
- _- a
I-H A
% **(
r\ O
O v
^ ;
__
— -: '
-....__...
!
- .- ... . ..- . .-. .
; : . i|''
1
"•'"•• ••^'•'•.B
tn\
iA A. :::.;..:•
0
' :--::.:,
o ::;;-;]:-,. •
• i .-,,...
' ' *
. ... ....
p.v: :•;•;::: i ;::.;: •" :.
•RRIC-PRIMARY SLUDGE
HASE 1 )
HASE 2 f 10-31% Ca(OH);
HASE 3 '
"" : : "r; ]"• T-:- - : • • •
i I i 1 - '
-rr r-'H-f-f- PI
~f'1 r T~l i ! it
-*~ — i"^t=C" zt
*Ti i "i r s«
: , i J ! 1 **
TTTrn-r.T
1 i ' i ! ' i
rrnTrr
RIMARY SLU
)% Ca(OH)2
/0 FeCb
i ] i i i : ' i
i ; ill1
1 i 1. : 1 !!:..:_
n^c
uut *
: —
i i i :
i ; ; . |_ , .
. .
. .
... . ....
z : : . : .' ::
.... -i-i- ^
~'Tnmm
-- !_.;. , - ..- , l : .
~~r~ ^"i" ' > : "
6
8
10
CORRELATING FACTOR-Od/W (KG-MIN/SQ M)
FIGURE 79. FILTER CAKE SOLIDS CONTENT VS. CORRELATING
FACTOR FOR ALL FERRIC-PRIMARY SLUDGE
PHASES
173
-------
A review of the ferric-primary solids retention time maintained
for the chemical treatment unit during each phase did not pro-
vide an explanation of the inconsistent results shown on Figure
78. The solids retention times during Phases 1, 2 and 3 were
approximately 4.5 days, 0.50 days and 0.25 days, respectively.
Only during Phase 1 was an excessive solids retention time main-
tained. The minimum solids retention time was maintained during
Phase 3 which corresponded to ferric-primary sludge consisting
of the minimum fraction of chemical solids. As shown on Figure
78, this sludge exhibited poor filtration dewatering character-
istics.
The curve on Figure 78, as shown, was arbitrarily located. It
must be recognized that the location of this curve affected the
relative costs of alum-primary and ferric-primary sludge de-
watering described in Section VII.
Figure 79 summarizes the relationship between filter cake solids
content and correlating factor for ferric-primary sludge de-
watering for all phases. Although the results are scattered,
the curve tends to be flat and indicates that a cake solids
content of 33-35% is readily achieved.
It is significant to note that the dotted line representing
primary sludge appeared below the data representing ferric-
primary sludge solids content. On Figure 79 this dotted line
appeared above all the data representing alum-primary sludge
solids content; thus, ferric-primary sludge clearly dewatered
to a higher cake solids content than did alum-primary sludge.
174
-------
CENTRIFUGAL DEWATERING
Alum-Primary Sludge
Phase 1 (High Alum; no Polymer)
Performance of the centrifuge on alum-primary sludge generated
during Phase 1 is shown in Table 40 and on Figures 80-81.
For a feed concentration of 23-24 g/H, 0.70-0.90 percent by
weight of dry sludge solids of anionic polymer (Dow AP-30) was
requir.ed to achieve total solids capture greater than 80 percent.
Highest solids capture using this polymer dosage was 98 percent.
Solids capture of only about 20 percent was achieved during
operation without polymer. Highest attainable cake dry solids
were 30-31 percent at very low (20 percent) solids capture.
Solids captures in excess of 40 percent resulted in cake dry
solids contents of 15-16 percent.
Phase 2 (High Alum; with Polymer)
Table 41 and Figures 82-83 illustrate the results of all centri-
fugal dewatering runs performed on alum-primary sludge during
Phase 2.
For a feed concentration of 21 g/H, 0.31-0.55 percent polymer by
weight was required to achieve solids capture greater than 85
percent. Highest solids capture using this polymer dosage was
94-99 percent. Solids capture of only 20-30 percent was achiev-
ed during operation without polymer.
As shown in Figure 82, an anomaly was observed. Using a poly-
mer dosage of 0.22-0.28 percent resulted in a decrease in solids
capture as the hydraulic loading rate to the machine decreased.
No reason can be offered for this inconsistency.
Cake solids relationships were essentially the same as those in
Phase 1. Highest attainable cake dry solids were 30-31 percent
at low (20 percent) solids captures, while solids captures greater
than 40 percent resulted in cake solids of 15-16 percent.
Phase 3 (Low Alum; with Polymer)
Results of all centrifugal dewatering runs performed on alum-
primary sludge generated during Phase 3 are illustrated in Table
42 and on Figures 84-85.
175
-------
TABLE 40
DATA SUMMARY OF CENTRIFUGAL DEWATERING OF
ALUM-PRIMARY SLUDGE FROM PHASE 1
PROCESS VARIABLES
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
ARPM
20
20
20
20
20
20
20
20
20
20
20
20
12
25
20
12
20
20
20
25
Feed
(g/D
22.5
22.5
22.5
22.5
22.5
22.5
22.5
22.5
22.5
22.5
22.5
22.5
22.5
22.5
22.5
23.9
23.9
23.9
23.9
23.9
Feed
(Vmin)
3.14
4.04
7.28
3.23
3.91
7.58
12.8
3.37
7.58
12.8
4.04
7.24
4.18
3.64
12.8
3.97
3.50
4.18
8.35
4.38
Polymer*
(%)
0
0
0
0.24
0.26
0.20
0.19
0.50
0.38
0.33
0.75
0.90
0
0
0.70
0
0
0
0
0
PERFORMANCE
Cake DS
(%)
30.1
28.4
27.1
19.4
13.9
18.1
16.3
15.6
14.7
15.2
16.7
15.9
23.1
31.5
16.0
28.9
29.8
29.9
27.9
30.8
Solids
Capture
(%)
21.0
20.6
19.3
33.5
39.9
23.4
51.8
58.6
48.7
42.2
97.7
96.1
22.8
21.9
82.3
19.9
21.2
20.7
18.2
20.7
Dow AP-30
176
-------
100
ill
O 80
ui
a.
~~
60
ill
DC
I
O 40
i
O
(O
< 20
O
O NO POLYMER;ARPM=12 TO 25
• 0.19-0.26% AP-30;ARPM=20
• 0.33-0.50% AP-30;ARPM=20
0.70-0.90% AP-30;ARPM=20
-ri-rrrr;
2.0
4.0
6.0
8.0
10
12
14
16
FIGURE 80.
FEED RATE — l/min
SOLIDS CAPTURE FROM CENTRIFUGAL DEWATERING OF ALUM-PRIMARY
SLUDGE — PHASE 1.
(High Alum; no Polymer)
-------
00
A25ARPM
• 20ARPM
12ARPM
10
20
30
40
50
60
70
i-i.
80
FIGURE 81.
TOTAL SOLIDS CAPTURE—WEIGHT PERCENT
CAKE SOLIDS CONTENT FROM CENTRIFUGAL DEWATERING OF
ALUM-PRIMARY SLUDGE — PHASE 1.
(High Alum; no Polymer)
H-'-H-t- •
H-4-H-f
l-r-i-T-t-T +
rt-
XT
90
100
-------
TABLE 41
DATA SUMMARY OF CENTRIFUGAL DEWATERING OF
ALUM-PRIMARY SLUDGE FROM PHASE 2
PROCESS VARIABLES
Run
NO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
ARPM
10
10
25
25
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Feed
(gA)
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
27.2
Feed
3.55
6.42
3.71
6.33
3.62
6.33
11.8
15.2
3.77
6.20
8.76
11.8
1.35
6.29
8.59
12.0
6.23
8.76
12.1
15.2
6.39
8.93
Polymer*
0
0
0
0
0
0
0
0
0.13
0.13
0.13
0.14
0.22
0.27
0.27
0.23
0.47
0.38
0.35
0.36
0.53
0.55
PERFORMANCE
Cake DS
25.4
28.4
29.9
29.3
30.5
27.4
29.6
28.0
16.0
16.7
16.6
17.1
16.4
15.2
14.4
16.4
15.0
15.2
13.7
14.9
14.8
15.0
Solids
Capture
24.5
21.9
22.6
21.4
19.4
20.8
20.3
19.2
32.6
32.4
30.1
28.4
39.2
36.8
72.0
78.9
93.0
58.9
85.0
93.3
93.3
93.9
kDow AP-30
Ccontinuedl.
179
-------
TABLE 41. (continued)
PROCESS VARIABLES
Run
No.
22
23
24
25
26
27
28
29
30
31
32
33
32
35
36
37
38
39
40
41
42
43
44
45
46
47
48
ARPM
20
20
20
10
10
25
25
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Feed
(g/4)
27.2
27.2
27.2
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
Feed
(Vmin)
8.93
11.9
16.2
3.64
5.73
3.71
6.51
3.75
6.06
11.6
14.9
3.97
6.06
8.53
11.6
4.04
5.66
8.09
11.6
5.93
8.22
11.7
14.8
5.95
8.59
11.8
14.6
Polymer*
(%)
0.55
0.48
0.40
0
0
0
0
0
0
0
0
0.14
0.12
0.09
0.08
0.19
0.28
0.25
0.22
0.38
0.33
0.31
0.28
0.47
0.46
0.43
0.43
PERFORMANCE
Cake DS
(%)
15.0
15.1
15.9
26.4
28.5
24.2
29.0
25.1
30.1
29.6
26.8
16.3
18.4
17.1
20.7
15.8
14.8
17.5
19.3
16.7
15.0
15.3
15.2
17.9
17.2
15.4
16.7
Solids
Capture
(%)
93.9
93.0
99.4
22.9
22.7
25.0
20.7
26.6
23.4
20.3
18.8
31.5
33.4
31.3
34.6
43.2
53.8
85.3
72.9
93.3
99.0
84.9
84.6
99.0
95.8
97.2
99.4
Dow AP-30
180
-------
00
100
g80
o
oc
uj
0.
h-
O 60
UJ
Ul
GC
3
K-
< 40
O
(0
o
o
(0
-J 20
^
5
FIGURE 82.
Oj
i!
NO POLYMER; ARPM=10 TO 25
0.08-0.19% AP-30;ARPM=20
0.22-0.28% A8-30;ARPM=20
0.31-0.55% AP-30;ARPM=20
4.0
6.0
12
14
16
8.0 10
FEED RATE — l/min
SOLIDS CAPTURE FROM CENTRIFUGAL DEWATERING OF ALUM-PRIMARY
SLUDGE — PHASE 2.
(High Alum; with Polymer)
-------
A 25ARPM
• 20ARPM
• 10ARPM
10
20
30
40
50
60
70
80
90
100
TOTAL SOLIDS CAPTURE—WEIGHT PERCENT
FIGURE 83. CAKE SOLIDS CONTENT FROM CENTRIFUGAL DEWATERING OF
ALUM-PRIMARY SLUDGE — PHASE 2 .
(High Alum; with Polymer)
-------
For a feed concentration of 45-47 g/£ and using a polymer dosage
of 0.21-0.25 percent, 90-93 percent solids capture was attained,
while a polymer dosage of 0.30-0.40 percent by weight resulted
in solids captures of 96-99 percent.
Using no polymer resulted in solids captures as high as 35-40
percent. Since a higher solids capture with no polymer was
achieved (compared to 20 percent during Phases 1 and 2), maximum
cake solids were slightly lower, at 27-28 percent. Cake solids
of 17-19 percent were achieved consistently at solids captures
above 50 percent.
FERRIC-PRIMARY SLUDGE
Phase 1 (High Ferric; no Polymer)
Results of all centrifugal dewatering runs performed on ferric-
primary sludge generated during Phase 1 are shown in Table 43
and on Figures 86-87.
For a feed concentration of 34 g/£, a polymer dose of 0.25-0.60
percent by weight was required to achieve solids recoveries in
excess of 80 percent. Highest solids capture using this polymer
dose was 96 percent. Solids capture of 47-58 percent resulted
from operation without polymer with feed concentrations of 34
g/H and 65 g/£. Feed concentration had no effect on solids
recovery.
Highest attainable cake solids were 40-44 percent at low solids
captures. At solids captures exceeding 80 percent, cake dry
solids were 22-28 percent.
