EPA-R2-73-231
JUNE 1973 Environmental Protection Technology Series
Pressure Filtration of Waste Water
Sludge with Ash Filter Aid
\
;
sssz
X PRO^
Office of Research and Monitoring
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
U. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental
degradation from point and non-point sources of
pollution. This work provides the new or improved
technology required for the control and treatment
of pollution sources to meet environmental quality
standards.
-------
EPA-R2-73-231
June 1973
PRESSURE FILTRATION OF WASTE WATER SLUDGE
WITH ASH FILTER AID
by
James W. Gerlich, P.E.
and
M. Daniel Rockwell
Assistant Superintendent
Water Pollution Control Plant
City of Cedar Rapids, Iowa
Program No. 11060 EZX
Project Officer
Ralph G. Christensen
U.S. Environmental Protection Agency
1 North Wacker Drive
Chicago, Illinois 60606
for the
OFFICE OF RESEARCH AND MONITORING
U. S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D. C. 20460
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402
Price $2.10 domestic postpaid or $1.75 QFO Bookstore
-------
EPA REVIEW NOTICE
This report has been reviewed by the Environmental
Protection Agency and approved for publication.
Approval does not signify that the contents necessarily
reflect the views and policies of the Environmental
Protection Agency, nor does the mention of trade names
or commercial products constitute endorsement or
recommendation for use.
ii
-------
ABSTRACT
Cedar Rapids, Iowa used pilot plant studies as an effective approach to
an economic solution to dewatering secondary digested sludge. After
piloting several dewatering processes, the pressure filter system was
selected and a full scale plant was constructed as the first major instal-
lation in the United States. During the course of the dewatering studies
it was observed that fly ash was an effective filter aid which cut chemi-
cal conditioning costs from about $20 per ton to $4 per ton dry solids.
The full scale plant was designed to utilize sludge ash from incinerated
filter cake, as well as power plant fly ash. The design of this facility
was based upon data collected in the operation of the pilot plant, which
data at that time represented nearly all of the analytical data available
on the operation of the pressure filter in the United States. The design
capacity is 28 tons of dry sewage solids for 16 hour operation at 48 per-
cent dry solids cake.
Performance data from the full scale plant was evaluated over a period of
approximately nine months. This data indicates that the full scale plant
is capable of operation at a greater capacity and efficiency than that pro-
jected from pilot plant data. Both fly ash and sludge ash were evaluated
as a filter aid, with and without chemicals. Economic evaluations were
made of operation and equipment.
This report was submitted in fulfillment of Project No. 11060 EZX under
the partial sponsorship of the office of Research and Monitoring, United
States Environmental Protection Agency.
iii
-------
CONTENTS
SECTION PAGE
I CONCLUSIONS 1
II RECOMMENDATIONS 3
III BACKGROUND AND INTRODUCTION 5
IV VACUUM FILTRATION PILOT PLANT
STUDIES 9
V PRESSURE FILTRATION OF SECONDARY
DIGESTED SLUDGE WITH FLY ASH
FILTER AID 29
VI BASIS OF DESIGN OF FULL SCALE PRESSURE
FILTER DEWATERING FACILITIES 55
VII FULL SCALE PRESSURE FILTER DEWATERING
FACILITY 69
VIII EVALUATION METHODS 83
IX RESULTS 95
X PROCESS EVALUATION 137
XI ACKNOWLEDGMENTS 151
XII GLOSSARY OF TERMS 153
-------
FIGURES
Figure No. Page No.
1 Vacuum Filter Leaf Tests - Correlation of the 11
Form Filtration Rate vs. Cake Formation Time
2 Vacuum Filter Leaf Tests - Form Filtrate Rate 12
vs. Form Time
3 Vacuum Filter Leaf Tests - Full Scale Filtration 12
Rate vs. Cycle Time
4 Vacuum Filter Leaf Tests - Full Scale Filtrate 13
Rate Curve
5 Vacuum Filter Leaf Tests - Cake Moisture vs. 14
Correlating Factor
6 Pilot Scale Vacuum Filter 16
7 Flow Schematic Pilot Scale Vacuum Filter 17
8 Test Results Pilot Scale Vacuum Filter 24
1/1 Fly Ash to Sludge Ratio
9 Test Results Pilot Scale Vacuum Filter 25
1.4/1 Fly Ash to Sludge Ratio
10 Test Results Pilot Scale Vacuum Filter 26
Filtration Rate vs. Fly Ash to Sludge Ratio
11 Test Results Pilot Scale Vacuum Filter Cake 27
Moisture vs. Fly Ash to Sludge Ratio
12 Laboratory Model Pressure Filter and Accessories 30
13 Laboratory Model Pressure Filter 31
14 Removing Cake from Laboratory Model Pressure 32
Filter
15 6" Diameter Filter Cake 32
vii
-------
Figure No. Page No.
16 Filtrate from Laboratory Model Pressure Filter 33
17 Schematic Flow Diagram Pilot Plant Pressure 36
Filter
18 Typical Filtrate Rate vs. Filter Time Curve 43
19 Typical Filtrate Suspended Solids vs. Filter 44
Time Curve
20 Sample Data Sheet for Pressure Filter 47
Performance
21 Filtration Rate vs. Filter Time for 4.5% 48
Sludge Solids
22 Filtration Rate vs. Filter Time for 5.0% 49
Sludge Solids
23 Filtration Rate vs. Filter Time for 5.5% 50
Sludge Solids
24 Filtration Rate vs. Fly Ash/Sludge Ratios 51
for 4.5% Sludge Solids
25 Filtration Rate vs. Fly Ash/Sludge Ratios for 51
5.5% Sludge Solids
26 Percent Cake Moisture vs. Filtration Time for 52
4.5% Sludge Solids
27 Percent Cake Moisture vs. Filtration Time for 53
5.5% Sludge Solids
28 Aerial Photograph of the Water Pollution 70
Control Plant
29 Section View of Pressure Filter Plate 75
30 Process Schematic and Plant Photographs 78
31 Photograph of Specific Resistance Meter 89
viii
-------
Figure No. Page No.
32 Diagram of Specific Resistance Meter 89
33 Resistance Meter Test Example 93
34 Sludge Ash Particle Size - Coulter Counter 100
Model T
35 Fly Ash and Sludge Ash Particle Size Analysis 103
Bahco Analyzer
36 Comparison of Sludge Ash vs. Fly Ash 106
Digested Sludge
37 Comparison of Sludge Ash vs. Fly Ash 106
38 Comparison of Sludge Ash vs. Fly Ash 107
39 Comparison of Sludge Ash vs. Fly Ash, 107
Raw Sludge
40 Comparison of Sludge Ash vs. Fly Ash, 108
Digested Sludge Domestic Wastes
41 Comparison Photographs of Fly Ash and 110
Sludge Ash 200X
42 Comparison Photographs of Fly Ash and 111
Sludge Ash 500X
43 Comparison Photographs of Fly Ash and 112
Sludge Ash 1000X
44 Ash and Chemical Conditioning Chart 116
45 Ash/Sludge Ratio Conditioning Chart 118
46 Chemical Costs vs. Percent Feed Solids 118
47 Filter Performance Curves 2.5% Solids 119
48 Filter Performance Curves 3.0% Solids 119
ix
-------
Figure No. Page No,
49 Filter Performance Curves 3.5% Solids 120
50 Filter Performance Curves 4.0% Solids 120
51 Filter Performance Curves 4.5% Solids 121
52 Filter Performance Curves 5.0% Solids 121
53 Filter Performance Curves 5.5% Solids 122
54 Filter Performance Curves 6.0% Solids 122
55 Filter Performance Curves 6.5% Solids 123
56 Filter Performance Curves, Composite 123
2.5 to 5.5% Solids
57 Typical Filter Pressure Chart 127
58 Sludge Cake Densities vs. Feed Solids and 128
Cycle Time
59 Phosphate in Filtrate 130
60 Yield vs. Cycle Time - Raw Primary Sludge 136
61 Process Costs 4j% Solids 140
62 Process Costs 5|% Solids 140
63 Process Costs 6j% Solids 141
64 Chemical Costs, Comparison of 143
Recommended and Actual
x
-------
TABLES
Table No. Page No,
I Typical Vacuum Filter Leaf Test Results 10
Test No. 3
II Average Characteristics Unconditioned Sludge 96
III Specific Resistance at 1012 cm'2 97
IV Comparative Ash Particle Size at Various Plants 101
V Chemical Analysis of Sludge Ash 101
VI Classification of Fly Ash Particles 103
VII Chemical Analysis of Fly Ash 104
VIII Comparison of Chemical Analysis of 109
Fly Ash and Sludge Ash
IX Average Filtrate Quality 125
X Phosphate Removal with Fly Ash 129
and Sludge Ash
XI Phosphate Removals by Pressure Filtration 131
and Chemicals
XII Summary of Raw Sludge Filtration Data 134
XIII Comparative Values Dewatering Raw 135
and Digested Sludge
XIV Process Costs 141
xi
-------
SECTION I
CONCLUSIONS
1. Pressure filtration of waste water sludges is an effective and eco-
nomical process.
2. Ash filter aid increases dewatering production and decreases chemical
costs.
3. Sludge ash available from incineration of sewage solids can be recy-
cled to the process as an effective filter aid.
4. Fly ash from a coal burning power plant is a more effective ash ad-
mixture than is sludge ash, however, haul and ultimate disposal is an
additional inconvenience,
5. A detailed pressure filter pilot plant program is of great value in
preparing for the process design. Good pilot plant data is reproducible
in a full scale plant which has been confirmed by this report. This data
may also serve as an operating aid for trouble-shooting problems that may
occur during operation of the full scale plant.
6. Some chemicals in combination with ash filter aid further improve
dewatering efficiencies at Cedar Rapids. Chemicals alone without ash
filter aid are relatively ineffective in the range of $20 per ton dry solids
versus $4 to $6 per ton in combination with ash at design conditions.
7. The full scale plant performed at greater capacity and efficiency than
was projected from the pilot plant studies.
8. The pressure filter is very effective in dewatering raw primary sludges,
This program is based upon secondary digested sludges, however, con-
siderable experimentation was done with raw primary sludge at the
termination of the basic program. More experimentation should be done
with raw primary, as well as raw secondary sludge dewatering, by an
extension of this program.
9. Filtrate from the pressure filter is low in suspended solids and phos-
phates and can be recycled without creating a significant load to the
treatment process.
-------
SECTION II
RECOMMENDATIONS
I. This program covered dewatering digested solids. An extracurricular
investigation of raw primary sludge was conducted which consisted of
extensive bench work and 60 pressure filter runs. The results were ex-
tremely encouraging. A more detailed investigation of the performance of
the pressure filter on raw sludge should be undertaken. This investiga-
tion should include raw primary, raw secondary and combined raw primary
and secondary.
2. The physical and chemical mechanism of ash should be further studied.
Some fly ash works much better than others. In some cases fly ash is a
better conditioner than sludge ash, while in other studies, the exact
opposite has been reported.
3. An optimization of sludge thickening as this relates to sludge de-
watering in a pressure filter should be undertaken.
4. An expanded investigation of polymer sludge conditioning should be
made to better define the limitations under pressure filtration. Consider-
able bench work was done at Cedar Rapids using many polymers. Some
showed satisfactory sludge conditioning characteristics in preparation
for filtration, but failed under actual pressure filtration. Favorable per-
formance was experienced with polymer C-7 on the vacuum filter, as an
example, but unfavorable experiences resulted when applied to the pressure
filter. Apparently polymer experiences related to low differential pressure
vacuum filters are not applicable to high differential pressure filters.
5. Some field work is necessary to measure the true demand of the
pressure filter. Experimentation at Cedar Rapids would indicate that
the true demand of the filter, particularly during early cake formation
is not well understood. It has been reported that some European pressure
filter processes slowly apply pressure and during initial cake formation
limit the rate of conditioned sludge applied. Experience at Cedar Rapids
indicated that better filter performance could be expected with extremely
high initial rate of application.
6. A better definition of the value of precoat would be of considerable
benefit as well as an evaluation of materials other than ash. The effect
of quantity and quality of precoat material, along with the effect of
pressure should be studied. The possibility of no precoat should also
be explored in more detail.
-------
7. Deterioration of sludge conditioning in both the mix tank and the
contact tank could have a direct relationship to filter performance.
Some earlier bench work at Cedar Rapids indicated a sludge deterioration
curve starting about 4 to 6 hours after conditioning, however, later bench
work could not confirm this for the actual filter operation. It is believed
this work could be directly related to improved design of the mix tank and
contact tank sludge conditioning phase.
-------
SECTION III
BACKGROUND AND INTRODUCTION
In 1966, the City of Cedar Rapids began testing ways of dewatering
primary and secondary digested sludge from its Water Pollution Control
Plant. At the outset of the dewatering investigations, laboratory work
soon confirmed that the sludge from the Cedar Rapids plant would be
difficult to dewater and that the filterability was unpredictable. Samples
drawn from the digesters indicated that the sludge characteristics varied
considerably. This was attributed to the complex and varied industrial
wastes which entered the plant along with other wastes from the City.
In terms of numbers, the BOD5 population equivalent at the time of this
program development was about 800,000 whereas the actual population
was about 113,000.
To evaluate the sludge dewatering equipment available and to apply it to
these complex wastes, numerous pilot plants were operated including two
sludge centrifuges, a gravity (plug) filter, vacuum filter and a pressure
filter. In addition, sludge samples were treated at a wet air oxidation
process representative laboratory. Some of these programs were a near
failure from the point of sludge conditioning and/or sludge dewatering.
The centrifuge showed some promise in the percent of total solids dis-
charged, however, the percent solids captured and the percent solids in
the concentrate was very poor. Following the centrifuge pilot programs
was a belt-type vacuum filter.
Preceding the vacuum filter pilot study, detailed laboratory work on di-
gested sludges was undertaken using the Buchner funnel tests to obtain
optimum chemical dosages for sludge conditioning. This work was
followed by leaf tests to determine the best yield and most suitable
filter media. These laboratory investigations indicated that the Cedar
Rapids sludge would not dewater at any reasonable cost. Chemical
costs alone were in the range of $20 per ton dry weight. Chemicals
used were combinations of ferric chloride, lime and every polyelectrolyte
available. Combinations of digested and raw sludge were also investi-
gated and this did not affect performance.
In January 1967 experimentation with some conventional and unconven-
tional filter aids was begun in an effort to reduce the chemical costs.
Some of the admixtures investigated are:
-------
1. Volcanic clay
2. Lime waste slurry from municipal water
softening plant
3. Clay fines frpm gravel Dlant 'sooil.jbile
4. Waste activated carbpn from*a local
industrial process
5. Pulverized recycled dried.grit cake from
waste treatment plant
6. Recycled grit and .fibrous material from
plant grit chambers'
7. Fibrous wastes from packing industry
8. Commercial diatomaeeous earth
9. Heated sludge to 140°F.
10. Fly ash from two power generation plants
Some of these materials were a complete failure. Others showed some
promise, but only fly ash performed satisfactorily. The fly ash used was
from a fossil fuel fired steam power generation plant where coal is fired
as a pulverized fuel. Fly,ash is in reality the total ash product of sus-
pended pulverized fuel (rather than only the precipitated ash product of
the stack as from many furnaces utilizing a traveling "grate for the com-
bustion area). Fly ash as a filter aid reduced the chemical costs to
approximately $4 per ton dry weight (compared to $20 per ton) and
delivered 28 percent cake solids from a vacuum filter. Details of the
vacuum filter study with fly ash sludge conditioning appears later in
this report.
In October 1967, a pressure filter was installed and operated for a five
month period using fly ash as a filter aid. No chemical conditioning
was used. The pressure filter-may be generally described as a plate
type consisting of 6 plates, 24 inches' in diameter having approximately
36 square feet of filtration area ,; operated at .a shut-off pressure of about
260 pounds per square inch (psi). The pressure-filter has been in service
in the United States for fifty plus years in industrial processes, but it
has only been in recent years-that its application.tp dewatering sewage
-------
sludge has been seriously considered. At this time the apparent attrac-
tion to the pressure filter was its success in Europe and England operating
without the use of chemicals in numerous installations and its ability to
produce a relatively dense sludge cake with extremely clear filtrate.
Analytical data on the performance of the pressure filter was nearly non-
existent at that time in this country and therefore the Cedar Rapids pilot
program was undertaken with very limited guidelines for evaluating the
process. After several early filter cloth failures, a cloth with suitable
characteristics was obtained and considerable operational data was
accumulated. It was recognized early during the pilot study that the
pressure filter offered numerous advantages which qualified this process
for consideration in the design of the proposed Water Pollution Control
Plant improvements for Cedar Rapids. Details of the pressure filter
pilot study and evaluation of a full scale process plant are included in
this report.
-------
SECTION IV
VACUUM FILTRATION PILOT PLANT STUDIES
Bench Scale - Vacuum Filter Leaf Test
The Cedar Rapids Water Pollution Control Plant is a two-stage high rate
trickling filter plant with separate sludge digestion. Incoming wastes
are high-strength from numerous wet industries such as meat packing,
metals, wet milling, and grain processing. The treatment plant covers
about 30 acres with all of the facilities under roof using thin-shell con-
crete domes. Flow averages about 22 million gallons per day with maxi-
mum treatment at 64 million gallons per day. The population equivalent
of organic loadings is approximately 800,000 persons. The plant has
been operating at a design overloaded condition for organic and solids
quantities for several years.
The feasibility of using fly ash as a filter aid in dewatering secondary
digested sludge at Cedar Rapids was confirmed through a sequence of
laboratory bench work and pilot plant studies. The pilot studies using
fly ash filter aid began with vacuum filtration and finished with pressure
filtration. A full scale sludge dewatering plant, using pressure filtration
was constructed as the result of these pilot studies.
The purpose of the first test was to study the feasibility of vacuum filtra-
tion of digested sludge from the second stage digester. A bench scale
leaf test was made on a mixture of secondary digested'sludge mixed with
chemicals and fly ash. The bench scale test equipment consisted of a
circular "leaf" with exactly 0%1 square foot area, upon which was supported
a piece of appropriate filter cloth. The test leaf was connected through
a filtrate receiver to a source of vacuum. During the test, the leaf was
immersed in the sludge mixture and filtration carried out for a timed
period. The leaf was then withdrawn from the sludge sample and per-
mitted to drain under vacuum for another timed interval. The cake formed
on the leaf was then taken off, and weighed both wet and after drying.
By making individual runs at a series of filtration times and drying times,
the relationship between cake formation time and yield was obtained.
From the same set of data, the effect of drying times on the final moisture
content was also established.
The sludge mixture consisted of 4000 ml of 5 percent by weight secondary
digested sludge, mixed with 500 grams of fly ash, 100 ml of 10 percent
by weight ferric chloride and 20 ml of 2 percent by weight C-7 polymer.
-------
Temperature was ambient, fly ash used had approximately 18 percent
moisture content. Mild agitation was performed to maintain a uniformly
suspended slurry. Results of this typical run No.3 are shown in Table I.
Filtration Results
The sludge mixture filtered fairly well, the cake discharged from the
media cleanly. The filtrate contained suspended solids that rapidly
settled (5 minutes) in a graduated cylinder. Figure No. 1 shows a
correlation of the Form Filtration Rate vs. Cake Formation Time. This
curve is from data obtained on test No. 3, Table I. Tests No. 1 and 2
were not plotted. On test No. 2 there was blinding and on test No. 1,
the medium was too open. Ideally, the slope of the curve shown on Fig.
No.l should be -0.5. The plot shows -1.0. It is believed that the
openings in the cake, after a short interval, were sealed off by the fines
in the sludge mixture. Essentially, the weight of the cake was the same
for all form times, because of this sealing off process. This doesn't
indicate that the mixture is not filterable but that faster cycle time should
be used, as the cake formation occurs only for a short period of time for
this sludge mixture.
TABLE I
Test Run No. 3 - Digester No. 7
Vacuum Filter 0.1 sq.ft., leaf type
Cloth - Nytex 413
Sludge Solids - 5%
Temperature - ambient
4000 ml. sludge
500 grams wet fly ash
410 grams dry fly ash
100 ml. Fe C13 at 10%
20 ml. polymer C-7 at 2%
Test Filtering Filtering
No. % Solids 7ac.In.Hg Time Min.
Feed Form Dry Form Dry
Cake Wt.
Filtrate Cake Grams Percent Dry Sludge Solids
1
2
3
4
5
6
7
10.8
10.6
11.5
11.1
11.0
10.8
25.5
26
25
24.5
24
24.5
24
25
24
24
24
24
,5 3 161
1 2 177
2 1 174
1 1 143
3 .5 1 68
Void
1.5 1.5 139
51.3
28.0
13.8
22.6
8.8
Run
1,7
1/4
1/4
5/16
3/16
3/8
3/16
64.2
68.4
91.8
65,0
93.6
58.4
24.2
25.9
30.5
23.2
29.4
21.2
62.3
62.1
66.8
64.3
70.2
63.8
20,
11.
6.
9.
4.
6.
5
1
1
8
2
0
10
-------
Figure No. 2 is a plot of Form Filtrate Rate vs. Form Time and the curve
was obtained in a manner similar to the plot for Figure No. 1. The slope
of this line is the same as the slope of the Form Filtration curve.
Figure No. 3 is a plot of the Full Scale Filtration Rate vs. Cycle Time.
This curve was obtained from Figure No. 1. Using a 0.8 scale-up
factor and air effective submergence of the filter drum of 25 percent, a
point on the full scale filtration curve was obtained as follows:
Form time = one (1) minute
Yield at one (1) minute form time = 38#/hr/sq.ft.
At 25 percent effective submergence, 25 percent of the filter area is
available to form a filter cake, therefore, the cycle time is:
1. minute = 4 minutes per revolution (MPR)
0.25 "
Full scale yield = 38 x 0.25 x 0. 8= 7. 6#/hr/sq.ft.
The slope of the full scale filtration curve is the same as the slope on
the form filtration rate curve Figure No. 1.
FORM FILTRATION V3 i-'O^M TIME
FLOCCULANTS & FLYASH INCLUDED
2
oc.
o
li.
10
iosU'fl Sluci.K' f.r. PKnU
Sliui i.' Solids Mixod JOOU n.I with
ii)i) .iidn.s fly ash. 100 ir,I 1'orri; Chlnndo
10 dr. wt. sfijution dud JO n.I - ?
ury w[. C-7 Pulyn Jr
1.0
FORM TIME
2.0
(minutes)
3.0
4.0
Figure 1
11
-------
FORM FILTRATION VS FORM TIME
FLOCCULANTS ft FLYASH INCLUDED
S.';-ond,\iv |ii.ii-st,'ci Sluil.i.1 :'.. . I'l.u ) O
>.' Skull.' Sn'ids MiXi-il -lODll i; I wl:-.
fillc) .|I,HV,S ll> .is! , IUO n:l IV'
10' iliy wl . su'uui ml W ir.l - .'
dry wt . ;j - '' IV»l\HUM
1.0 2.0
FORM TIME (minutes)
Figure 2
FULL SCALE FILTER RATE VS CYCLE TIME
4.0
Secondary Digested Sludg
5% Sludge Solids Mixed '
410 cjrams fly ash, 100 m
10% dry wt. solution and
dry wt. C-7 Polymer
NO CHEMICALS OR FLYASH
\L 2
2 3 4 56
CYCLE TIME (minutes/rev.)
Figure 3
12
7 8 9 10
-------
In order to size the filter station, it is necessary to know the full scale
filtration rate of the secondary digested sludge. This was obtained by
calculating the proportion by weight of the digested sludge in the mixture
and assuming the filter cake followed the same weight proportion. This
was calculated as follows:
4000 ml 5% by weight sludge 200 grams
410 grams fly ash 410 grams
100 ml FeCla 10% by weight 10 grams
620 grams
% digested sludge = 200 = 32%
620
The secondary digested sludge without additives was then plotted as
32 percent of the secondary digested sludge with additives.
The Full Scale Filtrate Rate Curve (Figure No. 4) was calculated and
plotted in a manner similar to Figure No. 3. The purpose of this curve
is to estimate the filtrate flow.
FULL SCALE FILTER RATE VS CYCLE TIME
~ 14
j: 13
o
« 10
& 9
~ 8
u
cc.
LE FILTRAT
o> c
o 4
tf>
_i
^ 3
1
-
-
: \
V
\
x
\
~
Tested 12-16-66
~ Basis Lab Test No. 3
Medium Mytex 413
Vacuum 25 Inch Hg
Test Leaf 0.1 sq. ft.
- Secondary Digested Sludge T.F. Plant
5% Sludge Solids Mixed 4000 ml with
500 grams fly ash, 100 ml Ferric Chloride
10% dry wt. solution and 20 ml - 2%
dry wt. C-7 Polymer
5 23
CYCLE TIME (minutes
Figure 4
\
i i
4 5
/ rev. )
13
-------
Figure No. 5, Cake Moisture vs. Correlating Factor is a plot of percent
cake moisture versus drying time divided by cake weight per unit area.
A good correlation was obtained. The purpose of this curve is to deter-
mine optimum moisture content relative to operating conditions.
8 Or-
13
U)
'o
6 75
O
o
65
I
O
UJ
60o
CAKE MOISTURE VS CORRELATING . FACTOR ( 9 d/w)
,, , I.,,: T .«( Si . i
M--ti|.ji: \yt--X -ill
.\I~LI tl:. 2 j iu_w!*;IJJ
.il l.-'cll U. I rig. fl.
.!..;.inu.ir, I .'i :- -,L..". Sim. : f.i. I'l.u.
. Sim: ;. SMiu.i Mi;0'U -Tl >> ;; 1 wit-.
'.'I'J jl.n: s f! .- .isn, MI) n I I'-'m;? (.; li
IT iir, wl. solution HIK! 2'J r:.l - >
'ir, wl. () -i I'olyn .-[
0
1,0 2.0 3,0
CORRELATING FACTOR (.6 d/w)
Figure 5
4.0
Summary
I. The mixture of digested sludge, fly ash and flocculants filtered very
well, giving a good discharge, with filter yields ranging to 20 Ibs. per
hour per square foot for short form times. The cake had a comparatively
low moisture content ranging from 62 to 70 percent. The filtrate appeared
to be fairly high in suspended solids by observation. It was suggested
that these fines be separated from the filtrate before the filtrate is return-
ed to the treatment system. The purpose of this is to prevent any possible
build up of non-separable substance in the treatment system. The material
in the filtrate settled out very rapidly.
