EPA-670/2-74-017
March 1974
Environmental Protection Technology Series
VALUATION AND DEMONSTRATION
OF THE CAPILLARY SUCTION SLUDGE
DEWATERING DEVICE
National Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
EPA-670/2-74-017
March 1974
EVALUATION AND DEMONSTRATION OF THE CAPILLARY
SUCTION SLUDGE DEWATERING DEVICE
By
T. E. Lippert
M. C. Skriba
Westinghouse Electric Corporation
Research and Development Center
Pittsburgh, Pennsylvania 15235
Program Element 1B2043
Project Officer
J. E. Smith, Jr.
Advanced Waste Treatment Research Laboratory
National Environmental Research Center
Cincinnati, Ohio 45268
NATIONAL ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
NOTICE
The Rational Enviroxnental Research Center —
Cincinnati baa rerleved this report and approved
its publication. Approval does not signify that
toe contents necessarily reflect the views and
policies of the U. 8. Environmental Protection
Agency, nor does mention of trade naaes or con-
nercial products constitute endorsement or recom-
nendation for use.
li
-------
FOREWORD
Man and his environment must be protected from the
adverse effects of pesticides, radiation, noise and other
forms of pollution, and the unwise management of solid
waste. Efforts to protect the environment require a
focus that recognizes the interplay between the com-
ponents of our physical environment—air, water, and
land. The National Environmental Research Centers
provide this multidisciplinazy focus through programs
engaged in
• studies on the effects of environmental
contaminants on man and the biosphere, and
• a search for ways to prevent contamin-
ation and to recycle valuable resources.
The research reported here was performed for the
Ultimate Disposal Section of the Advanced Waste Treatment
Research Laboratory to develop and demonstrate a new
method of activated sludge (concentrated pollutant stream)
dewaterlng. Since sludge hMHUwg and disposal represents
a significant part of the total wastewater treatment cost,
a new dewatering technique which promises higher cake
solids at higher loading rates and a cleaner filtrate at
reduced cost is very welcome.
A. W. Breidenbach, Ph.D.
Director
Rational Environmental
Research Center, Cincinnati
in
-------
ABSTRACT
A device for dewaterlng waste activated sludge that uses the capillary
suction in a porous belt has been demonstrated on a pilot scale test
unit. The system uses capillary action to initially dewater the sludge
and form a thin sludge cake which is then mechanically compressed in a
final step to obtain a still dryer cake. Both dewatering steps are
incorporated into a single system endless belt design. Sludge feed
capacities from 2 to 4.5 Ibs/hr ft2 (10 to 22 Kgs/hr m2) have been
demonstrated with the cake solids at discharge ranging from 15 to 18
percent. These results were obtained with a conditioned waste activated
sludge at a coagulant cost of approximately $4.00/ton ($4.40/metric ton).
The device has also been tested on an anaerobically digested and a
mixed primary-activated sludge.
This report was submitted in fulfillment of Research Contract 68-01-0094
under the sponsorship of the Environmental Protection Agency, Advanced
Waste Treatment Research Laboratory,
iv
-------
CONTENTS
SECTION PAGE
I CONCLUSIONS AKD RECOMMENDATIONS 1
II INTRODUCTION 2
III EXPERIMENTAL APPARATUS 5
IV DISCUSSION OF RESULTS 21
V REFERENCES 55
VI APPENDIX A - Error Analysis 56
VII APPENDIX B - A Chronological Summary Listing of
Test Data - Pilot Test Unit 60
VIII APPENDIX C - Economics of Capillary Dewatering of
Waste Activated Sludge 66
v
-------
FIGURES
No. Page
1 Schematic of Research Capillary Dewatering Unit .... 3
2 Laboratory Capillary Dewatering Unit .... 6
3 Capillary Dewatering Pilot Test Unit 7
4 Schematic of Demonstration Test Unit 8
5 Schematic of System Set-Up 9
6 Westinghouse Test Facility at the Long Road Sewage
Plant 11
7 Long Road Sewage Treatment Plant - Flow Schematic ... 12
8 Sludge Specific Resistance - Waste Activated Sludge . . 1^
9 Sludge Specific Resistance - Mixed Activated Primary . 15
10 Sludge Specific Resistance - Anaerobically Digested . . 16
11 Endless Belt Life Test Apparatus for Evaluating Porous
Media Swatch Samples 17
12 Porous Media Swatch Samples Before Entering Sewage . . 18
13 Sludge Capillary Dewatering Curves 24
14 Capillary Dewatering-Machine Capacity with Waste
Activated Sludge 25
15 Capillary Dewatering - Comparison of Digested and Mixed
Sludge Data with Design Curve for Waste Activated
Sludge 27
16 Capillary Dewatering - Calculated Machine Capacity . . 28
17 Capillary Dewatering - Calculated Machine Capacity . . 30
18 Capillary Dewatering - Calculated Machine Capacity . . 31
19 Capillary Dewatering - Calculated Machine Capacity . . 32
20 Correlation of Machine Apparent Capacity with Sludge
Specific Resistance for Two Coagulant Types 33
-------
FIGURES
No.
21 Schematic Arrangement of Swatch Test Apparatus for
Simulating Capillary Dewatering .............. 35
22 Belt Pore Size Effect on Capillary Dewatering Cake
Solids ........................... 36
23 Porous Felt and Reinforcing Scrim ............. 37
24 Five Basic Porous Material Swatch Samples Before and
After 30 Days Testing ................... 38
25 Swatch Samples After 3 1/2 Months ............. 39
26 Swatch Samples After 12 Months .............. **0
27 Photograph of Porous Belt From Pilot Unit After 12
Months of Testing ..................... k2
28 Capillary Dewatering - Final Cake Solids ......... W-
29 Effect of Coagulant Dosage on Final Cake Solids ...... ^5
30 Photograph Depicting Sludge Cake Detachment ........ 47
31 Effect of Screen Mesh Size on Solids Capture ....... k8
32 Capillary Dewatering - Sludge Cake Solids Capture ..... ^9
33 Capillary Dewatering - Sludge Cake Solids Capture ..... 51
34 Capillary Dewatering - Sludge Cake Solids Capture ..... 52
35 Capillary Dewatering - Sludge Cake Solids Capture ..... 53
vii
-------
TABLES
No . Page
1 Sludge Properties .................... 10
2 Sample Data Sheet ................... 19
3 Expected Uncertainty in the Calculated Quantities .... 20
4 Summary of Pilot Plant Capillary Dewatering System
Performance ....................... 21
5 Porous Material Physical Properties
1-A Expected Uncertainty in Measured Variables
1-B Pilot Test Unit - Summary of Data From May 22 Through
August 10, 1972 - System Installation and Shakedown . . 6l
2-B Pilot Test Unit - Summary of Data From August 16
Through October 4, 1972 - Base Line System Performance . 62
3-B Pilot Test Unit - Summary of Data from October 11
Through November 16, 1972 - Initial Study of Machine
Parameters on Waste Activated Sludge .......... 63
4-B Pilot Test Unit - Summary of Data From December 1
Through January 1, 1972-73 - Continue Study of Machine
Performance on Waste Activated Sludge ......... 6*4-
5-B Pilot Test Unit - Summary of Data From January 26
Through February 14, 1973 - Continue Study of
Machine Performance with Digested and Mixed Primary
Sludges ........................ 65
1-C Cost Estimate for Capillary Dewatering ......... 68
vlli
-------
ACKNOWLEDGMENTS
This project was conducted at the Westinghouse Research and Development Center
by Dr. T. E. Lippert, and Mr. M. C. Skriba, under the management of Mr. R. M.