Phase 2 (High Ferric; with Polymer)
Results of all centrifugal dewatering runs conducted on sludge
generated during Phase 2 are shown in Table 44 and on Figures
88-89.
For feed concentrations of 62 and 68 g/&, a polymer dose of
0.36-0.52 weight percent was required to achieve solids captures
of 90 percent and greater; highest capture using this polymer
dose was 99 percent. Solids captures of 40-50 percent resulted
from operation without polymer.
Inspection of Figure 88 shows that for one of the polymer doses
there is no curve drawn. Some data scatter is apparent for the
polymer range of 0.23-0.33 weight percent, but the points
essentially fall between a lower and higher polymer range.
183
-------
TABLE 42
DATA SUMMARY OF CENTRIFUGAL DEWATERING OF
ALUM-PRIMARY SLUDGE FROM PHASE 3
PROCESS VARIABLES
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
ARPM
10
10
25
25
20
20
29
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Feed
(gA)
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
44.6
Feed
(Vmin)
3.62
5.76
3.50
5.66
3.65
6.06
11.9
15.3
4.72
5.79
6.74
11.7
3.64
6.20
8.89
11.6
6.74
8.23
11.5
15.0
6.06
9.73
Polymer*
(%)
0
0
0
0
0
0
0
0
0.10
0.15
0.11
0.10
0.25
0.24
0.22
0.21
0.34
0.39
0.32
0.30
0.40
0.38
PERFORMANCE
Cake DS
(%)
26.6
25.3
27.6
26.6
27.3
25.6
27.0
25.7
20.1
16.2
18.0
19.3
16.8
17.5
17.0
16.1
18.7
18.6
18.0
18.0
19.0
18.6
Solids
Capture
(%)
38.3
36.7
38.1
36.2
37.7
36.6
34.0
34.7
37.8
98.2
53.5
45.6
91.3
92.8
83.7
74.2
95.7
99.3
93.6
85.5
96.3
96.0
Ccontinued)
Dow AP-30
184
-------
TABLE 42, (.continuedl
PROCESS VARIABLES
Run
No.
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
ARPM
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Feed
(g/£)
44.6
44.6
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
47.2
Feed
U/min)
11.2
15.1
4.04
6.20
11.6
15.2
3.91
6.06
9.03
11.6
3.77
6.20
8.76
11.5
6.60
8.49
11.7
14.4
6.00
8.89
12.0
15.5
Polymer*
0.39
0.40
0
0
0
0
0.13
0.13
0.12
0.11
0.25
0.25
0.24
0.24
0.36
0.36
0.24
0.33
0.65
0.42
0.47
0.36
PERFORMANCE
Cake DS
17.7
18.8
26.9
27.9
27.2
26.7
20.7
18.2
17.7
19.7
17.2
18.0
16.6
17.1
18.4
18.5
17.4
18.0
18.7
18.9
18.3
18.8
Solids
Capture
97.5
86.6
35.0
32.5
31.3
28.4
39.7
86.0
65.4
48.3
93.7
93.3
91.9
78.8
97.0
95.2
82.9
79.4
96.2
96.8
92.9
87.1
Dow AP-30
185
-------
100
00
CTi
g 80
DC
UJ
Q.
g
g 60
J,
DC
ID
S 40
CO
Q
II
O
(A
< 20
O
FIGURE 84.
NO POLYMER; ARPM=10-25
0.10-0.15% AP-30;ARPM=20
0.21-0.25% AP-30;ARPM=20
0.31-0.40% AP-30;ARPM=20
2.0
4.0
12
14
6.0 8.0 10
FEED RATE — l/min
SOLIDS CAPTURE FROM CENTRIFUGAL DEWATERING OF ALUM-PRIMARY
SLUDGE — PHASE 3.
(Low Alum; with Polymer)
16
-------
00
A 25ARPM
* 20ARPM
• 10ARPM
FIGURE 85.
30 40 50 60 70
TOTAL SOLIDS CAPTURE—WEIGHT PERCENT
CAKE SOLIDS CONTENT FROM CENTRIFUGAL DEWATERING OF
ALUM-PRIMARY SLUDGE — PHASE 3
(Low Alum; with Polymer)
100
-------
TABLE 43
DATA SUMMARY OF CENTRIFUGAL DEWATERING OF
FERRIC-PRIMARY SLUDGE FROM PHASE 1
PROCESS VARIABLES
Run
NO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
ARPM
12
25
20
20
20
5
12
20
12
12
12
12
12
12
12
12
12
12
12
12
Feed
65.1
65.1
65.1
65.1
65.1
33.7
33.7
33.7
33.7
33.7
33.7
33.7
33.7
33.7
33.7
33.7
33.7
33.7
33.7
33.7
Feed
U/min)
39.1
4.31
4.31
8.49
15.0
3.23
4.04
7.55
3.23
3.57
7.41
3.37
7.55
13.1
3.77
7.28
13.3
3.23
7.28
6.93
Polymer*
0
0
0
0
0
0
0
0
0.21
0.23
0.10
0.29
0.27
0.25
0.63
0.48
0.10
0
0
0.60
PERFORMANCE
Cake DS
33.1
33.7
34.6
40.4
44.1
36.9
36.2
39.4
29.0
23.0
24.4
21.5
27.4
24.5
25.5
22.4
26.6
32.5
41.4
21.8
Solids
Capture
52.0
52.5
52.7
48.5
47.4
55.5
54.1
51.5
63.4
65.1
73.1
96.4
90.0
67.7
91.6
88.2
63.7
58.1
45.4
92.6
Dow AP~30
188
-------
100
00
Ul
o
DC
UJ
Q.
g
UJ
oc
I
o
CO
Q
80
60
40
20
O
NO POLYMER; A RPM=5 TO 25
0.10-0.23% AP-30;ARPM=12
0.25-0.63% AP-30;*RPM=12
2.0
4.0
6.0 8.0
FEED RATE—l/min
10
12
14
16
FIGURE 86.
SOLIDS CAPTURE FROM CENTRIFUGAL DEWATERING OF FERRIC-PRIMARY
SLUDGE — PHASE 1
(High Ferric; no Polymer)
-------
vo
o
111
O
OC
UJ
D.
O
Ul
UJ
O
O
(0
o
o
CO
UJ
o
50
40
30
20
10
O 5ARPM
• 12ARPM
A20ARPM
• 25ARPM
10
20
30 40 50 60 70 80
TOTAL SOLIDS CAPTURE—WEIGHT PERCENT
90
100
FIGURE 87.
CAKE SOLIDS CONTENT FROM CENTRIFUGAL DEWATERING OF
FERRIC-PRIMARY SLUDGE — PHASE 1.
(High Ferric; no Polymer)
-------
TABLE 44
DATA SUMMARY OF CENTRIFUGAL DEWATERING OF
FERRIC-PRIMARY SLUDGE FROM PHASE 2
PROCESS VARIABLES
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
ARPM
10
10
25
25
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Feed
61.6
61.6
61.6
61.6
61.6
61.6
61.6
61.6
61.6
61.5
61.6
61.6
61.6
61.6
61.6
61.6
61.6
61.6
61.6
61.6
61.6
60.7
Feed
3.77
4.04
3.64
6.60
2.91
5.93
12.0
14.2
3.59
6.20
8.49
11.9
3.77
6.06
8.66
6.87
5.79
8.35
11.6
14.8
6.40
8.09
Polymer*
0
0
0
0
0
0
0
0
0.11
0.13
0.11
0.11
0.26
0.23
0.18
0.19
0.33
0.30
0.27
0.25
0.43
0.43
PERFORMANCE
Cake DS
31.7
32.7
34.5
32.4
33.4
32.1
33.4
33.3
31.4
28.5
27.6
29.3
25.8
24.6
27.5
27.6
25.4
25.6
26.8
27.8
26.7
25.5
Solids
Capture
53.3
52.4
49.2
49.2
53.3
50.9
47.4
43.8
49.5
58.8
50.9
54.8
77.4
71.0
62.0
48.4
47.1
81.0
73.8
52.1
90.9
86.9
Dow AP-30
CcontinuedL
191
-------
TABLE 44. (continued)
PROCESS VARIABLES
Run
NO.
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
ARPM
20
20
10
10
25
25
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
25
Feed
60.7
60.7
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
67.8
Feed
(A/min)
11.5
15.3
3.37
6.06
3.91
6.74
3.91
6.20
11.7
15.4
3.91
6.20
9.30
11.6
3.64
6.33
8.22
11.7
5.93
8.76
12.5
15.2
6.06
9.26
7.28
3.64
Polymer*
0.39
0.35
0
0
0
0
0
0
0
0
0.11
0.13
0.12
0.14
0.24
0.29
0.25
0.23
0.41
0.36
0.33
0.28
0.52
0.45
0.36
0
PERFORMANCE
Cake DS
26.4
27.3
32.0
31.9
31.0
31.4
32.5
30.3
30.7
31.9
26.4
25.9
26.7
27.2
23.0
23.0
23.2
24.0
24.2
25.0
25.7
24.9
25.6
24.8
23.5
31.8
Solids
Capture
72.7
61.0
48.8
49.1
48.0
44.7
49.3
46.8
42.5
40.6
53.0
42.3
43.0
41.6
88.3
88.8
80.5
61.9
92.0
91.4
89.8
76.1
93.0
96.8
98.8
46.5
Dow AP-3Q
192
-------
100
IO
CJ
Ui
o
oc
UJ
Q.
O
u"i
1
OC
I
<
u
(0
g
_j
o
(0
T NO POLYMER; ARPM=10 TO 25
0.11-0.19% AP-30; A RPM=20
0.23-0.33% AP-30; ARPM=20
m 0.36-0.52% AP-30; A RPM=20
40
20
FIGURE 88.
8.0 10
FEED RATE—l/min
SOLIDS CAPTURE FROM CENTRIFUGAL DEWATERING OF FERRIC-PRIMARY
SLUDGE —PHASE 2.
(High Ferric; with Polymer)
-------
Ul
o
oc
Ul
Q.
o
UJ
I
ui
O
O
(0
lu
s
40
30
20
10
4- !-!-'•
i
A 10ARPM
• 20ARPM
• 25ARPM
10
20
30
40
50
60
70
80
90
100
TOTAL SOLIDS CAPTURE—WEIGHT PERCENT
FIGURE 89. CAKE SOLIDS CONTENT FROM CENTRIFUGAL DEWATERING OF
FERRIC-PRIMARY SLUDGE — PHASE 2.
(High Ferric; with Polymer)
-------
Highest cake solids attained was 35 percent at total solids cap-
tures of 40-50 percent. At captures in excess of 60 percent,
cake solids were 24-26 percent.
Phase 3 (Low Ferric; with Polymer)
Results of all centrifuge runs conducted on ferric-primary sludge
generated during Phase 3 are shown in Table 45 and Figures 90-91.
For a solids capture of 80 percent, a polymer dosage range of
0.20-0.30 weight percent was found to be necessary, while a poly-
mer dosage of 0.34-0.58 weight percent resulted in solids cap-
tures greater than 90 percent. Highest solids capture was 94
percent at the polymer dose of 0.34-0.58 percent.
Solids captures of 45-50 percent was achieved without using poly-
mer. Cake solids of 33-37 percent at low solids capture were
generally achieved, while the cake solids decreased to 21-23 per-
cent at captures greater than 80 percent.
PRIMARY SLUDGE
Results of all centrifugal dewatering runs conducted on primary
sludge are shown in Table 46 and on Figures 92-93. From Figure
92, it is apparent that the use of anionic polymer was not
effective in assisting centrifuge performance. This tends to
indicate that a cationic polymer would have been the preferred
conditioner. However, time prevented additional evaluations
with cationic polymers.
Figure 92 shows a trend that was not present in any of the three
prior phases (on either type of sludge). Regardless of the feed
concentration (in this case 25 g/fc and 45 g/£), and using no
polymer, as the hydraulic loading decreased, solids captures in
excess of 60 percent were achieved. Cake solids ranged from 25
percent at 44 percent solids capture to 20 percent at 90 percent
solids capture.
The data generated during Phase 4 from primary sludge was in-
cluded in an analysis of centrifugal dewatering performance to
provide a clearer picture of the effects of chemical solids in
the sludge on centrifugal dewatering performance.