2. Full scale filtration of the sludge mixture tested is feasible. It is
our opinion that further leaf tests be run before a pilot plant study is
made. When optimum leaf test filtration results are obtained, based
upon an economic study of the various factors, then a short tern pilot
plant study would be feasible. Further leaf test studies of the fines in
the sludge mixture should be made.
3. The fines in the sludge mixture caused suspended solids in the fil-
trate and while not detrimental to filtration, limited the cake formation
by sealing off the openings in the cake. Further leaf tests on varying
dosages of fly ash and flocculants might indicate ways to lessen the
effect of the fines.
14
-------
4. It was noted that on Test 1 data sheet, using Nytex 415, the back
calculated feed solids are about 8, percent, Test 3 data sheet, Table I,
using Nytex 413, the back calculated, feed solids are about 11 percent.
The Nytex 413 has a slightly closer weave than the Nytex 415. This
would indicate the effect of fines or the size distribution of the fines.
Percent solids in the sludge admixture' was.probably 15 or 16 percent.
The Nytex 413 captured more of the finds than the Nytex 415 as indicated
by the back solids feed calculation.
5. It was not determined what proportion of fines in the filtrate originated
from the fly ash conditioner or the sewage sludge.
6. We believe that the filtration tests, .as reported herein, were well
wor th running. The material-is filterable but further bench scale tests
are in order to evaluate certain points. -This work was done in conjunc-
tion with the pilot plant study.
Pilot Scale Vacuum Filter
Following the, bench 'Scale vacuum- filtration leaf testing program reported
in the previous: section-of this report;'a pilot scale vacuum filter was
obtained jand operated for a period of 10 weeks. The basic objectives of
the pilot scale,vacuum,iilter6, and the flow schematic in
Figure 7. The vacuum filter was a continuous belt filter, 3 ft. diameter
by 1 ft. face. Two nylon belts were selected from previous laboratory
evaluations. Both were monofilame.nt nylon. One was a fine weave
designated by the manufacturer as Nytex 413 . The other a coarser
weave, designated as Nytex 415. The Nytex 413 was used exclusively
during the "test run as a convenience to data evaluation. Other filter
media may warrant consideration.'
15
-------
Vacuum and Filtrate
Pumps and Receivers
Operating Controls
"locculator Mixer
Sludge Mixing Tank
and Variable Speed
Feed Pump
-------
CHEMICAL
FEED
PUMPS
SCHEMATIC DIAGRAM OF PILOT PLANT
Figure 7
The sludge pump was a rotating cavity positive displacement pump, with
a variable speed motor drive. The pump was calibrated both by measur-
ing the displacement at the discharge side and the drawdown in the 225
gallon sludge tank. The displacement of the pump at the setting used
(No. 2) was 4.00+ 0.05 gallons per minute.
Chemicals were added by positive displacement piston type pumps with
variable stroke.
Chemicals and sludge were mixed in a flocculating tank by means of an
air diffuser. This method proved very satisfactory and causes the least
amount of shearing of floe while still maintaining adequate mixing.
Filtration Aids and Chemical Conditions
Fly Ash
The addition of fly ash to the sludge presented a problem and various
methods were used in an attempt to prepare a homogeneous mixture of
digested sludge and fly ash. The last method used, proved to be the
most successful. This method involved screening the fly ash through a
17
-------
coarse mesh window screen over the vortex described by the mixer in the
250 gallon mixing tank. Although this method has been most successful,
a combination of this coarser mesh screen and a finer mesh probably
would have yielded better results. The fly ash used consisted of many
basic oxides which upon addition to the digested sludge raised the pH
to approximately 8.4. Since the fly ash was stored out-of-doors, the
moisture content varied from 15 to 35 percent.
Polymeric Flocculants
The flocculant used in this test was a cationic polymer marketed by
Rohm and Haas as "Primafloc C-7" . This polymer had demonstrated
superior effectiveness as a flocculating agent in the laboratory over
other polymers used. Other polymers are now available which may be
more effective or as effective as the one used, but considering the
number of parameters which needed to be evaluated, it was felt that the
introduction of another parameter would not be justifiable considering the
length of time the pilot plant would be available. The polymer, a dry
powder, was dissolved in water before use. Stock solutions of 2 percent
by weight, were prepared. Dilutions of 0. 6, 0.4, 0.3, 0.2 and 0 .1 per-
cent were prepared from this solution for introduction into the sludge.
It was noted that the dilutions of polymer in the range of 0.1 to 0.01
percent are more desirable and full scale facilities should include chemi-
cal pumps capable of delivering the needed quantities of polymer at these
lower dilutions.
Other Chemical Additives
Ferric sulfate was used to a lesser extent in this study than the polymer
and was always used in conjunction with the polymeric flocculant. The
apparent reason for using ferric sulfate was to lower the dosages of
polymer needed to form an effective floe and not to replace it. Ferric
sulfate is not usually considered to be as effective as ferric chloride,
however, at increased dosages, it is still more economical considering
.the initial cost of 3 cents per pound versus approximately 10 cents per
pound for ferric chloride.
Ferric sulfate was purchased as "Ferrifloc", a dry powder, and dissolved
in water to form a 5 percent by weight solution, which was also the con-
centration used as the feed solution. Both ferric sulfate and the cationic
polymer form acidic solutions and all tanks and feed lines, and fittings,
should be coated to protect the metal. Sodium hydroxide was added to
the sludge on one occasion, to determine the dependence of sludge fil-
tration on pH. The pH which was 8.4 was raised to 9.4 and filtration
immediately ceased, that is, no dischargeable cake was obtained.
18
-------
Test Procedure
The vacuum filtration was performed on a batch basis. The holding tank
held approximately 225 gallons of sludge, which was enough for a 56 min.
run at a constant four gallons per minute feed rate. The fly ash was added
immediately and the run was started. The time between the drawing of
the sludge to the beginning of the run was never more than one hour. Dur-
ing the run, the cycle time was varied from approximately one minute to
four minutes depending upon cake dischargeability. Each cycle time was
timed by a stop watch. Samples of the filter cake were taken using a ring
exactly 0.1 square foot of area . The samples were then dried in the lab-
oratory at 102° C. to determine dry weight. The dry weight in grams was
then converted to pounds per square foot per hour by the following formula:
Cake Wt. in Grams x 10 sq.ft. x 60 Min/Hr. - Ibs./sq.ft./hr.
453.6 Grams/Lb. Min./Cycle
Many duplicate samples were taken at the same cycle time. This was
done because at times, the filter cake was not uniform across the filter
media . This condition is aggravated by the fact the filter face is only
one foot long, secondly, the digested sludge contained bits of matter
not representative of the whole which could give erroneous weights for a
Particular cycle time. Finally this method gives the operator a good esti-
mate of the consistency of operation and the precision of his technique.
Summary - Operating Procedure
General
Cloth
Length of Run
Condition of Cloth
Sludge
Filtering Minutes
Per Revolution
Va cuum
Nytex 413
56 minutes at 4 gallons per minute,
sludge wasted over overflow weir
on filter.
Good - three small holes developed
during la st week of run.
4 to 7% solids, 40-50% volatility
pH 7.0 + 0.2
Rate of filter feed - 4 gallons per min,
0.73-6.00 depending upon cake quality
22^25 inches Hg
19
-------
Percent Submergence 25% constant
Number of Wash 5
Headers Used
Chemicals Polymer C-7 only, dilution 0.1 to 0.6%
Ferric Sulfate 5% feed (when used)
Fly Ash Ratio 0.25 to 1.90 sludge
Fly Ash 15-35% moisture
Filtrate Clarity The suspended solids in the filtrate
ranged from 0.2% solids to 0.64%
with the average very nearly 0.5%
solids (5000 mg/1).
Cake thicknesses were measured by the metric system and the English
system. The thickness of the cake is not of great importance in itself/
because of its dependence upon moisture content and fly ash concentra-
tion.
Explanation of Data Sheet
Date Date of filter run
Percent Solids Dry solids (at 102°C.) content of
digested sludge
Fly Ash to Sludge Ratios of weight of dry solids of fly
Ratio ash to weight of dry solids of digested
sludge.
Lbs./Ton Polymer Pounds of polymeric flocculant per ton
dry weight digested sludge solids only.
Lbs./Ton Ferric Sulfate Pounds of ferric sulfate per ton dry
weight digested sludge solids only.
Percent Moisture Percent moisture of filter cake.
Lbs/Hr/Ft.2/Total Filter yield of all solids including fly
ash
Lbs/Hr/Ft.2/Sludge Filter yield in terms of sludge only
(fly ash not included)
20
-------
Sludge %
Date Solids
1967
4-12 7.71
4-18 7.40
4-18 7.74
5-17 fi.40
5-25 6.30
5-31 6.90
6-13 5.80
6-15 5.75
6-15 5.75
6-15 5.70
6-21 5.0
6-21 5.0
Fly Ash to Pounds/Ton Pounds/Ton Cycle
Sludoe Ratio Polymer Fej(SO4) j Tune _
0.706 6.90 - 4.00
2.62
1.59
0.253 11.00 - 6.40
3.75
2.75
1.66
1.16
0.359 9.33 - 6.50
4.92
2.67
1.42
0.531 8.64 44 3.00
1.71
0.974 4.34 44 6.50
3.93
2.62
2.06
1.66
0.760 2.16 54 3.92
2.7S
2.61
2.16
1.66
1.00 4.76 80 3,92
2.50
1.83
1.50
1.50
1.412 3.12 64.3 3.70
2.50
2.50
1.61
1.61
1.40
1.10
1.42 3.93 64.3 3.58
2,25
1,93
1.53
1,25
0.96 5.88 48.3 4.08
4.08
3.76
3.03
2.75
2,75
1.92
1.92
1.63
1.32
1-55 5. SI 3.05
3.05
2.13
2.13
1.72
1.4S
1.46
1.55 3.67 3.91
3.91
3 as
3. IB
2.15
2. IS
1.43
1.16
*
Moisture
65
67
68
77
76
72
76
76
75
75
73
73
72
72
64
64
62
63
61
66
57
59
58
60
67
68
66
66
66
_
-
-
-
-
.
-
_
-
-
-
-
70
71
71
68
72
72
67
68
67
70
57
S7
55
62
59
57
57
55
55
54
52
62
53
54
41
Lbs/hr/ft2
Total
4.5
6.52
10i96
7.12
8.19
9.05
13.21
11.44
4.46
4.69
4.84
10.10
6.64
9.36
4.02
5.04
5.47
7.77
7.92
2.17
2.61
2.97
3.18
4.21
6.13
7.97
9.97
10.80
10.80 "*
7.69
8.52
9.52
11.80
13.76
15.40
15,71
8.78
10.46
12.10
11.76
16,00
6.11
6.30
5.95
5.74
7.69
7.80
6.08
9.07
11.15
10.46
6.27
6.02
6.80
6.62
10.50
12.80
12.47
3.82
3.63
4.81
4.73
4.76
6.67
8.91
9.51
Lbs/hr/ft2
Sludae
2.63
3.82'
6.42
5.69
6.55
7.24
10.57
9.15
3.27
3.45
3.55
7.42
4.33
6.11
2.03
2.55
2.77
3,93
4.01
1.23
1.48
1.68
1.80
2.39
3.06
3.98
4.98
5.40
5,40
3.18
3.52
3.94
4.68
5.69
6.37
6.50
3.63
4.33
5.01
4.86
6,62
3.12
3.22
3.04
2.93
3.93
3.98
4.13
4.63
5.87
S.3S
2.45
2.35
2.66
2. 59
4.12
S.01
4.88
1.49
1.42
1.88
1.85
1.86
2.61
3.49
3.72
Discharge
C
C
C
B
B
C
C
C
B
B
C
C
A
A
A
B
B
B
C
B
B
B
C
C
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
C
C
C
C
C
C
D
D
B
B
B
B
C
C
C
B
B
0
C
C
C
C
C
Thickness
Inches
1/2
3/8
3/8
-
-
1/4
1/B
3/16
1/B
1/2
1/4
13/64
5/32
1/8
1/8
7/64
9/64
7/64
7/64
7/64
5/64
7/32
13/64
5/32
1/8
1/8
11/64
1/8
1/8
7/64
7/64
5/64
5/64
5/32
1/8
1/8
7/64
7/64
7/32
7/32
5/32
11/64
11/64
9/64
9/64
1/B
1/B
Dollar
Cost
6.90
11.00
9.33
9.96
5.66
3.78
7.16
5.04
5.85
7,33
5.51
3.67
21
-------
Sludge % Fly Ash to Pounds/Ton
Date Solids Sludge Ratio Polymer
6-22 6.6 0.951 4.17
6-28 5.20 1.02 6.93
6-29 6.6 1.07 4.09
6-29 6.6 1.07 2.73
6-30 4.0 1.89 8.03
6-30 4.0 1.89 5.3S
7-3 S.O 1.20 6.34
7-5 S.4 1.36 5.00
Pounds/Ion Cycle
Fej(SOj) 3 Time
3.05
3. OS
1.43
1.43
1.26
1.03
1.03
0.83
O.B3
0.73
0.73
3.20
3.20
2.45
2.20
2.20
2.13
1.50
1.45
3.07
3.07
2.17
2.17
1.47
1.47
1.47
1.27
3.08
3.08
2.13
1.45
3.03
3.03
2.10
2.10
1.43
1.43
1.03
1.03
3.05
3.05
2.10
1.46
1.46
4.43
2.52
2.16
2.16
1.50
1.50
1.07
1.07
3.00
3.00
2.17
2.17
1.50
1.50
1.03
1.03
% Lbs/hr/ft2 Lbs/hr/ft2
Moisture Total Sludge
58
60
59
64
60
58
58
sa
58
59
59
66
67
65
56
62
63
62
62
59
59
57
57
60
59
60
59
55
52
54
-
58
58
57
59
54
58
59
56
56
56
46
56
-
60
61
63
61
61
57
62
63
55
51
53
52
53
53
51
51
4.72
4.44
10.31
9.26
12.9
14.3
14.9
14.86
14.62
15.22
15.25
8.46
B.27
9.76
10.42
9.37
10.31
13.68
11.19
6.79
6.97
8.72
9.13
10.15
10.64
11.70
11.89
3.95
3.95
S.21
No Discharge
6.36
6.49
8.08
7.84
12.80
11.30
13.28
13.53
4.34
4.68
S.74
S.91
S.43
4.31
6,83
7.76
7.53
9.79
10.99
10.88
11.93
6.75
6.67
8.21
8.27
11.27
11.20
13.4
13.7
2.42
2.27
S.28
4.7S
6.62
7.33
7.64
7.60
7.49
7.79
7.82
4.23
4.13
4.88
5.12
4.63
5.15
6.84
5.59
3.27
3.36
4.21
4.40
4.90
5.13
5.65
5.74
1.90
1.90
2,51
2.20
2.24
2.79
2.71
4.42
3.91
.59
.68
.50
.62
.98
2.04
1.68
1.96
3.10
3.53
3.42
4.45
5.00
4.95
5.42
2,86
2.82
3.48.
3.50
4.77
4.74
5.68
5.80
Thickness
Discharge Inches
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
C
C
B
B
G
C
C
C
C
C
C
C
C
B
B
B
B
C
C
C
C
B
B
G
D
D
B
B
B
B
B
B
C
G
B
B
B
B
B
B
B
B
1/8
1/8
7/64
7/M
5/64
5/64
5/64
1/16
1/16
3/64
3/64
5/32
S/32
5/32
5/32
5/32
1/B
3/32
3/32
5/32
5/32
3/32
3/32
1/16
1/16
1/16
1/16
1/16
1/16
1/8
1/8
3/16
3/16
3/16
3/16
1/16
1/16
1/16
1/16
1/16
1/8
t/e
3/32
3/32
3/32
3/32
3/64
3/64
None
None
None
None
None
None
None
None
Dollar
Cost
4.17
6.93
4.09
2.73
8.03
5.35
6.34
5.00
22
-------
Discharge
A
B
C
D
F
Thickness
Chemical Costs
Since cake dischargeability is a critical
factor in filtration studies, cake dis-
charge was given a rating from A to F
and entered on the data sheets of the
valid filter runs.
Very good discharge - thick cake
Very good discharge
Satisfactory discharge, clean cloth
Incomplete discharge
No discharge
Thickness of filter cake in millimeters
or inches ,25.4 millimeters = 1 inch
Costs were evaluated on the following:
$1.00 per pound - Polymer
$0.03 per pound - Ferric Sulfate
A-SlUo_Sludge Ratio 1/1
iliustrates the results of four runs with an approximate ratio of
ash to sludge. The actual conditions are:
Line
1
2
3
4
Fly Ash Ferric Sulfate Polymer
Sludge Ratio Lbs /Ton Lbs/Ton
Chemical
Cost/Ton
6-29
6-28
6-13
5-25
1.02
1.02
1.00
0.974
80
44
4.09
6.93
4.76
4.34
$4.09
6.93
7.16
5.66
fc, e effectiveness of ferric sulfate is not apparent. These plots show
e range of yields and costs which may be found over a period of time.
23
-------
FILTRATION RATE VS CYCLE TIME
2 3 45.67
CYCLE TIME MIN./REV.
8
Figure 8
Fly Ash to Sludge Ratio 1.4/1
Figure 9 shows the result of four runs where the fly ash to sludge ratio
was approximately 1.4 to 1. The actual operating dosages are given
in the following table.
Line Date
Fly Ash/Sludge Ferric Sulfate Polymer
Ratio Lbs/Ton Lbs/Ton
Cost
1
2
3
4
7-5
6-21
6-15
6-15
1.36
1.55
1.412
1.412
_
-
64.3
64.3
5.00
5.51
3.12
3.93
$5.00
5.51
4.44
5.25
Figure 9 also shows the apparent effectiveness of ferric sulfate in increas-
ing filter yields at approximately the same cost as polymer only at this
higher fly ash to sludge ratio. Line 2 is characteristic of a high dose
fly ash. The steep slope shows the effect of fines sealing off the cake.
24
-------
FILTRATION RATE VS CYCLE TIME
(FLYASH SLUDGE RATIO I.4M)
8
7
X'V.
rr
a
04
x
en
CO
(Sewage Solids)
CYCLE TIME
3 4
MIN./REV
Figure 9
Sludge discharge at highest filtration rate (shorter cycle time) is apparent
also. Normally the thinnest dischargeable cake at the'shortest cycle time
will give the highest yield, however, a point of diminishing returns can be
reached at very fast cycle times, when the hourly yield will actually lower
somewhat. Practically the fastest cycle time is one minute per revolution.
Sludge Filtration vs. Fly Ash Ratio
Figure 10 illustrates the sludge filtration rate as a function of the fly ash
to sludge ratio. The points shown are filtration rates taken from their
respective graphs of filtration rate vs. cycle time. The cycle time of
2.00 minutes was chosen because it is a medium cycle time and repre-
sents a point usually well defined on all filtration curves. The points
used for this curve are as follows.
25
-------
Line
Date
Yield
Fly Ash/Sludge
Ratio
Cost
per Ton
1
2
3
4
5
6
7
8
7- 5
6-29
5-31
6-13
6-21
6-15
6-15
6-22
3.80
4.34
3.60
4.50
3.55
4.70
4.-60
3.85
1.36
1.07
0.760
1.00
1.55
0.956
1.415
0.951
$5.00
4,
3,
7,
5,
7.
09
62
16
51
33
5.07
5.60
The optimum fly ash to sludge ratio appears to be approximately 1.0 to
1.1/1. At this point, the sludge filters well, and gives agood dis-
chargeable cake of moderate moisture content, above this point (higher
fly ash ratio) filtration of sludge may improve, but the economic advan-
tages are diminished.
FILTRATION KATE VS FLYASH RATIO
6 r
I 5
H
J3.62
0
0.7
0.8
0.9 I.O I.! I.2 1.3 1.4
FL Y ASH/SLU DGE SOLIDS RATIO
Figure 10
1.5
1.6
Cake Moisture
Moisture content in the filter cake is a function of chemical dosages
and fly ash concentration. Figure 11 represents the range in moisture
content encountered during the test runs. It can be readily noted that
the addition of fly ash causes a significant decrease in moisture con-
tent of the filter cake.
26
-------
CAKE MOISTURE VS FLYASH RATIO
30
0,2
0.4 0.6 0,8 1,0 1.2 1.4
FLYASH/SLUDGE SOLIDS RATIO
Figure 11
1,6
1.8
2,0
Conclusions
1. Fly ash is a valuable sludge conditioner. Preliminary studies indi-
cate that its use lowered the total chemical costs of sludge conditioning
from $20 to $4 per ton dry sludge solids.
Optimum fly ash to sludge ratio for dewatering performance lies between
0.9 and 1.4. However, the optimum economic fly ash to sludge ratio may
be different due to costs of trucking and materials handling of ash and
chemicals, and other operating costs.
2. The value of ferric sulfate is not clear-cut. It appears to be an effec-
tive flocculant at higher (1.4/1) fly ash to sludge ratio, but its benefit is
not clearly defined at lower fly ash ratios. Additional investigation is
warranted in a full scale plant.
3. It has become clear that an efficient vacuum filtration system must be
under close control by the laboratory. Any automatic equipment which
might be incorporated in final plans for the close surveillance of sludge
characteristics should be evaluated.
27
-------
SECTION V
PRESSURE FILTRATION OF SECONDARY DIGESTED SLUDGE
WITH FLY ASH FILTER AID
Introduction
Prior to the pressure filter program the City of Cedar Rapids investigated
various other methods of dewatering bottom sludge from their secondary
digesters. A laboratory model pressure filter and a pilot scale pressure
filter were operated for a period of five months using fly ash as a filter
aid. The information collected during the initial study and presented in
the following report represented nearly all of the analytical data avail-
able on the operation of the pressure filter in the United States.
In October 1967, a pressure filter with a filtration area of 36 square
feet was installed with the necessary accessories at the Cedar Rapids
Water Pollution Control Plant to conduct a pilot plant study for dewater-
ing digested sludge. Fly ash from a local steam generation plant burning
fossil fuel was used as a filter aid. Operation extended over a five
month period with daily observations recorded. For a seven week period,
a two shift operation was in effect to obtain more data in the limited
time the equipment was available.
The pilot scale pressure filter may be generally described as a plate
type, consisting of a series of plates about 24 inches in diameter,
covered with a synthetic filter media. Sludge is center fed to the plates
at an operating pressure of about 260 pounds per square inch (psi) for a
Period of one to two hours. At the end of this period, the plates are
Parted and the filter cake is discharged by gravity and the cycle is
repeated. No conditioning chemicals were used, the only filter aid
being waste fly ash.
At the time the pilot scale pressure filter study was undertaken, the
Pilot scale vacuum filter study had been completed. The results of the
vacuum filter study have been presented earlier in this report. The
Purpose of the pressure filter study was to evaluate the pressure filter
as a practical method of sludge dewatering in light of the earlier pilot
Plant study using a 3 foot diameter prototype vacuum filter. The
vacuum filter study was considered successful and indicated that the
Cedar Rapids digested sludge could be economically dewatered by that
process. The basic objectives of the pressure filter study were as
follows:
29
-------
1. To determine the filterability of digested sludge at
Cedar Rapids.
2. To determine the fly ash filter aid requirements.
3. To evaluate the pressure filter performance versus the
vacuum filter performance reported earlier at Cedar Rapids.
4. To establish an estimate of cost for capital investment
and operation of a full scale sludge filtration facility.
The pilot plant equipment was obtained from the Beloit-Passavant
Corporation and was operated by personnel from the City Water Pollution
Control Plant, and the Howard R. Green Company, Consulting Engineers,
Cedar Rapids, Iowa.
Laboratory Model Pressure Filter
To determine the possibility of dewatering digested sludge using a
pressure filter, a laboratory model filter was first used as shown in
Figure 12.
Figure 12
30
-------
The unit consists of a pressure tank (1), which is a CO2 gas cylinder.
The gas cylinder, as a source of pressure for a laboratory scale experi-
ment, is merely a convenience.
Unit 2, the precoat tank, is used to mix fly ash and water, and then to
transfer it under pressure to the filter media at the beginning of the cycle.
This mixture precoats the filter media for protection and improved dis-
chargeability prior to applying sludge for filtration.
The feed tank, Unit 3, contains the sludge/fly ash mixture to be filtered.
Immediately after precoat, this mixture is applied to the filter at a pr«
sure of about 230 psi. The feed will continue for a predeterminec
of one-half to two hours. The cycle time is of little concern, as the
purpose of this experiment is to demonstrate whether a particular sludge
is filterable.
The pressure filter, Unit 4, Figure 12 and Figure 13, consists of a split
Plate, about 6 inches in diameter, with an elliptical shaped void
surrounded by filter cloth. Sludge is forced into this void under pressure
of 230 psi and builds up on the filter cloth. Filtrate is remove
machined grooves behind the filter cloth.
Figure 13 shows the laboratory pressure filter in operation at Cedar Rapids
Figure 13
Laboratory Model Pressure Filter
31
-------
Figure 14
Removing Cake From Laboratory Model Filter
When the cycle is completed, pressures are reduced and the split plate
is opened exposing the cake for removal as shown in Figure 14.
Figure 15
6" Diameter Filter Cake
A typical cake about 6 inches in diameter is shown in Figure 15. The
center thickness is about 3/4 inches tapering to less near the edges.
32
-------
Note the Clarity of the Filtrate in the Beaker
Sludge Left Beaker - Filtrate Right Beaker
Figure 16
33
-------
Pilot Model Pressure Filter
Description and Operation
At Cedar Rapids the source of sludge was from the secondary digesters.
Sludge from these units was conditioned with fly ash obtained as a by-
product from a fossil fuel power generation station.
The filter is a plate type, consisting of numerous plates in a vertical
position, operating on a carrier. Each plate face is covered with a filter
media, similar to materials used on a conventional vacuum filter. The
filter media used for this study was a synthetic fabric, monofilament
nylon. At the beginning of the filter cycle, all plates are closed and
held as a pressure seal. Feed to the filter media is through the center
of the plates.