Chamberlin, Manager of the Ecological Systems Department.
The support of the project by the Environmental Protection Agency and the many
useful contributions made by Dr. J. E. Smith, Jr., Project Officer, is greatly
appreciated.
The bulk of the experimental testing was conducted at the Westinghouse test
facility located at the Long Road Sewage Treatment Plant, Penn Hills, Pittsburgh,
Pennsylvania, under the capable guidance of Mr. R. F. Watson. The cooperation
of the sewage plant personnel, Mr. Ronald Young, Mr. Ernest O'Malley, and
Mr, Matthew Drop and the cooperation of the Penn Hills Municipal Authorities
is greatly acknowledged.
Sampling and analysis throughout the test program were performed by Mr. M. J. Testa,
Mr. J. W. Parks, and Mr. C. G. Slater. The typing of the manuscript was done by
Miss Darlene Kuszyk.
iac
-------
SECTION I
CONCLUSIONS AND RECOMMENDATIONS
A device for dewatering waste activated sludge that uses the capillary
suction in a porous belt has been demonstrated on a pilot scale test
unit. The device was also tested on an anaerobically digested and a
mixed primary/activated sludge. Feed capacities from 2 to 4.5 Ibs d.s./
hr ft2 (10 to 22 Kgs/hr m2) were achieved with the cake solids at
discharge ranging from 15 to 18 percent for activated sludge. Machine
capacities more than twice these values may be possible. It was found
that the device can be operated without coagulant addition at a penalty
to solids capture. With chemical addition of the polyelectrolyte at
10 Ibs/ton (5 Kgs/metric tons) or about $4.00/ton ($4.40/metric ton)
a cake solids capture of 95 percent can be obtained when the machine is
operated at 2.0 Ibs/hr ft2 (10 Kg/hr m2) . Operating at higher machine
capacities, i.e., 4.0 Ibs/hr ft2 and higher, it appears that the ferric
chloride conditioner yields more economical system operation. Overall
machine operation was found to depend on chemical addition, sludge solids
loading and screen mesh size.
This study has led to the development of a sound engineering understanding
of the operation of this device including the advancement and development
of specific design curves applicable to the particular waste activated
sludge used. It is recommended that supporting data from other plants on
waste activated sludge be developed to verify these results. Furthermore,
additional data and analysis is required to optimize system capacity
with respect to both solids loading and chemical addition. Present data
Indicates that such an analysis could lead to developing machine capacities
as high as 20 Ibs/hr ft2 (100 Kg/hr m2). Additional screen evaluations
are needed to establish if screen type and design significantly effect
the capillary dewatering rate and sludge cake detachment (i.e., prefer-
ential adherence to the compression roller as opposed to staying attached
to the screen). Operation of a full scale field unit is required in
order to establish realistic operating and total cost figures.
- 1 -
-------
SECTION II
INTRODUCTION
Frequently a sludge must be dewatered before it can be satisfactorily
disposed of, and the dewatering process is one of the most costly
involved in the handling and disposal of sewage solids. The quantities
of sludges to be handled and the difficulties in handling and disposing
of these sludges increase directly with the quality of the treated
wastewater product.(1) The Federal Water Pollution Control Act Amend-
ments of 1972 provide for improved wastewater quality including secondary
treatment or its equivalent for municipal discharges by 1977, (2)
In most cases this will require treatment plants to include biological
treatment although in some it may only mean upgrading existing processes
with chemical addition. Either alternative will cause more sludge to
be produced, and this sludge will probably be more difficult to process.
Sludges which present a problem to mechanically dewater include waste
activated and chemical sludges. This is partly because of their
diluteness and partly because of their gelatinous nature. Presently
very few municipalities will dewater only a waste activated sludge,
even after first thickening it. It is almost always first combined with
a primary sludge, even though it is well known that the presence of waste
activated sludge increases the cost of dewatering primary sludge.
Conventional dewatering equipment is unable to dewater straight waste
activated sludge unless prohibitive quantities of chemicals are added.
The Westinghouse Electric Corporation over the last three years has been
Involved in developing a unique method to dewater dilute sludges. This
work over the last year and a half has been supported under EPA contract
specifically to answer the need for an economical mechanical dewatering
system for use with unthickened waste activated sludge. Although thickened
activated sludge and other sludge types were also evaluated, the proposed
system should produce a truckable cake and/or a cake capable of under-
going combustion with minimum fuel requirements.
The system under development uses capillary action to initially dewater
the sludge and form a thin sludge cake and it then employs mechanical
compression to obtain a still drier cake. Both these dewatering steps
are incorporated into a single system, endless belt design. A schematic
of the device is shown in Figure 1. The endless belt portion of the device
is actually comprised of two belts; a belt made of a porous material
approximately 1/2 in. (13 mm) thick and a fine mesh screen belt.
To operate this device, the sludge is generally preconditioned, and then
pumped onto the top at one end of the screen belt in a thin layer while
both the porous and screen belts move towards the opposite end. The
porous belt provides the capillary action for dewatering while the
screen belt carries the sludge and serves to minimize clogging of the
- 2 -
-------
Sludge
Cake
Sludge
Compression
Roller
Sludge Feed
Capillary Dewatering Section
Sludge
Cake
u>
I
Belt & Screen Motion
mutt Hit Itmimiiiiill
Screen
Wash
Cake
Filtrate Porous
Belt
Screen
Belt
Fig. 1 -Schematic of research capillary dewatering unit
-------
belt pores by the sludge solids. The porous belt, screen belt, and
sludge are in direct contact which allows the free water associated with
the sludge to be drawn preferentially by capillary forces through the
screen belt into the porous belt. The sludge solids remain on the
screen. By the time the sludge has reaqhed the end of the capillary
section of the machine, it has been transformed from its initial liquid
form to a semi-solid thin sludge cake containing typically from 6 to 10
percent solids by weight. At this point, the sludge cake is passed
under a compression roller causing additional water to be extracted. It
has been found that the final solids of the cake depend on many factors
but a 16 to 20 percent cake is typical for a waste activated sludge.
This compression roller also serves as the sludge cake removal mechanism.
The sludge cake after being compressed attaches preferentially to the
roller surface and is lifted away from the screen belt. This provides
for an effective method of detaching the thin sludge cake and thereby
maintaining a high solids recovery. The water that was extracted from
the sludge is contained within the porous belt and is subsequently
squeezed-out and the belt and screen returned to the feed position.