DISCUSSION
Alum-Primary Sludge
Figures 94-95 illustrate the effect of the fraction of chemical
solids in the sludge on the dewaterability of the sludge. On
Figure 94, percent solids recovery, without polymer addition,
was plotted versus hydraulic loading. Noting that Phases 1 and
2 represent sludge generated at virtually the same chemical
195
-------
TABLE 45
DATA SUMMARY OF CENTRIFUGAL DEWATERING OF
FERRIC-PRIMARY SLUDGE FROM PHASE 3
PROCESS VARIABLES
Run
NO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
ARPM
10
10
25
25
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Feed
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
46.6
Feed
(Vmin)
3.91
6.20
3.77
5.93
3.77
6.20
11.7
15.2
3.77
6.33
8.62
11.5
3.78
6.21
8.51
11.3
6.16
9.18
11.8
15.3
6.23
8.37
Polymer*
0
0
0
0
0
0
0
0
0.17
0.14
0.13
0.13
0.25
0.28
0.27
0.20
0.42
0.38
0.37
0.36
0.58
0.55
PERFORMANCE
Cake DS
37.1
35.9
33.4
33.0
34.6
32.8
34.0
35.5
21.5
22.2
23.0
24.1
22.0
20.9
22.5
21.9
22.6
22.7
22.5
23.5
22.6
21.2
Solids
Capture
50.9
48.5
49.4
49.5
49.5
46.9
46.6
44.2
78.1
74.5
73.5
64.5
93.8
76.7
70.5
76.6
88.0
75.9
75.2
70.3
89.0
86.6
(.continued)
Dow AP-30
196
-------
TABLE 45. (continued)
PROCESS VARIABLES
Run
No.
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
ARPM
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Feed
(g/M
46.6
46.6
58.4
58.4
58.4
58.4
58.4
58.4
58.4
58.4
58.4
58.4
58.4
58,4
58.4
58.4
58,4
58.4
58.4
58.4
58.4
58.4
Feed
(Vmin)
11.6
13.6
3.38
6.21
11.8
15.1
3.85
6.21
8.78
11.9
3.59
6.08
8.37
11.9
8.24
8.64
9.86
15.0
6.21
8.64
12.3
15.3
Polymer*
(%)
0.49
0.34
0
0
0
0
0.04
0.11
0.12
0.15
0.23
0.28
0.27
0.22
0.37
0.37
0.29
0.35
0.54
0.48
0.37
0.30
PERFORMANCE
Cake DS
(%)
22.6
25.0
34.6
35.7
35.8
36.9
24.8
23.3
23.5
26.4
20.9
21.3
24.6
23.6
22.3
22.5
24.0
24.9
20.5
22.1
23.2
23.2
Solids
Capture
(%)
82.7
73.9
49.1
46.5
46.0
44.3
65.2
84.5
70.0
68.8
89.7
88.8
82.1
77.6
87.7
79.8
72.2
74.6
93.0
90.0
83.1
78.6
Dow AP-30
197
-------
100
1U
u
oc
UJ
0.
g
UJ
i
DC
CO
O
CO
1
80
60
O 40
20
NO POLYMER; A RPM=10 TO 25
0.04-0.17% AP-30; A RPM=20
0.20-0.30% AP-30; ARPM=20
0.34-0.58% AP-30; ARPM=20
2.0
4.0
6.0 8.0 10
FEED RATE—l/min
12
14
16
FIGURE 90.— SOLIDS CAPTURE FROM CENTRIFUGAL DEWATERING OF FERRIC-PRIMARY
SLUDGE — PHASE 3.
(Low Ferric; with Polymer)
-------
vc
VD
111
O
QC
Ill
I
1
8
(0
a
UJ
O
40
30
20
10
—-
A 25ARPM
20ARPM
10ARPM
FIGURE 91,
10 20 30 40 50 60 70 80
TOTAL SOLIDS CAPTURE—WEIGHT PERCENT
CAKE SOLIDS CONTENT FROM CENTRIFUGAL DEWATERING OF
FERRIC-PRIMARY SLUDGE — PHASE 3.
(Low Ferric; with Polymer)
90
100
-------
TABLE 46
DATA SUMMARY OF CENTRIFUGAL DEWATERING OF
PRIMARY SLUDGE FROM PHASE 4
PROCESS VARIABLES
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
ARPM
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Feed
44.9
44.9
44.9
44.9
44.9
44.9
44.9
44.9
44.9
44.9
44.9
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
27.8
27.8
Feed
3.86
6.36
11.5
6.21
4.05
4.50
7.64
6.96
6.96
10.3
13.4
3.75
6.06
11.4
14.8
3.37
3.52
6.24
5.05
6.24
11.6
11.6
14.5
15.5
Polymer*
0
0
0
0.07
0.20
0.52
0.06
0.27
0.64
0.08
0.17
0
0
0
0
0.23
0.40
0.20
0.36
0.78
0.14
0.33
0.24
0.50
PERFORMANCE
Cake DS
27.1
25.7
25.7
21.7
22.9
19.3
19.7
19.6
19.1
20.6
21.2
26.7
25.4
24.5
24.2
21.9
22.3
21.4
21.6
21.1
21.8
20.3
20.4
21.6
Solids
Capture
70.5
63.6
56.4
95.5
76.9
87.4
65.1
69.9
72.9
62.5
59.7
63.0
43.0
44.0
50.0
75.0
62.0
58.0
65.0
62.0
66.0
52.0
54.0
53.0
Dow AP-30
200
-------
100
KJ
O
UJ
o
DC
UJ
Q.
O
UJ
1
oc
o
CO
Q
80
60
40
20
—i ..i—,
NO POLYMER; A RPM=20
0.06-0.78% ANIONIC POLYMER; A RPM=20 i i
,.. - t . -l.-i-.-.H-l
, , I-U-.UI1
2.0
4.0
6.0 8.0
FEED RATE—l/min
10
12
14
16
FIGURE 92. SOLIDS CAPTURE FROM CENTRIFUGAL DEWATERING OF PRIMARY SLUDGE
-------
10
o
K)
Ul
o
oc
111
Q.
I- 40
g
UJ
ui
o
o
(0
O
i
2
30
20
10
20nRPM
10
20 30 40 50 60 70 80
TOTAL SOLIDS CAPTURE—WEIGHT PERCENT
90
100
FIGURE 93. CAKE SOLIDS CONTENT FROM CENTRIFUGAL DEWATERING OF
PRIMARY SLUDGE.
-------
o
CO
UJ
O
CC
UJ
CL
uj
I
CC
o
9
8
I
20
ALUM-PRIMARY SLUDGE — PHASE 1
ALUM-PRIMARY SLUDGE — PHASE 2
ALUM-PRIMARY SLUDGE — PHASE 3
PRIMARY SLUDGE — PHASE 4
6.0 8.
FEED RATE -
16
FIGURE 94. SUMMARY OF SOLIDS CAPTURE FROM CENTRIFUGAL DEWATERING OF
ALUM-PRIMARY SLUDGE FROM ALL PHASES.
-------
K>
O
O
DC
111
Q.
O
m
J
8
i
8
S
O
40
30
20
10
O ALUM-PRIMARY SLUDGE
O ALUM-PRIMARY SLUDGE
A ALUM-PRIMARY SLUDGE
PHASES 1
PHASES 2
PHASES 3
PRIMARY SLUDGE — PHASE 4
10
20
30 40 50 60 70 80
TOTAL SOLIDS CAPTURE—WEIGHT PERCENT
90
100
FIGURE 95. SUMMARY OF CAKE SOLIDS CONTENT FROM CENTRIFUGAL DEWATERING
OF ALUM-PRIMARY SLUDGE FROM ALL PHASES.
-------
dosage, that Phase 3 represents sludge generated at a lower chem-
ical dosage, and that Phase 4 represents primary sludge, it was
found that as the fraction of chemical solids present in the
sludge decreased, the maximum percent solids recovery increased
for any given hydraulic loading. Also, note that there was
little difference in the solids recovery in the first two phases
where the chemical solids fraction in the sludges were virtually
the same. All results shown on Figure 94 represent tests per-
formed using no polymer during centrifugation.
A similar relationship was observed for cake solids content, as
shown on Figure 95. As the fraction of chemical solids present
in the sludge was reduced, cake solids content increased for a
given solids capture. Results from Phase 1 and Phase 2 were
nearly identical.
Figure 96 shows how the maximum observed solids recovery dras-
tically decreased as the fraction of chemical solids in the
sludge increased. Again note the little difference in the re-
sults of Phases 1 and 2. All results shown on Figure 95 rep-
resent tests conducted on the centrifuge using no polymer.
Figure 97 is a graphical illustration of the effect of the
chemical solids fraction on cake solids content. The summary
of results plotted on Figure 97 shows that, for any level of
solids capture, cake solids content decreased as the fraction
of chemical solids in the alum-primary sludge increased. The
effect of chemical solids fraction on cake solids content was
especially apparent at low levels of solids capture. At high
levels of solids capture, cake solids content was relatively
insensitive to the fraction of chemical solids present in the
sludge since all sludge constituents were efficiently captured
and exhibited themselves in the cake. At lower levels of
solids capture/ however, cake solids content was sensitive to
the fraction of chemical solids present in the feed sludge due
to the classification characteristics of the solid bowl centri-
fuge.
Ferric-Primary Sludge
Figures 98-99 show the results of all centrifuge runs performed
on ferric-primary sludge using no polymer. All curves for
ferric-primary sludge shown on Figure 98 are flat with little
difference between them. The results obtained for primary
sludge were plotted on Figure 98 and indicated that solids re-
covery for primary sludge was generally higher than for ferric-
primary sludge at equivalent centrifuge loadings.
Figure 99 shows cake solids content versus solids capture.
Again, there is little difference in the results of all three
ferric-primary sludges; at 40-45 percent solids captures, cake
solids of 40 percent attained. In all cases, cake solids
205
-------
UJ
o
oc
UJ
Q.
I-
g
UJ
oc
UJ
o
o
UJ
oc
(/>
Q
o
5
X
<
30
20
FIGURE 96.
10 20 30 40 50
CHEMICAL SOLIDS IN SLUDGE—WEIGHT PERCENT
EFFECT OF CHEMICAL SOLIDS CONTENT OF
ALUM-PRIMARY SLUDGE ON MAXIMUM TOTAL
SOLIDS RECOVERY WITHOUT POLYMER.
206
-------
30% TOTAL SOLIDS RECOVERY
50% TOTAL SOLIDS RECOVERY
95% TOTAL SOLIDS RECOVERY
0 10 20 30 40 50
CHEMICALS SOLIDS IN SLUDGE—WEIGHT PERCENT
FIGURE 97. EFFECT OF CHEMICAL SOLIDS CONTENT OF
ALUM-PRIMARY SLUDGE ON CAKE SOLIDS CONTENT
207
-------
o
00
80
ui
o
cc
UJ
o.
g eo
UJ
I
UJ
cc
o
(0
o
8 20
B
FIGURE 98.
O FERRIC-PRIMARY SLUDGE - PHASE 1
_P FERRIC-PRIMARY SLUDGE - PHASE 2
A FERRIC-PRIMARY SLUDGE - PHASE 3
• PRIMARY SLUDGE - PHASE 4
2.0
4.0
6.0 8.0
FEED RATE—l/min
10
12
14
16
SUMMARY OF SOLIDS CAPTURE FROM CENTRIFUGAL DEWATERING OF
FERRIC-PRIMARY SLUDGE FROM ALL PHASES.
-------
O
VO
UJ
O
QC
UJ
Q.
g
UJ
I
o
O
(0
o
O
(0
UJ
*
3
40
30
20
10
r
FERRIC-PRIMARY SLUDGE — PHASE 1
FERRIC-PRIMARY SLUDGE — PHASE 2
FERRIC-PRIMARY SLUDGE — PHASE 3
PRIMARY SLUDGE — PHASE 4
30
40 50 60 70 80
TOTAL SOLIDS CAPTURE—WEIGHT PERCENT
90
100
FIGURE 99. SUMMARY OF CAKE SOLIDS CONTENT FROM CENTRIFUGAL DEWATERING
OF FERRIC-PRIMARY SLUDGE FROM ALL PHASES
-------
attained on ferric-primary sludge were higher than those attained
on primary sludge, for a given solids capture.