Initially, the filter media is precoated with fly ash. The rate of appli-
cation used at Cedar Rapids was 14 pounds wet fly ash per 36 square
feet filter area. Ash moisture ran about 30 percent due to wet-down at
the power generating station prior to open truck transportation and out-
door storage at the Water Pollution Control Plant. The dry rate of
precoat was:
14 pounds ash x 70% solids = 0.27 pounds ash/square foot
36 square feet filter
This precoat is an arbitrary figure empirically determined on the basis
of protecting the filter media and improving dischargeability. Precoat
was applied at a pressure of 40 psi.
Sludge was Placed directly from the digesters into a mixing tank and
conditioned with fly ash. This tank contained a mechanical agitator
and a sloped bottom which held the mixture in a homogeneous state
during the filtering cycle. Fly ash in the desired quantity to obtain a
given sludge/fly ash ratio was fed into this tank by a screw conveyor.
This mixture is pumped to the filter feed tank to begin a filtering cycle.
Air pressure applied to the filter feed tank transfers the mixture to the
filter.
The sludge filtering cycle begins simultaneously with the ending of the
precoat cycle and runs for a predetermined time or pressure. The mix-
ture of digested sludge and fly ash was applied at an initial pressure
of 100 psi and increased to 260 psi where this pressure is held until
the sludge filtering cycle is terminated. The time lapse from start (100
psi) to beginning of maximum operating pressure (260 psi) was
34
-------
approximately 20 minutes (+ 2 minutes). The filtering cycle was termina-
ted at various time intervals, usually increments of one hour.
Filter pressure was developed by applying compressed air to the filter
feed tank, which in turn, forced the sludge/fly ash mixture into the
filter press at the established pressure. The same air source was used
to maintain the filter tank mixture in a homogeneous state during the
filtering cycle by means of agitation from a diffused air header on the
bottom of the tank.
To complete a cycle, the sludge feed line is closed, and the wet sludge
filter core is blown back to the storage tank. A reversing hydraulic
operation of the filter plate carrier causes the filter plates to part, thus
releasing the cake by gravity. The filter cake was caught in a wheel-
barrow and weighed. In a full scale plant, the cake would be discharged
to a conveyor. Following the removal of the sludge cake, the plates
are closed, the filter is precoated, and the cycle is repeated.
A schematic flow diagram of the Cedar Rapids pilot plant is shown in
Figure 17.
Pilot Filter
Filter
Operating Pressure
Design Pressure
Capacity
Closure
Hydraulic Power Pack
Precoat Tank
24" diameter nominal, carbon steel plates,
gasketed with Teflon impregnated, braided
asbestos. Dressed with polypropylene fil-
ter medium. (This can be changed to any
other desired medium.)
225 to 260 psig (pounds square inch gauge)
300 psig
Six filter chambers
Filtration Area: 36 square feet
Cake Space: 2.25 cubic feet, if"
thick cake
Hydraulic cylinder 7" x 30 strokes at4500 psi
Operated by compressed air at 100 psig
60 gallon capacity - 22" diameter by
48" high, 200 psig water pressure
35
-------
FLY ASH FEED
SCREW CONVEYOR
FLY ASH
MIX TANK
CJD
fi
WEIGH FILTRATE
ASH *
~HtO
A i-*
1 J, ' J FitTta
.^i .
TRANSFER
PUMP
4 FILTRATI
FIIC »
F
ftESSUM
FILTER
i i
E
1 H
i J
0 0
CAKl
DIICMAROC
WtlflM SLUDOC CAHC
SCHEMATIC FLOW DIAGRAM OF PILOT PLANT PRESSURE FILTER
Figure 17
36
-------
Feed Tank 300 gallon capacity - 36" diameter by
72" high, 300 psig water pressure
Sludge Mix Tank 48" diameter by 42" high open top tank
polyester fiberglass with agitator,
30 RPM, 1 horsepower, gear motor
Air Compressor 2 stage, 2 cylinder, with 10 horsepower
motor. Air receiver 120 gallon capacity
for 300 psi water pressure.
Sludge Transfer Pump Centrifugal, horizontal
Operational Experiences
No mechanical problems of significance occurred during the four months
pilot plant testing period.
Considerable difficulty was encountered with the first two filter media
due to extremely high grease content of the digested sludges. The high
grease content of the digesters occurred during a period of internal
reconstruction of the grease recovery system at a major meat processing
plant. This problem was not immediately obvious until filter clogging
began to occur. The early media was a yarn type, multifilament. Micro-
scopic examination of the media, even after washing with grease cutting
detergents, indicated that the grease particles could not be removed from
the twists of the yarn configuration. The first media lasted forty cycles
(80 hours). The second media of similar physical characteristics lasted
twenty-six cycles (52 hours). When clogging occurred, it happened
almost instantaneously.
First Media - No. PP1Y1362, plain weave; multifilament
polypropylene, thread count of 60 (warp) and 27 (woof),
weight of 9.78 oz. per sq. yd., porosity of 8.5 cfm
Second Media - No. PP2Y2362, Oxford weave, multifilament
polypropylene, thread count of 72 (warp) and 22 (woof),
weight of 11.8 oz. per sq. yd. , porosity of 10-14 cfm
For the third media, it was decided to use a monofilament nylon which
then performed satisfactorily the remainder of the testing period of
160 cycles, or in excess of 200 operating hours. At the end of the
37
-------
testing period, the m on of i lament media had no evidence of clogging or
deterioration.
' Third Media - No. NY1S9611, plain weave; monofilament
nylon, thread count of 107 (warp) and 101 (woof), weight
of 3 .3 8 oz. per sq . yd., porosity of 582.7 cfm
Sludge Characteristics
Sludge to be dewatered is bottom sludge from heated digesters and
originates as thickened sludge from primary settling tanks, and second-
ary settling tanks following two-stage trickling filters (intermediate and
final).
Sludge was removed directly from the digesters and conveyed to the
pilot plant pressure filter site in a three compartment tank. Each com-
partment held approximately 150 gallons which was sufficient for one
cycle, or three cycles per load. Sludge was used from all eight digesters,
however, the majority of the sludge used for testing came from a second-
ary digester operated in series with a primary digester. In the future,
all digesters will be operated in series as primary and secondary units,
or stage digestion.
General Sludge Characteristics
Total Solids 4 - 7%
Volatile Solids 47 - 57%
pH 6.8-7.3
Volatile Acids 200 mg/1 as Acetic Acid
Alkalinity 2800 - 3500 mg/1 as CaCO3
An extremely high ratio of industrial wastes to domestic wastes exists
at Cedar Rapids. This is reflected in the range of raw sludge charac-
teristics which often is also reflected in digester operation and its
digester bottom sludge. This ratio of industrial to domestic wastes is
in the range of 5 to 1 as measured in (BOD) population equivalent. In
addition to domestic wastes from a population of approximately 110,000
persons, there are industrial wastes from meat processing; grain proc-
essing to starch, syrup, and cereal products; dairy operations; metal
plating and over 100 miscellaneous smaller industries.
38
-------
An example of extreme changing sludge characteristics during the pilot
plant evaluation, which would not be normally anticipated, came from
the meat processing industry. Municipal ordinances restrict the grease
content of wastes discharged to the sewer to 200 mg/1. To comply with
this ordinance, a major meat processing firm undertook extensive in-plant
modifications, at which time considerable quantities of grease were by-
passed to the waste treatment plant. This resulted in poor settling in
the primary clarifiers and poor sludge thickener operation. The digester
operation was frequently disrupted creating a low quality bottom sludge
with high grease content. The early pressure filter runs were somewhat
erratic because of this, and the filter media blinded due to high grease
content. Blinding of the media was a very sudden occurrence, rather
than a progressive one. However, some progressive blinding character-
istics were observed in the precoat operation.
Filter Run Grease Percent
No. (% Total Dry Solids)
34 5.21
37 8.16
39 13.3
41 16.3
50 8.0
70 16.6
100 3.1
106 1.9
108 1.7
110 1.0
The grease content which caused so much trouble was reduced by improve-
ments at the industrial source, and thereafter the filter operation was
more uniform and dependable.
Fly Ash Characteristics
Fly ash used as sludge conditioner in this pilot plant study was obtained
from a local fossil fuel fired, steam power generating station. At this
plant, coal is pulverized and blown into the furnace for suspended com-
bustion. The total ash is principally in the form of fly ash removed
through cyclone collectors. This material is spherical in shape with
about 95 percent passing a 200 mesh sieve.
At the present time, and during the pilot plant study, fly ash was trans-
ported in open dump type trucks from overhead bunkers. To move this
39
-------
material through the City, it was necessary to wet down the load. Conse-
quently, the fly ash was received with varying moisture values. Outdoor
storage also affected the moisture content. This varied from 23 to 35 per-
cent moisture as it was applied to the sludge mixture. Fly ash, in dry
condition as it is recovered at the power plant, has a tendency to "float"
and is difficult to mix into the sludge mass.
For pilot plant operation, the moisture content was not critical because
proper allowances could be made in the calculations, however, in a full
scale plant, consideration should be given to receiving the ash in a dry
condition.
Ash was hand fed through a screw conveyor to the sludge conditioning tank
The quantities were pre-calculated on a dry solids ratio after total solids
determinations were made on both ash and sludge.
Automation of material handling facilities for a full scale plant would
minimize pre-calculations and ash handling.
Laboratory Control
Primary consideration is given in laboratory control to the accurate and
speedy evaluation of the dry solids content of the sludge, fly ash and
cake. From this information most operational control data are determined.
The principal control data are the fly ash to sludge ratio and specific
sludge yield.
To accomplish the above evaluations it was necessary to divide the
laboratory control program into two phases, on site and laboratory anal-
ysis. To determine sludge dry solids concentrations and fly ash moisture
content on the site, an Ohaus moisture determination balance was used.
The moisture balance is a ten gram balance with optical projection read
out to a hundredth (0.01) of a gram. Drying is done by an infrared lamp
which has time and heat level controls to adjust for differing samples.
Operation of the balance consists of measuring 10 grams of the sample
in a tared aluminum dish and setting the heat and timer controls. The
timer shuts off the heater after a predetermined length of time. When
the sample has been dried, solids content is read directly to the nearest
tenth of a percent (0.10%). Drying time is determined by plotting weight
against time at various heat levels until a time and heat level is found
which gives the fastest drying time without burning the sample. By using
this balance it was possible to obtain the solids content of a sludge in
20 minutes and the solids content of fly ash in 5 minutes.
40
-------
Infrared
Heater
Timer
Heat Control
Tare Control
OHAUS MOISTURE BALANCE
The accuracy of the moisture balance was +_ 4.0 percent of laboratory
results. However, laboratory data are necessary to completely evaluate
the facilities.
Two samples of the sludge were taken for each pilot run. These were
composited in the laboratory and triplicate analyses for solids were
performed. The sludge was also analyzed for grease content. Grease
was determined by drying with magnesium sulfate monohydrate and
extracting in a Soxhlet apparatus with petroleum ether.
A sample of fly ash and sludge admixture was taken and analyzed in the
same manner as the sludge. This enabled back calculation of the fly ash
and sludge contents. These samples confirmed the fly ash to sludge
ratios obtained by dividing the dry fly ash weight by the weight of the
dry solids in the sludge. This confirmed that there was adequate mixing
in the sludge conditioning tank and that the sludge and fly ash samples
were representative.
41
-------
Two cake samples were taken by quartering the entire cake and completely
mixing one quarter. The two samples of the mixed cake were .then com-
posited in the laboratory and analyzed for moisture in triplicate. All
moisture samples were analyzed as residue by evaporation at 102° C-
after drying 16 to 24 hours in a forced draft oven to constant weight.
Filtrate samples were grab samples taken during the run to determine the
relative suspended solids concentrations. Suspended solids were deter-
mined by filtration through a glass fiber filter mat in a Gooch crucible.
The suspended solids present in the filtrate appeared to be essentially
precoat which was forced through the filter under the initial filtering
pressure.
The filtrate was analyzed also for BOD and COD. The BOD samples were
seeded to ensure an active aerobic bacterial population. Both the BOD
and COD tests were performed according to the 12th Edition of Standard
Methods. Good correlation was obtained between the BOD and COD,
where the BOD equaled 0.6 COD.
Filtrate Characteristics
The filtrate flow rate was monitored by using a container and a stop watch.
The filtrate quantity was then measured in a graduated cylinder and the
flow rate computed.
Filtrate flow rate observation began at the end of one minute and contin-
ued to the end of the cycle. At one minute for the higher fly ash ratios
(2.00 to 1), the flow rate is approximately 0.15 gallons per minute per
square foot. At lower fly ash ratios of approximately 1.5, the filtrate
rate is 0. 10 to 0.12 gallons per minute per square foot. At the end of
the cycle, the filtrate rate is about 0.007 gallons per minute per square
foot. A typical filtrate flow rate versus time is given in Figure 18.
The filtrate from the pressure filter is nearly free of suspended solids.
At the start of each cycle, the filtrate contains approximately 400 mg/1
suspended solids but this rapidly diminishes to less than 10 mg/1 sus-
pended solids. It was determined by the physical characteristics and
microscopic analysis that the small amount of suspended solids present
primarily originated from the fly ash precoat and were not fines from the
sludge itself.
42
-------
FILTRATE RATE VS TIME
0.12
0.12
c
1
0.08
o
0>
0.06
UJ
cc
UJ
CC
I-
0.04
0.02
40 60 80 100
TIME (minutes)
Figure 18
Figure 19 illustrates the suspended solids content at varying times during
the filter run. Note the rapid decrease in suspended solids to nearly zero
milligrams per liter, Because of the low suspended solids content of the
filtrate, it can be discharged directly to the biological treatment unit.
43
-------
400[
FILTRATE SUSPENDED SOLIDS VS. FILTER TIME
i '1
8 9
TIME
_1_.
10 II 12 13 14-30
(minutes)
40
50 60
General Filtrate Analysis
PH
Suspended Solids
Total Solids
BOD
COD
Color
Figure 19
8.9 - 10.2
<10 to 400 mg/1
2000 to 3600 mg/1
180 to 1200 mg/1
312 to 1700 mg/1
Pale Yellow
The high pH of the filtrate results from the fly ash which contains several
soluble alkaline metallic oxides such as calcium, magnesium, potassium,
and sodium oxides.
Since the pressure filter acts as an efficient solids-liquid separator, the
BOD in the filtrate is dissolved organic material. The BOD of the filtrate
varies with the general conditions of the digesters. During the test period
the BOD of the filtrate remained at the lower values of 180 to 300 mg/1
for extended periods of time. At intervals the BOD of the filtrate increased
upward to 1200 mg/1. This value in itself is not significant, however,
44
-------
the range of values is significant and probably reflects increased hydrau-
lic and organic loadings to the digesters during periods of unusual
industrial wastes loads to the treatment plant. The pounds of filtrate
BOD to secondary treatment at the highest observation is such a small
part of the total that it is insignificant to this study.
Test Results
The evaluations and conclusions in this report are based upon 222 complete
pressure filter runs on Cedar Rapids digester sludge. Sludge was with-
drawn from all digesters and every effort made to obtain representative
samples for dewatering.
Considerable operating difficulties were experienced for a time due to
unusual and varying characteristics of industrial wastes received at
this plant. Particular problems occurred when wastes extremely high in
animal fats adversely affected the digester operation and in turn reduced
the filterability of the digester bottom sludge. Because of these problems
some of the early filter data were erratic and did not classify well. This
was frustrating to the operation of the pilot plant, however, it did reflect
the potential problems characteristic of a plant operating with complex
industrial wastes. These pressure filter operating problems resulting
from sludge with high grease content were very similar to earlier ones
experienced with the pilot vacuum filter at this plant. Fly ash used as
a filter aid was obtained from a local power generation plant and was
considered to be of uniform quality during the period of study. This was
confirmed by analysis by the power company's laboratory at the beginning
and end of the study.
Most work was performed by a two-man team. After the initial startup, it
required about 20 minutes of cycle preparation, which consisted of pre-
coating the filter, transferring the ash/sludge mixture, and charging the
pressure filter. During the cycle, the second batch was prepared by one
man, while the other man prepared laboratory data, or was free to perform
assignments elsewhere.
The filter was large enough to render an approximate evaluation of the time
required to operate a full scale facility. This equipment did not require
constant operating attention. The total time spent at operation was dic-
tated by the laboratory evaluations desired. It is assumed that a full
scale plant would shut down automatically at the end of a predetermined
cycle. The cake would be discharged and a new cycle started at the
convenience of the operating personnel.
45
-------
An operating data sheet was prepared for each filter run and was processed
through the laboratory to include all of the desired data for that particular
run. A copy of the data sheet is shown in Figure 20.
The initial step was to precoat the filter for protection of the media and
to improve cake discharge. The precoat rate is an arbitrary figure and was
determined to be 0.27 pounds ash per square foot as previously calcu-
lated in this report.
Filter runs were made over a range of digester solids varying from 3 .2 to
7.6 percent, with the majority determinations in the range of 4.5 to 5.5
percent. It has been observed at Cedar Rapids that it is difficult to
consistently produce digester bottom sludge above 5.5 percent. Sludge
thickening should be considered in a full scale plant facility on digested
sludge preparatory to dewatering.
Fly ash to sludge ratios are all expressed on the dry sludge solids basis.
These ratios were varied over the range of sludge characteristics. This.
was done in an attempt to demonstrate the optimum fly ash/sludge ratio
irrespective of material handling economics. The ratio was carried
considerably above that which would be economical for municipal sludges.
Filter runs were classified into groups having similar characteristics of
percent digester solids and fly ash/sludge ratios. These values were
plotted to establish performance curves for Cedar Rapids' sludge and to
ultimately develop design criteria to evaluate a full scale pressure filter
sludge dewatering facility. This objective and procedure was similar to
the earlier evaluation of sludge dewatering using a pilot plant vacuum
filter.
The primary objective was to obtain a filtration rate (yield) in terms of
pounds of dry sludge solids, per hour, per square foot of filter media.
Performance curves were established plotting the filtration rate (yield) in
terms of pounds per hour, per square foot, versus time in hours for various
sludge solids densities. Figures 21, 22 and 23 are for 4.5, 5.0 and 5.5
percent sludge solids.
A general observation indicates that as the fly ash/sludge ratio is increased
to a given level, the filtration rate decreases. The optimum yield is the
highest yield at the lowest fly ash/sludge ratio consistent with acceptable
cake discharge and mechanical protection of the filter media. For this
reason, a considerable amount of field interpretation is necessary and
the plotted performance values are not an answer in themselves.
46
-------
;EDAR RAPIDS PRESSURE FILTER PILOT PLANT
Date
Digester #
PH
Quantity of Sludge
Run# Time Start
% Solids
% Volatile
Gallons Vol. Acids
Time End
Avg.
Avg.
mg/1
Remar cs:
£ly Ash: % Solids Avg..
Weight Wet x (%S ) = Weight Dry
+ Sludge: %Solids % Volatile
Fly Ash/Sludge pH
Precoat: Lbs. Fly Ash (Wet) x (%S ) = Lbs. Dry
Filtering: Lbs. Pressure High to low. Length of run rrrs .
Fjjtrate; Turbidity TS mg/1 VS %
COD ppm. BOD ppm. Susp. Solids ppm.
pH G aliens Gal. /SF/Hr.
Filler Cake; % Solids % Volatile
Weight Wet x (%S ) = Weight Dry
Dry Cake Density -^2.25cf. Ibs./cf. Total Gallons Filtered
Yield (Lbs. Dry Cake-^-36 SF) Ibs./SF Media
Filter Rate Ibs./SF/Hr.
Remarks:
Figure 20
47
-------
FILTRATION RATE VS CYCLE TIME
(45% SLUDGE SOLIDS)
2 0 r
--' i o
u.
o ° 9
w 0.8
° 0 6
cri 0 5
CD
_J
O 4
Q
ill 0 3
>-
0 2
0 0
-1.8:1 FLYASH /SLUDGE RATIO
A 1.5:1
O 1.8:1
2.1:1
0 5
I 0
I 5
2 0
2 5
3 0
TIME
HOURS
Figure 21
Cake discharge characteristics were observed at the end of each filtering
cycle and evaluated to the fly ash/sludge ratio. As the filter plates are
opened, the formed cake should break clean from the filter media and drop
away as a single mass.
It was observed that satisfactory cake discharge characteristics could be
maintained at fly ash/sludge ratios of 0.5/1 and with adequate precoat on
the filter media satisfactory cake discharge should be maintained at much
lower fly ash/sludge ratios.
48
-------
CO
CD
:.o i-
o
CO
x
tr
1.0
0.9
0.8
00,7
I
0,6
0.5
Q O.4
0,3
0.2
FILTRATION RATE VS CYCLE TIME
AT VARYING F L Y A S H / S LU D G E RATIOS
(5.0% SLUDGE SOLIDS)
°'5
1.0
TIME
1.5 2.0
HOURS
2,5
3.0
Figure 22
Plotting filtration rate versus fly ash/sludge ratio for a given percent
sludge solids develops a reverse curve. As the fly ash/sludge ratio is
increased the filtration rate is increased until an optimum ratio is reached,
A further increase in the fly ash/sludge ratio causes a decrease ui the
filtration rate (yield).
49
-------
2.0
- 0,9 i-
CO
"- 0.7 -
o:
go.e
0.4
UJ
>-
0.3
0.2
FILTRATION RATE VS CYCLE TIME
AT VARYING FLYASH / SLUDGE RATIOS
(5.5 % SLUDGE SOLIDS )
-2.1.1
O 1.2:1
A 1.5:1
.81
A 2.1:1
0.5
1,0
1,5 2.0
TIME (HOURS)
2.5
3.0
Figure 23
The optimum fly ash/sludge ratio appears to be less than 1.8/1, as
shown in Figure 24 for 4.5 percent solids and Figure 25 for 5.5 percent
solids for varying filter runs from 0.5 to 2,0 hours.
At the 0.5 hour (30 minutes) filter interval it was difficult to obtain a
truly representative sample for analysis because the filter cavity was
not completely filled with cake before beginning a new cycle. From an
operational point of view the filter cavit£ should be completely filled
with cake before beginning a new cycle and therefore, any filtering
period of less than 0.5 hour (30 minutes) would not be practical for
design purposes.
50
-------
FILTRATION RATE VS FLYASH/SLUDGE RATIO
AT VARYING FILTRATION TIME
(4,5% SLUDGE SOLIDS)
1 . C.
. 1.0
6
n
v °<8
LC
O
^ 0.6
./i
m
0.4
a
_j
bJ
> 0 2
,^^-
0.0 1
/
. '(
r
y^
f^~
(
C
o *c
'0 c
-o -c
\
) 0.75 HOUR
) 1 . 0 HOUR
) 1 ,5 HOUR
)2,0 HOUR
}3,0 HOUR
1
i
1 J
1.00 1.25 1,50 1.75 2.00 2.25 2.50
FLYASH / SLUDGE RATIO
Figure 24
FILTRATION RATE VS F L Y AS H / S L U D G E RATIO
AT VARYING FILTRATION TIME
(5 ,5 % SLUDGE SOLIDS)
0.8
a:
o
-
co
CD
Q
_J
LJ
0.6
0.4
0.2
0.0
1,00 1.25 1,50 1.75 2,00 2.25 2,50
FLYASH / SLUDGE RATIO
Figure 25
51
-------
Percent cake moisture decreased at a rapid rate with an increase in
filtration time up to approximately one hour. A very slight decrease
in cake moisture values was experienced after about one hour of
filtration.
This data is given as Percent Cake Moisture vs. Filtration Time in
Figure 26 for 4.5 percent sludge and Figure 27 for 5.5 percent sludge.
Note the similarity of curves for varying fly ash/sludge ratios.
The rate of development of cake quality is about the same for all fly
ash/sludge ratios, however, the cake moisture value improves with
an increase of fly ash/sludge ratio. Yield is expressed as percent
of sludge solids only.
% CAKE MOISTURE VS FILTRATION TIME
AT VARYING PL YASH / SL UDGE RATIOS
(4,5 % SLUDGE SOLIDS )
0.0 0,5 1.0 1.5 2.0 2.5 3.0
Figure 26
52
-------
% CAKE MOISTURE VS FILTRATION TIME
AT VARYING FLYASH / SLU DGE RATIOS
( 5.5 % SLUDGE SOLIDS }
90
30
,0 0.5
1,0 1,5 2.0
TIME HOURS
2,5 3,0
Figure 27
53
-------
SECTION VI
BASIS OF DESIGN FOR FULL SCALE
PRESSURE FILTER DEWATERING FACILITIES
Introduction
Pilot test data compiled during the pilot scale pressure filter program
was interpreted as a basis of design for a full scale pressure filter
plant. During the period of study and of in-plant evaluations to plan
solids handling facilities at Cedar Rapids, Iowa, numerous systems
were evaluated and cost estimates prepared. The primary objective of
these investigations was to determine the most economical system for
dewatering sewage sludge at Cedar Rapids.
At the time of the original submittal to Research and Development for
grant consideration in July 1967, pilot work had been completed using
a vacuum filter and fly ash filter aid. The grant offer was made in
March 1968 and during that time a pilot model pressure filter had become
available and was operated at Cedar Rapids. Economic evaluations of
handling power plant fly ash were less favorable than those of incinerated
sludge ash and it was determined to install sludge incineration and re-
cycle sludge ash as a filter aid.
For the application at Cedar Rapids the pressure filter and the multiple
hearth incinerator was the most economical combination. Evaluating
capital investment and full operating costs, the pressure filter-inciner-
ator combination was estimated to be 10.3 percent less cost per ton of
dry sewage solids dewatered and incinerated over the vacuum filter-
incinerator combination. These cost comparisons are given later in this
section. Other dewatering systems considered were even less favorable
in cost comparison. Factors other than cost which influenced the selec-
tion of the pressure filter over the vacuum filter are also stated later in
this section. Some of these favoring the pressure filter selection are:
less building space, less operator attendance, closed system with fewer
odor problems, drier cake, greater capture and clear filtrate, less power
requirements, expandable capacity of filter unit.