The practice of using capillary action for solid liquid separation is
not new (3,4), nor is the application of belt and sludge presses to
sludge dewatering.(5) However, the capillary suction device described
above is an improvement over prior systems in that it incorporates both
these process operations in an arrangement that allows effective sludge
dewatering and cake detachment with a minimum of energy expended. More-
over, the results of this study provide an improved technological basis
of the design and performance of this capillary suction device.
The scope of this program has included both an evaluation of the basic
mechanical functions of the device as well as a determination
of specific design relationships. Machine performance on a pilot scale
system was evaluated on three sludge types; the evaluation included
first a determination of overall machine performance, such as system
capacity and final cake solids, and secondly an evaluation of certain
design parameters such as screen mesh size, capillary belt pore size,
and compression roller force.
-------
SECTION III
EXPERIMENTAL APPARATUS
CAPILLARY DEWATERING APPARATUS
Two experimental capillary dewatering test units have been operated.
The first unit, shown in Figure 2, is a small scale system approximately
4 inches (10.1 cm) wide and 2 1/2 ft. (0.76 m) long (1 lit/min max.
sludge feed rate). During the course of this program, this unit was used
to generate the basic data for evaluating the mechanical operation of
the system and determining parametric interactions. The second unit
(or pilot unit), shown in Figure 3, is about 1 ft. (0.30 ra) wide and
5 ft. (1.5 m) long and has six times the capacity of the smaller unit.
This unit was built later and operated at a local sewage treatment plant
to determine overall system operation and to substantiate results obtained
from the smaller unit.
A schematic layout of the pilot test unit is given in Figure 4. Originally,
this unit was designed so the porous belt and screen belt would track
over separate drive rollers. This was later changed so that both the
porous belt and screen belt track over the aft drive roller. The porous
belt is driven by a combination of the tension force exerted around the
aft roller and the drive force imparted to the belt by the set of
compression rollers indicated in the schematic. This drive arrangement
allows a minimum tension force on the porous belt thereby eliminating
any belt stretch problems. The set of belt compression rollers also
serve as the method of extracting the belt water (cake filtrate). Three
water spray nozzles are located beneath the aft drive roller and are used
to wash the screen belt which, at this location, is separated from the
porous belt. The cake filtrate and wash water streams are maintained and
sampled separately. The porous and screen belts are driven from a 1/2 hp
(40 Kg-m/sec) variable speed motor using chain and sprocket drives.
Screen material used in the program was a dacron polyester supplied by
Wire Cloth Enterprises, Pittsburgh, Pa. Porous belt material was a nylon,
needle punched felt of approximately 7 oz/ft2 density (2.2 Kg/m^)
supplied by Orr Felt Company, Piqua, Ohio.
The general arrangement of the test unit, sludge and chemical feed
pumps is given in Figure 5. The sludge was drawn from the sewage plant
at the return line using a positive displacement 2.0 gal/min (7.5 lit/min)
pump and a 2-inch (52 mm) line. Sludge for the dewatering unit was
drawn from this line via a smaller (0-6 lit/min) pulsation pump. The
unused sludge was circulated back to the primary treatment tank. This
arrangement provided fresh sludge continuously. Coagulant was prepared
in stock solutions and pumped from a holding tank into the sludge line
on the output side of the pulsation pump. There were approximately 3 ft.
(1 m) of 3/4 inch (19.0 mm) tubing in which the sludge and coagulant
were mixed before spilling into the feed tray. The feed tray utilized a
splitter plug to dampen flow variations, thus assuring a uniform sludge
layer onto the dewatering unit.
- 5 -
-------
Fig. 2 - Laboratory capillary dewatering unit
-------
Fig. 3 - Capillary dewatering pilot test unit,
- 7 -
-------
I
03
Belt And Screen
Adjustment Feed
Reservoir
Capillary
Dewatering
Section
Sludge
Compression
Roller
Compression Rollers
(Alternate Position)
Scale 1cm ±.13 Meters
Fig. 4- Schematic of demonstration test unit
-------
SO
I
Coagulant
Feed Tank
Sludge Bypass
To Drain
Moyno
Sludge Pump
t
IL
Feed
Pump
Test Unit
Cake Filtrate
to Drain
Sludge
Feed
Pump
Screen Wash
Water to Drains
Bypass to Drain
Sludge Cake
after Dewatering
Input Sludge from Treatment Plant
Flow
Meter
House Water
Fig. 5-Schematic of system set-up
-------
LONG ROAD SEWAGE PLANT FACILITY
Available for the study of municipal sludges was a Westlnghouse operated
test facility, shown in Figure 6, located at the Long Road Municipal Sewage
Treatment Plant. The plant, located in Perm Hills, Pennsylvania, uses the
spiral diffused aeration activated sludge process to treat 1 mgd (4 million
liters/day) of municipal sewage. Figure 7 is a schematic of the treatment
process and illustrates the flow of waste. Recycle activated sludge from
the final clarifier is wasted through a primary clarifier. Sufficient
activated sludge solids are wasted so that a sludge age of three days can
be maintained in the aerator. The combined sludge solids from the primary
clarifier are digested and dewatered by a vacuum drum filter. Available
for this study, therefore, for processing by the capillary dewatering unit
were activated, mixed primary, and digested sludges. Table 1 below lists
the average sludge characteristics, which were compiled from data taken
during this test program. Plant influent BOD and suspended solids averaged
150 and 180 mg/1 respectively.
It should be noted that this activated sludge is relatively high in solids
content. They carry an unusually high aeration cell miss concentration and
achieve good settling characteristics. Because of the high solids content,
the sludge may be partially stabilized aerobically in the aeration cell
resulting in low values of mlvss. This plant feed also is almost exclusively
municipal in character with practically zero industrial load which may account
for the relatively low dissolved solids level. Tap water TDS in this area
normally is in the range of 200 mg/1.
Table 1
Total solids (mg/1)
Total volatile
solids (mg/1)
Dissolved solids
(rag/1)
Dissolved volatile
solids (mg/1)
PH
Sludge specific
resistance
(sec2/gm) x 10~8
Sludge type
Activated
19,545
10,620
840
200
6.9
17.3
Mixed
primary
40,480
23,310
1,572
646
6.3
40.0
Digested
48,235
23,425
1,280
860
7.1
3.89
- 10 -
-------
Fig. 6 - Westinghouse test facility at the Long Road Sewage Plant
-------
NJ
I
Anaerobic
DFgester
1st Stage
Preaeration
2nd Stage
Sludge
Holding
Aeration Tank
Chlori nation
Tank
Com mm uter
Secondary
Clarifier
Aeration Tank
Vacuum
Filter
Return Activated Sludge
Waste Activated Sludge
Filtrate From Vacuum Filter
:Treated Water
To Stream
Fig. 7-Long Road Sewage Treatment Plant - flow schematic
Waste
Solids
"Trucked
To Land
Fill
-------
To further characterize these sludges, specific resistance determinations
were made on each sludge type according to the procedure described in
Reference 6. The specific resistance values recorded in the table above
are for no coagulant addition. Figures 8, 9, and 10 show the specific
resistance values of these sludges as a function of coagulant dosage.