Figure 100 summarizes the effects of the fraction of chemical
solids present in the ferric-primary sludge on centrifuge per-
formance. Although there was a considerable amount of scatter,
the results indicate that (at the same level of solids recovery)
cake solids tended to increase as the fraction of chemical solids
in the sludge increased. The effect of chemical solids fraction
on cake solids content was especially apparent at low levels of
solids capture.
Constituent Captures
Studies were conducted during each phase to determine volatile
solids, metal, and phosphorus recoveries on each centrifuge run.
This information was collected in order to determine the -relation-
ship between constituent capture and solids capture.
Figures 101-102 show the results of the correlations of percent
volatile solids capture and total solids capture for all alum-
primary sludge and ferric-primary sludge runs. Volatile solids
recovery was found to be directly related to total solids re-
covery for alum-primary and ferric-primary sludge centrifugal
dewatering.
Figures 103-104 show the results of the correlations between
percent metal capture and percent total solids capture for all
alum-primary sludge and ferric-primary sludge phases, respec-
tively. Although some data scatter occurred, reasonably con-
sistent results were achieved. The results from any one phase
did not differ significantly from the results of the other two
for either chemical-primary sludge. Aluminum was not captured
well at low total solids captures (6 percent aluminum capture
at 20 percent total solids capture), while iron capture was
more efficient (35 percent iron capture at 20 percent total
solids recovery). Iron recovery of 90 percent was achieved at
70 percent total solids recovery, but 90 percent aluminum re-
covery was not realized until 92 percent total solids recovery
was effected. Clearly, iron capture was much greater than alum-
inum capture at any level of total solids capture.
Figures 105-106 show the results of the correlations between
phosphorus recovery and total solids recovery for alum-primary
and ferric-primary sludge dewatering runs for Phases 2 and 3.
Similar results were not determined for Phase 1 since phosphorus
analyses were not performed. The results plotted on Figure 105
showed a relationship similar to those for volatile solids.
Only a relatively minor amount of scatter occurred. Phase 3 re-
sults showed slightly more scatter than those of Phase 2, but
this did not significantly affect the shape or location of the
curve. It is evident from Figure 106 that a considerable amount
210
-------
"Trr
50% TOTAL SOLIDS RECOVERY
70% TOTAL SOLIDS RECOVERY
80% TOTAL SOLIDS RECOVERY
dip 95% TOTAL SOLIDS RECOVERY
F~l i ' III ; | i ili r ; i : ' i i | '
i-l~U
"1 r
10 20 30 40 50
CHEMICAL SOLIDS IN SLUDGE — WEIGHT PERCENT
60
FIGURE 100. EFFECT OF CHEMICAL SOLIDS CONTENT OF FERRIC-
PRIMARY SLUDGE ON CAKE SOLIDS CONTENT.
211
-------
100
u
cc
<
O
111
O
O
LU
O
CC
UJ
O.
-PHASE 1
PHASE 2
PHASE 3
20
20 40 60 80
PERCENT TOTAL SOLIDS CAPTURE
100
FIGURE 101. VOLATILE SOLIDS CAPTURE VERSUS TOTAL SOLIDS
CAPTURE FOR CENTRIFUGAL DEWATERING OF
ALUM - PRIMARY SLUDGE,
212
-------
100
UJ
QC
o
h-
Ul
D
O
O
UJ
o
DC
UJ
0.
3 PHASE 1
O PHASE 2
PHASE 3
20
20
40
60
80
100
PERCENT TOTAL SOLIDS CAPTURE
FIGURE 102. VOLATILE SOLIDS CAPTURE VERSUS TOTAL SOLIDS
CAPTURE FOR CENTRIFUGAL DEWATERING OF
FERRIC-PRIMARY SLUDGE.
213
-------
100
Ul
cc
I
g
UJ
h-
I
O
o
Ul
o
DC
UJ
Q.
PHASE 1
PHASE 2
PHASE 3
80
60
40
20
20
40
60
80
100
PERCENT TOTAL SOLIDS CAPTURE
FIGURE 103. ALUMINUM CAPTURE VERSUS TOTAL SOLIDS
CAPTURE FOR CENTRIFUGAL DEWATERING OF
ALUM-PRIMARY SLUDGE.
214
-------
100
UJ
cc
fc
S
UJ
O
O
UJ
O
cc
UJ
0.
O PHASE 1
§ PHASE 2
A PHASE 3
20
20
40
60
80
100
TOTAL SOLIDS CAPTURE - WEIGHT PERCENT
FIGURE 104. IRON CAPTURE VERSUS TOTAL SOLIDS CAPTURE FOR
CENTRIFUGAL DEWATERING OF FERRIC-PRIMARY
SLUDGE.
215
-------
100
in
cc
0.
g
K
UJ
H
(0
z
o
o
UJ
o
DC
UJ
0.
80
60
40
20
O PHASE 2
A PHASE 3
20
40
60
80
100
PERCENT TOTAL SOLIDS CAPTURE
FIGURE 105. PHOSPHORUS CAPTURE VERSUS TOTAL SOLIDS
CAPTURE FOR CENTRIFUGAL DEWATERING OF
ALUM-PRIMARY SLUDGE.
216
-------
100
o
o
UJ
o
cc
UJ
0.
80
60
40
20
D PHASE 2
A PHASE 3
20
40
60
80
PERCENT TOTAL SOLIDS CAPTURE
100
FIGURE 106. PHOSPHORUS CAPTURE VERSUS TOTAL SOLIDS
CAPTURE FOR CENTRIFUGAL DEWATERING OF
FERRIC-PRIMARY SLUDGE.
217
-------
of scatter occurred. This was due to analytical problems en-
countered in analyzing for phosphorus in ferric-primary sludge.
Inspection of Figure 106 shows that Phase 3 data contributed
heavily to the scatter problem. Points from Phase 2 were rela-
tively consistent; these points were used to plot the curve on
Figure 106.
In spite of the problem experienced with scatter, if the curves
on Figures 105-106 are compared to the aluminum and iron capture
curves on Figures 103-104, it can be seen that, for either alum-
primary or ferric-primary sludge, the phosphorus capture and
the respective metal capture curves were similar. This tends
to indicate that the metal-phosphate complex constituent retained
its identity in terms of capture or rejection during centri-
fugation.
The results of the correlations developed for both volatile
solids and phosphorus recovery versus total solids recovery for
primary sludge are summarized on Figures 107-108. Figure 107
indicated the same relationship that was observed for the vola-
tile solids correlations developed for both alum-primary and
ferric-primary sludge, namely, volatile solids capture versus
total solids capture was on a near one-to-one basis. Figure 108
shows the phosphorus capture correlation developed for primary
sludge. Note that this curve has the same general shape as the
aluminum recovery and phosphorus recovery curves developed from
alum-primary sludge.
218
-------
100
UJ
cc
£
g
UJ
D
z
O
o
UJ
O
DC
UJ
QL
80
60 —
40
20
20
40
60
80
100
PERCENT TOTAL SOLIDS CAPTURE
FIGURE 107 VOLATILE SOLIDS CAPTURE VERSUS TOTAL SOLIDS
CAPTURE FOR CENTRIFUGAL DEWATERING OF
PRIMARY SLUDGE.
219
-------
100
111
cc
t
s
I-
UJ
o
o
I-
UJ
UJ
Q.
80
60
40
20
20
40
60
80
100
PERCENT TOTAL SOLIDS CAPTURE
FIGURE 108- PHOSPHORUS CAPTURE VERSUS TOTAL SOLIDS
CAPTURE FOR CENTRIFUGAL DEWATERING OF
PRIMARY SLUDGE.
220
-------
PRESSURE FILTRATION DEWATERING
Filter press studies commenced at the end of Phase 3 and then
progressed into Phase 4. More meaningful data was generated in
Phase 4 as Phase 3 proved to be primarily a shake-down period.
Results of all filter press test runs are presented in Table 47.
An inspection of the results listed in Table 47 permit several
generalized conclusions. For any specific sludge from Phase 4,
filter press operation with thin filter cakes resulted in shorter
press cycle times, higher filtration rates and cakes of greater
solids concentration than resulted from press operation with
thicker cakes.
Similar conclusions may be drawn from a review of the alum-primary
sludge filter press results from Phase 3 operations. Shorter
cycle times and higher filtration rates resulted from pressure
filtration with thinner cakes. Generally, cake solids contents
were equal for each cake thickness with the exception of run #10-
24-74-1 where the thicker cake exhibited a higher cake solids
than the thinner cake.
When comparing the three different types of sludges, some ob-
servations concerning performance can be made. When filter
pressing primary sludge, the longest cycle times were required,
but the cake solids were the lowest. Alum-primary sludge genera-
ted at a low alum dosage falls between the other two sludges,
both in terms of cycle time and cake solids.
A filtration test in a filter press gives the time versus filtrate
volume curve, and the dry solids yield and percent solids at the
final filtration time. It is useful to be able to predict the
moisture content of the cake if the form time were longer or
shorter than the time selected. A predictive equation for cake
moisture content can be developed from a dry solids and total
solids balance around the filter press from time t to the final
filtration time t,:
Dry solids: D + (AV)(C) = D^ (la)
Total solids: W + (AV) (Ap) = W-j^ (Ib)
From Equations la and Ib, the cake solids content at time t can
be calculated:
221
-------
TABLE 47
SUMMARY OF PRESSURE FILTRATION RESULTS
Uncondi-
Date Type tioned
Run of Feed Cone.
No. Sludge (g/JZ.)
12-3-74-2 Alum1 33.5
12-3-74-3 Alum' 32.1
12-3-74-4 Alum1 24.3
12-4-74-5 Alum' 33.0
|sj 11-19-74-1 Primary 57.3
IO 11-10-74-2 Primary 57.3
11-21-74-3 Primary 28.9
11-22-74-4 Primary 28.9
10-23-74-1 Alum2 49.7
10-23-74-2 Alum2 49.7
10-24-74-1 Alum2 30.2
10-24-74-2 Alum' 30.2
Condi —
tioned
Feed
Cone.
(g/ • )
41.8
38.9
32.8
35.2
59.7
66. 7
49.2
51.2
54. 5
59.1
39.7
40. 5
Lime
Condi-
tioning
(wt. %)
37
26
36
33
37
44
48
65
30
45
33
45
Filtration'
Cake Rate
Thickness Cycle Time3'" Cake Solids6 Cake Area Front Back
Front Back Front Back Front B'ack Front Back (Kg/hr- (kg/hr-
(cm) (cm) (min) (min) (%) (%) (sq cm) (sq 'cm) sq m) sq m)
3.5 2.7 104 87 26 27 455 : 494 .020 .023
3-5 2.7 352 302 29 33 455 , 494 .007 .008
3-5 2.7 262s 197 28 28 455 ! 494 .008 .010
3.5 2.7 142 116 26 26 455 494 .017 .022
3-5 2.7 369 222 34 42 455 494 .009 .014
3-5 2.7 340 266 25 33 455 494 .006 .010
3.5 2.7 650' 365 41 41 455 ': 494 .003 .007
3-5 2.7 491'' 415' 43 38 455 ' 494 .005 .005
2-9 2.7 302 259' 33 33 506 ! 494 .009 .010
2.9 2.7 149 108 35 34 506 494 .021 .026
2.9 2.7 494' 266 37 34 506 494 1 .005 .009
2.9 2.7 223' 179 37 36 506 494 ! .012 ; .014
'Alum dosage in flocculating clarifier was 15.6 nn Al/
2Alum dosage in flocculating clarifier was 8.1 m« Al/
'Cycle time includes 20 minutes for press down-time
during discharge and reloading
""All cycle times at 1/20 five minute rate
This cycle time never reached, thie time extrapolated
from filtrate volume versus time cuve
Cake solics recalculated for all estimated cycle times
'Filtration rate for sludge solids only; conditioning
chemicals subtracted out
-------
Solids content (%) S _ 100D D, - (AV)(C)
c - =1
W W-L - (AV) (Ap)
where V = volume of filtrate at time t
S = percent dry solids in the cake
C
D = total solids in frames after V has been
collected
W = total mass in frame after V has been
collected
D, = total solids at end of filtration (V, has
been collected)
W.. = total mass at end of filtration (V, has
been collected)
AV = .final filtrate volume (V,) minus filtrate
volume at form time t (V7
C = conditioned feed concentration )g/l)
Ap = difference in feed and filtrate densities
(g/D
The above equation is not rigorous unless concentrations and
densities are based on final filtrate volume. However, it gives
good approximations near the final filtering times. If SG at t^,
D,, W,, V,, Ap, and C are known, the value of S at any V can
be found. If t versus V has been recorded in trie course of the
filtration test, then S versus t may be predicted.