The full scale pressure filter sludge dewatering facility was designed on
the basis of experiences with the pilot scale pressure filter and with the
attitude that the performance of the pressure filter could be improved
upon. The equipment and the process was intended to be sufficiently
flexible to allow a wide range of ash/sludge ratio to be evaluated,
55
-------
and also to perfect other approaches to sludge conditioning for improved
dewatering such as chemical admixtures.
Any deviation from the intended program or process which may be con-
ceived, developed or demonstrated, and which would improve the
performance, should be considered and undertaken in both the plant
design and in the plant study to follow construction. Modifications to
the process, if successful, might accomplish one or more of the
following:
1. Decrease operating time
2. Improve the quality of the product
3. Decrease the unit operating cost
4. Encourage the development of new methods and new
processes
5. Give additional meaning to the Research and Development
program
Filter Loadings
As discussed in Section V, Pilot Scale Pressure Filter, there appears
to be an optimum fly ash/sludge ratio for maximum filter yield. As
evidenced by Figures 21 and 22, this optimum ratio appears to be less
than 1.8 parts fly ash to 1 part sludge at the lower sludge solids con-
centrations of 4.5-5.0 percent. However, at 5.5 percent sludge solids
concentration, the optimum ratio of fly ash to sludge appears to be 1.5
to 1. (Figure 23). The apparent optimum ash/sludge ratio determined by
the pressure filter performance may not be the optimum ash/sludge
design ratio when evaluated with the total sludge dewatering and dis-
posal process. Factors influencing the selection of the ash/sludge
design ratio which are not equated in the pressure filter performance
data include:
1. Material handling facilities
2. Volume of feed ash
3. Relative moisture in the cake product
4. Cycle time
56
-------
5. Volume of ash/sludge product to be dried or
incinerated
6. Unit heat requirements of mixture for drying
7. Unit heat value of mixture for incineration,
and others
To provide continuity to the pilot plant study and this report, we sized
the pressure filter and accessories on the basis of the performance at
a fly ash/sludge ratio of 1.5 to 1, and reported out alternate performance
at 1 to 1 fly ash/sludge ratio, and also 1 to 1 organic ash/sludge ratio.
Referring to either Figure 23 or Figure 25, at a fly ash/sludge ratio of
1.5 to 1, yield is found to be 0.75 pounds per hour per square foot of
dry sludge solids at 1.5 hours' filtration time. Figure 27 indicates that
the cake moisture would be 52 percent (cake solids 48 percent) which
by normal sludge dewatering standards is a very dry cake. Further
reference to Figure 27 shows that it would be impractical to extend the
filtration time beyond 1.5 hours because the moisture content of the
cake would not further diminish in proportion to the time used. On the
other hand, the filtration time might well be decreased to about one
hour and not appreciably increase the moisture content of the filter cake.
For basic design, a 1.5 hour filtration time was used.
Between filtration cycles, during which time the cake will be discharged
and preparations made for a new cycle of operation, an allowance of
30 minutes was made. This is a generous amount of time, however, it
does not greatly affect the total cost of operation if it is estimated to
be 10 minutes or 30 minutes. Unlike the vacuum filter, the pressure
filter does not require constant attendance. It is, therefore, possible
for the operator responsible for the pressure filter to be performing other
duties between filter cake discharges, and he may elect to delay recy-
cling immediately after the predetermined 1.5 hour filtering cycle. Or,
the operator may quickly discharge cake and recycle in five or ten
minutes and return to other duties to be delayed on the next cycle. The
total hours of filtration would remain the same at 16 hours per 24 hour
day.
Sake Production and Handling
It was planned to have the pressure filter and accessories as completely
automated as was practical and to monitor most functions from a central
console. Based upon operating experiences with the pilot scale plant,
the method of controlling the filtration cycle would be:
57
-------
1. At preset time interval of filtration. At the end of this time, filtration
would cease and the operator would discharge the cake and recycle the
process.
2. At preset pressure value. Theoretically, pressure should build up at
an increasing rate as the filter cake is formed and could be used to deter-
mine the end of the filtration cycle when the filter cavity is filled. This
theory of progressive pressure build-up might work in the filtration of a
quality controlled manufactured product, however, it does not appear to
be a useful control in the filtration of sewage sludges.
3. A preset rate of filtration. By this method the filtrate discharged
from the pressure filter is metered and the filtration cycle is terminated
at a given rate of filtration, such as 0.01 gallons per minute per square
foot of filter media. A variation of this would be to meter the total volume
of filtrate and the filtration cycle would terminate when a given value of
total gallons was registered. This facility should be provided on all
installations. Where it is not used to control filtration time, it may be
used to monitor the physical condition of the filter media.
Time, pressure, and metered flow information will be desirable in the
operation of the equipment and, therefore, all three methods might be
readily available. This installation will plan to utilize the first and
third methods. The time interval will be adjustable to changing condi-
tions or requirements of sludge, ash, cake moisture and other admixtures.
The rate of filtration will be metered to determine the end of the cycle.
At the end of the filtration cycle, the operator would normally wish to
discharge the cake promptly and recycle. However, with the pressure
filter it will not be necessary for the operator to do this promptly if he
is engaged in other activities. Unlike other filtration equipment, the
pressure filter may be turned off at any time during the filtration process
and left unattended for an indefinite period with no deleterious effect on
operation or cleanup.
At the completion of filtration cycle, the operator standing at the control
panel will prepare to discharge the cake. The procedure will be similar
to that of the pilot filter. Sludge pumping to the filter will be terminated
(if this was not done automatically as indicated above), the unfiltered
sludge remaining in the filter core and feed lines will be discharged to
the feed tank, conveyor equipment will be made to start, and the operator
will proceed to the pressure filter.
Cake discharge is to be observed by the operator after activating a con-
troller to open the filter plates. The cake will discharge freely and the
58
-------
large disc-shaped sludge cake will be sheared into small pieces by a
series of steel cables strung across the receiving hopper. The broken
filter cake will be transported by a conveyor system to the incinerator
and then to ultimate disposal.
If additional sludge is to be dewatered, the operator will close the filter
plates and recycle the process from an operating console. The sequence
of'filter precoat, sludge filtration, and cake discharge will then be
repeated.
Filtrate Disposal
Disposal of pressure filter filtrate should present no problem. Filtrate
from the pressure filter is nearly free of suspended solids and those
present originate from the fly ash, rather than fines from the sludge solids.
For a short period of time at the beginning of the filtration cycle filtrate
rate may be relatively high. For the size of this facility, the filtrate rate
may be about 650 gallons per minute for a few minutes. Within 10 minutes
this value may decrease to about 25 percent of this higher rate and then
continue to decrease to less than 10 percent of the maximum rate when
the filtration cycle is half completed. Refer to Figure 19 for a typical
filtrate rate of flow curve versus time. Filtrate may be handled in one of
several ways.
1. Filtrate may be discharged to a relatively small holding tank having
10 to 30 minutes detention and then discharged at a lower rate directly
to the biological (secondary) treatment units. The first 10 minutes of
pressure filtration will produce approximately 33 percent of the total
filtrate volume from the cake and at 20 minutes approximately 58 percent
will be removed. (Figure 18.) Thus, with about a 7,000 gallon detention
tank, the filtrate may be discharged at a rate of about 100 gallons.per
minute to the plant for secondary treatment.
2. Filtrate may be recycled back into the dewatering process as makeup
water for the filter precoat with excess quantities discharged through the
plant for secondary treatment.
3. Filtrate may be discharged to sludge thickening facilities where these
are available and where the high pH of the filtrate might be an aid to
sludge thickening.
Filtrate quantities should be metered on a continuous basis as an instru-
ment of filter operating control. Filtrate rate of flow and total filtrate
Per time interval information can be used to adjust the filtration cycle
and ash/sludge ratio.
59
-------
Ash Requirements and Handling
Fly ash was available from two sources. Both sources are steam genera-
ting stations firing fossil fuel and are located approximately three miles
from the treatment plant.
Fly ash should be hauled in closed trucks and blown into storage. Stor-
age facilities should be provided for a minimum of 2 to 4 days' ash
requirements. This will provide an adequate supply of ash to continue
operation until other solids disposal arrangements can be made, should
either ash delivery or cake disposal systems be temporarily out of
service . Ash feed may be fed to the slurry mixing tank using volumetric
feeders or controlled volume materials handling equipment. An adjustable
speed screw conveyor may be calibrated with sufficient accuracy to
handle the large volume of ash required. For this installation, gravimetric
feeders were used to measure ash to the screw conveyor.
Solids Preconditioning and Thickening
Solids should be preconditioned in a sludge thickening facility prior to
pressure filter feed conditioning. The pressure filter pilot study was
conducted without sludge thickening because this facility was not avail-
able. However, sludge receiving facilities are necessary to control the
application to the sludge dewatering equipment and these facilities may
well serve a dual purpose. Any degree of sludge thickening will further
improve the performance of the pressure filter. It was anticipated that
the filter yield would be materially increased with thickened sludge.
Detention times in the digested sludge thickeners should be relatively
long, in the range of 2 to 4 days1 detention. With 5.5 percent solids
entering the thickener, it might be possible to consistently come out
with 8 to 10 percent solids. Small quantities of chemicals might further
reduce the ash/sludge ratios required. It was planned to do research at
this facility to determine the minimum ash/sludge ratio to obtain the
maximum filter yield at the least cost.
Filter Cake Disposal
Cake discharged from the pressure filter having a moisture content of
only 50 percent can be transported and disposed of as landfill or soil
conditioner, providing it is pulverized for the latter use. Cake disposal
in this condition would be limited to the City landfill, and sparsely
populated areas for bulk dumping. Considerable quantities of pulverized
cake might be used as a soil conditioner and fill material by the Parks
60
-------
Department operating many municipal parks/ golf courses and,recreational
areas. However, even at maximum utilization of this material by the
Parks Department, a large volume will have to be disposed of by other
programs.
To reduce the volume of material to be handled for disposal, and to
provide an in-plant source of ash for conditioning, incineration facilities
were installed. The proposed process will provide for disposal of solids
at one or more points in the dewatering process to best serve the interest
of both the process operator and ultimate disposal as follows:
1. Filter cake may be discharged in part or in total volume where soil
conditioning might require maximum nutrient value or where landfill
might require maximum bulk.
2. Filter cake may be dried in part or in total volume where the product
might be stored for future use, or where wet filter cake might be offen-
sive, and in this manner maintain maximum nutrient value and minimum
bulk handling.
3. Filter cake may be incinerated in part or in total volume, thus mini-
mizing the bulk handling requirements and utilizing the resultant organic
ash as a sludge conditioner and filter aid. This will minimize ash
handling requirements as well as solids handling requirements.
It was planned to both dry and incinerate by using a multiple hearth
incinerator with provisions for reverse air flow when only drying is
desired. This type of equipment affords the greatest flexibility of
operation and with a reasonably low unit product cost. More detailed
cost analysis appears later in this report. Modern facilities include
gas cleaning devices to meet the local air pollution codes. The minimum
operation and maintenance requirements of a multiple hearth incinerator
are compatible with the minimum operating attention demanded of the
pressure filter used as the primary equipment for sludge dewatering ahead
of incineration. This combination will demand the minimum operator
attention and may be operated by the same personnel.
Other systems evaluated and cost comparisons made for Cedar Rapids
include:
1. Vacuum filter and rotary drier
2. Vacuum filter and flash drier
61
-------
3. Vacuum filter and furnace incinerator
4. Vacuum filter and multiple hearth incinerator
5. Wet air oxidation and vacuum filter
6. Pressure filter and multiple hearth incinerator
a . Fly ash filter aid
b. Organic ash filter aid
Summary of Design Data
Digester Solids
Total Solids 56,000 Ibs ./day
Volatile Solids 26,400 Ibs./day
Percent Solids in Sludge 5.5
Volume of Sludge Withdrawn 122,000 gals./day
Pressure Filter Loadings
Fly Ash Organic Ash
Ash/Sludge Ratio 1.5:1 1:1 1:1
Filtration Time per Cycle Hours 1.5 1.0 1.0
Total Cycle Time Hours 2.0 1.25 1.25
Total Filtration Time Hrs./Day 16 16.7 14.4
Cake Moisture Percent 52 60 50
Yield Filter Cake Lbs./Sq.Ft./Hr. 0.75 0.65 0.75
Filtrate Suspended Solids Nil Nil Nil
Chemical Required None None None
62
-------
Other Plant Data
Population 1968 Present 113,000
1988 Future 172,000
BOD Equivalent 810,000
Plant Flow Average, MGD 28.6
Wet Weather,MGD 64.4
Estimate of Costs
The original estimates of cost of construction were based upon the
August 1968 costs and were limited to the solids handling facilities
directly related to sludge dewatering and incineration. For the conven-
ience of this report, those 1968 costs have been scaled upward to
August 1972 costs based upon the Engineering News-Record Cost Index
of cost trends for construction and increased equipment costs. During
the period since the original cost estimate (1968) and the present (1972),
construction costs have increased about 54 percent and equipment costs
about 20 percent. Other costs, such as operation, have inflated at
least 4 percent per year.
The cost of operating the proposed pressure filter and incinerator was
estimated from the power, fuel, materials and labor required at the
design quantity and quality of sludge to be processed. These costs
were determined with an electrical rate of 1.8 cents per kilowatt hour;
off-peak natural gas rates of 4.8 cents per therm for the first 500 therms
and 3 .85 cents per therm over. Labor was determined on the basis of
continuous attendance. Maintenance costs were based upon known cost
of repair and replacement with consideration given to the equipment manu-
facturer's stated minimum estimates.
63
-------
Ash/Sludge Ratio
Hours of Operation
Power Pressure Filter
Power Incinerator
Fuel Incinerator
Fuel Deodorize Stack Gases
Material Handling:
Fly Ash to Plant
Incinerator Ash Disposal
Filter Cloth Replacement
Equipment Maintenance
Operating Labor
Total Estimated Operating Costs
Cost per Ton Dry Solids (Sewage)
Fly Ash Organic Ash
1.5:1
16
0.48
1.46
0.98
1.89*
1.25
0.15
0.52
2.75
1:1
1:1
16.7 14.4
$0.60 $ 0.70 $ 0.54
0.40
0.18
1.68 1.70
1.06
0.24
1.26*
0.97 0.34
0.16
0.56
2.75
0.14
0.56
2.75
$10.08 $ 9.54 $ 6.45
* The Power Company did not charge for the delivered fly ash,
however, a minimum transportation charge likely would be
charged in most cases. To compare processes on an equal
basis an estimated charge is added for transportation.
Assuming amortization of bonds at 4f percent interest for 20 years, the
capital cost per ton of dry sewage solids would be $20.00 based upon
the estimated 1972 construction cost of $2,570,000. The total cost
per ton dry sewage solids, including capital investment and operation,
would be as follows:
64
-------
Fly Ash/Sludge Ratio Organic Ash Ratio
Ash/Sludge Ratio 1.5:1 1:1 1:1
Capital Investment $20.00 $20.00 $20.00
Operation & Maintenance 10.08 9.54 6.45
$30.08 $29.54 $26.45
Valuations of Other Solids Handling Systems
During the period of planning solids handling facilities at Cedar Rapids,
numerous systems were evaluated and cost estimates prepared. For
the application at Cedar Rapids, the pressure filter and multiple hearth
incinerator combination is the most economical and offers many additional
advantages which are difficult to equate to a dollar value. Because of
the great amount of extra work done in the plant operation, in the labora-
tory, and in design calculations, it is warranted to at least report briefly
on the systems considered. It should be clearly understood that these
evaluations are based upon sewage solids conditions at Cedar Rapids,
and it is not intended to be a recommendation for other installations,
nor does it imply criticism of those existing.
At the time of the original submittal to Research and Development for
grant consideration, pilot plant work had been completed using a vacuum
filter. This was in July 1967. The grant offer was not made until March
1968. During that time a pilot model pressure filter became available
and was operated at Cedar Rapids and is the main subject of this report.
To effect a fair and equal basis for comparison of costs for several
Sludge dewatering systems, all previous cost estimates are projected
to 1972 construction costs based upon the Engineering News-Record
Index of Cost trends as previously stated. All sludge thickening facil-
ities were omitted from these comparisons, and all other accessories
were made to be exactly equal. As an example, all product haul was
limited to three miles whether it be wet cake or incinerator ash. All
costs of both the vacuum filter and the pressure filter were based upon
performance at a 1:1 ash/sludge ratio because most of the vacuum filter
Work was conducted at this optimum ratio.
The estimated costs per ton of dry sewage solids were determined from
the following sources.
65
-------
1. All equipment costs are budget costs quoted by various manufacturers.
2. Building costs to house the equipment are estimated by the Engineer
and are minimum building requirements.
3. Fuel and electrical costs are based upon equipment requirements and
Cedar Rapids rates.
4. Labor costs are at plant rates.
5. Labor for operation and maintenance was determined by the Engineer
and was considered approximately equal for all installations. This is
conservative as it is reasonable to suspect that the vacuum filter main-
tenance costs are considerably greater.
Many advantages and disadvantages have been proposed by many authors
and equipment representatives for each and every item of equipment. The
merits of these are too of ten shaded by prejudice, and at best have little
real value. Conclusions and observations from actual pilot plant perform-
ance using both a vacuum filter and a pressure filter at Cedar Rapids are
shown on the following page.
66
-------
Operating Cost
Amortize Dewater Dewater and Total
Construction 20yrs,@4|% Only Dry or Incin. Cost
Costs Per Ton Per Ton Per Ton Per Ton
Vacuum Filter
and Rotary Drier $2,135,000 $16.55 $11.25 $25.70* $42.25*
Dry
Vacuum Filter &
Flash Drying
Furnace 2,155,000 16.75 13.05 23.00* 39.75*
Dry
Vacuum Filter &
Flash Drying
Incinerator 2,600,000 20.40 13.05 20.05** 40.45**
Vacuum Filter &
Multiple Hearth
Furnace 2,080,000 16.15 11.25 16.80** 32.95**
Pressure Filter &
Multiple Hearth
Furnace
(Fly Ash Aid) 2,570,000 20.00 5.00 9.54** 29.54**
Pressure Filter &
Multiple Hearth
Furnace
(Organic Ash Aid) 2,570,000 20.00 4.12 6.45** 26.45**
* Dry cake approximately 12% moisture
** Incinerated ash
67
-------
Near full time attendance required for
one operation
Va cuum
Filter
Yes
Is a cleanup shift required? Yes
Can one operator perform all shift duties? Yes
Odors produced from operation Considerable
Odor control accessories required
due filter
Chemical required
Ash additive required
Cake, percent solids
Filtrate, suspended solids, mg/1
Cloth life, estimated hours
Power requirements, with
accessories, KWH
Can filtration area be increased?
Operating Costs:
Capital investment for equipment
Capital investment for building
Incinerated costs per ton dry sewage
solids
Yes
Yes
Yes
33-38
2000-6400
2000-4000
1430
No
50% more
$32.95
Pressure
Filter
No
No
Yes
Nil
No
No
Yes
40-55
10-400
6000-8000
1080
Yes
20% more
$26.45
Obviously, there are many other items of comparison, some significant
to treatment requirements, and many of little importance. The most
important consideration, and the one which is often disguised by report-
ing it in part rather than in total, is how much will the total cost be for
both capital investment and complete operation. As reported earlier,
the total cost per ton of dry sewage solids disposed of strongly favors
the pressure filter.
68
-------
SECTION VII
FULL SCALE PRESSURE FILTER DEWATERING FACILITY
Treatment Plant Description
The full scale Solids Handling Facility at Cedar Rapids was planned in
1968, bids were received and contracts awarded in December 1969, and
construction was completed in the spring of 1971. During the remainder
of 1971 the sludge dewatering and incineration processes were checked
out. The full scale Research and Development program for sludge de-
watering with ash filter aid was carried out during the calendar year
1972, and is the subject of this report.
To better understand the size and operation features of the sludge de-
watering facility, it would be well to briefly discuss the basic water
Pollution control plant from which the sewage sludges originate. An
aerial view of the treatment plant is shown in Figure 28. The original
plant construction was done in 1929 and at that time the plant served
only that portion of the Cedar Rapids population served by sanitary sewers.
Now the plant serves all of the metropolitan area, including Cedar Rapids,
Parts of the Town of Hiawatha, and numerous suburban residential areas.
Primary and secondary biological treatment and separate sludge digestion
was completed in 1933, with chemical-mechanical separate industrial
waste treatment also constructed at that time. Expansion of all phases
of the treatment processes, along with the addition of intermediate stage
trickling filters was made in 1955. Separate sludge thickening and super-
natant oxidation facilities were completed in 1965. A second major plant
expansion, whereby the plant capacity was doubled to the present cap-
acity, was completed in 1966-67.
The entire facility, except the final clarifiers, is covered with thin-shell
domes and the exhaust air ozonated for atmospheric control. Over 8 ac.
(378,500 sq.ft.) of process units are under cover to assure high effi-
ciency during cold weather and to control the potential source of odors.
Five separate air handling facilities evacuate the air from under the
domes at a rate of over 120,000 cfm and deliver this air to a detention
chamber after ozonation. The ozone installation represents the largest
installation of its kind on waste water treatment with a capacity of
150 pounds per day. The largest domes are 167 feet in diameter.
69
-------
Figure 28
The plant processes are:
Primary
Secondary
Screen chambers
Raw lift pumping
Grit removal
Preaeration
Primary clarification
Plastic media roughing filters (first stage)
Intermediate clarifiers
Final rock filters (second stage)
Final filters
Separate sludge digestion, raw sludge thickening, supernatant oxidation,
digested sludge dewatering and incineration.
Flow
28.6 MGD average
64.4 MGD maximum
70
-------
Organic Raw BOD 135,000 pounds per day
Effluent BOD 8,500 pounds per day
Population Approximately 113,000 persons
Equivalent BOD approximately 800,000 persons
The digested sludge dewatering and incineration facility was completed
in 1971. It consists of parallel trains of sludge dewatering facilities
followed by multiple hearth sludge cake incineration. The two sludge
dewatering trains consist of 50 foot diameter sludge holding tanks,
sludge pumping and metering, ash feeders, sludge-ash mix tank, sludge-
ash contact tank, conditioned sludge feed pumping, pressure filter,
filtrate removal, cake storage bunker, followed by a multiple hearth
sludge incinerator.
General Description of the PressjjreFUter Process
Digested solids are periodically pumped from the secondary digesters to
the solids holding tanks. Bottom solids from the sludge holding tanks are
removed by pump on a demand basis and sent through preconditioning
facilities to the pressure filters where the sludge is dewatered.
The sludge preconditioning facilities consist of sludge grinders; mix tanks
where ash and/or chemicals are added as a filter aid; contact tanks where
slow mix of filter aids and flocculation occur; and variable rate sludge
pumps to feed the filter.
The pressure filters consist of a series of plates covered with a nylon, or
similar, filter cloth. Sludge is pumped through the filter leaving a solids
deposition on the filter cloth. This solids formation/ or filter cake, is
periodically discharged at the end of the filter cycle. The filter cycle
may be 1 to 2 hours depending upon numerous variables.
Cake discharged from pressure filter is broken and then fed to the multiple
hearth furnace for drying or complete incineration. Filtrate removed from
the solids in the pressure filter is discharged back through the plant for
further treatment.
The multiple hearth furnace may be operated as a cake drier, or as an
incinerator. As a drier, the cake will be removed from the bottom and
used as a soil conditioner at about 12 percent moisture. As an incinerator
the combustible materials in the cake will be destroyed and the ash remain-
ing will be disposed of as a low grade soil conditioner. The volume of
71
-------
incinerated ash will be a small fraction of the volume of dried cake. The
complete process is observed from a panelboard, and operating records
established. The basis of design is:
Total solids to digester-lbs/day 92,000
Total solids from digester-lbs/day 56,000
Volatile solids from digester-lbs/day 26,400
Percent solids in sludge 5.5
Gallons sludge withdrawn per day 122,000
Period operation-hrs/week 112
Ash to sludge ratio 1.5 to 1
Filtration time per cycle-hours 1.5
Cake moisture - percent 52
Yield filter cake-lbs/sq.ft./hr 0.75
Cake production with fly ash:
Sludge-lbs/day 56,000
Fly ash-lbs/day 84,000
Total solids-lbs/day 140,000
Moisture 52%-lbs/day 152,000
Total cake-lbs/day 292,000
Operating schedule-hrs/week 112 (plus startup and
shutdown)
Incinerator feed rate-Ibs/hr 18,250
Pressure Filtration Technology
It is not the purpose of this report to expound upon the theory of filtra-
tion, or specifically pressure filtration. There are volumes of literature
published on the theory and practice of sewage sludge filtration, most
all of which relates to vacuum filtration, with an occasional comment on
pressure filters or variable volume filters. In this country there is a Very
limited reference to pressure filtration theory and practice. This project
was researched and developed by the use of pilot scale filters, starting
with bench work and progressing to a small sized, production model
pressure filter pilot plant as previously described. At that time there
was no readily available analytical data in this country on the operation
of the pressure filter. However, if there had been other detailed per-
formance data available, the pilot plant approach for the development
of the dewatering facility would still have been the recommended
approach.
Pressure filter knowledge was not sufficient to provide information on
which to predict or calculate performance. It is still a fact that the
72
-------
state of present knowledge is very limited and it is necessary to examine
each material in the laboratory and, if possible on the pressure filter
pilot plant, to determine its filtration characteristics, to obtain some
indication of filter performance, and to provide basis to calculate a full
scale plant design. A brief review of the pressure filtration process may
be of interest. The object of any sludge dewatering process is to recover
the solid content in a concentrated form suitable for disposal or further
processing such as incineration, and to dispose of a filtrate relatively
free from suspended solids. There are different kinds of cake filters,
however, they can be conveniently grouped under four headings.
1. Vacuum filters, batch or continuous.
2. Pressure vessel filters, vessels operated
under pressure containing elements upon
which the cake is formed.
3. Pressure filters, filter presses, plate and
frame, or recessed plate.
4. Variable volume filters, variable chamber,
tube press.
The pressure filter which is the subject of this report is item three (3).