Several different coagulants were evaluated. This data was used to set
approximate coagulant levels for the sludges used in tests on the capillary
dewatering device.
BELT-LIFE TEST APPARATUS
As part of this program, a continuing life test evaluation on several swatch
samples of different porous media was made using the apparatus shown in
Figure 11. Photographs of the swatch test samples are given in Figure 12.
The various samples were attached to the endless metal belt. The apparatus
was partially filled with activated sludge (about 0.3m depth) and the swatch
samples tracked through the sludge. The sludge was continuously aerated and
also changed approximately every three days. A small compression roller was
situated such that each swatch sample would pass under the roller and be
slightly compressed. This was done in order to approximately simulate the
mechanical loading expected in the actual capillary dewatering device. This
test apparatus was operated about one year.
TEST PROGRAM AND PROCEDURE
The test program on the pilot scale capillary dewatering unit was conducted
at the Westinghouse facility located at the Long Road Sewage Plant. This
test program included operation of the capillary dewatering device on activated,
mixed primary-activated, and anaerobically digested sludges. The experimental
design was organized such that an overall machine performance and operation
could be determined as well as an evaluation of specific system operating
parameters such as coagulant dosage effects, belt speed, roller compression
force, etc. Day to day test runs generally lasted from 2 to 6 hours. More
than 300 total hours were accumulated in the operation of the test unit.
Start-up procedure involved preselecting belt speed, sludge feed rate, and
coagulant dosage. The compression roller force was selected (except in
certain specific tests) such that satisfactory operation of the compression
station could be maintained. This generally involved manipulating the roller
spring deflection to achieve the best cake detachment efficiency possible,
while, at the same time, obtaining the highest cake solids. These judgements
were based on prior experience.
Test data was taken only after sufficient time (usually 15 to 30 minutes)
was allowed for attaining equilibrium operation. This same procedure for
data taking was practiced whenever any of the operating parameters were varied.
Routine test measurements included sludge feed rate, coagulant feed rate,
sludge cake dry solids, screen wash water flow rate, cake filtrate flow, and
compression roller spring deflection. In addition, the feed sludge was sampled
and analyzed for total, volatile, dissolved solids and dissolved volatile
solids. These constituents were analyzed in accordance with procedures outlined
- 13 -
-------
Total Solids Coagulant Date
18(400mg/je C-31 10/71
23,300mgft C-31 8/71
15,800mg/l FeCf 2/72
CXJ
2X10
2 4 6 8 10 12 14 16 18 20
Coagulant Concentration (Ibs/ton)
C-31 Polyelectrolytex 1.0
FeCI3 x 10-1
Fig. 8-Sludge specific resistance - waste activated sludge
(Kgs/tonlm) =2.0 Ibs/ton)
-14 -
-------
CM
O)
o
c
ro
U
O>
10
8
6
8
10
7
Total Solids
38,739 mg/l
Coagulant
ST-260
Polyelectrolyte
Date
9/31/71
I , I L_l
J L
0 5 10 15 20 25 30 35 40 45 50
Coagulant Concentration (Ibs/ton)
Fig. 9-Sludge Specific Resistance - Mixed Activated - Primary
(Kgs/ton(m)=2.0 Ibs/ton)
- 15 -
-------
CVJ
c
CD
^U
*o
10
8
6
10
10
8
Total Solids
35,476mg/e
Coagulant
ST-260
Polyelectrolyte
Date
9/31/71
J_
1
J L
J
0 5 10 15 20 25 30 35 40 45 50
Coagulant Concentration (Ibs/ton)
Fig. 10-Sludge Specific Resistance - Anaerobically Digested
(Kgs/ton)m)=£0 Ibs/ton)
-16-
-------
Fig. 11 - Endless belt life test apparatus for evaluating porous
media swatch samples.
-------
1
t
£
Fig. 12 - Porous media swatch samples before entering sewage.
-------
Date:12/4/72
Sludge Characterization
Type sludge Activated
Total solids 19.500 mo/1
Run: 1204
Cross Reference Nos.: 1766190-91
Machine Parameters
Sludge feed rate 6.5 1/min
Total dissolved solids 9800 mg/1
Dissolved solids 800 mg/1
Belt-sludge contact time 3.5 sec.
Coagulant type C-31
Dissolved volatile solids 125 mg/1
pH 6.9
Coagulant concentration
Run time
18.7 ///ton
start 217.7 finish 219.2
Sample
No.
930
931
932
933
934
935
936
937
938
939
*Roller st
Inches
Along
Belt
5
10
15
20
25
30
35
40
50
mple for tc
Sludge Cake Dewatering
Characterization A
Before After Other
Comp . Comp . g/mln
T.S.
5.9
8.3
9.8
10.4
10.6
10.7
11.0
11.0
11.0
tal soli
T.S.
14.5
ds balanc
T.S.
112.8
e.
Filtrate
Characterization
Belt Water Screen Wash
Water
SKffii,
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
T.S.
2920
Quan.
tt/min.
T.S.
Machine Performance
Solids Loading
(Ibs/ft2)
.0045
.0045
.0045
.0045
.0045
.0045
.0045
.0045
.0045
.0045
Solids
Recovery
%
88.6
92.2
Solids Loading
Rate
(Ibs/hr ft2)
4.63
4.63
Compression Roller
Load—in. deflection in springs 1/2 Inch (12.4 #/in)
Pickup efficiency
Belt squeegee_
NOTES: 300 mesh screen
Table 2 - Sample Data Sheet (after sample analysis)
-------
in the Standard Methods for the Examination of Water and Wastewaters or
EPA approved alternate methods. Sludge pH was also occasionally monitored.
A sample data sheet is shown in Table 2. The cake solids capture (indicative
of cake process water quality) was determined by two methods; first by
determining for a given time increment the actual dry sludge solids discharged
from the compression roller and comparing this with the measured solids in
the feed stream, and secondly by calculating cake solids capture from a mass
balance on the feed stream, sludge filtrate and wash water flows. These
two methods generally agreed to within a few percentage points.
An estimate of the error involved in the measurement and reproducibility of
each of the basic test parameters is given in Table 1-A in Appendix A. The
cumulative effect of these errors on the various calculated quantities is
given in Table 3 below. Further discussion of the error analysis procedure
including some sample calculations is also given in Appendix A.