The equation may be used to predict the cake solids (S ) versus
V relationship that would have been obtained had the press been
allowed to run longer than the actual termination time of the
test. The authors' experience indicates that this extrapolation
should not be made if the filtration rate at the end of the test
has dropped to below 1/20 of the average rate during the first
five minutes. To predict the relationship between S and t for
times greater than t,, the t versus V curve must first be
extrapolated.
Figures 109 illustrates the use of the equation for a primary
sludge filter press run on 11-20-74. Note how the cake solids
increased very little once the form time pertinent to 1/20 of
the five minute rate had been reached.
If the filtration rate were calculated for different cake dry
solids on the same filter press run, curves similar to those
223
-------
UJ
U
DC
UJ
o.
40
35
30
25
20
O
UJ
I
9
d
CO
cc 15
O
UJ
10
2.8 cm CAKE THICKNESS
3.3 cm CAKE THICKNESS
DENOTES FILTRATE RATE
EQUAL TO 1/20 OF 5
MINUTE RATE
100 200
FORM TIME • MIN
300
FIGURE 109.
PRESSURE FILTER FORM TIME VS. CAKE SOLIDS
CONCENTRATION FOR PRIMARY SLUDGE.
224
-------
shown on Figures 110 through 112 would be generated. On Figure
110, two different filter press runs on alum-primary sludge are
illustrated. In all cases, note the very rapid drop in filtra-
tion rate once 1/20 of the five minute rate had been reached;
at this point and beyond, cake solids increased only slightly
for a large loss in filtration rate. Figure 110 also shows the
effect of cake thickness and lime dosage. Decreasing cake thick-
ness and increasing lime dosage improved filtration rate.
The curves shown on Figures 110-112 can also be used to assist in
selection of an operating point. By examining the curve on Figure
110, for a cake thickness of 3.5 cm (1.4 inch) and a feed concen-
tration of 33 g/Jl, it can be seen how drastically the filtration
rate decreased after 25 weight percent cake solids was attained,
and that 30 weight percent cake solids was unattainable.
Figures 111 and 112 generally showed the same pattern. Once 1/20
of the five minute filtrate production rate had been attained,
filtration rate dropped markedly; also, thinner cakes provided
higher filtration rates on a given sludge. On Figure 112, re-
sults for press runs on primary sludge were plotted. Results
were consistent for cake thickness, however, results were incon-
sistent for level of lime conditioning. In this case, an in-
crease of 7 percentage points in lime conditioning dose caused
a drop in the filtration rate (for the same feed solids). Since
the run using 44 weight percent lime was not conducted the same
day as the one with 37 weight percent lime, the reason for the
decrease in filtration rate could be attributed to different
sludge characteristics. It is felt that the test with 44 per-
cent lime conditioning was an overdose (for example, Figure 8
shows that any lime conditioning in excess of 30 weight percent
on that sludge was of no value in further reduction of specific
resistance).
Some general observations can be made concerning pressure filtra-
tion of alum-primary sludge and primary sludge. When pressing
alum-primary sludge (generated at both a high and low alum dosage)
lime conditioning in the range of 24 to 25 percent by weight was
required; however, for primary sludge a higher range of 37 to 65
percent was required. On alum-primary sludge (high dosage), a
decrease in the cake thickness from 3.5 to 2.7 cm (1.4-1.1 inch)
resulted in a filtration rate increase of from 14 to 29 percent;
cake solids generally increased about 2 percentage points for
this cake thickness decrease. For alum-primary sludge generated
at a low alum dosage, a much smaller cake thickness difference
(2.9 cm to 2.7 cm {1.1-1.0 inch}) generally resulted in filtra-
tion rate increases of from 11 to 24 percent, while cake solids
showed little difference. On primary sludge, the cake thickness
difference had the most significance. A cake thickness decrease
from 3.5-2.7 cm (1.4-1.1 inch) resulted in filtration rate
increases of from 56 to greater than 100 percent, while cake
solids increased as much as 8 percentage points for this cake
225
-------
4.5
4.0
DC
I
O
UJ
i
z
o
I
3.5
3.0
2.5
2.0
1.5
1.0
.5
O t = 3.5 cm; FEED
; LIME =26%
O t = 2.7 cm; FEED
—- LIME = 26%
• t = 3.5cm; FEED
---- LIME = 33%
• t = 2.7 cm; FEED
_ LIME = 33%
32.1 g/l
32.1 g/l,
33.0 g/l,
33.0 g/l,
?£ DENOTES FILTRATE
RATE EQUAL TO 1/20
Of 5 MINUTE RATE
0 10 20 30
CAKE SOLIDS - WEIGHT PERCENT
FIGURE 110. SUMMARY OF PRESSURE FILTRATION RESULTS
FOR ALUM-PRIMARY SLUDGE — PHASE 4.
(High Alum; with Polymer)
226
-------
2.25
t = 2.9 cm; FEED = 30.2 g/l,
LIME = 33%
t= 2.7 cm; FEED = 30.2 g/l,
LIME = 33%
10
20 30
CAKE SOLIDS - WEIGHT PERCENT
FIGURE 111. SUMMARY OF PRESSURE FILTRATION RESULTS
FOR ALUM-PRIMARY SLUDGE - PHASE 3.
(Low Alum; with Polymer)
227
-------
tr
z
O
z
g
I
1.0
FEED = 57.7 g/l,
37%
FEED = 57.7 g/l,
37%
FEED = 57.7 g/l;
44%
FEED = 57.7 g/l;
44%
= 3.5 cm;
LIME =
= 2.7 cm;
LIME =
= 3.5 cm;
LIME =
= 2.7 cm;
LIME =
10
20
30
40
50
CAKE SOLIDS - WEIGHT PERCENT
FIGURE 112. SUMMARY OF PRESSURE FILTRATION RESULTS
FOR PRIMARY SLUDGE.
228
-------
thickness decrease. Filtration rates were lowest for primary
sludge, generally about one-half the rates for alum-primary
sludge generated at either alum dosage. Cake solids were higher
for primary sludge, generally 8 to 9 percentage points above
alum-primary sludge generated at a high alum dosage.
229
-------
SECTION VII
DEWATERING PERFORMANCE AND COST COMPARISON
One of the key objectives of this study was to compare the
application of solid bowl centrifugation and vacuum belt filtra-
tion to the dewatering of chemical-primary sludges generated
under various pilot plant operating modes. The purpose of this
Section is to present a comparative summary of performance
achieved and to establish the relative costs for these two
dewatering techniques for specific chemical-primary sludges.
The approach used to develop the comparative performance and
costs for these two dewatering techniques involved:
1. The selection of a municipal wastewater with typical
concentrations of BOD5, TSS and phosphorus and,
2. a prediction of the characteristics of the sludges
produced from chemical treatment of this wastewater
using alum or ferric chloride, each at two specific
levels of phosphorus removal.
The prediction of sludge characteristics was formulated based
on the various analyses made during this study. To review the
relative costs of the two dewatering approaches, plant sizes
of 20,000, 40,000 and 200,000 cu m/day were selected. A detail-
ed summary of assumptions made for this performance and cost
analysis is contained in Appendix G.
Table 48 describes the predicted sludge characteristics that
results from chemical treatment of a typical municipal waste-
water using either alum or ferric chloride to remove phosphorus.
From the various analyses made during this study, the dewatering
characteristics of the sludges described in Table 48 were pre-
dicted. These predictions were used to compare performance be-
tween dewatering methods, chemical coagulant and degree of phos-
phorus removal. Table 49 contains a summary of comparative
dewatering performance results.
The results in Table 49 indicates that dewatering ferric-primary
sludge would require more chemical conditioning and more machine
capacity than that for alum-primary sludge. Cake solids contents
produced from either method were significantly higher when
230
-------
TABLE 48
SUMMARY OF SLUDGE CHARACTERIZATION FROM
CHEMICAL TREATMENT OF MUNICIPAL WASTEWATER
Sludge
Characterization
Sludge production,
Kg DS/cu m
Volatiles, wt. % of
total
Inerts, wt. % of
total
Chemical, wt. % of
total
Thickened feed sludge
Percent P
Alum
80
0.221
69.1
12.2
18.7
4.6
Removal with Primary Coagulant
95
0.247
68.8
10.9
27.3
3.8
Ferric
80
0.228
67.2
11.8
21.0
5.6
Chloride
95
0.258
59.2
10.4
30.4
5.4
to dewatering,
wt. % TS
231
-------
TABLE 49
SUMMARY OF DEWATERING PERFORMANCE
FOR CHEMICAL-PRIMARY SLUDGES
De water ing
Performance
Vacuum Filtration
1. Chemical conditioning lime
10
(jj
10
2.
3.
4.
dose, % ,2«
Filter yield, Kg/hr-sq m* '
Cake ...
Solids content, % TS, \
Volatile content, %u;
Solids capture, %
Percent P Removal with Primary Coagulant
Alum
80
25
13.5
25.5
55
97-99
Ferric Chloride
90
25
8.6
25.6
49
97-99
80
30
10.2
34-5
52
97-99
90
30
8.6
34.5
46
97-99
Centr i f uguation
1.
2.
3.
4.
Chemical conditioning polymer
dose , % / _ .
Loading, Jlpm
Cake m
Solids content, % TSJ.'
Volatile content, %
Solids capture, %
0.3
8.7
18
69
95
0.3
8.7
18
62
95
0.35
6.7
22
69
90
0.35
6.7
22
62
90
(1)
(2)
(3)
Including chemicals
Excluding chemicals
Solid bowl at 15.2 cm x 31.4 cm
-------
dewatering ferric-primary sludge as compared to alum-primary
sludge.
The results summarized in Table 49 also indicate that the solids
content of the filter cakes produced by vacuum filtration of
either chemical-primary sludge were substantially greater than
those produced by centrifugation. Cake volatility, however,
was lower for the filter cakes due to lime conditioning required
for filtration dewatering as compared to polymer conditioning
required for centrifugal dewatering.
An inspection of the results summarized in Table 49 indicates
that vacuum filtration rates for either chemical-primary sludge
produced with 95 percent phosphorus removal were lower than for
sludges produced with 80 percent phosphorus removal. Filter
cake solids contents were not significantly affected by level
of phosphorus removal between 80 to 95 percent. Deterioration
in machine capacity was not observed for centrifugal dewatering.
Generally, operation at 95 percent phosphorus removal as com-
pared to 80 percent removal resulted in a greater sludge pro-
duction and in a cake of lower volatility.
The results given in Tables 48 and 49 were used to establish
cost information for vacuum filtration and centrifugal de-
watering alternatives for specific chemical-primary sludges.
The results of the cost analysis are summarized in Tables 50-53.
As shown, the cost analysis was performed for two levels of
phosphorus removal. This was done in order to study the implic-
ations of the level of phosphorus removal attained from chemical
treatment of raw wastewater on the costs of sludge disposal.
Many of the analyses developed from the results obtained during
this pilot plant study indicated that the thickening and dewat-
ering characteristics of chemical-primary sludges deteriorated
as the fraction of chemical solids contained in the sludge in-
creased. It was reasoned, therefore, that the sludge disposal
costs associated with 95 percent phosphorus removal would be
higher than the sludge disposal costs associated with 80 percent
phosphorus removal.
Estimated incineration costs were developed in addition to de-
watering costs. This was done in order to account for the var-
iation in the characteristics (moisture content, volatile con-
tent) of the cakes produced by the dewatering machines.