It may have been described by numerous names, such as plate and frame
press, or filter press. Literature has reported that about 1900, a number
of plants in England installed sludge filter presses, principally those
plants using chemically precipitation type of treatment processes. Plants
were also reported at Providence, R.I. and Worcester, Mass, in the
United States. These units were reported to consist of a number of cast
iron recessed plates fitted in a frame between two head blocks which
were forced together by screw jacks. The plates were 36" in diameter
with corrugated faces for drainage channels and covered with filter cloth.
Sludge was fed to these cloth covered recesses (chambers) through a 6"
center opening by pump pressure or through an ejector using compressed
air. Filtrate passing through the cloth was drained from individual spigots
from each chamber.
Houston, Texas installed two 120 plate presses in 19X9 to dewater acti-
vated sludge which operated at a maximum 100 pound pressure. Operation
Was discontinued after a short period of time due to cloth problems and
high costs.
73
-------
Literature indicates that the early pressure filter installations were not
very successful due to premature blinding of the media, excessive leak-
age contributing to poor housekeeping, low capacities, high labor re-
quirements to remove the cake, long filtration periods, and poorly fabri-
cated plates which frequently failed. European interests stimulated in the
late 1950's has lead to the present more efficient pressure filter. This
has been followed in the past few years with interest in the United States
for further development and improvements. An example of more recent
improvements is the design to provide a separate plate gasket to assure
longer media life and less leakage. The earlier pressure filters, and
some of the present units, use a filter cloth sack which covered the entire
plate and also served as the gasket on tho joints of the plate. The filter
media life was often controlled by the mechanical wear, or pressure wear
as the plates were held together. Most often the cloth failed as a gasket
long before it deteriorated as a filtering media. Separate plate gaskets
control leakage if properly installed and adjusted and minimizes house-
keeping. This gasket arrangement also greatly extends the useful life of
the filter media. Details of a typical pressure filter plate are shown in a
cross-section view in Figure 29.
There are a number of other forms of solids-liquid separation systems,
some of which have a wide range of application, and some of which have
often been used in domestic and industrial sludge dewatering. In the
case of waste treatment sludges where the material contains a high pro-
portion of fine particles, the separation of the solids-liquid phase is
extremely difficult, and the force required to displace the liquid phase
and replace it with sludge particles to produce a cake formation increases.
In this application the use of applied pressure as in the use of the pres-
sure filter to develop sufficient force to cause the separation of solids-
liquid offers a distinct advantage. The proper sludge preconditioning,
selection of an efficient filter media, controlled pressure development
will improve the efficiency of sludge dewatering beyond that normally
found in other systems. The product of pressure filtration is an increase
in solids concentration in the cake with a decrease in suspended solids
content of the filtrate. Compared to other systems for sludge dewatering
the advantages for pressure filter may be summarized as follows.
1. The quantities of sludge conditioners are usually reduced.
Conditioning materials may be of the least expensive variety (ash)
as was the case at Cedar Rapids.
74
-------
UrmUer«d Sludge
Sludge Cake
J Filtrate
RHClH >i I View
,MUj !: ' .
i)\ pl.M'1
Figure 29
75
-------
2. Filtration efficiency is maintained even when there is a sub-
stantial range of sludge characterises, and is improved over most
systems at normal filterability conditions.
3. Increased solids content of the filter cake. The low moisture
content is important where cake is incinerated and often will assure
autocombustion of the incinerator feed, thus saving supplemental fuel
cost.
4. The filtrate is considerably lower in suspended solids and
BOD, and thus minimizes the loadings back to the plant for treatment,
or other problems of ultimate disposal.
5. Operation and manpower requirements are minimal, and con-
siderably less than most systems. The attending operators may perform
other duties during most of the period of filtration. This is not true of
several other systems. Operators not skilled in mechanics can easily
be trained to operate this filtration process.
6. Maintenance is minimal due to relatively few moving parts and
standard supporting equipment such as pumps and mix mechanisms.
7. Filter cakes are more easily disposed of due to lower moisture
content and smaller volumes. The cake may be conveniently stored in
bunkers under the filter and removed by standard conveyor equipment and
at continuous discharge rate thus overcoming any small effect of the
batch process.
8. The costs based upon amortized capital costs are competitive
with other processes. The initial capital costs may be higher than some
other systems, however, the minimum maintenance costs extended over
a significant period of useful life, and the low operating costs make the
pressure filter system highly attractive.
The usual methods for evaluation of the filterability of a sewage sludge
is the Buchner funnel test and the vacuum leaf test in the laboratory.
The data is then applied to the standard filtration equations and established
theory. This conventional laboratory investigation and filtration theory
does not lend itself to pressure filtration because the majority of the
pressure filter designs provided for a fixed volume cake formation. The
cake is of a fixed thickness and occupies a fixed volume between the
plates.
76
-------
Cake formation also takes place under several varying conditions. At the
beginning of the filtration cycle cake is formed at a high rate at relatively
constant flow with the pressure increasing as the cake formation is
developed. This is followed later in the cycle with a decreasing flow rate
to a very small flow rate at a constant pressure drop. There are other
variances from the standard filtration process which encourages the inves-
tigation using a pilot filter study to establish design criteria.
The present day pressure filter equipment offers many advantages and an
efficient and economical system for dewatering sewage sludge. With a
minimum of investigative study and pilot plant work, meaningful data can
be obtained to size and evaluate a full scale pressure filter dewatering
system.
Detailed Description of the Process
The process schematic and photographs of the facility are shown on Figure
30. Digested solids are removed daily from one or more of the secondary
digesters and alternately transferred to one of the two 50 foot diameter
sludge holding tanks. Each tank has about 1\ days' detention and sludge
thickening mechanism. There was no previous experience that digested
sludge thickening could be assured at this plant, however, bottom sludge
removal mechanism was necessary so gravity sludge thickening with the
picket design was installed in anticipation of some degree of sludge
thickening. These holding tanks may be overflowed to a scum well for
scum removal and may be decanted if thickening occurs. The scum may
be preconditioned with chemicals and sent through dewatering, or it may
be pumped back through the plant for further treatment.
Sludge from the bottom of the holding tank is pumped to the mix tank
which is the beginning of the dewatering process. In the pump discharge
line is a nuclear density gauge to measure and record the sludge density
in terms of percent total solids. Also included in this line is a magnetic
flow meter to register the volume of liquid flow pumped to the mix tank.
With these two instruments a total accounting of the sludge pumped to the
process can be accurately determined at all times. Both the nuclear den-
sity gauge and the magnetic flow meter are registered on the instrument
panel in the operating room. Installed in the pump discharge line to the mix
tank is a sludge grinder. The purpose of this grinder is to emulsify the
sewage sludge solids to a uniform consistency and thus improve precon-
ditioning and filterability of the sludge liquid applied to the pressure
filter. The sludge grinders have replaceable cutting plates and blade
assemblies and may be externally cleaned. Periodic cleaning may be
necessary and may be detected by observing the pressure drop across
77
-------
SCHEMATIC SLUDGE DEWftTERING 8 INCINERATION
Control Center
Pressure Filter
Figure 30
78
-------
the grinders. Bypass valving and piping has been provided to bypass the
nuclear density gauge, the sludge grinders and the magnetic flow meters,
should the units be taken out of service for repair.
The mix tank is designed to blend the incoming sewage sludge with the
filter aid which may consist of incinerated sewage sludge ash, fly ash
from a local power plant, chemicals, or any combination of these. The'
mix tank is a horizontal steel tank, approximately 120 cu. ft. capacity,
equipped with two speed agitators designed to thoroughly mix the sludge
slurry and the ash and chemical admixture. Ash is fed to the mix tank by
screw conveyor and is accurately measured by gravimetric feeders. Chem-
icals are added to the mix tank from a metered ferric chloride feed pump
and from a lime slaker using a magnetic flow meter for control. The ad-
mixtures are controlled from the instrument control panel at the operators'
room. Ash dust is controlled with spray showers in the mix tank and the
operator may observe the slurry mixing process through a plexiglass
observation door in the top of the tank.
Slurry overflow from the mix tank is gravity fed into the contact tank
where it is further mixed at a very slow rate to provide ample contact
between the sludge and the admixtures of ash and/or chemicals. This
tank is also of horizontal steel construction with a slow speed agitator
to maintain uniform consistency of the slurry. The tank has a large volume
of about 600 cubic feet to assure an adequate supply of slurry feed to the
pressure filter during each cycle of the filter operation. The contact tank
is equipped with liquid level controls for starting and stopping sludge
pumping and the mixing system. When the liquid level in the contact tank
fails to a predetermined low level, the mix tank operation will start and
a supply of slurry will be fed into the contact tank. As the pressure
filter feed is satisfied, the liquid level in the contact tank will rise to
a predetermined high level and shut the mix tank operation off. Slurry
from the contact tank is pumped to the pressure filter by two filter feed
pumps. These pumps are of a special design and are variable speed,
variable capacity, variable pressure and are automatically controlled to
decrease the output capacity as the pumping head increases until it
reaches a maximum stall pressure of approximately 225 pounds per square
inch, which is the normal high operating pressure of the pressure filter.
This pump is hydraulically driven from a power pack which furnishes the
hydraulic pressure necessary to stroke the piston.
The filter feed pump discharges to a surge tank to smooth the flow and to
minimize pulsating pressures in the feed line to the pressure filter.
79
-------
The pressure filter operates on a process cycle of about one to two hours
duration depending on the density of the sewage sludge and the dewater-
ing characteristics of the sludge slurry fed to the pressure filter. Prior
to the beginning of the filter cycle the pressure filter is precoated to pro-
tect the filter cloth from blinding or clogging due to possible grease
content or fine particles, and to form a shear plane between the cloth and
cake to assure free and clean discharge of the cake. Filter precoat is a
mixture of ash carried to the filter cloth in recycled filtrate water which
is stored during each filter cycle for this purpose. Ash is charged into the
precoat tank by a gravimetric feeder and conveyor system. All of these
operations are sequenced automatically from the operator's control station
by selecting the filtration cycle button.
Immediately after precoat the sludge feed to the pressure filter is auto-
matically applied through motorized control valves to begin the filtration
cycle, Sludge feed to the filter is from both the front and the rear ends
and the rate of feed at the beginning of the cycle is very high and dimin-
ishes as the cake is formed. To accommodate this high initial rate of
feed at the beginning of the filtration cycle, an equalization tank which
was charged with conditioned sludge during the previous filtration cycle,
and two filter feed pumps will discharge into the pressure filter. As the
filtration cycle continues and the filter feed rate decreases, first the
equalization will drop out of the system, and later one of the feed pumps
will trip out of service. The second filter feed pump will continue to
supply the feed to the filter at a decreasing rate until the filtration cycle
has been completed.
The duration of the filtration cycle may be determined by time, pressure,
or rate of filtrate flow. When the filtration cycle is determined by time,
this may be one to three hours duration based upon previous filtration
experience. It is not anticipated that the filtration cycle will be governed
by pressure, however, should this be the choice, the filtration cycle
would continue until a predetermined pressure level was developed, at
which time filter feed would cease or would continue for a predetermined
number of additional minutes.
Usual operation will be controlled by rate of filtrate wherein the filtration
continues until a predetermined rate of filtrate flow is observed across
a V-notch weir. An audio or visual signal is operated by a liquid level
sensor at the weir.
At the end of'the filtration cycle, the filter core feeding the individual
filter plates remains filled with wet, soft sludge slurry. This soft core
must be removed prior to opening the filter plates to discharge the cake.
80
-------
This is accomplished by applying compressed air to the core at one end
of the filter thus blowing the wet sludge core back to the contact tank
through an energy dissipating cyclone. The energy dissipating cyclone
is called a core expansion tank which discharges the wet slurry to the
contact tank and vents the air to the atmosphere. At the completion of
the core blow operation, the filter plates may be opened at any time con-
venient to the operator. Core blow and opening the filter plates for cake
discharge is controlled from the hydraulic control console located near the
filter. When the filter is opened each plate is progressively moved by a
mechanical plate shifter located at the top of the filter and which con-
tinues to operate automatically unless stopped by the operator. The filter
cake is discharged to the bunker below the pressure filter and is sheared
into smaller particles as it passes shear cables stretched across the open-
ing below the filter.
It is desirable that the operator be in attendance during the time the filter
cake is being discharged. The operator should be certain that the cake
has dropped free of the filter plates and should observe the condition of
the filter cloth media, the plate gaskets, and other physical features of
the filter. Should any large cake fragments hang up on the filter media
the operator should remove these using a wooden paddle or similar
approved device. The total time required to discharge the filter cake
and recycle the process is in the range of 5 to 10 minutes. This is the
time the operator spends at the filter.
Filter cake is discharged to a cake bunker under the filter and is conveyed
to the multiple hearth incinerator by a series of cake conveyors and ele-
vators. The cake bunker conveyor has variable speed control which serves
as a cake feed rate control for the incinerator. All other conveyors moving
cake to the incinerator are constant speed. If the incinerator is not in
operation and it is still desirable to dewater sludge, an alternate discharge
is provided to a truck for conveyance of wet sludge cake to a suitable
disposal area. Conveyors are controlled from the operating console and
are interlocked for proper sequence of operation.
The incinerator is a multiple hearth type, having nine hearths of 25 ft.,
10 inch O.D. Temperature is maintained with automatic temperature con-
trols by self combustion of the sludge cake and by auxiliary fuel fired
burners using either digester or natural gas. Incinerated ash is discharged
through a grinder to elevated ash storage hoppers. Combustion gases are
removed through a high-energy wet Venturi scrubber operating at about
40 in. W.C. The system also has a deodorizing afterburner system. The
incinerator is totally automatic and monitored at the operator's control
Panel.
81
-------
Two (2) pressure filters are installed, each having 83 chambers of
approximately 2.5 cubic feet volume per chamber, for a total of 207
cubic feet volume. The filters have 3400 square feet of filter area
each and are equipped with an expanded frame for a future increase to
approximately 100 chambers. The filter plates are carbon steel con-
struction, epoxy coated. The filter media is a monofilament polypropylene
with a stainless steel mesh backup screen and a carbon steer drainage
screen. The filter operates at approximately 225 psig filtration pressure.
The plates, during operation, are held closed by hydraulic cylinders
operating with an oil pressure at approximately 5,000 psig. The pressure
filter plates are gasketed with asbestos Teflon gasket material and the
filter media is caulked in place with cotton caulking cord.
82
-------
SECTION VIII
EVALUATION METHODS
Sampling
There were five basic sampling points for the evaluation of a pressure
filter:
1. Sludge feed
2. Ash feed
3. Sludge-ash mixture
4. Filter cake
5. Filter filtrate
Sludge Feed
A sludge feed grab sample was collected by the operator from the discharge
of the transfer pumps twice per operating cycle. The operator determined
the sludge solids content of each sample using an Ohaus moisture bal-
ance in order to adjust the ash feed. The remainder of the sample was
composited equally with the other sample for that cycle, refrigerated,
and sent to the laboratory for analysis.
Ash Feed
Ash feed was analyzed on an infrequent basis. Samples were obtained
from the ash feeders. A device was fabricated to collect the ash from
a predetermined area of the gravimetric feeder belt. This ash sample
was weighed and the quantity checked against the calibrated feed rate
for the feeder belt. The device cut an ash sample one foot long in size
and functioned on the principle of an ordinary cookie cutter.
Sludge-Ash Mixture
Sludge-ash mixture was sampled at the intake of the filter feed pumps.
These samples were handled in the same manner as the sludge feed.
Solids analysis on these by the operator confirmed the ash feeder settings.
(ash/sludge ratio).
Ash/Sludge Ratio = % Sludge Mix -1
% Sludge Feed
83
-------
Example:
Sludge = 5% Ash + Sludge = 17%
Ash/Sludge Ratio = 17% -1=3.4-1 = 2.4
5%
These results were later confirmed by the laboratory.
Filter Cake
The filter cake was sampled by grabbing representative pieces of cake
when the cake was discharged and broken on the shear cables at the
top of the cake storage bunker. These samples were sealed in glass
jars.
Filtrate
Filtrate samples were collected from the weir box adjacent to each filter.
Three grab samples of filtrate were taken during each cycle, composited
and refrigerated for laboratory analysis.
Plant Measurements
All important measurements, such as pressures, ash weight, sludge vol-
ume, sludge density, time, etc. were monitored continuously by appropriate
instrumentation. All measurements were recorded and integrated in the
operators' control room.
Sludge Flow
Sludge flow was monitored by integrating magnetic flow meters.
Sludge Density
The percent of total solids of the sludge feed was monitored by nuclear
density meters. The output (0-16 percent solids) was displayed on the
chart as the sludge flow.
Lime Feed
Gallons of lime feed were measured by integrating magnetic flow meters.
The concentration of lime was maintained at 0.4 pound CaO per gallon
of water by appropriate adjustment of feed water rotameters and lime feed
transmission at the slaker.
84
-------
Therefore, pounds of lime feed equals gallons (from integration) times
0.4 pounds per gallon.
Ferric Chloride
Ferric chloride feed was by positive displacement feed pumps with cali-
brated discharge integrated from time of operation.
A 40 percent ferric chloride solution was used. Therefore, pounds of
ferric chloride equals gallons solution times 4.75 pounds ferric chloride
per gallon.
Ash Feed (Precoat and Admixture)
Ash quantities were metered by use of gravimetric belt feeder equipped
with totalizers.
Filtrate
Total filtrate from the plant was measured by a 60° V-notch weir on the
discharge of the filtrate storage tank. The depth over the crest was
monitored by a bubble tube sensor.
Instrument Maintenance
All instrumentation was routinely calibrated by a qualified instrument
technician.
Gravimetric feeders were calibrated monthly (minimum) by either weighing
a timed discharge (Lime Feeders) or by removing material from a measured
area of the weigh belt and checking against theoretical discharge (Ash
Feeders).
Laboratory Analysis
The following routine laboratory analysis was performed.
Sludge Feed
Analysis Method
Total Solids Dried at 104° C. overnight
Volatile Solids Ignition of dried residue at 550° C.
85
-------
PH
Grease
Volatile Acids
Alkalinity
Ammonia
Organic Nitrogen
Sludge +Ash (Mixture)
Total Solids
Volatile Solids
PH
Filter Cake
Total Solids
Volatile Solids
Nitrogen
Filtra te
PH
Total Solids
Volatile Solids
Turbidity
Glass electrode
Dried using magnesium sulfate extracted
with hexane in a Soxhlet extractor
Determined by column chromotography
according to the 13th Edition of Standard
Methods
Titration of the sample to pH of 4.5
using 0.5N H2SO4
Analyzed by distillation into boric acid
and titration with standard acid
Kjeldahl method using mercuric oxide
as catalyst
Same as above
Same as above
Same as above
Same as above
Same as above
Same as above
Same as above
Same as above
Same as above
Determined using a Hach Model 2100A
turbidimeter
86
-------
BOD,
Total Alkalinity
Color
Conductivity
COD
Total Phosphate
Dilution method (three dilutions per
sample) initial and final dissolved
oxygen concentrations measured with
a Weston and Stack Model 350 dissolved
oxygen meter with stirrer
Titrated with 0.03N H2SO4 to a pH of
4.5
Determined using a Hach color compar-
itor
Measured using a Labline Lectromho
conductivity bridge with a cell constant
of 1.0
Measured basically the same as Standard
Methods and except samples were auto-
claved for two hours at a pressure of
15 psig and titrated with 0.25N FeSO4
Determined according to Standard Meth-
ods. Potassium persulfate was used as
the oxidant. Stannous chloride and
ammonium molbdate were used for color
development
Determination of Specific Resistance
The filterability of a sludge can be defined as the ease at which the sludge
9ives up water. While it is possible to talk about the filterability of
sludge using subjective terms such as difficult, average, etc., it is
desirable to be capable of objectively describing a numerical value to a
sludge to give us a meaningful value for operation.
method of evaluating a sludge is by determining its specific resist-
ance. Specific resistance can be viewed as the reciprocal of filterability
Wiere a high value of resistance means a poor value for filtration and
vice versa. The theory and development of specific resistance and its
relationship to pressure dewatering have been developed by Dr. Oswald
of the Passavant Werk, Germany. This background material may
87
-------
be found in the dissertation "Entwicklung Eines Wirtschaftlichen Verfahrens
Fur Die Beseitigung Von Abwasserschlammen" by Dr. Oswald Busse.
Specific resistance has proved valuable at Cedar Rapids in day to day
operation. By using the specific resistance an operator can:
1. Determine whether a sludge is conditioned properly
2. Evaluate and optimize new methods of conditioning such
as organic polymers
3. Evaluate the filterability of the unconditioned sludge
All these evaluations can be accomplished using a bench scale filter.
Specific resistance is measured by using a small pressure filter which is
generally referred to as a resistance meter.
Mechanical Description of the Resistance Meter
The resistance meter is essentially a pressurized Buchner funnel. The
filter is composed of a stainless steel body and support screen for a
standardized filter paper that serves as the filter media. Meter acces-
sories required are a 250 psig compressed air supply, a 100 milliliter
buret and stand, stop watch, and convenient glassware as indicated in
the photograph, Figure 31. A functional diagram of the resistance meter
is shown in Figure 32.
A standardized, or calibrated, filter paper is necessary to assure a uniform
filter media . This is wetted and placed in the meter on the support screen.
A 100 milliliter sample of the sludge to be tested is measured and placed
in the resistance meter. The top of the unit is secured and 250 psig pres-
sure is applied. The effluent from the meter then drains into the 100 mil-
liliter buret, which was initially filled to the lower 100 ml mark before
the test. As the filtrate is collected in the buret, readings of filtrate
quantity and time are noted and recorded.
In the case of an unconditioned sludge which has a high resistance to
filtration the total filtrate is read after two minutes and at one minute
intervals thereafter for a period of twenty minutes, or until the filtration
has ceased and air blow-by occurs with air passing through the filter.
88
-------
r
RESISTIVITY
MT IEH
Y
Figure 31
PRESSURE
RELEASE VALVE
100ml
BURETTE
o
o
BEAKER
it
REGULATOR
COMPRESSED
RESISTIVITY METER SCHEMATIC
Figure 32
89
-------
In testing a condtioned sludge which yields water quite readily if the
filtrate volume reaches the 75 ml mark (25 ml filtrate) within one minute,
filtrate volume readings are then taken every 10 seconds. These readings
are taken until blow-by occurs or an elapsed time of 200 seconds passes.
Following are two examples of resistance meter tests, one with an uncon-
ditioned sludge and another with a conditioned sludge.
Unconditioned Sludge pH = 7.4 % Solids = 6.0 T = 28.0°C.
Filtration Time
Minutes Seconds
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
0
60
120
180
240
300
360
420
480
540
600
660
720
780
840
900
960
1020
1080
1140
1200
Volume
Milliliters
0
1.3
2.3
3.3
4.1
4.9
5.6
6.2
6.8
7.4
8.0
8.5
9.0
9.5
10.0
10.4
10.9
11.4
11.8
12.2
12.5
Seconds/
Milliliter
46.2
52.2
54.5
58.5
61.2
,3
7
64,
67,
70.6
73.0
75.0
77.6
80.0
82.1
84.0
86
88
89
91
93.4
96.0
90
-------
Conditioned Sludge pH =9.5 A/S =1.5 CaO =0.5%
Ferric Chloride = 0.5% % Solids = 6.0%
T = 29.0°C. '
Filtration Time Volume Seconds/
Seconds Milliliters Milliliter
0 0
10 8.6 1.2
20 16.5 1.2
30 22.6 1.3
40 27.7 1.4
50 32.4 1.5
60 36.6 1.6
70 40.4 1.7
80 Blow-by
Analysis of Data
The equation for specific filter resistance is
r = 2 x p x Fj x b
n x c
Where r = specific resistance in
p- differential pressure2 in dynes/cm2
Fi = filtration area in cm 2
b = slope of line obtained in test by plotting seconds/milliliter
versus milliliters
n - filtrate viscosity in poises
c = grams of dry solids/grams liquid for 1 gram sludge
In practice the above equation reduces to
r = k x b/c
Where k = 2 x p x Fi2
n
When the pressure and filtration area are held constant, k then becomes
^ function variable only with temperature which governs the filtrate
viscosity.
91
-------
Therefore the constant k can be obtained from a table of viscosities of
water versus temperature.
The remaining factors b and c are obtained by determining the density of
sludge and the slope of the line obtained in the test by plotting seconds
per cc versus cc.
In the case of an unconditioned sludge, Figure 33 shows a graphical
solution of the slope b of 4.4. For this sludge the percent solids was
6.0, c = 0.06/0.94 = 0.064, temperature = 28.0° C. for a k of 5.315,
therefore, r equals k x b/c = 5.315 x 4.4/0.064 = 366 x 1012cm~2.
Also in the case of the conditioned sludge as shown in Figure 33,
b = 0.020, k again is 5.315 and c = 0.064. Therefore, r = 5.315 x
0.020/0.064 = 1.66 x 1012cm-2.
Interpretation of Specific Resistance
In the two previous examples the specific resistance of an unconditioned
sludge 366 x 10l2cm~2, and a conditioned sludge 1.66x 1012cm~2 have
been determined. This data becomes meaningful when it becomes known
what value of specific resistance a sludge must have to filter well.
Generally speaking, at Cedar Rapids a conditioned sludge must have a
specific resistance of less than 20 x 10l2cm~2 and preferably less than
10 x 10^2cm"2 to filter well. Knowing this limit we can optimize chemi-
cal or ash dosages using the bench model pressure filter much more
easily and rapidly than a trial and error method on the full scale filter.
92
-------
lOOi-
84, 10
60,4.6
UNCONDITIONED SLUDGE
SLOPE « * 4,4
366xlOIZcirfz
30
I I
v-
2.0
CONDITIONED SLUDGE
SLOPE « . 0.020
1.7, 40.4
"r" « 1,66 x I012 cm2
1.0
I I
20 30 40 50
Volume (cm3)
Figure 33
93
-------
SECTION IX
RESULTS
Sludge Characteristics
The source of digested sludges for dewatering was eight heated digesters
operated in parallel. Each digester has a capacity of approximately
600,000 gallons. Six digesters have some sort of gas mixing while two
gas holders have none. Hydraulic retention time of the digesters is
approximately 12 to 16 days and the organic loading is approximately
0.08 to 0.10 pounds volatile solids per 1000 cubic feet.