Table 3. EXPECTED UNCERTAINTY IN THE CALCULATED QUANTITIES
Parameter
Typical value
1. Sludge feed rate
a. Activated sludge
Activated sludge
b. Digested sludge
c. Mixed primary
2. Machine capacity
a. Activated sludge
Activated sludge
b. Digested sludge
c. Mixed primary
3. Solids capture
a. Activated sludge
b. Digested sludge
c. Mixed primary
4. Coagulant concentration
Level
2.4 1/min
6.5 1/min
3.0 1/min
3.5 1/min
2.0 Ibs/hr ft2
5.0 Ibs/hr ft2
5.0 Ibs/hr ft2
5.0 Ibs/hr ft2
90%
85%
85%
4 Ibs/ton
(2 Kg/metric ton)
Expected maximum
deviation
±2.4%
+3.5%
+6%
±6%
+4.1%
+4.8%
+7%
+7%
+5.3%
-4.1%
+7%
-6.4%
+7%
-6.4%
+10%
(1 Ib/hr ft2 21 5 Kg/hr m2)
- 20 -
-------
SECTION IV
DISCUSSION OF RESULTS
SUMMARY OF CAPILLARY SUCTION DEVICE SYSTEM PERFORMANCE
The performance of most mechanical sludge dewatering devices is measured by
at least four parameters; system capacity (Kgs d.s./hr m^), final cake solids
(%), solids capture (%) and total costs ($/Kgs d.s.). Following the convention
used for rotary vacuum belt filters only the working area of the belt is used
in the calculation of machine capacity.
Sufficient test data has been generated from this program, both from the small
lab test unit and from the larger pilot unit, to indicate how these specific
performance parameters are influenced by changes in the various operating and
machine design variables. For example, trends have been determined indicating
how certain screen parameters affect sludge solids capture and therefore
coagulant costs or similarly, how the sludge solids loading affects dewatering
rate and thus machine capacity. A summary listing of all the data obtained
from the pilot unit is given in Tables 1-B through 4-B in Appendix B. These
data reflecting these parametric interactions are discussed in subsequent
sections of this report. However, before examining this data, a general
summary of machine performance is presented in order that the overall program.
results can be succinctly identified.
Table 4 below summarizes the performance data obtained with the pilot test unit
while operating at the Long Road Sewage Plant on waste activated, anaerobically
digested, and a mixed primary/activated sludges. This unit was operated for
approximately nine months and more than 300 actual operating hours were accumu-
lated during this study.
Table 4. SUMMARY OF PILOT PLANT CAPILLARY DEWATERING SYSTEM PERFORMANCE
Sludge type
WAS
WAS
WAS
WAS
WAS
DIGST
MIXPRIM
Machine capacity
(Ibs/hr ft2)
2.0
2.0
Coagulant
cost
($/ton)
0
4.00
Final cake
solids
(%)
16-19
17-19
(C-31, Polyelectolyte)
3.0
4.3
0
6.40
16
15
(Ferric Chloride)
4.5
4.00
14-15
(C-31, Polyelectrolyte)
5.2
10.00
16-18
(Ferric Chloride)
5.4
(Fe
10.00
rric Chloric
14-15
e)
Solids
recovery
«>
60-75
95
50-65
91
80
88
85
(Kgs/m2 = 0.205 lbs/ft2)
($/Ton = $1.I/metric ton)
- 21 -
-------
The results recorded in Table 4 typify machine performance on the various
sludge types at several coagulant dosage and sludge feed rates. The cost
figures reported in the table represent coagulant requirements only since
the capital and operating costs of this test unit are not indicative of a
full scale system. Estimates of this cost for a full scale system are given
in Appendix C. Coagulant costs were taken as $.AO/lb ($.88/Kg) for the poly-
electrolyte and $.05/lb ($.ll/Kg) for the ferric chloride.
As shown in the table, the test unit was operated both with and without
chemical conditioning. With the activated sludge at a feed rate of 2.0 Ibs
d.s./hr ft2 (10 Kgs/hr m2) and a coagulant demand (polyelectrolyte C-31)
corresponding to $4.00/ton d.s. ($4,40/metric ton), final cake solids ranged
around 18 percent with solids recovery as high as 95 percent. An 18 percent
cake represents about a 90 percent reduction in the sludge volume and a cake
of "truckable" consistancy. Final cake solids did not appear to depend signifi-
cantly on coagulant addition but did decrease slightly when the machine feed
rate was increased from 2.0 to 4.6 Ibs/hr ft2 (10 to 22 Kgs/hr m2). Cake
solids recovery, however, was found to depend strongly on coagulant addition
and to a lesser extent on feed rate. With no coagulant addition, solids
recoveries were found to vary from 60 to 75 percent at the lower feed rates
and measured as low as 50 percent at the higher rates. With coagulant addition,
cake solids capture improved dramatically; to 95 percent with the polyelectrolyte
at the lower feed rates. Comparable solids recoveries are possible at the high
feed rates but at an increased coagulant cost.
In general, it was found that the ferric chloride coagulant yielded better
overall performance i.e., dryer cake and better cake detachment than the poly-
electrolyte. To achieve comparable solids recoveries with the ferric chloride
conditioner, dosage rates around 200 Ibs/ton (100 Kgs/metric ton) are indicated.
Reported data on the mixed primary-activated sludge was limited to just one feed
rate and only the ferric chloride conditioner. A final cake containing 15 per-
cent solids with 85 percent solids recovery was achieved. This is a ferric
dosage rate of 246 Ibs/ton (123 Kgs/metric ton) with a machine loading of 5.4
Ibs/hr ft2 (26 Kgs/m2 hr).
2
With the anaerobically digested sludge, and a machine loading of 5.4 Ibs/hr ft
(26 Kgs/m^ hr), solids recoveries in excess of 90 percent were attained with a
ferric chloride dosage of approximately 300 Ibs/ton (150 Kgs/metric ton). The
final cake solids varied from 16-18 percent. Assuming a 5 cent/lb (11 cents/Kg)
cost for the ferric chloride, chemical costs for dewatering the mixed primary and
the digested sludges ranged from $10.00 to $15.00/ton ($11 to $l6/metric ton),
respectively.
Further consideration is given to the test data in Table 4 in subsequent sections
where it is shown how the machine capacity, final cake solids and cake capture
relate to the various machine operating and design variables. Test data obtained
from the smaller lab unit and from the belt life test unit is also discussed.
- 22 -
-------
MACHINE DESIGN AND SYSTEM CAPACITY
The physical size of this dewatering device is obviously determined by the
physical size of the capillary zone. Thus, the efficiency of the capillary
dewatering in forming the sludge cake represents an important design consider-
ation. With activated sludge, it has been learned through experience that an
8 percent (solids) cake is needed at the end of the capillary zone before
passing into the sludge compression zone. With wetter cakes, operational
problems with the compression roller may be encountered. Some examples of
quantitative measurements made of cake formation along the capillary zone are
given by the sludge dewatering curves shown in Figure 13. The data is for a
waste activated sludge with 2 percent solids that was conditioned by the
addition of 40 mg/1 of a cationic polyelectrolyte. This data was obtained by
determining the cake solids from sludge samples taken at various positions along
the length of the capillary section of the small test unit. Two sets of data
are shown, one for a solids loading of 0.005 lbs/ft2 (0.024 Kgs/m2) and the
other for a loading of 0.012 lbs/ft2 (0.58 Kg/m2). As seen, the two sets of
data exhibit a similar trend in that the rate of water extraction is highest
initially and decreases exponentially as the sludge becomes concentrated. The
curve for the lower solids loading shows a 10 percent cake is achieved in less
than a 10 second sludge-belt contact time while an 8 percent cake in 20 seconds
is indicated for the higher solids loading condition. Thus, both the cake
solids and the time needed to dewater the sludge depend on the sludge solids
loading. A system optimization must consider therefore both these parameters.