Inspection of the annual cost data shown in Tables 50-53 in-
dicated the following conclusions:
233
-------
TABLE 50
COST SUMMARY OF VACUUM FILTRATION OF
ALUM-PRIMARY SLUDGE
ro
CO
1 ' — — -^— — ^— ^_^^^___^__________
Cost
1000 Dollars
DEWATERING
Capital cost
Annual power cost
Annual labor cost
Annual chemical cost
Annual maintenance cost
Annual amortized cost
TOTAL ANNUAL COST
TOTAL COST, Dollars/Metric TDS
INCINERATION
Capital cost
Annual O&M cost
Annual amortized cost
TOTAL ANNUAL COST
TOTAL COST, Dollars/Metric TDS
TOTAL DISPOSAL COST
80 Pei
Plant
20,000
270
5
8
20
4
28
65
40
1512
76
154
230
143
rcent P Re
Size - ci
40,000
266
9
32
40
4
27
112
35
1512
81
154
235
73
jmoval
i/m day
200,000
664
28
63
200
14
68
373
23
3024
248
308
556
34
95 Per
Plant
20,000
389
8
8
22
7
40
85
47
1512
80
154
234
130
cent P
Size -
40,000
351
13
31
45
7
36
132
37
1512
90
154
244
68
Removal
cu in/day
200,000
1087
52
62
223
26
111
474
26
3938
317
401
718
40
Dollars/Metric TDS
183
108
57
177
105
66
-------
TABLE 51
COST SUMMARY OF VACUUM FILTRATION OF
FERRIC-PRIMARY SLUDGE
lo
to
tn
80 Percent P Removal
Cost
1000 Dollars
DEWATERING
Capital cost
Annual power cost
Annual labor cost
Annual chemical cost
Annual maintenance cost
Annual amortized cost
TOTAL ANNUAL COST
TOTAL COST, Dollars/Metric TDS
INCINERATION
Capital cost
Annual O&M cost
Annual amortized cost
TOTAL ANNUAL COST
TOTAL COST, Dollars/Metric TDS
TOTAL DISPOSAL COST
Plant
20,000
313
6
8
25
5
32
76
46
1096
45
112
157
95
Size - cu
40,000
306
11
32
49
4
31
127
38
1096
38
112
150
45
m/day
200,000
809
37
65
247
18
82
449
27
2520
98
257
355
21
95 Percent P Removal
Plant
20,000
388
8
8
28
7
40
91
48
1260
49
128
177
94
Size - cu
40,000
339
15
30
56
6
35
142
38
1298
46
132
178
47
m/day
200,000
1088
54
64
208
26
111
535
28
2646
102
270
372
20
Dollars/Metric TDS
141
83
48
142
85
48
-------
TABLE 52
COST SUMMARY OF CENTRIFUGATION OF
ALUM-PRIMARY SLUDGE
Cost
1000 Dollars
80 Percent P Removal
95 Percent P Removal
Plant Size - cu in/day
Plant Size - cu m/day
20,000 40,000 200,000 20,000 40,000 200,000
DEWATERING
Capital cost 276 276 686
Annual power cost 7 13 40
Annual labor cost 9 34 68
Annual chemical cost 22 45 224
Annual maintenance cost 6 6 21
ro Annual amortized cost 28 28 70
TOTAL ANNUAL COST 72 126 423
TOTAL COST, Dollars/Metric TDS 45 39 26
INCINERATION
Capital cost 1512 1575 4095
Annual O&M cost 91 116 441
Annual amortized cost 154 160 417
TOTAL ANNUAL COST 245 276 858
TOTAL COST, Dollars/Metric TDS 152 86 53
TOTAL DISPOSAL COST
343
8
8
26
7
35
84
47
1890
117
193
310
172
343
14
28
53
7
35
137
38
1890
150
193
343
95
852
54
68
264
28
87
501
28
4253
551
443
984
55
Dollars/Metric TDS
197
125
79
219
133
83
-------
TABLE 53
COST SUMMARY OF CENTRIFUGATION OF
FERRIC-PRIMARY SLUDGE
K>
Ul
Cost
1000 Dollars
DEWATERING
Capital cost
Annual power cost
Annual labor cost
Annual chemical cost
Annual maintenance cost
Annual amortized cost
TOTAL ANNUAL COST
TOTAL COST, Dollars/Metric TDS
INCINERATION
Capital cost
Annual O&M cost
Annual amortized cost
TOTAL ANNUAL COST
TOTAL COST, Dollars/Metric TDS
TOTAL DISPOSAL COST
Dollars/Metric TDS
80 Percent P Removal
Plant
20,000
339
7
7
28
7
35
84
51
1575
79
160
239
144
195
Size - cu
40,000
339
11
24
56
7
35
133
40
1827
90
186
276
83
123
m/day
200,000
842
45
59
281
28
86
499
30
3024
221
308
529
32
62
95 Percent P Removal
Plant
20,000
339
8
8
32
7
35
90
48
1575
87
160
247
131
179
Size - cu
40,000
339
13
28
63
7
35
146
39
1575
100
160
260
69
108
m/day
200,000
844
53
68
316
28
86
551
29
3938
326
401
727
39
68
-------
1. It was costlier by 10-20 percent to centrifugally de-
water and 15-25 percent to vacuum filter dewater alum-
primary or ferric-primary sludge produced from chemical
treatment aimed at 95 percent phosphorus removal as
compared to 80 percent phosphorus removal. This was
caused by the increased quantities of sludge and a
generally more difficult sludge to dewater for the 95
percent phosphorus removal case.
2. There was no significant difference between centrifugal
and vacuum filtration dewatering costs for either alum-
primary or ferric-primary sludge produced from chem-
ical treatment aimed at 95 percent phosphorus removal.
The total costs (dewatering plus cake incineration)
were 25-30 percent higher when centrifugation was used
as compared to vacuum filtration.
3. Centrifugal dewatering was 10-15 percent more expensive
than vacuum filtration of either alum-primary or ferric-
primary sludge produced from chemical treatment aimed
at 80 percent phosphorus removal. The total costs
(dewatering plus cake incineration) were 10-35 percent
higher when centrifugation was used as compared to
vacuum filtration for the alum-primary sludge case.
The total costs (dewatering plus cake incineration)
were 30-45 percent higher when centrifugation was used
as compared to vacuum filtration for the ferric-
primary sludge case.
The conclusions listed above indicate the increased sludge
handling costs realized as a result of chemical treatment of
municipal wastewater to effect 95 percent phosphorus removal as
compared to lower levels of phosphorus removal, i.e., 80 per-
cent. The reasons for the increased cost are the deteriorated
sludge dewatering properties which are documented in this re-
port and the increased quantities of waste sludge generated
with operation at 95 percent phosphorus removal.
The conclusions also indicate that centrifugal dewatering was
generally costlier than vacuum filtration dewatering partic-
ularly when the cost of cake disposal (i.e., incineration) was
considered in the economic analysis.
Figures 113-116 summarize the total dewatering costs and total
disposal costs (including incineration) on a unit cost basis
(dollars per metric ton of dry solids) for the various process
and machine options studied. When viewed on the basis of unit
costs, the results on Figures 113-116 indicate the following
conclusions:
238
-------
to
CO
UJ
I
g
50
CO 40
QC
30
20
0- HIGH ALUM DOSAGE, VACUUM FILTER
HIGH ALUM DOSAGE, CENTRIFUGE
LOW ALUM DOSAGE, VACUUM FILTER
LOW ALUM DOSAGE, CENTRIFUGE
20
30 40 50 100
PLANT SIZE - 1000 CU M/DAY
200
.4
FIGURE 113. COMPARATIVE COSTS OF CENTRIFUGAL AND VACUUM
FILTRATION DEWATERING OF ALUM-PRIMARY SLUDGE.
-------
8
o
£
u
60
50
40
30
20
O HIGH FERRIC CHLORIDE DOSAGE, VACUUM FILTER
& 'HIGH FERRIC CHLORIDE DOSAGE, CENTRIFUGE
• LOW FERRIC CHLORIDE DOSAGE, VACUUM FILTER
A LOW FERRIC CHLORIDE DOSAGE, CENTRIFUGE
20
30 40 50 100
PLANT SIZE-1000 CU M/DAY
200
FIGURE 114. COMPARATIVE COSTS OF CENTRIFUGAL AND VACUUM
FILTRATION DEWATERING OF FERRIC-PRIMARY SLUDGE
-------
to
8
HIGH ALUM DOSAGE, VACUUM FILTER
HIGH ALUM DOSAGE, CENTRIFUGE
LOW ALUM DOSAGE, VACUUM FILTER |
LOW ALUM DOSAGE, CENTRIFUGE
20
30 40 50
100
200
PLANT SIZE - 1000 CU M/DAY
FIGURE 115. COMPARATIVE TOTAL COSTS OF CENTRIFUGAL AND VACUUM FILTRATION
DEWATERING PLUS INCINERATION OF ALUM-PRIMARY SLUDGE.
-------
10
I
o
DC
LU
200
to
O
O
-------
1. There was no significant difference in dewatering costs
for centrifugal dewatering of sludge produced from
chemical treatment aimed at phosphorus removal of 80
percent versus 95 percent. This conclusion was indicat-
ed whether alum or ferric chloride was used as the
primary coagulant.
2. Alum-primary sludge produced from 95 percent phosphorus
removal was 10-15 percent more costly to vacuum filter
dewater than alum-primary sludge produced from 80 per-
cent phosphorus removal. There was no significant
difference in these costs for the case of ferric-primary
sludges.
3. There was no significant difference between centrifugal
and vacuum filtration dewatering costs for alum-primary
sludge produced from 95 percent phosphorus removal.
The same was the case for ferric-primary sludge pro-
duced from 95 percent phosphorus removal.
4. Centrifugal dewatering was 10-15 percent more expensive
than vacuum filtration of alum-primary sludge produced
from 80 percent phosphorus removal. When dewatering
ferric-primary sludge produced from 80 percent phos-
phorus removal, the centrifugal dewatering approach
was approximately 10 percent more costly than the
vacuum filtration approach.
5. There was no significant difference in total costs
(dewatering plus cake incineration) for centrifugal
dewatering of sludge produced from 80 percent versus
95 percent phosphorus removal. This conclusion was
indicated whether alum or ferric chloride was used as
the primary coagulant.
6. There was no significant difference in total costs
(dewatering plus cake incineration) for vacuum filtra-
tion dewatering of sludge produced from 80 percent
versus 95 percent phosphorus removal. This conclusion
was indicated whether alum or ferric chloride was used
as the primary coagulant.
7. The total costs (dewatering plus cake incineration)
were approximately 25 percent higher when centrifuga-
tion was used as compared to vacuum filtration of alum-
primary sludge produced from 95 percent phosphorus
removal. The cost differential increased to approxi-
mately 30 percent for the case of ferric-primary
sludge produced from 95 percent phosphorus removal.
243
-------
8. The total costs (dewatering plus cake incineration)
averaged approximately 20 percent higher when centri-
fugation was used as compared to vacuum filtration of
alum-primary sludge produced from 80 percent phosphorus
removal. The cost differential increased to approxi-
mately 40 percent for the case of ferric-primary sludge
produced from 80 percent phosphorus removal.
9. Operating costs comprise 60-80 percent of the dewater-
ing costs for either machine dewatering either chemical-
primary sludges.
10. Operating costs comprise 40-60 percent of the total
(dewatering plus cake incineration) costs for either
machine dewatering either chemical-primary sludge.
244
-------
REFERENCES
1. Burns, D. E., and Shell, G. L., "Physical-Chemical Treatment
of a Municipal Wastewater Using Powdered Carbon," EPA
Contract Number 14-12-585, Report Number EPA-R2-73-264.
2. Burns, D. E., et al., "Physical-Chemical Treatment of a
Municipal Wastewater Using Powdered Carbon II," EPA Contract
Number 68-01-0183, First Draft.
3. Sharpies, Division of Pennwalt Corporation, Warminster,
Pennsylvania, private communications (November 1975).
4. Standard Methods, 13th Edition, American Public Health
Association,Inc. (1970).
5. Shriver, Division of Envirotech, Harrison, New Jersey,
private communication (June 1974).
6. Johnson Progress, Limited, Great Britain, private communica-
tion (June 1974).
7. Eimco Process Machinery Division, Envirotech Corporation,
Technology Files (1973) .