Sludge feed to the digesters consists of raw sludge from primary clarifiers
and secondary sludge from plastic media roughing filters and rock final
filters.
Digested sludge is pumped more or less continuously to sludge holding
tanks at the dewatering site. These holding tanks tend to stabilize the
input to the dewatering process and provide some thickening.
Table II shows the average characteristics of the underflow of the sludge
holding tanks to the process for the months of February through August
1972. The table shows that the solids concentration of the digested
sludge was reasonably stable except for the months of June and July
where large amounts of waste lime sludge from the water purification
Plant were received. During this period, the solids content increased
to 7 percent and the organic content expressed as percent volatility
decreased from 45 to 53 percent to 33 percent volatile.
Other parameters of digestion remained relatively constant. An increase
in volatile acids was experienced in April due to a temperature recovery
in the digesters after an extremely cold winter.
Filterability of Sludge
Digested sludge at Cedar Rapids is unusually difficult to dewater. As
mentioned previously, European experience has indicated that a sludge
With a specific resistance value of 150 x 10l2cm-2 is an average sludge
as far as the "degree of difficulty" in dewatering. Values of specific
resistance for digested sludge at Cedar Rapids are all considerably
higher than this value, typically possessing specific resistance values
of 300 to 700 x 10l2cm~2. Higher resistance values indicate greater
difficulty in, or resistance to, dewatering.
95
-------
PH
Table II
Average Characteristics Unconditioned Sludge
February to August 1972
Feb. Mar. Apr. May June July Aug. Av.
7.1 7.2 7.2 7.3 7.3 7.4 7.5 7.28
Total Solids 3.6 3.5 3.6
Percent
3.5 7.0 6.0 5.0 4.6
Volatile
Solids
Percent
53.8 51.9 53.2 45.0 33.6 36.9 44.7 45.6
Volatile Acids 49
mg/1 as
Acetic Acid
41 410
62
50
77 73 108
Total Alka-
linity mg/1
as CaCO3
Ammonia
mg/1 as N
Organic
Nitrogen
as N
Grease- %
of Total
Solids
4000 2900 2700 3600 3850 3800 4050 3557
846 594 604
838 944 993 1049 838
1464 1411 1338 1123 1272 1424 1493 1360
9.32
5.1 4.0 4.6 6.6
96
-------
Complicating the picture is the fact that sludges at Cedar Rapids are not
consistent. Digested sludge is always difficult but can be more or less
difficult at various times.
One of the problems arises from the fact that the sludge varies consider-
ably in the solids content. Because the value for specific resistance is
inversely proportional to solids content, sludges with low solids content
are generally more difficult to dewater than the same sludge which has
been thickened. For example, if a 2 percent sludge, which has an
r= 600 x 1 0*2 cm"2 were thickened to 6 percent it would then have an r
value of less than 200 x lO^cm"2 providing, of course, that the basic
nature of the sludge was not modified.
The basic pattern could be observed at Cedar Rapids as indicated in Table III.
Table III
Specific Resistance at 10
Percent Solids r (Specific Resistance) x
2.43 772
3.33 571
5.34 342
6.68 236
As can be noted by the above table, the specific resistance is inversely
proportional to the solids content. Unfortunately, however, in practice
a 2 percent sludge cannot be handled as a mere volumetric dilution of a
4 to 6 percent sludge. Generally speaking, low solids sludge (2-3 per-
cent) appears to contain a higher proportion of fines and requires a greater
proportional amount of conditioning chemicals than a higher solids content
sludge of 5.5 percent or greater.
Also it has been found that specific resistance of sludge may vary con-
siderably at Cedar Rapids even though the percent solids remain relatively
constant as shown in the following continuation of Table III.
Percent Solids r (Specific Resistance) x 10I2cnT2
5.8 366
5.3 342
5.5 308
5.7 333
5.6 443
6.0 361
97
-------
In the above table it can be seen that for the samples shown, while the
percent solids are relatively uniform the specific resistance varied
appreciably. For example, a 5.5 percent sludge had an r = 308 x 1012cm"2
while a 5.6 percent sludge had an r = 443 x lO^cm"2, a variance of 44
percent.
This variance in filterability is not unexpected considering the widely
varying nature of the wastes as far as quality and quantity received at
Cedar Rapids.
The most important ramifications of this varying nature of the sludge is
that conditioning must
1. Be dependent on the solids content of the sludge, i.e., quantities
(ash and chemicals) must vary as the percent solids vary.
2. Conditioning must be adjusted to the worst possible condition.
Because the conditioning charts are based on conditioning the'worst
possible sludge/ a large quantity of sludge is often over-conditioned.
In the data that will be presented later in this report, there will be in-
stances where two sludges with the same solids content and essentially
identical conditioning will have widely varied yields. The problem can
be traced to the varying filterability as shown by specific resistance
value in the table above.
Sludge Ash Characteristics
The source of sludge ash for dewatering was the ash from the multiple
hearth incinerator used for the incineration of the cake from the pressure
filter. Initial startup of the pressure filters was made with fly ash from
a local power plant, consequently the initial characteristics of the sludge
ash probably were a composite of fly ash and sludge ash, however, with
the continued wasting of ash the sludge ash characteristics stabilized
within a short period.
The physical and chemical properties of sludge ash are in all probability
quite unique from installation to installation, depending upon the non-
volatile portions of the sludge dewatered, chemical additives (lime,
ferric chloride and/or polymers), temperature of combustion and the method
of combustion (multiple hearth or fluidized bed).
98
-------
At Cedar Rapids the sludge has at times a high inorganic fraction repre-
senting the discharge of waste lime from the lime soda softening process
at the municipal water purification plant. No alum is used or discharged
from the water purification plant.
Chemical additives in addition to sludge ash used for conditioning sludge
are ferric chloride and lime.
The temperature of combustion ranges from 1200° F. to 1400° F.
Physical Properties of Sludge Ash
The physical properties of sludge ash are as follows:
Specific Gravity 2.63 - 2.78
Bulk Density 50 Ibs. per cu. ft.
Color Yellow
Mean Particle Size 29 mm (22 - 40}
The particle size was determined by Process Measurements Section,
Research Laboratory Branch, Control System Division of the Environmental
Protection Agency with the cooperation of the Taft Center, Cincinnati.
Particle size analyses were made using two different techniques, one
being the electronic sensing of particle volume with the Coulter Counter
Model T, and the second, aerodynamic sizing with the Bahco Micro-
particle Classifier.
In the following table are listed the average particle sizes of sludge ash
as determined by the Coulter Counter Model T for nine samples of sludge
taken from January 3, 1972 to May 10, 1972.
% of Particles Size in Millimeters x 10~3
less than Given Average of
Size Nine Samples Range mm x 10"3
25
50
75
90
14.5
29.1
50.7
74.3
9.5
22
38
60
- 18
- 40
- 72
- 100
In Figure 34 is shown the results of a typical particle size determination
by the Coulter Counter Model T .
99
-------
PARTICLE SIZE ANALYSIS
N
to
Q
1 01
T. <
o S
UJ H
? W
z
> H-
< en
3 J
5 .
3 ^
O u
o
tr
LU
100
90
80
70
60
5 0
40
30
20
10
w>l_UL/\JL_ I->»JII
COULTER COUNTER
MODEL T OX°
X
O
X
rr
/^
/
O
0/°
X
o^
c^pS
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
3 4567 8910 20 30 4050 100
MICRON DIAMETER ( u )
Figure 34
Particle size analysis performed by the Bahco Microparticle Size Analyzer
are given in Table IV. Along with ash from the Cedar Rapids installation
the table includes data on ashes from Mill Creek/ Lake Tahoe, and
Kansas City.
This table shows that for the incinerator ashes from Mill Creek/ Lake
Tahoe and Kansas City approximately 70 percent of the particles were
28 microns or less, for Cedar Rapids only 47.5 percent of the particles
were less than 28 microns.
100
-------
Table IV
Classification of Ash Particles
by Bahco Microparticle Size Analyzer
Effective .Particle Cedar Rapids Kansas City* Lake Tahoe* MiUCreek*
^Size (10-3 mm) % % % %
0.00 - 0.92 1.49 1.38 2.22
0.92 - 1.60 3.0 2.95 2.71 3.70
1.60 - 2.97 5.0 5.70 7.53 7.81
2.97- 7.71 12.5 9.05 9.83 21.86
7.71-12.19 9.5 14.56 10.95 26.52
12.19-20.75 10.5 22.41 16.20 22.46
20.75-24.72 4.0 9.21 10.13 6.83
24.72-27.30 3.0 8.83 4.28 8.60
28.38 52.5 25.8 36.99
* Sludge Conditioning With Incinerator Ash
Smith, Hathaway, Farrell, Dean, AWTR1
EPA -Cincinnati, Ohio
Chemical Analysis of Sludge Ash
In Table V is shown the chemical analysis of sludge ash samples taken
in January, May and September 1972. The samples are somewhat similar,
however, the September ash sample had more calcium which would indicate
the use of a greater amount of lime and less of the R2O3 group of oxides.
Table V
Chemical Analysis May 24. 1972
Phosphate pentoxide,
Silica, SiO 2
Ferric oxide, Fe2O3
Alumina, A12O3
Titania, TiO2
Lime, CaO
Magnesia, MgO
Sulfur trioxide, SO3
Potassium oxide, K2O
Sodium oxide,
Undetermined
101
-------
Silica Value - 27.77
T250 Poises: *1950° F.
* less Than
Loss on Ignition at 1000° C.
Metals by Atomic Absorption -%
Ca
Fe
Mg
Zn
Cy
Cr
Cd
Na
K
pH of 1% Solution
Fly Ash Characteristics
16.50% wt. dry
January 1972
12.10
3.63
0.70
0.61
0.10
0.04
<0.01
0.27
0.40
9.5
September 1972
20.30
3.09
0.07
0.57
0.09
0.04
0.00
Not determined
Not determined
10.4
Fly ash for the project was obtained from a local power plant. The fly
ash was the residue from combustion of pulverized coal. This residue
consisted of a combination of ash collected by cyclone type dust collec-
tors and electrostatic precipitators. The ash was transported dry from the
power plant to the solidshandling plant by truck and pneumatically con-
veyed into the ash storage bins.
The physical properties of the ash are:
C olor
Sp'^ctfic gravity
Bulk density
Black
2.3
43 .6 Ibs. per cu. ft.
Analysis of the particle size of fly ash was performed in the same manner
as sludge ash using both the Coulter Counter and the Bahco type analysis.
Coulter analysis of the particle size is shown below, and in Figure 35.
Percent Less Than Given Size
25
50
75
90
10
-3
mm
11
20
33
42
102
-------
PARTICLE SIZE ANALYSIS
BANCO ANALYZER
to
N
tn
> I-
80
70
60
50
40
30
2 .
D z
°3 20
CE
10
_L
I
I
I
I
I
I
I
0 2
8 10 12 14 16 18 20 22 24 26 28 30 32
MICRON DIAMETER (u)
Figure 35
Fly ash particle size was also determined by using the Bahco apparatus.
Results are shown in Table VI.
Table VI
Classification of Fly Ash Particles
by Bahco Microparticle Size Analyzer
Effective Particle Size (10"3 mm)
0.00 - 1
1.60 - 2
60
97
2.97 - 7.71
7.71 - 12.19
12.19 - 20.75
20.75 - 24.72
24.72 - 28.30
28.30
Percent
2.5
7.5
20.5
16.5
17.5
5.5
5.0
25
Chemical analysis of fly ash is shown in the following table.
103
-------
TABLE VII
Chemical Analysis of Fly Ash -
Phosphate pentoxide, P2O5 0.85
Silica, SiO2 53.90
Ferric oxide, Fe203 17.00
Alumina, A12O3 11.93
Titania, TiO2 0.57
Lime, CaO 6.80
Magnesia, MgO 1.65
Sulfur trioxide, SO3 2.30
Potassium oxide, K2O 4.32
Sodium oxide, Na2O 0.39
Undetermined 0.29
100.00
Silica Value - 67.93
T250 Poises: 2355° F.
Loss on Ignition at 1000°C. 14.5%
As can be noted from Table VII, fly ash is mainly composed of silica, iron
oxide and alumina.
Comparison of Sludge Ash and Fly Ash as a Conditioning Agent for Sludge
Extensive testing was performed using the resistance meter in comparing
the effects of sludge ash and fly ash as a conditioning aid for the pressure
filtration of sludge, both digested sludge and raw sludge.
The reason for the initial testing was to attempt to explain the apparent
lack of correlation between the pilot plant study and the performance of
the full scale filter. The pilot plant had operated without the use of any
other conditioning agent except fly ash, the full scale plant could not
successfully operate at any reasonable sludge/ash ratio without using
ferric chloride and lime.
Consequently a program of study was set up comparing the two different
ashes. Experimentally the ashes were compared by securing a sample of
digested sludge, determining the solids content of the sludge and then
dosing liter samples with varying amounts of ash, up to an ash ratio of
3 to 4.
104
-------
In Figures 36 to 40 are shown an example of the testing. Figure 36 repre-
sents a sample of digested sludge with a solids content of 2.5 percent.
The initial specific resistance of the sludge was 772, At an ash/sludge
ratio of 1.3 with fly ash the specific resistance was lowered to 78 x 10*2
cm"2 while at an ash/sludge ratio of 3.95 for sludge ash the specific
resistance was 67 x 1012cm~2 comparable values.
Figure 37 represents a sample of digested sludge with a solids content of
5.5 percent and an initial specific resistance of 308 x lO^cm-Z. in this
case an ash/sludge ratio of 1.5 for fly ash gave a specific resistance of
12.3 x lO^cm"2 while an ash/sludge ratio of 4.0 for sludge ash gave a
specific resistance of 14.1 x lO^cm"2, again essentially the same values.
Figure 38 represents a sample of digested sludge with a solids content of
5.6 percent and a specific resistance of 443 x 10^2cm~2. Here a value
of 1.0 ash/sludge ratio for fly ash has a specific resistance of 63.2 x
179 199
lO-^cm , comparable to a value of 61.0 x lO^cm"^ for an ash/sludge
ratio of 2.5 for sludge ash.
The above data indicates quite strongly that fly ash is 2.5 to 3.0 times
more effective for conditioning digested sludge per unit weight of ash
than sludge ash.
Additional work was performed in comparing the conditioning effect of the
two ashes with other sludges. If the ashes performed in the same manner
with other sludges then any individual peculiarities of the digested sludge
at Cedar Rapids would be ruled out.
Figure 39 shows a comparison of fly ash and sludge ash as conditioning
agents for raw sludge at Cedar Rapids. Here again it can be seen that
fly ash was consistently better than sludge ash.
Figure 40 shows a comparison of fly ash and sludge ash as conditioning
agents for a digested sludge obtained from a waste water treatment plant
operated by the City of Cedar Rapids which handles strictly domestic
sewage. Here again fly ash performs better than sludge ash.
In summary, the fly ash obtained from a local power plant at Cedar Rapids
performs significantly better as a sludge conditioning agent than ash
obtained from the incineration of sludge at Cedar Rapids.
105
-------
LU
O
300
200
I 0 0
5 0
tr
u
ul
a I 0
LJ
O.
01
DIGESTED SLUDGE
MAIN PLANT
"r" WITH NO ASH =772
% S = 2.5
i
_L
_L
O
J_
0,5 i.O 1.5 2.0 2 . 5
ASH/SLUDGE RATIO
Figure 36
3,0 3.0 4.0
2 0 0 r
0 0
LU
O
z
I-
co
01
LJ
cr
o
i
u
LJ
a.
tn
5 0
0
DIGESTED SLUDGE
MAIN PLANT
"r" WITH NO ASH =308
% S * 5.5
.....!.. _ _j i i
" UDGE ASH
0,0 0.5 1,0 1.5 2.0 2,5 3,0
ASH/SLUDGE RATIO
3.5 4,0
Figure 37
106
-------
£OU
100
UJ
u
2 50
h-
cn
UJ
cr
o
u.
u 1 0
UJ
a.
CO
,5
It
;-
0.0
. **Sw /SLUDGE ASH
\ A<
\^^^^ A
^.
"
^FLYASH
\;
v°
X
DIGESTED SLUDGE ^
MAIN PLANT
"r" WITH NO ASH =443
% S = 5. 6
i i i i i i . .. . j
0,5 1.0 1.5 2,0 2,5 3,0 3.5
ASH/SLUDGE RATIO
200
1 0 0
(STANCE
en
O
CO
UJ
a:
o
iZ
o | 0
UJ
Q_
,, 5
k_
0
0
Figure 38
-
-
X^^^X*^ .r-SLUDGE ASH
X^ "^-S^/^
^VQ ^^^^A
x.
>,. /-FLYASH
^o
RAW SLUDGE- MAIN PLANT
"r" WITH NO ASH = 196
% S = 3,1
l i i i l i l l
0 0,5 1.0 1.5 2,0 2.5 3.0 3,5 4.0
ASH / SLUDGE RATIO
Figure 39
107
-------
2 0 0 r
I 0 0
5 0
UJ
o
en
to
LJ
o:
o
U.
o 1 0
UJ
Q.
0
0.0
SLUDGE ASH
DOMESTIC SLUDGE
INDIAN CREEK PLANT
"r" WITH NO ASH= 142
% = 7. 0
FLYASH
l
l
i
I
i
0.5 1.0 1.5 2.0
ASH/SLUDGE
2.5 3.0
RATIO
3.5
4.0
Figure 40
As previously stated, a substantial difference exists between the perform-
ance of fly ash and sludge ash as a conditioning agent for sludge. Signi-
ficantly this is a problem which is unique at Cedar Rapids. Studies by
others have found differences between sludge ash and fly ash but in reverse
order, i.e., fly ash having little effect or sludge ash being much better.
At Cedar Rapids the fly ash is a good conditioning agent significantly
better than sludge ash, therefore, it appears that the fly ash and possibly
the sludge ash at Cedar Rapids are unique and results cannot be directly
applied to other cities.
Fly ash and sludge ash can differ in three fundamental ways, chemically,
by size, and by shape. Chemical analyses have been presented for both
sludge ash and fly ash but are given in Table VIII for comparison.
108
-------
Table VIII
Comparison of Chemical Analysis of Fly Ash and Sludge Ash
Results % by Weight on Ignition at 1000° C.
Fly Ash Sludge Ash
Phosphorous pentoxide ?2O5 0.85 4.41
Silica SiC>2 53.90 22.96
Ferric oxide Fe2C>3 17.00 7.49
Alumina A12O3 11.93 6.27
Titania 11203 0.57 0,31
Lime CaO 6.80 50.60
Magnesia MgO 1.65 1.62
Sulfur trioxide SOs 2.30 4.42
Potassium oxide K2O 4.32 0.78
Sodium oxide Na2O 0.39 0.59
Undetermined 0.29 0.55
100.00 100.00
Chemically fly ash and sludge ash are quite dissimilar. Fly ash is
approximately 50 percent silica and sludge ash is approximately 50 per-
cent calcium oxide. Other components such as ferric oxide and alumina
are present in differing quantities. The effect of these higher levels
of iron and aluminum salts in fly ash is not known and deserves further
research.
Particle Size
The range of particle sizes for fly ash and sludge ash have been given
in previous sections. Sludge ash particle analysis is shown in
Figure 34, and fly ash analysis is shown in Figure 35. Fly ash has a
smaller particle size than does sludge ash. As has been shown in
previous sections, ash fractions of smaller size (<45pt) are significantly
far more effective for sludge conditioning than ash of larger size(>45/x).
Further research is needed to further clarify the effect of particle size
on sludge filtration.
Particle Shape
Photo micrographs of sludge ash and fly ash were obtained by a scanning
electron microscope.
109
-------
Sludge Ash 200X
Fly Ash 200X
Figure 41
110
-------
Sludae Ash 500X
Fly Ash 50OX
Figure 42
111
-------
Sludge Ash 1000X
Fly Ash 1000X
Figure 43
112
-------
A small amount of each sample was vacuum coated with a thin conductive
layer of first carbon, then gold. The coated samples were then examined
by a scanning electron microscope. On the preceding pages are compar-
ison photographs of fly ash and sludge ash taken at magnifications of
200x, 500x, and 1000 x. Figures 41, 42 and 43.
From these photomicrographs it appears that fly ash tends to be more
spherical than sludge ash. The sludge ash is more irregular and appears
to agglomerate. In comparing particles of each ash at equal magnifications,
fly ash appears larger than sludge ash although both Coulter Counter and
Bahco Microparticle Classifier show otherwise. This apparent particle
size problem may be attributed to the fact that the sludge ash agglomerates.
In laboratory work the sludge ash tends to clump in water and resists
dispersion much more than does fly ash. This tendency would be charac-
teristic of a material which agglomerates.
Effect of Particle Size on Specific Filter Resistance
Concern has been expressed that, in sludge filtration processes using
recycled sludge incinerator ash as a filter aid, that the recycling of ash
through the incinerator may result in a shift of ash particles to the smaller
particle sizes. This concern is based upon the premise that the fine
fraction of ash is not effective in sludge conditioning and may in fact be
deleterious.
Consequently sludge ash samples from Cedar Rapids were split up on an
air jet sieve into the following fractions.
0-15 microns
15-32 microns
32-63 microns
63-90 microns
These fractions were subsequently used to condition a digested sludge
from Wiesbaden Biedrich which had a specific resistance of 159x 10^2cnr2.
All the samples were mixed with ash equivalent to a 1.0 ash/sludge ratio,
or 42.1 grams ash per liter, for this 4,21 percent solids sludge. These
ash/sludge mixtures were then tested for specific filter resistance. The
following results were obtained.
Fraction (microns) A/S r x 1012cm"2
0-15 1.0 29.4
15-32 1.0 46.5
32 - 63 1.0 62.4
63 - 90 1.0 67.0
Original mixtureallfractions 1.0 60.4
113
-------
Here it was observed that a decreasing ash particle size enhances the
ability of the ash to condition the sludge. Sample conditioned with
fractions of ash below 32 microns had less filter resistance than the
original mixture and the most effective fraction of ash was that fraction
below 15 microns.
The preceding data was obtained in Germany by Dr. O. Busse, Pas savant
Werk, during an earlier evaluation of the Cedar Rapids process.
In Cedar Rapids the sludge incinerator ash was mechanically separated
into three fractions.
0-45 microns
45 - 106 microns
> 106 microns
These fractions of sludge ash were used to condition digested sludge
from Cedar Rapids. Ash to sludge ratios of 2.1 were prepared and then
tested for specific resistance.
Type of Ash Size A/S r
None 0 467
Sludge ash 0- 45 microns 2.1 77.1
Sludge ash 45-106 microns 2.1 110
Sludge ash :> 106 microns 2.1 112
Sludge ash All sizes
(Original) 2.1 88.3
Fly ash All sizes 2.1 44.7
These results at Cedar Rapids verify .the work done in Germany. Again
the smaller fractions of ash are more effective in conditioning sludge.
However, the smallest fraction of sludge ash is still not as effective
as fly ash as indicated above. Unfortunately, equipment was not avail-
able for fractionating sludge ash below 45 microns. It may be possible
that smaller fractions of sludge ash may have performed as well as fly
ash since 78 percent of fly ash has a particle size less than 32 microns.
It appears that diminution of ash with the subsequent increases in the
fraction of fines with recycle through an incinerator is not a problem.
The opposite condition of the fines being eliminated might pose a real
problem. Operation with recycle of incinerator ash at Cedar Rapids has
not apparently had a significant effect upon ash particle sizes.
114
-------
Conditioning Digested Sludge with Sludge Ash and Chemicals
Quite early in the operation of the pressure filter it became apparent
that the digested sludge could not be conditioned satisfactorily with
sludge ash alone. Consequently, in January of 1972 Dr. Oswald Busse
of the Passavant Werk, West Germany, came to Cedar Rapids to assist
in a series of tests to determine how the digested sludge could be con-
ditioned for satisfactory filter performance.
Dr. Busse's testing program consisted of dosing various samples of
sludge with ash/ ferric chloride, and lime and determining the specific
resistance of the conditioned sample. Adequate conditioning was reached
when the conditioned sludge sample had a specific resistance below
r = 20 x 1012cm~2.
Based upon this testing program and previous experience, Dr. Busse
made the following recommendations for ash, ferric chloride and lime
dosages which are still basically in use today.
Figure 44 shows the dosage chart used for operation. Dosages are based
upon pounds of conditioning agent (lime, ferric chloride or ash) per
100 gallons. The dosages are presented in this manner because the
sludge is generally pumped from the holding tanks through conditioning
to the contact tank at 100 gallons per minute.
Ferric chloride dosages range from 2.2 pounds liquid (0.88 pounds actual)
per 100 gallons at 2 percent solids to 3.3 pounds liquid (1.32 pounds
actual) per 100 gallons at 15 percent solids. These dosages on a pound
of dry ferric chloride per ton dry solids basis range from 103 pounds per
ton at 2 percent solids to 21 pounds per ton at 15 percent solids. Ferric
chloride costs at Cedar Rapids are $0.0625 per pound delivered.
Lime dosages range from 6.2 pounds per 100 gallons at 2 percent solids
to 7.5 pounds per 100 gallons at 15 percent solids. This is equivalent
to 743 pounds per ton dry solids at 2 percent to 120 pounds per ton dry
solids at 15 percent. Lime costs at Cedar Rapids are $0.0114 per pound.
Another way of expressing the recommended ash quantities for properly
conditioned sludge is shown in Figure 45. It is apparent that below a
sludge concentration of about 5.5 percent the ash/sludge ratio must be
increased in relation to the decreased sludge concentration to assure a
Properly conditioned sludge for filtration,
115
-------
UJ
190
180
170
60
o
o I 30
CO
-1 I I 0
x 100
en
< 90
80
70
60
o
p 3,0
>« 2.5
2.0
en
CO
o ll5
E
tr
UJ
^ 1,0
,5
CEDAR RAPIDS
SLUDGE CONDITIONING RECOMMENDATIONS
V1-
7.5
UJ
7 n w °
'' O (_)
^
^ fO
65° £
6<52 uJ
>. s
6.0
CD
5,5
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
PERCENT OF SLUDGE SOLIDS
Figure 44
116
-------
Equivalent costs of these chemical dosages are shown in Figure 46.