By developing a family of these sludge dewatering curves, a basis was established
for designing and operating the capillary section of the machine. The results
of this data for the activated sludge is given by the curve in Figure 14. This
curve shows machine capacity as a function of solids loading. The curve is for
a 2 percent waste activated sludge conditioned with 40 mg/1 cationic polyelect-
rolyte and, furthermore, assumes an approximate 9 percent cake at the end of
the capillary section. This curve was developed from data obtained from the
small lab unit while the data points shown represent actual operating data from
the pilot unit. The majority of these data points reflect machine performance
using the polyelectrolyte conditioner. One data point is shown however, in which
the ferric chloride conditioner was used.
The data in Figure 14 show that the highest machine capacities are achieved at
the lowest solids loading i.e., when the sludge is applied in a very thin layer.
Rates in excess of 6.0 Ibs/hr ft2 (30 Kgs/hr m2) are indicated corresponding to
a solids loading of 0.004 lbs/ft2 (0.019 Kgs/m2). It was found, however, that
when the solids loading was decreased below 0.003 lbs/ft2 (0.014 Kgs/m2) the
sludge layer was so thin that it would not distribute uniformly over the screen
belt and therefore surface area utilization was diminished. This would represent
then a lower operational limit on the sludge solids loading. An upper operational
limit on solids loading occurred when the sludge layer became too thick to be
effectively compressed by the roller. This occurred at about a loading of 0.03
lbs/ft2 (0.15 Kgs/m2).
-------
16
14
0
Solids Loading
0.005 (Ibs/ft2)
Solids Loading
0.012 (Ibs/ft?)
Sludge: Waste Activated
10 20 30
Belt-Sludge Contact Time (sec)
Fig. 13- Sludge capillary dewatering curves
z z
40
-------
7.0
r5.0
§4.0
s.
CO
2.0
1.0
FeCI3 - 127 IbsAon
Curve Developed from
Lab Test Unit, Fig. 2
operating Data from Pilot
Unit,
Polyelectrolyte - 4.0|bs/
ton
0.005 0.010 0.015 0.020 0.025
Solids Loading (Ibs/ft2)
0.030
Fig. 14- Capillary dewatering-machine capacity with waste activated sludge
(kgs/m2=0.205 Ibs/ft2, kgsAon(m) =2.0 fos/ton)
-25 -
-------
Of fundamental importance, the curve in Figure 14 shows that the actual
capillary dewatering phenomena is governed by several different rate limiting
mechanisms which depend on the sludge solids loading. At very low solids
loadings the sludge cake is thin and its resistance to dewatering is minimal.
Hence, the controlling factors are the rate at which water is absorbed by
capillary action into the belt and the total quantity of water that must be
transferred. As the solids loading is increased (i.e., a thicker sludge cake),
the cake resistance per unit of belt surface area increases in greater propor-
tion than the solids weight. Therefore, the dewatering rate decreases. This
continues until a point is reached where a pore flow is established in the
cake and further increases in solids loading no longer produce a proportionate
increase in cake dewatering resistance. This represents the minimum point
observed in the curve. Further increases in solids loading i.e. increase in
dry solids weight, result in increasing machine capacity. This continues until
the hydraulic capacity of the porous belt is exceeded.
Sufficient data was not obtained to develop design curves for the mixed primary
and the digested sludges. It can be argued, however, that the design curve
for these sludges should be very similar to that of activated sludge if these
latter sludges are similarly conditioned i.e., conditioned to allow free water
drainage. This observation is somewhat supported by the limited data that was
obtained for the mixed primary and the digested sludges as seen when these data
are compared with curve developed for the activated sludge case, Figure 15. The
significance of this result is that the design of the capillary section may be
independent of the initial solids concentration of the sludge providing (1) that
the system can be designed to handle the water hydraulics at the desired solids
loading and (2) the free water in the sludge can be extracted by capillary forces.
The first qualification applies to very dilute sludges while the second is a
limiting condition on the more concentrated sludge types.
The practical utility of the design curve in Figure 14 is two fold; first it
serves initial design purposes for determining the physical size of a unit for
a given plant size and secondly, the curve can be used to adjust machine
operating conditions i.e., belt speed to maximize capacity in any off-design
condition.
The relatively high machine capacities that are indicated by the data in Figure
14 are the result of minimizing cake dewatering resistance by controlling sludge
solids loading i.e., cake thickness. Another method of decreasing cake resis-
tance and therefore increasing machine capacity is by increased coagulant dosage.
This is seen from the data given in Figure 16 where the capacity of the pilot
test unit was determined as a function of coagulant dosage for a fixed solids
loading. The reference data points, shown unshaded, are machine capacities
calculated by sampling the sludge layer at different positions along the length
of the machine and determining the effective belt surface area needed to dewater
the sludge to about a 9 percent cake. This is the same basis used to develop
the curve in Figure 14. The results from Figure 16 show that when coagulant
dosage is increased from 4 to 20 Ibs/ton (2-10 Kg/metric ton) of the polyelec-
trolyte, machine capacity is increased from 4.6 to a maximum of 8.0 Ibs/hr ft
(40 Kgs/hr m2) but a further increase in coagulant diminishes capacity. Likewise,
- 26 -
-------
7.0
6.0
5.0
3.0
U
03
Q.
co
O
Q)
2.0
1.0
— Curve Developed from
Lab Test Unit, Fig. 2.
• Operating Data from Pilot
Test Unit, Anaerobically
Digested Sludge, FeCL
at 170 Ibs/ton
• Operating Data from Pilot Test
Unit, Mixed Primary, Fed
at 172 Ibs/ton
3
Note: Capillary Dry Cake
Solids Ranging from 9.5 to 11.5
0.005 0.010 0.015 0.020
SolidsLoading(lbs/ft2)
0.025 0.030
Fig. 15-Capillary de watering - comparison of digested and mixed
sludge data with design curve for wasted activated sludge
(Kgs/m
2
Ibs/ft, Kgs/ton(m) =2.0lbs/ton)
-27-
-------
11
10
9
2
1
Sludge: Waste Activated
Solids Loading: 0.0046 (Ibs/ft?)
Coagulant: Polyelectrolyte-
Cationic
Note: Calculations Based on a Sludge
Cake Dry Solids of 9 Percent.