245
-------
APPENDIX A
DISSOLVED AIR FLOTATION DATA REDUCTION PROCEDURE
Sample Calculation
1. Data collection
(a) Recycle liquid temperature, 20.0°C
(b) Flotation chamber calibration 60.0 ml/cm
(c) Initial slurry height, 0.167 m
(d) Feed concentration, 16.5 g/£
(e) Feed volume, 400 m£s
(f) Recycle volume, 600 mis
(g) Recycle pressure, 4.57 Kg/sq cm
(h) Barometric pressure, 0.896 Kg/sq cm
2. Calculations
(a) Initial slurry concentration
= (16.5 g/A) (0.4001) = fi 6Q /Jt
Co (0.400£ + 0.600*) 6>6° g/*
(b) Solids loading rate (SLR)
(6.60 g/£) (1000 H/cu m) (0.167 m) (60 min/hr)
(4.1 min) (1000 g/Kg)
SLR = 16.1 Kg/sq m-hr
(c) Full-scale flotation rate (FSFR)
FSFR = (0.667*) (16.1 Kg/sq m-hr)
FSFR =10.8 Kg/sq m-hr
(d) Hydraulic loading rate (HLR)
(10.8 Kg/sq m-hr) (1000 g/Kg)
u; (6.60 g/A) (1000 2,/cu m)
HLR = 1.64 cu m/sq m-hr (includes recycle)
*Scale-up factor
246
-------
(2) Bulk separation rate (BSR)
(330 mfc/min)(60 min/hr) (1000 &/cu m) (0.5*)
BSR (60 mfc/cm) (100 cm/m) (1000 fc/cu m)
BSR = 1.65 cu m/sq m-hr
(e) Air to solids ratio (A/S)
g air = (0.129 g/*)(0.600fc) - (0.0211 g/A)
(0.6005,) = 0.0647 g
solids = (16'5 g/M(0-400£) = 6<6° g
A/S = 0.980%
*
Scale-up factor
247
-------
WEIGHT OF AIR DISSOLVED IN
WATER AT 4.57 Kg/sq cm
Sg, G/L at 1 ATM
15 0.0266
16 0.0260
17 0.0255
18 0.0252
19 0.0247
20 0.0244
21 0.0240
22 0.0236
23 0.0233
24 0.0229
25 0.0226
of Air -8,
P = barometric pressure, Kg/sq cm
248
-------
800
600
o
UJ
x
UJ
o
2 400
DC
UJ
200
ULTIMATE
15 MINUTES
LIQUID TEMP = 20°C
FEEDCONC. = 16.5g/l
FEED VOLUME = 400 mis
RECYCLE VOLUME = 600 mis
CHAMBER CALIBRATION = 60 mis/cm
INITIAL HEIGHT = 0.167 m
RECYCLE PRESSURE = 4.57 Kg/Sq.cm
BAROMETRIC PRESSURE = 0.896 Kg/Sq. cm
2345
SEPARATION TIME - MIN.
FIGURE A1.— SAMPLE PLOT OF TYPICAL DAF THICKENING DATA.
249
-------
APPENDIX B
SPECIFIC RESISTANCE DETERMINATION PROCEDURES
I. Equipment Required
A. Laboratory vacuum pump capable of at least 51 cm. Hg
vacuum, fitted with bleed valve and vacuum gauge.
B. Vacuum hose
C. Fifty mi or 100 ml graduated cylinder specially adapted
so vacuum can be applied to it.
D. Seven point zero (7.0) cm diameter Buchner funnel
fitted with appropriate rubber stopper so funnel can
be tightly inserted in graduated cylinder.
E. Whatman GF/A filter paper
F. Stop watch
II. Set-up
A. Insert Buchner funnel into graduated cylinder
B. Connect vacuum hose between vacuum pump and graduated
cylinder
III. Procedure
A. Place filter paper in Buchner funnel and wet.
B. Turn vacuum pump on to evacuate water from filter paper;
following this, open bleed valve full so no vacuum is
applied to funnel.
C. Add approximately 200 m£ sludge sample (already con-
ditioned if conditioning is desired) to Buchner
funnel.
D. Quickly close bleed valve so vacuum of 50.8 cm Hg is
applied to funnel.
E. When desired vacuum level is reached, note filtration
volume and start stop watch.
250
-------
F. Record time and filtrate volume as test progresses,
As this is done, a chart as shown below will be
developed.
Filtrate Corrected
Time Volume Filtrate Volume -9-y
(sec) (mfc) (Vc) (mfc) _ Vj_
0 V0 V0-V0
v, Vl.Vo
V2 V2-VQ -9-2
Specific Resistance Calculation
2 2
where - R = specific resistance, sec /g
P = filtration pressure, g/sq cm
A = filter area, sq cm
V = filtrate viscosity*, poise
c = feed solids concentration, g/cu cm &
b = slope of-0-/V versus V curve, sec/cm
_2
*assume viscosity = 1.002 x 10 poise if capture
exceeds 99%
Sample Calculation
P = 686 g/sq cm
A = 38.5 sq cm
y = 1.002 x 10~2 poise
c = 53.2 g/£
b = slope = .039 sec/cm6
= (2) (686) (38. 5)2 (Q 2q)
(.01002) (.0532) w-o:"
R = 1.49 x 108 sec2/g
R = 1.46 x 10" cm/g
251
-------
Ul
to
30 40 50
FILTRATE VOLUME - mis
FIGURE B1. SAMPLE PLOT OF SPECIFIC RESISTANCE DATA.
-------
APPENDIX C
LABORATORY ANALYSES AND SCHEDULED TESTS
ANALYTICAL PROCEDURES FOR CHEMICAL ANALYSES
TOTAL PHOSPHORUStORTHOPHOSPHATE
Standard Methods, 13th Edition, pp. 526, 532-534. Total
phosphorus is determined by the persulfate digestion method.
Orthophosphate is determined by the ascorbic acid method.
Soluble total phosphorus and soluble orthophosphate are defined
as that phosphorus or phosphate material that passes through a
0.45 micron membrane filter.
TOTAL HARDNESS
Standard Methods, 13th Edition, pp. 178-184
METHYL ORANGE ALKALINITY
Standard Methods, 13th Edition, pp. 370-376
BOD
Standard Methods, 13th Edition, pp. 489-495
COD
Standard Methods, 13th Edition, pp. 495-499.
Note: Soluble BOD and soluble COD are defined as material
passing through a 0.45 micron membrane filter.
EH
The pH will be determined by using a laboratory pH meter.
TURBIDITY
Turbidity of all samples will be determined with a Hach Chemical
Company, Model 2100, laboratory turbidimeter. This device is
a nephelometer (90 degree light scattering) which is standardiz-
ed using standards manufactured by Hach Chemical Company. Tur-
bidity will be reported in NTU.
253
-------
SUSPENDED SOLIDS
Suspended solids are defined as material retained on a 0.45
micron membrane filter. Concentrated samples of chemical-sewage
sludge are filtered through glass mat filter pads (Type GFB). A
calibrated 30 mi syringe is used to measure sludge sample volumes,
The Standard Methods/ 13th Edition, pp. 537-538, procedure for
Nonfiltrable Residue is used. Volatile and ash content of
chemical sludges is determined according to Standard Methods,
13th Edition, p. 536.
IRON
Iron is determined using an atomic absorption analysis on the
filtrate generated during the inerts analysis. A known volume
of filtrate is diluted using distilled water.
"Techtron Atomic Absorption Manual"
ALUMINUM
Aluminum is determined using an atomic absorption analysis on
the filtrate generated during the inerts analysis. A known
volume of filtrate is diluted using distilled water containing
ionization depressant KC1.
"Techtron Atomic Absorption Manual"
CALCIUM
Calcium is determined using an atomic absorption analysis on
the filtrate generated during the inerts analysis. A known
volume of filtrate is diluted using C02-free water, containing
ionization depressant KC1 and Lanthanum as a releasing agent.
"Techtron Atomic Absorption Manual"
INERTS
Inerts are determined by digesting the ash with 70 mfc of 2:1
HCliHNOs. Following the digestion, filtration through a 0.45
micron filter is performed. All material retained on the filter
is reported as inert material.
254
-------
SCHEDULED ANALYTICAL TESTS OF PROCESS
STREAM COMPOSITE SAMPLES
Water Quality Parameter
Turbidity
Suspended Solids
Total and Soluble Phophorus
Chemical Oxygen Demand
Soluble Chemical Oxygen Demand
Biochemical Oxygen Demand
Soluble Biochemical Oxygen Demand
PH
Methyl Orange Alkalinity
Al
Fe
Raw
Wastewater
a
a
a
a
a
a
a
a
a
Parameter
Chemical
Treatment
Effluent
a
a
a
a
a
a
a
a
a
b
b
Dewatering Runs
Routine Sludge Inventories Total Solids
Vacuum Filter and Pressure Filter
Feed
Feed and Lime
Filtrate
Cake
Centrifuge
Feed
Centrate
Cake
Total Solids
Total Solids
Total Suspended Solids
% Solids, % Volatile,
% Ash, % Inert,
% Fe or Al, % P
Total Solids
Total Solids
% Solids, % Volatile
% Ash, % Inert,
% Fe or Al, % P
d
d
d
d
d
d
d
a - Routine analysis on 48 hour composites
b - Rarely
c - As used or discarded
d - On demand
255
-------
APPENDIX D: SAMPLE JAR TEST DATA SHEET
Company .
Address _
Test No
Date Tested .
By
Location.
Material
Object ol Test .
Solids Consisting ol.
Liquids Consisting ol
RAW SAMPLE
TURBIDITY, JTU
SUSPENDED SOLIDS, mg/l
pH, Units
Color, Units
Sample Size
Container Size
Temp °C
Mix Time Flash
Floe.
Settling
. Win. @.
. Min. @.
. Min.
.RPM
. RPM
SAMPLE NUMBER
mg/ml
mg/ml
mg/ml
Dosage, mg/l
Addition. ml
Dosage, mg/l
Addition, ml
Dosage, mg/l
Addition. ml
Dosage, mg/l
Addition. ml
Interlace Height
(In In. ia x Min.) Mm
Min
Mm
Min.
Mm
Min.
Height ol Sample In
Height Alter Min.
Difference In.
Rate In /Min.
Rise Rate (gpm/sq ft) = (Rate)(0.312)'
1
2
3
4
5
6
NOTES
• Includes 0.50 Scale Up Factor
SUPERNATANT
CHARACTERISTICS
Turbidity. JTU
Suspended Solids, mg/l
Color, Units
pH, Units
Sludge Generation, mg/l
Ibs/MG
Sludge Volume mis
%
rSl Urn
Form E-3022
256
-------
APPENDIX E
SLUDGE PRODUCTION DETERMINATIONS
PHASE 1A
= 0.864 Q {TSS + (P) (3.94) mg Al PO./mg P) + (Al)
(2.89 mg Al (OH) 3/mg Fe) } r
Q = 1.40 Ips
TSS = 66 mg/£
P = 4.4 rag/A
Al dosage = 14.7 mg/5,
Alr = Al dosage - (Al used in Al POi, )
Al used in Al PO, = 4'4 ?9 P x '87 "9 A1 = 3-8 mg Al
* mg P 5!