Chemical costs range from $14.89 per ton at 2 percent solids to $2.69
per ton at 15 percent solids. The optimum solids concentration appears
to be near 6 percent. At this point the cost per ton for chemicals is
$5.48 and the ash/sludge ratio is at a minimum 1.5/1. These dosages
and costs are essentially theoretical and will be compared to actual
costs later in the report.
These costs also are not necessarily typical for sludge dewatering with
pressure filtration but are dependent upon specific resistance of the
sludge and the nature of the sludge ash.
Filter Performance Curves
Filter performance curves were established plotting the filtration rate
(yield) in terms of pounds per hour per square foot, versus time in hours
at various sludge solids concentrations. Figures 47 to 56 show filter
Performance for feed sludge solids ranging from 2.5 percent to 6.5 per-
cent.
Sludge conditioning requirements were determined with bench studies
using the resistivity meter prior to actual filtration in the pressure filter.
It was obvious that proper sludge conditioning was necessary to assure
good filter operation and a satisfactory cake product. On the other hand,
if good sludge conditioning was obtained then the process performed very
well at a given level of production and that production was fairly predict-
able over a broad range of raw solids input. Referring to the filter per-
formance curves, Figures 47 to 55, it is apparent that production (yield)
has an optimum ash/sludge ratio for each sludge density. The optimum
yield is at the lowest ash/sludge ratio which will assure good sludge
conditioning as determined on the bench using the resistivity meter. As
an example, at 5.5 percent solids, an ash/sludge ratio of 1.1/1 is
slightly better than an ash/sludge ratio of 1.5/1 as shown in Figure 53.
However, as the sludge density is decreased to 2.5 percent solids, the
ash/sludge ratio is necessarily increased to 4.1/1 as shown in Figure 47
and the optimum ash/sludge ratio is still the lowest ratio which will pro-
vide good sludge conditioning.
Maintaining good sludge conditioning, the yield, expressed as pounds
Per hour per square foot of filter media increases as the sludge solids
concentration increases as indicated by Figure 56, which is a best-fit
composite curve from all of the computer curves of Figures 47 to 55.
117
-------
5.0
r
O
4,0 A
O
cc
LU
o
Q
n
(ft
~^
i
CO
3.0
\
2.0
1.0
\
V
-I - I...
DIGESTED SLUDGE
O-
J L
J I ! L
.0
6789
PERCENT OF SLUDGE SOLIDS
10 II 12 13 I*
Figure 45
AVERAGE CHEMICAL COSTS
EQUIP MFGRS. RECOMM. DOSAGE
46 8 10 12
PERCENT SLUDGE SOLIDS
Figure 46
118
-------
FILTER PERFORMANCE
~r
.3
0 4:1 Log |Q YIELD= -0,1837 (TIME) t(-0.04573)
A 4,5:1 Log IftjYIELD= -0.1342
-------
FILTER PERFORMANCE
.3
.2
A/S RATIO
O 3:1 Log |Q
A 3.5: 1 Log YIELD=- 0.1761 ,(TIME)+0.02527 :
Yl ELD =-0.1725 (TIME) tO.05403.
.0
1.5 2.0
TIME (hours)
Figure 49
FILTER PERFORMANCE
2.5
j
3.0
*- \(
?! o
..0^
*- a
3 -O
o
j: 7
^ '
4
3
.2
i
5
^^^/»^^r
A/S RATIO
0 2.5:1 Log|0
A 3.5:1 Log|0
> 1
O
5,^0
k.'»_ ^^^^^
^>^^
n y *»*'
u / 4
3.5: I-/
YIELD = -0.1700
YIELD = -O.I648
0 1
SOLIDS= 4%
D^_
k^*^O_
"^-^
8 N\
(TIME)t0.08548
(TIME)tO.OI589
5 2
FIME (hours)
2.5: \-j
[<
^>^
0 2
5
£^
5 3.
Figure 50
120
-------
FILTER PERFORMANCE
Figure 51
FILTER PERFORMANCE
A 1.51 Log
0 2:1 Log
|Q
|0
YIELD =-0,2127
YIELD =-0,1894
TIME)+0.27987
(TIME)+0,I8048
1.0
1,5
TIME
2.0
(hours)
2,5
3.0
Figure 52
121
-------
FILTER PERFORMANCE
o
-C
Figure 53
FILTER PERFORMANCE
£
C
t
t
.9
.8
.7
.6
.5
.4
.3
.2
i
*** t
A/S RATIO
0 |:l Log|0
A |.5:l Log
~-£
<"*<*?'»i A C
/ '"*''"* t
^-1.5:1
YIELD = ~O.I57I
jQ YIELD = -0.1803
0
T
SOLIDS = 5.5%
3^^O
k. ^>^***1**- r
i.
^*V-. f^***1*11*-
A "*^
(TIME)t0.22520
(TIME)t 0.18972
^"^^ ^C
"""""s^
5-^*^^^
k<-^^ C
~"^^
^^
5 2.0 2,5 3.
IME (hours)
i
^^
£ ,C
cr
O Q
^ '*
H -B
? 7
U>
-^ '6
Q -5
_i
UJ
r~ A
*
2
k\
\
^
^
'5:I7
W
s^ss
s^H
SOLIDS' 6%
SA A
A/S RATIO
A 1.5:1 Log|0
0 2.1 Log|0
~~^^^^(
A ""^Hfc.
^_C C^Sw
0
YIELD = -0.237^
YIELD = -0.1003
/ O"OJ*^
2.)
t (TIME) + 0.3635«
(TIMEJ+0.05876
»i II ^.
^*^^ A ~m C
O ^^
3
^^1
5 1.0 1.5 2.0 2.5 3.
TIME (hours)
Figure 54
122
-------
FILTER PERFORMANCE
2.0
(hours)
3.0
Figure 55
FILTER PERFORMANCE-VARYING SLUDGE DENSITY
3.0
Figure 56
123
-------
Filter yield increases in almost direct proportion to increased raw sludge
solids density at a ratio of about 1.8/1, that is, if the sludge density is
increased times 2, the resulting filter yield is increased times 1.8.
It is apparent that properly conditioned sludge filters satisfactorily at
both low and high sludge densities and that yield is principally influenced
by time to completely fill the cavity forming the predetermined cake vol-
ume. In practice it became apparent that the limiting factor to improving
yield was the ability to pump conditioned solids to the pressure filter in
sufficient capacity to develop the maximum rate of filtration. Each system
consisted of two pumps and some experimentation was performed using
three and four pumps simultaneously on one system.
Filtrate Characteristics
The pressure filter is a very efficient solids liquid separator with effi-
ciencies of separation practically at 100 percent. During the months of
February to August 1972 the filtrate averaged 74 mg/1 suspended solids
while the sludge solids averaged 4.60 percent solids giving a percent
removal of suspended solids of 99.99984 percent.
During the pilot study it was noted that during a run the suspended solids
in the filtrate were relatively high initially (400 mg/1 approx.) and then
decreased to very low values during the run. No variance in suspended
solids could be noted during a run on the full scale filter. Possible
explanations for this difference may lie in two areas, the initial pressure
on the pilot plant was a full 225 psig while on the full scale filter the
pressure started at a low value (approximately 40 psig) and gradually
increased to 225 psig. The full 225 psig pressure may have forced sus-
pended solids through the thin precoat on the pilot plant while this did
not occur on the full scale filter.
The BOD and COD of the filtrate are due primarily to dissolved organics
such as volatile acids and are not necessarily affected by the filtering
process. It is possible that some dissolved COD may be removed by
the process of absorption, however, this was not investigated on the
full scale filter and deserves further research. Limited bench work
indicates that conditioning with fly ash alone may remove as much as
50 percent of the dissolved COD. This removal was not noted in samples
conditioned with sludge ash alone.
Filtrate quality is shown in Table IX.
124
-------
Table IX
Average Filtrate Quality
Suspended
Solids
mg/1 '
Total
Solids
mg/1
Volatile
Percent
Solids
Turbidity
Feb. Mar. Apr. May Tune July Aug. Average
25 94 103 60 84 65 88 74
5000 4900 5600 6200 6400 6000 6800 5842
11.2 15.6 23.7 20.8 26.0 30.2 25.1 21.8
18
20 29 19 17 22 27 21.7
Color Cobalt 62 61 69 89 107 123 132 91.8
Platinum
Units
Conductivity
Micromhos/
7883 7520 7914 7483 6570 5800 5600 6967
cm-
COD 623 706 1310 847 1014 720 780 857
mg/1
BOD 5
mg/1
PH
329
901
387 398
504
12.1 12.4 12.4 11.9 11.7 11.5 11.6 11.9
125
-------
Sludge Cake Characteristics
Sludge cake is the discharge from the pressure filter and is about 58 inches
diameter, 1^ inches thick, weighs about 200 pounds. Each pressure filter
produces 83 cakes per cycle which are discharged to a storage bunker.
Shear cables are positioned across the top of the bunker to break these
large cakes into smaller pieces. Cake is conveyed through a series of
drag conveyors and elevators to the incinerator. At normal operation
cake moisture content is in the range of 36-38 percent and the appearance
is dense, dry and textured. Extreme care should be exercised in select-
ing proper materials handling equipment to transport this cake to disposal.
It is essential that minimum transfer and handling of the cake be performed
from the filter to ultimate disposal. Excessive shear, breaking, dropping,
grinding, or any form of handling results in plasticizing the cake to the
consistency of dense putty or clay-like material. At these properties the
plasticized cake material tends to be extremely sticky even at 62-64 per-
cent solids and causes many problems due to poor release from drag and
screw conveyors. The crumbly, dry, textured character of the original
sludge cake can be maintained with minimum transfer and handling.
Grinders were originally installed at Cedar Rapids in the incinerator feed
system, however, these have been removed and are undesirable with this
type of cake. By the time the cake has traveled through the materials
handling system to the incinerator it has broken into small pieces of
i inch to 2 inch size. These pieces are generally in the form of balls
due to the rolling action of the incinerator feed screw conveyor. Balls
probably would not develop if this final conveyor were of a different
design. However, the shape of the feed particles is not significant to
the incinerator performance.
The specific weight of the discharged filter cake varies from about 107
to 114 pounds per cubic foot dependent upon moisture content and ash
ratio. Under normal operating conditions the moisture content of the
sludge cake varies only a few percentage points, therefore, the greatest
influence on the specific weight of the cake is the ash content (ash/sludge
ratio). The specific gravity of organic ash is in the range of 2.6 to 2.8.
Laboratory determinations of specific gravity using water medium was
2.63; using hexane medium, was 2.78.
Cake discharged from the filter to the cake storage bunker is sheared by
a series of cables which tends to bulk the broken cake. However, the
impact of additional cake falling on the pile of previously discharged
cake tends to cause recompaction due to the plastic properties of the
sludge cake. Bulked filter cake was determined to be about 47 pounds
per cubic foot, but this was for cake not exposed to impact and compres-
sive forces due to dropping and stacking. That low figure might be valid
126
-------
for some conditions of limited cake storage where bulking is the major
influence. Cake discharged from the bottom of the hopper under com-
pressed conditions of superimposed loads of 10 to 12 feet depth had a
specific weight of 83 pounds per cubic foot. This was determined by
weighing many fixed volume samples in the bunker conveyor.
Conditioned sludge applied to the pressure filter forms a cake structure
which almost immediately serves as the principal filter media. The
filter cloth has served only as a base structure for this development, and
after the cake has formed to 1/16 or 1/8 inch thickness, the influence of
the filter cloth is negligible to the continued filter performance. As the
cake continues to form and the filter void is filled, the total resistance
increases rapidly developing full pressure differential in possibly 30-40
rriinutes as shown by a typical filter pressure chart, Figure 57.
Typical Filter Pressure Chart
0 l5 I
TIME (HOUR)
Figure 57
127
-------
Continued full pressure differential is desirable to assure that most of
the free water has the time to travel through the cake and underdrainage
system to be disposed of as filtrate. The noteworthy point after the
development of the full cake formation is that prolonged high pressure
does very little to further dewater the cake. The moisture content may
range only a few percentage points over a relatively wide span of sludge
feed densities; and the cake moisture content is even less influenced by
prolonged dewatering time. An example of this is found in Figure 58
where sludge cake densities are shown for different percent feed solids
and cycle time. The curve is a best fit line using arithmetic probability
points shown as dark points of average solids groupings.
75 r-
70
g 65
O
en
60
55
o
ir
ui
o.
50
45
o o
o o
0 o
PILOT PLANT DESIGN CONDITION 48% TOTAL SOLIDS
1 1
1
1
1
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5
PERCENT OF TOTAL SOLIDS
Figure 58
Grease and nitrogen determinations were made on composite-samples for
each shift of operation. This consisted of three composite samples per
day made up from each filter run per shift. The average monthly analysis
is as follows:
April May June Tulv August
Nitrogen mg/1 1.83 1.58 1.16 0.86 0.87
Grease mg/1 2.40 1.36 1.80 1.70
128
-------
Phosphate Removal
There has been increasing emphasis upon the removal of phosphorous
from waste. It then becomes very important that the dewatering of
phosphate rich sludges should produce a filtrate low in phosphorous.
Phosphorous generally is removed by precipitation using three commonly
available cations, A1+3, Fe+2, or Fe+3, and Ca+2 .
In the process of sludge dewatering Fe+3 and Ca+2 are present as coag-
ulants. Presumably because of the negative oxidation reduction potential
of the sludge a portion of the iron is reduced to Fe+2. Consequently
three precipitates of phosphorous could be possible; ferrous phosphate
(Fe3(PO4)2), ferric phosphate (FePC^), and calcium phosphate or more
properly calcium hydroxyapatite
Phosphorous removal through the pressure filter was investigated by
using the resistance meter on the bench and spot checking actual
phosphorous removal through the full scale filter.
Bench Scale Studies
Since most of the data from the pressure filter was obtained when the
sludge was conditioned with lime and ferric chloride, some bench work
was also performed using the resistance meter and conditioning only
with ash. In this test a portion of the same sample was conditioned
with fly ash using ash/sludge ratios of 0, 1.0, 1.5,2.0, 2. 5 and 3.0.
Separate portions were also conditioned with sludge ash using the same
ash/sludge ratios. The samples were then filtered through the resistance
meter and the filtrates were analyzed for total phosphate. The results
are graphically displayed in Figure 59 and listed in Table X.
Table X
Fly Ash _ Sludge Ash
Phosphate
A/S Ratio as P
O(noash) 25.
1.0
1.5
2.0
2.5
3.0
12.
10.
10.
8.
7.
Percent
Removed
PH
6 mg/1
3
4
0
6
6
51.
59.
60.
66.
70.
9
4
9
4
3
8.
8.
8.
8.
8.
1
4
4
5
6
Phosphate Percent
as P Removed
25.
12.
10.
8,
6.
6.
PH
6 mg/1
3
4
3
8
2
51.
59.
67.
73.
75.
9
4
6
4
8
8.3
8.3
8.5
8.6
8.7
129
-------
The removal of phosphate was nearly identical for samples treated with
fly ash and the samples treated with sludge ash.
PHOSPHATE IN FILTRATE
A Samples conditioned with SLUDGE ASH
O Samples conditioned with FLYASH
0.5
1.0 1.5 2.O
ASH/SLUDGE RATIO
2.5
3.0
Figure 59
Full Scale Studies
During the course of the study several runs on the full scale filter were
made using only fly ash and no chemicals (ferric chloride and lime).
The phosphate levels in the raw sludge and the removals are:
Run No.
1 Sludge
Filtrate
2 SJudge
Filtrate
Total
Phosphorous as P
516
8.75
528
18.5
Soluble
Phosphorous as P
100.5
2.5
73.5
4.5
Percent
Removed
91.3
74.8
130
-------
The percent phosphorous removal was based upon the soluble phosphorous
in the sludge and total phosphorous remaining'in the filtrate. Presumably
the insoluble phosphorous in the sludge is principally discharged with
the sludge cake.
Percent Removal = Soluble P Sludge - Total P Filtrate = 100.5-8.75 = 91.3%
Soluble P Sludge 100.5
Additional data on phosphorous removal was obtained when ferric chloride
and lime were also used for conditioning along with ash. This data is
presented in Table XI.
Table XI
Phosphate Removals by Pressure Filtration
Using Lime, Ferric Chloride and Sludge Ash as Coagulants
Sludge
Sludge
Date
7-20-72
7-24-72
7-25-72
7-26-72
7-27-72
7-27-72
7-27-72
8- 3-72
8- 7-72
8- 9-72
8-10-72
Unconditioned
Total
500 mg/1
592
564
786
520
546
514
442
980
446
1150
Soluble
*
*
*
*
*
*
it
96
91.5
104
142
Conditioned
Total
1204
1596
1448
1752
1124
1256
1528
1732
1348
1656
2890
Soluble
*
*
*
*
*
*
*
96
131
134
172
Filtrate
Total
2.4
16.8
16.8
11.8
16.8
14.8
19.8
13.2
12.3
8.75
3.65
Soluble
*
*
*
*
*
*
*
< 0.1
< 0.1
*
*
* Not determined
The percentage reduction of phosphorous realistically should be based
upon the soluble phosphorous in the unconditioned sludge versus the
total phosphorous in the filtrate. This will give a somewhat lower
percentage of removal than a percentage removal based upon total phos-
phate in the conditioned sludge versus total phosphate in the filtrate.
Therefore, the comparison of the two processes of phosphate removal ,
ash only conditioning versus ash plus chemical conditioning, may be
difficult unless the basis of comparison is the same. That is, the
comparison of total and soluble phosphate.
131
-------
Percentage removals of phosphorous through the pressure filter averaged
86.7 with the soluble phosphorous in the sludge averaging 108.4 mg/1,
and the filtrate total phosphorous averaging 14.4 mg/1. On a total
phosphate basis the percent removal was 97.8.
Filtration with Fly Ash
During the course of the study a program was developed to operate the
full scale pressure filters using only fly ash as a digested sludge condi-
tioner. The limited work that was done affirmed the fact that fly ash
alone (without chemicals) was sufficient to dewater digested sludge.
This also confirmed the conclusions of the original pilot plant report
wherein only fly ash was used as a sludge conditioner.
However, the use of fly ash posed many problems in materials handling
and conveying. Fly ash tends to fluidize and with the ash handling equip-
ment used at Cedar Rapids the accurate control of admix quantities was
impossible. Our experience would indicate that gravimetric feeders are
not adequate for feeding materials which fluidize.
Another problem which occurred was conveying the filter cake. Although
the percent cake solids averaged 55.5 percent for all runs, the fly ash
cakes were virtually impossible to convey. After being conveyed from
the bottom of the bunker the cake turned into a sticky mass which refused
to fall off the cake elevator flights. Therefore, only limited work was
performed using the full scale filter.
Below is the summary of the best filter runs using fly ash as a conditioning
agent.
Percent Solids = 5 percent
Ash/Sludge 2.57
Cycle Time 1.80 hours
Yield 0.57
No. Run 4
Other runs which were made at lower percent solids in the feed sludge
were dosed with extreme amounts of ash (A/S =4.5 to 6.2) because of
the fluidizing of the fly ash.
Filtrate
Filtrate from dewatering, using fly ash, showed great differences from the
filtrate using sludge ash and chemicals. Not enough runs were made to
draw any generalizations.
132
-------
Following is a comparison of filtrate from one day of operation.
Fly Ash .
Conditioned
pH
Suspended Solids
Color
COD
mg/1
Units
mg/1
mg/1
9.3
148
50
480
250
Sludge Ash and. Chemicals
Conditioned
10.4
43
150
830
470
It can be seen that the filtrate appears to have a lower COD, BOD and
color when the sludge is conditioned with fly ash rather than sludge ash
and chemicals .
Dewaterinq of Raw Primary Sludge
The filterability of raw sludge as compared to digested sludge at Cedar
Rapids is generally much better. Specific resistance values for raw sludge
range from values of 120 to 300 x lQl2cm-2/ with an average value of
approximately 150 x lol2cm-2 .
Because of better filterability of raw sludge, a program was set up to
develop chemical and ash dosage requirements so that raw sludge could
be run on the full scale filter.
Bench testing and testing using the six inch pilot pressure filter indicated
that satisfactory performance could be obtained using the chemical dosages
for digested sludge and using a 1/1 ash/sludge ratio at all solids concen-
trations.
Accordingly, in-plant arrangements were made to empty one storage tank
and to pump raw sludge from a gravity thickener to the empty storage tank.
First filter runs with raw sludge were made November 28, 1972, and during
the month of December over 60 filter runs were made on raw sludge. The
runs were generally quite satisfactory.
During the month of December, the raw sludge pumped to the sludge hold-
ing tanks averaged 3.7 percent solids. During this same period, sludge
from the holding tanks to the process averaged 5.78 percent solids, and
much clear supernatant could be decanted from the holding tanks.
133
-------
Table XII
Summary of Raw Sludge Filtration Data
5%
Average Ash/Sludge 1.20 range 1.56 to 1.00
Chemical Cost $/ton 6.26range $7.01to4.67
Yield lb,/sq.ft./hr. 0.68 range 0.97 to 0.30
Cycle Time hrs . 2.03 range 1.33 to 3. 75
Number of runs - 27
6%
Average Ash/Sludge 1.19 range 1.44 to 0.68
Chemical Cost $/ton 5.17 range $5. 61 to 4.71
Yield Ib./sq.ft./hr. 0.90 range 1.41 to 0.29
Cycle Time hrs. 1.75 range 0.83 to 3.83
Number of runs - 22
7%
Average Ash/Sludge 1.13 range 1.47 to 0.91
Chemical Cost $/ton 4.97 range $5.29 to 4.05
Yield Ibs./sq.ft./hr. 0.92 range 1.30 to 0.39
Cycle Time hrs. 1.69 range 1.0 8 to 3.25
Number of runs - 8
In Table XII, the results for the filtration of raw sludge are given. Data
are given for 5, 6 and 7 percent solids. Other data f or 3, 4, 8 and 9 per-
cent solids are not given. These were insufficient filter cycles to interpret
a curve. As can be noted, yield, cycle time and chemical costs all im-
proved as the percent feed solids increased from 5 to 7 percent.
Yield versus cycle time is shown in Figure 60 for raw sludge. Yields are
all quite good compared to digested sludge, basically due to low ash
ratios of approximately 1.2.
Filter Cake
Pressure filter cake averaged 54, 56 and 58 percent solids for filter feed
solids concentrations of 5, 6 and 7 percent solids respectively. Volatile
solids content of the cake was higher than for digested sludge, averaging
25, 27 and 29 percent for 5, 6 and 7 percent feed solids respectively.
These values compare to 11-13 percent volatile solids at comparable de-
watering conditions with digested solids. These levels of volatility were
much higher than that necessary (approximately 11 percent) to provide
134
-------
autocombustion in the incinerator. No difficulty was experienced in
conveying the raw sludge filter cake.
Filtrate
The most significant difference in filtrate quality caused by filtration of
raw sludge was the very high BOD's and COD's of the filtrate. In
Table XIII is given comparative values of filtrate COD's and BOD's for
digested sludge and raw sludge during the two month period of November
and December 1972. These high BOD's are the result of the formation of
soluble volatile acids in the sludge. These volatile acids depressed the
pH of the raw sludge to values of approximately 5.0 (+0.2). The filtrate
pH averaged 11.6 due to sludge conditioning with lime.
Filtrate suspended solids showed no basic difference than the filtrate from
digested sludge. Filtrate samples from all runs averaged 79 mg/1. During
this same period suspended solids from the digested sludge filtrate aver-
aged 90 mg/1. As shown in Table XIII there was no significant difference
between filtrates.
Table XIII
Comparative Values Dewatering Raw and Digested Sludge
COD mg/1 BQD5 mq/1 BOD5/COD (%)
Sludge Digested Raw Digested Raw Digested Raw
Maximum 1080 13,500 740 10,300 70 '94
Minimum 270 2,000 140 1,700 41 72
Mean 510 7,080 300 5,700 56 81
Median 470 6,600 250 5,200 58 81
Comparative Values of Suspended Solids in Filtrate
Type of Sludge Maximum Minimum Mean Median
Raw 430 mg/1 22 mg/1 79 mg/1 53 mg/1
Digested 345 mg/1 20 mg/1 90 mg/1 61 mg/1
135
-------
u. I
o .9
>? .8
5 .7
-------
SECTION X
PROCESS EVALUATION
Economic Evaluation
Basis for Economic Evaluation
The labor requirements were determined for sludge dewatering and sludge
incineration for both operation and maintenance. These labor costs were
based upon actual wages paid for 1972 including an allowance for fringe
benefits such as insurance, vacation, workmen's compensation, retire-
ment, etc. Hourly rates were $4.58 for the operator and maintenance
shift foremen and $4.17 for assistant. Management or laboratory costs
are not included.
The sludge dewatering plant was bid in December 1968. To be consis-
tent throughout this report, all costs have been interpreted to 1972 costs.
Earlier in this report the pilot plant evaluation costs predictions made in
1968 were interpreted to current 1972 costs, therefore the capital costs
for the construction of the sludge dewatering facility for which bids were
received in December 1968 have also been interpreted to 1972 using the
Engineering News-Record index of cost trends. The equipment costs in-
crease for that period has been interpreted to be 20 percent. The capital
costs used in this evaluation are the actual bid costs for December 1968
scaled upward as outlined to reflect what these costs might have been for
1972.
Chemical costs are those actually paid for pebble lime and liquid ferric
chloride. Power costs were assumed to be $0.02/KWH for filtration
although it may actually be less due to a favorable power demand-power
consumed ratio. Power consumed for each filter is separately metered,
however, total power for the dewatering and incineration processes is
also metered and is the basis for determining power company billings.
High power demand charges for the incinerator high energy off-gas
scrubber distort the power billings and make it necessary to interpret
the filtration power charge as indicated above. Power consumed for
filtration was measured by the separate watt-hour meters during observed
filtration cycles and was determined to be 40 KWH per hour of actual
filtration.
Cedar Rapids has two pressure filters and all labor costs are determined
for the operation of two units on a full time basis. All costs were reduced
137
-------
to dollars per filter hour operation so that the final cost would be evaluated
to filtration capacity regardless of the operating capacities experienced
during the study.