10 20 30 40
Coagulant Concentration (Ibs/Ton)
50
60
Fig. 16- Capillary dewatering - calculated machine capacity
(kgs/m2 =dW5 bs/ft2, kg$/ton
-------
a decrease in coagulant concentration from 4 Ibs/ton (2 Kg/metric ton) to
zero decreases machine capacity to 3.0 Ibs/hr ft2 (15 Kgs/hr m2).
Similar data obtained in a second series of tests would indicate that an even
more dramatic increase in machine capacity can be realized by increasing
coagulant dosages. Capacities more than twice those suggested by the data
in Figure 16 were calculated i.e., capacities as high as 20 Ibs/hr ft2 (100
Kgs/hr m ). In this second case, the basis for the calculation was an 8
percent cake at the end of the capillary section. This calculated data is
shown in Figure 17 for both the use of polyelectrolyte and ferric chloride.
It must be cautioned that the calculated quantities are based on sludge samples
taken along the length of the capillary zone. Because of the expontial nature
of the sludge drying curves (see Figure 13) relatively small discrepancies in
the cake solids near the equilibrium value can represent large incremental
changes in the equivalent length of the capillary zone. This length value
figures directly into the capacity calculation. Actual operating data at
these high capacities is needed to verify the calculations.
A similar trend with coagulant dosage and machine capacity was obtained for
the mixed primary and the digested sludges, Figures 18 and 19. In these cases
a maximum four fold increase in machine capacity was calculated.
All of the above data were obtained for a relatively low solids loading. The
effect of coagulant dosage on capacity at a higher solids loading is unknown.
Thus, data at other sludge solids loadings with and without chemical additions
is needed to determine how these curves are affected. This information would
provide then a more comprehensive basis to optimize machine capacity. Further-
more, it remains to be determined if the reduced capital cost realized by
doubling the machine capacity with large coagulant addition exceeds the higher
chemical costs.
The behavior of machine capacity with coagulant concentration is not unlike that
observed with the sludge's specific resistance (a measure of sludge dewaterability).
During the course of this investigation, the specific resistance measurements
were made for each sludge type and the data are presented in Figures 8, 9, and
10. The data for the mixed primary and the digested sludges were obtained after
the sludges had been conditioned with a polyelectrolyte coagulant. As can be
learned from the figures, increasing the coagulant concentration improved
dewaterability in each case i.e., decreased specific resistance. This suggests
then, that the data reported in Figures 17, 18, and 19 can be combined using
specific resistance as a normalizing parameter. Such a correlation was made
for activated sludge using the data obtained for both the C-31 polyelectrolyte
and the ferric chloride i.e., the data from Figure 17. The results are shown in
Figure 20 where the machine capacity (Ibs/hr ft2) is plotted against sludge
specific resistance. Based on the limited data obtained, it appears that machine
capacity can be directly related to sludge specific resistance for learning the
effect of increasing coagulant dosages. A similar comparison for the other two
sludge types could not be made, since specific resistance values were not determined
for the use of ferric chloride with these sludges.
-29-
-------
30
Sludge: Waste Activated
Solids Loading: 0.0046fbs/«
Coagulant • poJyelectrolyte
A Fed3
20
15
QC
10
Note: Calculations Based on a Sludge Cake Dry Solids of 8 Percent
100 200 300 400
10 20 30 40
Coagulant Addition (Ibs/ton Dry Solids)
fig. 17-Capillary dewatering - calculated machine capacity
2 = 0.205lbs/ft2, Kgs/ton(m) =^0Ibs/ton)
300
50
600FeCL
60C-3I
-------
Sludge: Anaerobically Digested
Solids Loading: 0.006lbs/ft2
Coagulant: Fed.
9
8
7
6
5
4
3
2
1
Note: Calculations Based on a Sludge Cake
Dry Solids of 10 Percent Before Roller
100 200 300 400 500
Coagulant Concentration (Ibs/ton)
600
700
Fig. 18-Capillary de water ing - calculated machine capacity
(Kgs/m2 = 0..205 Ibs/ft2, Kgs/ton (m) =2.0 Ibs/ton)
- 31-
-------
20
CM
IE 15
i/i
CO
a-
o
o>
.£ 10
Sludge: Mixed Waste Activated and Primary
Solids Loading: 0.006Lbs/R2
Coagulant:
Note: Calculations Based on a Sludge
Cake Dry Solids of 9 Percent
Before Roller
i I
100 200 300
Coagulant Concentration (Ibs/ton)
Fig. 19-Capillarydewatering - Calculated machine capacity
(Kgs/m2 = 0.205 Ibs/ft2, Kgs/ton (m) =2.0 Ibs/ton)
400
-32-
-------
26
24
22
~ 20
esi^
£ 18
Il6
>%
O
O
12
10
£ 8
CO
6
4
2
0
Sludge: Waste Activated
Solids Loading: 0.0046 Ibs/ft2
Coagulant
* FeCI3
• Polyelectrolyte
Note: Machine Capacity Based on an 8 Percent Cake
— Solids Before Roller
1
1
10
20
30
—8 2
Sludge Specific Resistance xlO (Sec /gm)
Fig. 20-Correlation of machine apparent capacity with
sludge specific resistance for two coagulant types
-------
A second consideration in the design of the capillary section is selection
of the porous belt material. One parameter that was expected to affect cake
dewatering is the belt average pore size. Decreasing belt pore size theoret-
ically increases the capillary pressure and therefore produces a dryer sludge
cake. Experiments were conducted to simulate the capillary dewatering section
using sintered metal plates in which the pore sizes could be accurately con-
trolled. The test apparatus is shown in Figure 21. Five such plates were
tested corresponding to pore sizes of l/2)j, 5y, lOy , 40y and lOOy (y= micron).
These results are shown by the solid curve in Figure 22 where the sludge cake
solids after 2.0 minutes dewatering (sufficient time to attain the final
equilibrium) are plotted against pore diameter. The data is for a fixed sludge
solids loading. The curve shows that final cake solids increased with decreased
pore sizes, as expected, as long as the pores were above a lOy size. Below 10u,
however, final cake solids decreased markedly. At these lower pore sizes,
surface effects apparently inhibited wetting of the pore spaces and the capillary
action was diminished. Thus, there is an optimum pore size (about lOy) for
obtaining maximum cake solids. This maximum was about a 12 percent cake for the
particular solids loading indicated.
In addition to the sintered metal plate data, data from three other porous
media materials are included in Figure 22. This data falls in line with the
porous plate results. In view of the only slight improvement In sludge cake
solids with the denser porous media and since an 8 percent cake at the end of
the capillary section is satisfactory for machine operation, pore size alone is
not an overriding consideration in specifying belt requirements. Pore sizes of
lOy or greater are adequate. It should also be noted that denser porous material
(smaller pore size) costs more.