Alr = 14.7 - 3.8 = 10.9
- (0.864X1.40) {66+ (4. 4) (3. 94) + (10. 9) (2. 89) }
= 13.9
day
PHASE IF
= .0864 Q {TSS + (P) (4.87 mg FePO^) + (Fe ) (1.89 mg Fe
(OH)3/mg Fe)} r (2)
Q = 1.39 ftps
TSS = 66 rag/A
P = 4.4 mg/fc
Fe dosage = 28.3 mg/Jl
Fer = Fe dosage = (Fe used in FePOi»)
Fe used in FePOu = 4'4 m? p x 1'81 m9 Fe = 8-° ™<* Fe
I mg P Jl
Fer = 28.3 - 8.0 = 20.3 mg/Jl
= (.0864) (1.39) {66+ (4.4) (4.87) + (20. 3) (1. 89) }
= 15.0 ^2_DS
day
257
-------
PHASE 2A
= .0864 Q {TSS + (P) (3.49) + (Alr) (2.89)} (1)
Q = 2.56 £ps
TSS = 61 mg/5,
P = 5.7 mg/5,
Al dosage = 15.9 mg/£
Al = Al dosage - (Al used in A1PCK)
Al used in AlP0lt= 5'7 ?? P x '81 ^ Al = 5'° ^ Al
x/ ing P x<
Al = 15.8 - 5.0 = 10.9 mg Al
r
. = (.0864) (2.56) {61+ (5. 7) (3. 94) + (10 . 9) (2. 89) }
Y = 25 4 Kg DS
^5'4 day
PHASE 2F
N
^JT = -0864 Q {TSS + (P) (4.87) + (Fer) (1.89)} (2)
Q = 2.56 £ps
TSS = 61 mg/l
P = 5.7 mg/£
Fe dosage = 32.8 mg/£
Fer = Fe dosage = (Fe used in FePO )
Fe used in FePO,= 5-7m? P x 1'*1 m* l& = 10'3mg Fe
mg P
Fer = 32.8 = 10.3 = 22.5 mg Fe/i
= (.0864) (2.56) {61+ (5. 7) (4. 87) + (2 . . 5) (1. 89) }
= 29.2
_
day
PHASE 3A
{TSS + (p) (3.94) + {Al) (2.89)} (1)
Q = 2.47 Ips
TSS = 108 mgA
P = 5.4 mg/X,
Al dosage = 8.1 mg/£
Al = Al dosage - (Al used in AlPOi, )
TM ^ • «nr>^ 5-4 mg P .87 mg Al 4.7 mg Al
Al used in A1PO.,= - g-2 — x - ^jg ,. = - ^ -
x mg c Hi
Al =8.1-4.7= 3.4 mg Al
r I
258
-------
Kf "S = (.0864) (2.47) {108 + (5. 4) (3. 94) + (3. 4) (2. 89)}
Y - 29 7 Kg DS
~ y*7 day
Phase 3F-I
= .0864 Q {TSS + P (4.87) + (Per) (1.89)} (2)
Q = 2.44 £ps
TSS = 88 mg/i
P = 5.2 mg/i
Fe dosage = 22.8
Fe = Fe dosage - (Fe used in FePOi*)
Fe used in FePO^ = 5'2 fff P x l'B1 ™g Fe = 9'4 ** Fe
Fe = 22.8 - 9.4 = 13.4 mg Fe
r ^
= (.0864) (2.44) {88 + (5. 2) (4. 87) + (13. 4 ) (1 . 89) }
- 29 1 Kg DS
~ y>1 day
PHASE 3F-II
= .0864 Q {TSS + P (4.87) + (Fer) (1.89)} (2)
Q = 2.47 Jlps
TSS = 115 mg/£
P = 5.5 mg/£
Fe dosage =10.9 mg/&
Fe = Fe dosage - (Fe used in FePOi» )
Fe used in FePO, = 5'5 P x l'*1 e 10'0 Fe
Fe = 10.9 = 10.0 = 0.9 mg Fe/£
= (0.0864) (2.47) {115 + 5.5 (4.87) + (0.9) (1.89)}
, 30.7
day
PHASE 4A
= (.0864) Q {TSS + (PX3.94) + (Al)(2.89)} (1)
_ . . r
Q = 2.49 ips
TSS = 120 mg/A
P = 5.8 mg/i
Al dosage = 15.6 mg/H
Al = Al dosage - (Al used in A1PCU)
»i j • TVTT^ 5.8 mg P .87 mg Al 5.0 mg Al
Al used in A1PO* = - - — x - p~ = - 1 -
Al = 15.6 - 5.0 = 10.6 mg Al
r • a
= (.0864) (2.49){120 + (5. 8) (3. 94) + (10. 6) (2. 89) }
= 37.4 «2_D&
day 259
-------
APPENDIX P
SLUDGE CHARACTERIZATION DETERMINATIONS
Sample Calculation
PHASE 3F
TSS = 108 mg/fc
Voltailes = 82%
P = 5.4 mg/Jl
Fe dosage = 13.3 mg/&
Constituents Present
Volatiles = (108 mg/i) (.82) = 89 mg/£ VSS
Inerts = (108 mg/£)(.18) = 19 mg/£ Inerts
FePO, = (5.4 mg P)(4.87 mg FePO.)= 26
Fe(OH)3 = (3.5 mg Fe)(1.89 mg Fe(OH)3) = ? mg/J, Fe (OH)3
TOTAL = 141 mg/fc
Constituents Present by Percent
Volatiles = 89/141 = 63
Inerts = 19/141 = 14
23% chemical
, mn ,w
Fe(OH)3 = 7/141 = _ 5
100
260
-------
APPENDIX G
COST ANALYSIS ASSUMPTIONS
This Appendix contains a detailed summary of the assumptions
made in performing the cost analysis discussed in Section XII.
I. WASTEWATER
A. Flows
1. 20,000 cu m/day
2. 40,000 cu m/day
3. 200,000 cu m/day
B. Strength
1. BOD5 = 200 mg/i
2. TSS = 200 mg/i
3. VSS = 160 mg/SL
4. Phosphorus = 8 mg P/&
II. CHEMICAL TREATMENT
A. Primary coagulants
1. Alum (8.25% as A1203)
2. Ferric chloride (43.4% as FeCl3)
B. Phosphorus removal levels
1. 80 percent phosphorus removal
2. 95 percent phosphorus removal
C. Primary coagulant doses
1. Alum
a. 12 mg Al/fl, at 80 percent phosphorus removal
b. 22 mg Al/A at 95 percent phosphorus removal
2. Ferric chloride
a. 22 mg Fe/fc at 80 percent phosphorus removal
b. 40 mg Fe/A at 95 percent phosphorus removal
261
-------
D. Sludges produced (by calculation)
1. Alum-primary
a. 80 percent phosphorus removal
1. 0.221 Kg DS/cu m
2. 69.1 percent volatile solids
3. 12.2 percent inert solids
4. 18.7 percent chemical solids
b. 95 percent phosphorus removal
1. 0.247 Kg DS/cu m
2. 61.8 percent volatile solids
3. 10.9 percent inert solids
4. 27.3 percent chemical solids
2. Ferric-primary sludge
a. 80 percent phosphorus removal
1. 0.228 Kg DS/cu m
2. 67.2 percent volatile solids
3. 11.8 percent inert solids
4. 21.0 percent chemical solids
b. 95 percent phosphorus removal
1. .258 Kg DS/cu m
2. 59.2 percent volatile solids
3. 10.4 percent inert solids
4. 30.4 percent chemical solids
III. SLUDGE THICKENING
A. Alum-primary sludge
1. Underflow = 4.6 percent total solids at 80 percent
phosphorus removal
2. Underflow =3.8 percent total solids at 95 percent
phosphorus removal
B. Ferric-primary sludge
1. Underflow = 5.6 percent total solids at 80 percent
phosphorus removal
2. Underflow = 5.4 percent total solids at 95 percent
phosphorus removal
262
-------
IV. SLUDGE DEWATERING
A. Vacuum filtration
1. Alum-primary sludge
a. 80 percent phosphorus removal
1. Cycle time = 5.0 minutes
2. Lime conditioning dose = 25 percent
3. Filtration rate = 13.5 Kg/hr-sq m (exc.
chemicals)
4. Cake = 25.5 percent dry solids
5. Cake = 55 percent volatiles
b. 95 percent phosphorus removal
1. Cycle time = 5.0 minutes
2. Lime conditioning dose = 25 percent
3. Filtration rate = 8.6 Kg/hr-sq m (exc.
chemicals)
4. Cake = 26.5 percent dry solids
5. Cake =49 percent volatiles
2. Ferric-primary sludge
a. 80 percent phosphorus removal
1. Cycle time = 3.3 minutes
2. Lime conditioning dose = 30 percent
3. Filtration rate = 10.2 Kg/hr-sq m (exc.
chemicals)
4. Cake = 34.5 percent dry solids
5. Cake = 52 percent volatiles
b. 95 percent phosphorus removal
1. Cycle time = 3.3 minutes
2. Lime conditioning dose = 30 percent
3. Filtration rate = 8.7 Kg/hr-sq m (exc.
chemicals)
4. Cake =34.5 percent dry solids
5. Cake = 46 percent volatiles
B. Centrifugation
1. Alum-primary sludge
a. 80 percent phosphorus removal
1. Total solids capture =95 percerft
2. Flow at P600 =8.7 Apm
3. Polymer dose =0.30 percent
4. Cake = 18 percent dry solids
5. Cake = 69 percent volatiles
263
-------
b. 95 percent phosphorus removal
1. Total solids capture = 95 percent
2. Flow at P600 =8.7 £pm
3. Polymer dose = 0.30 percent
4. Cake = 18 percent dry solids
5. Cake = 62 percent volatiles
2. Ferric-primary sludge
a. 80 percent phosphorus removal
1. Total solids capture = 90 percent
2. Flow at P600 = 6.7 Jipm
3. Polymer dose = 0.35 percent
4. Cake = 22 percent dry solids
5. Cake = 69 percent volatiles
b. 95 percent phosphorus removal
1. Total solids capture = 90 percent
2. Flow at P600 = 6.7 £pm
3. Polymer dose = 0.35 percent
4. Cake = 22 percent dry solids
5. Cake = 62 percent volatiles
C. Operating schedule (nominal)
1. Vacuum filtration
a. 15 minutes start-up
b. 30 minute shut-down
c. 8 hrs/day at 20,000 cu m/day
d. 16 hrs/day at 40,000 cu m/day
e. 16 hrs/day at 200,000 cu m/day
2. Centrifugation
a. 15 minute start-up
b. 15 minutes shut-down
c. 8 hrs/day at 20,000 cu m/day
d. 16 hrs/day at 40,000 cu m/day
e. 16 hrs/day at 200,000 cu m/day
V. SLUDGE DISPOSAL
A. Multiple hearth incineration
1. Fuel
a. Fuel selected = fuel oil
b. Fuel value = $2.80/106 BTU
2. Time of operation
For all cases, it was assumed the furnace would
operate for the same time period as the respec-
tive unit(s) of dewatering equipment.
264
-------
3. Design basis
Envirotech technical data
VI. AMORTIZATION, INTEREST, POWER
A. Amortization
20 year period
B. Interest rate
8.0 percent
C. Electricity costs
a. $.0275/kwh at 20,000 m3/day
b. $.025/kwh at 40,000 m3/day
c. $.020/kwh at 200,000 m3/day
265
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-79-055
2.
4. TITLE AND SUBTITLE
CHEMICAL PRIMARY SLUDGE THICKENING
AND DEWATERING
3. RECIPIENT'S ACCESSIOr*NO.
5. REPORT DATE
July 1979(Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
David DiGregorio, J. Brian Ainsworth, and
Keith J. Mounteer
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Envirotech Corporation
Eimco Division
Salt Lake City, Utah 84110
10. PROGRAM ELEMENT NO.
1BC611 SOS 1 Task A/18
11. CONTRACT/GRANT NO.
68-03-0404
Cin.,OH
Municipal Environmental Research Laboratory—
Office of Research and Development
U. S. Environmental Protection Agency
Cincinnati/ Ohio 45268
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer: Roland Villiers (513)684-7664
16. ABSTRACT
This report presents the results of a ten month study of the thick-
ening and dewatering characteristics of chemical-primary sludges. Alum-
primary and ferric-primary sludges were produced in parallel trains of a
pilot plant operated using a municipal wastewater. Each chemical treat-
ment unit was operated under several coagulant doses during the four
phases of this study resulting in the production of several chemical-
primary sludges with distinct characteristics.
Gravity thickening and dissolved air flotation thickening results
for each chemical-primary sludge are presented. Gravity thickening was
evaluated using continuous, pilot scale gravity thickeners; dissolved
air flotation thickening evaluations were performed using batch, bench-
scale equipment. Sludge dewatering evaluations were performed for all
chemical-primary sludges using a pilot scale bowl centrifuge, vacuum
belt filter and filter press.
The report presents correlations developed relating performance of
each unit operation to specific characteristics identified for each
chemical-primary sludge. An economic analysis of centrifugation and
vacuum belt filtration of each chemical-primary sludge is presented.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Phosphorus, Sludge, Sludge disposal,
Sewage treatment, Centrifuges,
Dewatering, Thickening,
Filter press, Economic analysis,
Phosphate, Alums
b.lDENTIFIERS/OPEN ENDED TERMS
Phosphorus sludges,
Phosphate removal,
Wastewater treatment,
Ferric chloride, Air
flotation, Vacuum
filter,
COSATl Field/Group
13B
18. DISTRIBUTION STATEMENT
Release to public
19. SECURITY CLASS (ThisReport)
Unclassi fled
21. NO. OF PAGES
288
20. SECURITY CLASS (TMspagej
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
266
u s MvimMEin HUNTING OFFICE 1979 -6 5 7-060/54Z9
------- |