Operation Foreman
Assistant
Man Hours/Hr. Filtration $/Hr. Filtration
0.90 4.13
0.33 1.23 1.37 5.50
Maintenance Foreman 0.17
Assistant 0.33
0.50
0.76
1.37
2.13
1.73
Labor costs per filter ( -*- 2 units)
Power costs 40 KWH/hr. filtration x $0.02
$7.63
$3.82/hr. filtration
0.80/hr. filtration
Chemical costs ferric chloride ($130.00/ton) 0.065 /pound
lime ($ 22.75/ton) 0.0114/pound
Capital costs 20 years at 4|%
5^ days per week - 286 working days/year
Building $ 417,000
Equipment 1,255,000
$1,672,000 = $19.10/hr+
units = $9.55/
hr. filtration
No cost allowance has been made for fly ash haul to the plant or product
ash from the plant. This cost may vary considerably from a small cost
as in the case at Cedar Rapids, to a significant cost if the fly ash was
purchased and the haul distance was great. The basic operation is with
incinerated sludge cake ash recycled for sludge conditioning. However,
as a part of this report and to follow up on the original pilot plant study
wherein fly ash was used as a sludge conditioner, power plant fly ash
was transported and used as a sludge conditioner. The fly ash was
obtained from the same power generating plant as was used in the pilot
plant study. The ash was obtained at no cost. Transportation was ar-
ranged with a contract hauler using a bulk cement transport trailer truck.
Fly ash was loaded by gravity chute at the power plant, transported
about three miles distant, unloaded by blowing into the ash storage bins.
The cost for transporting fly ash filter aid was $4.00 per ash ton. The
filter performance using fly ash filter aid averaged about $6 to $8 per ton
dry sewage solids as the cost of hauling fly ash to the process. Product
138
-------
ash haul to disposal would vary considerably with local conditions and
would be less significant with organic ash than with fly ash filter aid.
At Cedar Rapids it was possible to dispose of the excess ash on the site,
therefore haul-out costs were not accounted.
Process Costs
The total cost for pressure filtration is composed of four separate costs.
These costs are labor for maintenance and operation, power, chemicals
for conditioning, and capital investment.
Labor costs were determined, as described before, to be $3.82 per hour
per filter. Power costs were determined in the same manner to be $0.80
per hour per filter. Likewise capital costs were determined to be $9.55
per hour. In the compilation of the data each cost was determined as
follows:
Cost/ton = cycle time (hours) x cost factor
tons dry solids/filter
Cycle time represents the time to complete a filter run. Added to filtra-
tion time, approximately 0.4 hour is required to discharge the cake,
refill the filter with filtrate, and precoat, and is referred to as turn
around time.
Cycle time (hours) = filtration time (hours) + 0.4 hours turn around
The cost facfor represents the dollar per hour value placed upon each cost.
Chemical costs were determined for each cycle by measurement of actual
quantities of lime and ferric chloride used to condition the sludge.
i>
Figures 61, 62 and 63 show cost data for 4|%, 5|%, and 6|% solids
respectively. Only operating costs, which includes both power and
labor, capital costs and total costs are shown.
These three cost curves were determined by using costs for each run
which were determined by a computer, then taking these values and
performing a linear regression analysis based upon the relationship log
cost versus log yield.
139
-------
PERCENT OF TOTAL SOLIDS*
l.D
£ 1.4
qr
< 1.2
1 i.o
X.
A 8
""" .6
3 -4
UJ
> -2
0
(
\
\
\
\
\
OPER
COST
3
i
%
\
\
V
ATING ^
S
\
\
CAPIT
\r
[~ COSTS -7
\. V
^
^>^^
-c
n /
' c
TOTAL
*c
>-....c
i-»mm^
COSTS If
:*^,^r
c
^~ 5
iJCL. CHEI
J^^ri
_ I
r
^ICAL CO
C!TC:
i :&=saa»a
3 5 10 15 20 25 30 35 40 45 5
COSTS (8/Ton.)
Figure 61
PERCENT OF TOTAL SOLIDS
TOTAL COSTS INCL. CHEMICAL COSTS
CAPITAL rn_
\1 COSTS
20 25 30
COSTS (B/Ton.)
Figure 62
140
-------
PERCENT OF TOTAL SOLIDS
1,6
£l.4
cr
4J 1,2
2 1.0
.... __
\
V j fv.
\ i I x
... %, CAPITA I J*
9- t-
i
\ ' 1
-TOTAL COSTS INCL. CHEMICAL COSTS-
OPERATING
COSTS
20 25 30
COSTS (fi/Ton,)
Figure 63
The following correlation coefficients were obtained.
Percent Solids Operating Cost Capital Total
45
4*
5*
6*
-0.985
-0.991
-0.988
-0.985
-0.991
-0.988
-0.875
-0.925
-0.865
The range of cost experience is shown in Table XIV.
Table XIV
Process Costs
Maximum Minimum Average
Operating
Capital
Total (including chemical)
Operating
Capital
Total
6i%
Operating
Capital
Total
11.89
24.05
43.64
9.32
19.27
34.05
6.85
14.17
28.44
2.84
5.87
17.72
2.65
4.54
14.80
1.47
3.04
11.22
5.83
12.05
26.83
4.69
9.71
21.69
3.83
7.91
18.20
50
141
-------
Negative correlation coefficients mean an inverse correlation. The cor-
relation coefficient can have values from -1 to +1 where an absolute value
of 1 means a perfect correlation whether the correlation be direct (+1) or
inverse (-1). A value of zero indicates no correlation or simply a random
scattering of points.
All correlation coefficients shown above are negative and closely approach
1 in absolute value. As can bo expected, costs related to hours operation
decrease as productivity or yield increase, and as can be seen from the
graph, these costs decrease in a predictable manner.
The correlation coefficient for total costs are lower because of the seem-
ingly lack of correlation of chemical dosages (costs) with yield. Correla-
tion coefficients for chemical costs were developed in the same manner
as the other costs. The correlation coefficients were 0.122, 0.217 and
0.005 for 4|, 5| and 6| percent solids respectively. These are very low
and poor correlation coefficients.
Chemical costs showed poor correlation for several reasons, they are:
1. Varying specific resistance of digested sludge.
2. Lack of sufficient conditioned sludge pumping
capacity to the filter.
3. Over-ranging of chemical feeders.
The first two reasons are interrelated. As has been brought out in previous
sections, chemical (lime and ferric chloride) dosages were set up to properly
condition sludge with high resistance. Since the sludge varied consider-
ably in resistance, some sludge was over-conditioned, While over-condi-
tioning the sludge should have resulted in a higher yield by giving a higher
filtration rate, the maximum filtration rate of the filter is mechanically
limited by the capacity of the two filter feed pumps. Later it was demon-
strated that using all four filter feed pumps for one filter significantly
decreased filtration time (up to 50%), by meeting the true capacity of the
filter for sludge. Because of this mechanical limit, chemical conditioning
was only partially effective in raising yields.
The other problem which occurred was 'the oversizing of the chemical feeders
specifically the ferric chloride feeder. Because of this oversizing, the
feeder operated in the lower 20 percent of its capacity where accuracy
was at a minimum.
142
-------
CHEMICAL COSTS (AVER.)
I 5
I 4
o
5
LJ
I
O
8
/ACTUAL OPERATING COSTS
X:
1 ' i ' ' i i i i t i i i
0 2 4 6 8 10 ' 12 1415
PERCENT SLUDGE SOLIDS
Figure 64
Figure 64 shows average chemical costs per ton of dry solids versus
percent solids. Also shown is. the chemical cost curve for the proposed
costs developed from work on the resistance meter by Dr. O. Busse. As
can be seen from the graphs, operating costs were somewhat higher than
theoretical yet followed the cost curve very closely.
Optimization of Costs
From the figures present it becomes quite apparent that real cost savings
can be made by pre-thickening of sludge solids. Thickening the incoming
sludge from 3 to 6 percent decreases chemical costs at Cedar Rapids from
$10.00 per ton dewatered to $6.00 per ton dewatered. Ash requirements
also decrease from 3 pounds per pound of sludge solids to li pounds per
pound of dry solids. See Figure 45. Obviously the less volume of the
filter taken up by ash means more capacity for sludge. Cost advantages
for pre-thickening decrease for influent solids in excess of 6 percent.
143
-------
All costs developed in this section, especially those for chemicals, are
specifically for Cedar Rapids and do not have general application, and
do not form a sound basis of comparison with other processes. Reference
is made to the previous section of this report wherein chemical costs with-
out ash on a vacuum filter at Cedar Rapids could not be lowered below
$20.00 per ton because of the high specific resistance of the sludge.
Process Evaluation
Having operated the pressure filter pilot plant at an early date when no
analytical data was available for evaluation and guidance, and having
developed the design criteria for a full scale plant from those pilot
studies, Cedar Rapids has observed the process performance without
bias to other installation performances. As stated previously, the pilot
plant performed reasonably well with good data correlation, however, the
pilot process was hand controlled and should reasonably be expected to
perform well. The full scale plant process is highly automated and there-
fore, reflected some of the malfunctions associated with equipment failure
from numerous interrelated units.
Some unsatisfactory process performance was experienced at Cedar Rapids
during the early days of start-up and check-out. Much of this was related
to inadequate training of the operating personnel, particularly those of the
equipment manufacturer. This is mentioned here because so many visitors
to this facility during more than a year of start-up and check-out prior to
acceptance were aware of this problem. It is understandable that the first
major installation would offer some degree of challenge on start-up and that
experience would make the next one easier.
Looking backward to the expectations of the pilot plant study, and to the
process performance observed in the full scale plant, it is obvious that
the pressure filter process is dependable, economical, and offers distinct
advantages over other forms of sludge dewatering based upon Cedar Rapids
sludge. The process has achieved a higher quality product with greater
capacity in the full scale plant than was predicted by the pilot plant study.
Cake quality is maintained over a broad range of sludge solids concen-
tration, particularly at the low solids level (2.5-4.5%) which were not
previously experienced in the pilot studies.
Incineration of Sludge Cake
The primary subject of this report is dewatering of digested sludge by
pressure filtration. The process at Cedar Rapids is dependent upon
144
-------
recycled organic sludge ash as a sludge conditioner and filter aid. It
would, therefore, seem appropriate to briefly review the incineration of
the pressure filter cake and the recycle process. As previously described,
cake is discharged from the pressure filter to a storage bunker and trans-
ported through a series of conveyors to the incinerator.
The cake incineration process starts with the bunker feed conveyor which
is a variable speed drag conveyor. This speed control on the bunker dis-
charge serves to regulate the rate of cake feed to the incinerator and is
manually set by the operator. Automatic temperature controls regulate
the auxiliary fuel feed, if required, to maintain hearth temperatures.
The incinerator is a conventional multiple hearth furnace previously
described.
When pressure filter performance is maintained close to design ash/sludge
ratios of 1.5/1, it has been observed that furnace performance can be
controlled so that the cake feed is autocombustible with 11 to 13 percent
volatile solids content. Under normal filter operation the moisture con-
tent of the cake falls in a narrow range of values regardless of the raw
sludge feed density, therefore, the major factor in controlling the per-
centage of volatile solids in the cake is the ash/sludge ratio. As the
ash/sludge ratio increases, the volatile content of the cake decreases.
Numerous heat balances have been programmed on the furnace operation,
however, these are not the subject of this report and it will suffice to
summarize that the filter cake from digested sludge will burn without
auxiliary fuel when furnace feed conditions are held within a reasonable
range (+ 50%) of the design ash/sludge ratio and cake feed rate.
The internal conditions of the furnace are significantly different while
burning cake with recycle ash. The internal atmosphere is extremely
dusty even under minimum excess air control and good balance. A con-
siderable amount of ash is picked up in the gas stream in the upper drying
hearths and carried with the off-gas. This added particulate volume
makes an adequate scrubber system even more important. Cedar Rapids'
incinerator off-gases are directed to a wet venturi high energy scrubber
operating at about 30 inch water column pressure drop. Scrubber water
is returned to the treatment plant and contains 4,000 - 10,000 mg/1
suspended solids. This seemingly large quantity of fine ash carried
out in the scrubber water was of considerable concern for some time.
The general philosophy for successful pressure filtration with ash filter
aid was that the fine ash particles were extremely important to develop-
ing a good dewatering cake. The Germans had experienced and reported
145
-------
this requirement. In the early periods of check-out the apparent shortage
of finer ash in the recycled conditioning ash was considered a factor of
poor filter performance. After considerable bench work with and without
the fine ash carried out in the scrubber water it was determined that the
absence of this fine ash did not influence sludge conditioning and filter-
ability.
Incinerator ash passes through a hammer mill grinder which originally
contained 1/16 inch screens. These were later enlarged and ultimately
increased in size to f inch. It has been demonstrated at Cedar Rapids
that, contrary to the European practices, grinding recycled ash is not
necessary, and for ordinary protection of the process equipment simple
screening for foreign and oversized material would be satisfactory.
Considerable difficulty has been encountered with incinerated ash in
storage and feeding equipment. Both organic ash and fly ash readily
fluidizes and compacts over a narrow range of external influence.
Elevated ash storage bins were equipped with customary bin vibrators to
assure even flow to the outlet feeders. It was necessary to employ a
trial and error adjustment of both frequency and magnitude of vibration to
get the correct adjustment of the external influence to assure satisfactory
flow of ash to the feeder. Until the correct combination of magnitude of t
impact and frequency was determined it was unpredictable whether the
output ash from the storage bin to the feeder would fluidize or compact.
At times it would fluidize at one set of conditions and at other times it
would compact. Temperature of ash leaving the incinerator is normally
about 250° F. and this causes no problems in the elevated storage bins.
However, due to process upsets the ash temperature may go to 500-700
degrees F. for short periods. These high ash temperatures into a partially
filled bin apparently caused a temperature turbulence and a fluidized con-
dition which would progressively work through the ash in storage to the
outlet and thus upset the feeder performance.
Proper precoat is essential to satisfactory filter performance and precoat
pressure is an important control for assuring good cake formation. Normal
precoat pressure for a clean filter media starts at about 25 psig and ranges
upward to 40 psig. Above this operating range it appears that gradual
deterioration of filter performance often occurs. To operate within the
precoat pressure range it is necessary to have properly conditioned raw
sludge applied to the filter and to have had the proper quantity of ash
precoat.
146
-------
Improperly conditioned raw sludge will cause some buildup on the media
and thus gradually raise the successive precoat pressures. Sometimes
this problem can be corrected by dropping the quantity of precoat from
150 to 50 pounds thus forming a new cleavage plane encouraging the
discharged cake to remove some accumulated precoat. Improper quanti-
ties of precoat may cause a gradual buildup whereby the excess ash
creates a shear plane distant from the filter media so that when the cake
breaks away excess precoat remains on the media.
An operating experience which has reduced the precoat pressure, or at
least prolonged the need for a complete filter wash is that of extending
filter fill cycles. During the fill cycle filtrate water is pumped into the
filter and by continuing recirculation of the overflow it is sometimes
possible to perform some degree of washing. This practice is not totally
successful, but it may reduce the precoat pressure by 10 psig which may
extend the need for a complete filter wash for a few days.
Estimated filter runs between media wash is about 100 runs, or 150 to
200 filter hours of operation. The method of washing a pressure filter
could stand much improvement. The wash rod with a series of high
pressure nozzles originally furnished is no longer used. A more success-
ful system has been a single high pressure nozzle with a broad discharge
operating at about 750 pounds pressure. Commercial grade detergent is
used in the washing solution,
Equipment Evaluation
From the beginning of check-out considerable equipment difficulties were
experienced with the full scale plant. These are equipment problems
beyond those normally expected or experienced during start-up and check-
out. Many of these unusual problems related to poor workmanship and
fabrication details of the pressure filter and accessories equipment. These
problems were greatly compounded by manufacturers' personnel inexperienced
in detailed process and equipment performance. Cedar Rapids became a
training center for these people which accounts for the fact that it took
almost 18 months to approach a reasonable performance. Numerous
modifications to the mechanical features of the equipment have been
made, along with the electrical and process controls. Some of the poor
fabrication quality control, such as poor machining and assembly,
resulted in failures and were subsequently repaired or replaced with
improved equipment.
147
-------
Some of the major problems experienced, or which are still of concern
are mentioned here for purposes of objective reporting. It is not the
intent to criticize the equipment supplier at this late date, because in
fact, periodic improvements are still being made to the equipment at
Cedar Rapids by Beloit-Passavant as of January 1973. The equipment
supplier has continued to respond to the problems as they have occurred
and we have every reason to tnink that this cooperation will continue.
Pressure filter plate warpage has been a major problem from day one.
Some of this has been attributed to the early days of operation when the
plates were inadequately shimmed, however, adequate shimming of the
stay bosses has not totally eliminated this problem. Warpage occurring
in the plate diaphragm transfers bending to the plate frame which in turn
accelerates plate gasket deterioration due to warped gasket seating plane.
Plate alignment also affects gasket life wherein poor alignment, or reposi-
tioning of the plates each cycle, causes a reshaping of the plate gasket
leading to premature failure. Plate gasket life is short from whatever the
cause, and is still unsatisfactory in our opinion. .
Cake and incinerated ash grinders were originally installed because the
German process recommendations stressed this requirement. However,
experience at Cedar Rapids confirmed that cake grinding was undesirable
and that incinerated ash grinding was unnecessary. Filter cake broke and
crumbled sufficiently in the conveying system for delivery to the incinerator.
Cake passing through the grinder tended to knead to a plastic like ball and
under high moisture content of about 55 percent cake into the grinder came
out more as an extrudable paste. The cake grinder screen was first removed,
then the rotor as well, so that in effect no grinder exists. The incinerated
ash grinder was originally a 1/16 inch screen. This fine screen had a
tendency to clog even though the ash particles were much smaller. At
present the screen is f inch and serves as a trap metal safety device with
the rotor operating to assure through-put. No measurable effect has been
detected in ash particle size.
Sludge nuclear density meters are of considerable value to this process to
have an immediate sludge density reading to guide trie operator in setting
ash feed for proper ash/sludge ratio. The density meters originally in-
stalled did not hold calibration. This seems to be the major complaint of
most sludge nuclear density meters. New meters are on order, and some
calibration techniques developed in the field by plant personnel will be
used which have proven more dependable than those practiced by the meter
manufacturers.
148
-------
Ash feeders used at Cedar Rapids are of the gravimetric type. These
were installed for accurate proportioning of ash. Experience in handling
both fly ash and organic ash would suggest that this refinement is not
necessary and that volumetric feeders would be satisfactory and probably
more trouble free. Feeder problems are associated with the unpredictable
bulking and compaction characteristics in storage. It is difficult to cover
both extreme properties of ash in the performance range of the storage
bin activators and ash feeders.
Ferric chloride feeders were very inconsistent throughout the program,
however, this was primarily due to improper sizing and not really a fault
of the equipment. The feeders were oversized and operated on the low
range where repeatability was poor.
Materials handling equipment, particularly filter cake conveyors were
the cause of many problems. The primary cause of nearly all conveyor
problems was the fact that no experience had existed in discharging
filter cake to bunker storage. The bulk density of broken filter cake was
greatly underestimated at about 48 pounds per cubic foot. This is
reasonable for filter cake as it is sheared and bulked. However, in
bunker storage the cake recompacted dropping into the bunker so that the
bulk density was about 83 pounds per cubic foot at the drag conveyor
discharge.
Filter feed pumps are hydraulic driven ram pumps having a variable cap-
acity, variable head characteristic. Pumps having these characteristics
are desirable as the filter cycle pressure develops and the input dimin-
ishes, however, a more suitable primary pump should be provided to
meet the early demands of the filter, particularly on a filter installation
as large as Cedar Rapids. Considerable experimentation was carried
out by the plant operating personnel wherein filter performance was
greatly improved by placing all four filter feed pumps on one filter rather
than the normal two units. Prolonged slow feed rates to the pressure
filter to form satisfactory cake development is uncertain from our observa-
tions with sewage sludges and ash precoat. Where a precoat system is
not used, it may be desirable and even necessary to slowly develop a
cake on the cloth to provide a protective zone so that the solids are not
driven into the cloth upon increasing pressure. With a precoat system
the protective zone is already established and through-put should be as
rapid as possible. Increased filter supply could also be accomplished
by increased equalization tank capacity, but in that system some prob-
lems of maintaining air volume, level controls, solids suspension, and
mechanical controls offset the advantages of additional direct pumpage
to the filter..
149
-------
The mix tanks have mechanically failed in many ways from broken shafts,
bearing failures, gear misalignment, stuffing boxes, blades falling off
and others. Many of these failures have been attributed to poor shop
workmanship and probably could happen to any manufactured product at
some time. Modifications and repairs have been made and are contin-
uing and hopefully will be eliminated with better design and fabrication.
The pressure filter has been in service in this country for fifty plus years
in industrial processes. It has only been in recent years that its applica-
tion to sewage sludge dewatering has been seriously considered in Europe
and more recently in the United States.
The pressure filter is nearly the same age as mechanical type waste water
treatment yet it has taken almost as many years to develop a serious in-
terest to its application to sewage sludge dewatering.
Equipment and process improve with experience of people in specific
applications, therefore, it is reasonable to have confidence that the
pressure filter application to sewage sludge dewatering will be improved
with time and that specific experience.
At Cedar Rapids the pressure filter process has performed well, with many
of its problems solved by the human beings who created those problems
in the first place.
150
-------
SECTION XI
ACKNOWLEDGMENTS
The participation and contributions of Mr. George W. Milligan, P.E.,
Water Pollution Control Plant Engineer, and Mr. Sabry Kamhawy,
Chief Chemist, are particularly noted and were vital to the preparation
of this report.
A special thank you is extended to the entire staff of plant operators,
chemists, and maintenance personnel who contributed and cooperated
in the success of this program.
The Honorable Mayor and City Council have been active in this program
from the preliminary planning through the final report as has the
Director of Public Services. These dedicated and concerned people
authorized the program which otherwise would never have been under-
taken.
Donald J. Canney, P.E., Mayor
Commissioners
John D. Oberthien Richard L. Phillips
Stanislavs Reinis Harold G. Schaefer
Richard H. Jensen, P.E., Director of Public Services
Appreciation is extended to the Iowa Electric Light and Power Company,
Cedar Rapids, for their civic spirit in providing fly ash at no cost to
the City. Particular appreciation is extended to Mr. Lou Stolba, Super-
intendent of the 6th Street Power Station for his generous cooperation,
often at great inconvenience to him and his staff to make the fly ash
deliveries possible.
Mr. Ralph G. Christensen, Project Officer, EPA Region V, Chicago,
and Mr. James E. Smith, Jr., Technical Assistant, EPA, AWTRL,
Cincinnati, during numerous program conferences offered experienced
advice and technical support and for this, a sincere thanks.
151
-------
SECTION XII
A/S
BOD 5
CC (cc)
CFM (cfm)
COD
Cu. Ft.
Ft.
Gals/Min/Sq.Ft.
Gals/SF/Hr.
In.
In. W.C.
KWH
Lbs. (or#)
Lbs./Cu. Ft.
Lbs./SF/Hr.
(*
mg/1
ml
mm
MPR
%
PH
Psig
r
r
Rev.
S.S.
Sq. Ft.(S.F.)
T.F.
T.S.
V.S.
Wt.
GLOSSARY OF TERMS
Ash to Sludge Ratio
5 Day Biochemical Oxygen Demand
Cubic Centimeter
Cubic Feet per Minute
Chemical Oxygen Demand
Cubic Feet
Feet
Gallons per Minute per Square Foot
Gallons per Square Foot per Hour
Inches
Inches Water Column
Kilowatt Hour
Pounds
Pounds per Cubic Foot
Pounds per Square Foot per Hour
Micron
Milligrams per Liter
MillilUer
Millimeter
Minutes per Revolution
Percent
Hydrogen Ion Concentration
Pounds per Square Inch Gauge
Correlation Coefficient
Specific Filter Resistance (10l2Cm-2)
Revolution
Suspended Solids
Square Feet
Trickling Filter
Total Solids
Volatile Solids
Greater than
Less than
Weight
153
AU.S. GOVERNMENT PRINTING OFFICEH973 546-308/4 1-3
-------
i
Accession Number
W
2
Subject Field &. Group
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
I Organization
City of Cedar Rapids, Iowa
Title
Pressure Filtration of Waste Water Sludge With Ash Filter Aid
]Q Authors)
James W. Gerlich, P.E.
Howard R. Green Company
Consulting Engineers
Cedar Rapids, Iowa
16
21
Project Designation
11060
EZX
Afofe
edar Rapids, Iowa
22 Citation
Environmental Protection Agency report number,
EPA-R2-73-231, June 1973.
Descriptors (Starred First)
Sludge Dewatering
Ash Filter Aid
Pressure Filtration
Identifiers (Starred First)
27
Abstract
Cedar Rapids, Iowa used pilot plant studies as an effective approach to an eco-
nomic solution to dewatering secondary digested sludge. After piloting several dewatering
Processes, the pressure filter system was selected and a full scale plant was constructed.Dur-
In9 the course of the dewatering studies it was observed that fly ash was an effective filter aid.
The full scale plant was designed to utilize sludge ash from incinerated filter cake, as
as power plant fly ash. The design capacity is 28 tons of dry sewage solids for 16 hour
n at 48 percent dry solids cake.
Performance data from the full scale plant was evaluated over a period of approximately
lne months. Both fly ash and sludge ash were evaluated as a filter aid, with and without chemr
^s- Economic evaluations were made of operation and equipment.
Some of the conclusions are: (1) Pressure filtration of waste water sludges is an effective
economical process. (2) Ash filter aid increases dewatering production and decreases chemi-
Cal costs. Sewage sludge ash from an incinerator can be recycled to process; power plant fly
ash is even more effective sludge conditioner. (3) A detailed pilot plant program is of great
value in design of a full scale plant. (4) Some chemicals in combination with ash filter aid
urther improve dewatering efficiencies.
Abstractor
I. W. Gerlich, P.E,
Institution
Howard R. Green Company
(F*EV
SEND. W.TH COPV Or DOCUMENT. TO, WATER. RESOURCED £NT |F ^NFC-MAT 'ON CENTER
: WASHINGTON. D. C. 30240
* r.POI I 970-360-831
------- |