A third consideration in designing the capillary section is belt material
integrity i.e., belt life. Two factors have been considered; (1) belt structural
strength and (2) belt biological stability. The mechanical tensile strength of
the porous belt is maintained by inlaying a "scrim" ( a reinforcing plastic or
cloth screen) along the length of the belt. Figure 23 shows a photograph of one
felt porous belt and reinforcing scrim. This scrim serves as the load carrying
member and will tolerate tension loads as high as 2500 psi without failing.
To evaluate the biological stability of various porous belt materials, swatch
samples of these belt materials were attached to the tracking belt of the life
test unit (Fig. 11) and allowed to interact with activated sludge. A small
mechanical compression load was also applied to each swatch sample in an attempt
to simulate the mechanical-biological interactions that would occur in the
operation of the capillary dewatering machine. Five different basic types of
porous materials were initially selected for evaluation. They were polypropylene
felt, wool felt, polyurethane foam, dacron felt and a rayon viscose felt.
Several grades of the different felts and foam were tested, totaling some 15
swatch samples altogether. A photograph of the five basic porous material types
before testing began and after 30 days is shown in Figure 24. Shortly after the
first 30 days, both the rayon viscose and wool felts were destroyed i.e..
completely fell apart. At this time, a nylon felt was substituted. A plctorical
history of the remaining swatch samples is given in Figs. 25 and 26 which show
-------
Plexiglass
Mold
Sludge
co
i
Screen
Fig.21- Schematic arrangement of swatch test apparatus for simulating capillary dewatering
(cm = 0.3937 in)
-------
16
15
1 1 10
A
>» 6
jro
•a. ^
2-
, i
• I I I I I III I I I I I ' ••
Sludge.- Waste Activated
Solids Loading; 0.013 (Ibs/ft2)
o Sintered Metal Plates
• Material I5p, Foam
• Material 70p, Polypropylene
A Material 100M, Nylon
0.2 0.5 1.0
I 5 10 20
Belt Pore Size (micron)
50 100 200
Fig. 22 - Belt pore size effect on capillary dewatering cake solids
(kgs/m2=0.205»bs/ft2)
- 36-
-------
Fig. 23 - Porous felt and reinforcing scrim.
-------
POLYPROPYLENE
CO
CD
AFTERI30 DAYS IN SEWAGE
POLYURETMANE FOAM
Fig. 24 - Five basic porous material swatch samples before
and after 30 days testing.
-------
I
POLYPROPYLENE
POLYURETHANE 10/50
POLYURETHANE 10/900
POLYURETHANE 3/900
CAPILLARY MEDIA
SWATCH SAMPLES FROM
LIFE TEST APPARATUS
AFTER 3.5 MONTHS
POLYURETHANE FOAM 10/900
Fig. 25 - Swatch samples after 3.5 months
- 39 -
-------
POLYPROPYLENE
POLYURETH»NE 10/50
•OL TUBE THANE I O/»00
POLYURETM»N£ 3/900
SWUTCH StHPLES F H 0 N
•E TEST »PP»R»TUS
AFTER 12 MONTHS
1NE F 0« M 10/900
Fig. 26 - Swatch samples after 12 months
-------
the samples after 3 1/2 and 12 months testing, respectively. The three felt
materials appear in reasonable shape even after 12 months but the foams would
probably not be acceptable for use. Capillary dewatering tests after 3 1/2
and 12 months and a strength test (tension) after 12 months were performed on
each sample. Results of the dewatering tests were inconclusive but the data
showed that each of the swatch samples including the foams retained a wicking
capability even after 12 months of testing in the life unit. It is probable
that 12 months exposure in the life test unit corresponds to about 24 months
of equivalent operating time on the actual dewatering device. This extra-
polation is based simply on the number of cycles per day the life test apparatus
imposes on each sample in comparison to the cycles per day demanded by the
dewatering unit.
Results of the tension test done on both new and used belt samples (after 12 1/2
months in life test unit) are given in Table 5 below.
Table 5
Material
Polypropylene
Nylon
Dacron
Polyurethane
Tensile strength (ultimate)
New (psi) 12 1/2 mo.
2500
1440
1600
75
800
1200
1000
65
% of Original
strength
32
84
62
86
It should be noted that tensile strength measured reflects that of the scrim
support and not that of the belt material. The polyurethane had no scrim support.
The scrim materials used for the polypropylene and Dacron felt materials are
uncertain but believed to be polyester. The nylon material used a nylon scrim.
In addition to the life test evaluations, periodic examinations were made of the
porous belt used on the pilot test unit. Figure 27 shows a photograph taken of
the belt after 12 months of testing with the pilot unit (approximately 300 hrs
of actual test hours). Some surface abrasion has occurred as evident by the
partial unstranding of the fibers. This has not deterred machine operation.
Some discoloration is also noted. Based on this study, it would appear that
the porous felt belt materials should endure at least one year operation.
COMPRESSION ROLLER DESIGN - FINAL CAKE SOLIDS
The function of the compression roller is to: (1) increase sludge solids from
8 percent to over 16 percent and (2) detach the sludge from the screen belt.
Sludge solids loading (solid weight/unit area) and compression roller force
(force/unit width of belt) are two parameters found to influence final sludge
cake solids. These effects were measured in separate tests using the small lab
unit where the actual roller compression force was directly measured with a
strain gauge. These tests were done with the waste activated sludge. Similar
-------
Fig. 27 - Photograph of porous belt from pilot unit after
12 months of testing.
-------
data was also obtained from the pilot test unit using all three sludge types.
In this case the magnitude of the roller compression force was determined by
measuring the spring deflection (the springs to which the roller is mounted).
It was found that cake solids increased when the compression force was increased
and increased when the solids loading (cake thickness) was decreased. The
normal force load range studies corresponded to approximately a 15 to 45 psi
average (1-3 atm). A third parametric effect was also identified which
corresponds to the time of contact between the sludge and compression roller.
This sludge contact time is directly related to the linear belt speed, roller
diameter and wrap angle between the belt and compression roller i.e.,
= !§.
fceff V,
b
where t ,.,. = contact time between the sludge and roller during compression
r = radius of the roller
0 = wrap angle
V, = linear belt velocity
It was found that by increasing t ._ (keeping solids loading and compression
force constant) a dryer cake was produced. Combining the three parameters
according to the observed experimental effect, a roller compression parameter
was formulated that adequately correlates the data. The parameter is given by
[F • t /S.L.] where F is the roller compression force in Ibs/in, S.L. is the
sludge solids loading in lbs/ft2 and t ff is in seconds.
The correlation established between cake solids and the roller compression
parameter is shown in Figure 28. In each case, the wrap angle 0 was taken as
20° on the lab unit and 10° on the pilot unit.
The data for the activated sludge suggests that an approximately linear relation-
ship exists between final cake solids and the roller compression parameter over
the range of the test data. Increasing the compression parameter from 100 to 400
increased cake solids from 14 to 20 percent.
The two single data points shown for the anaerobically digested sludge suggests
that the correlation developed for the activated sludge is not applicable to this
sludge. Only one data point is available for the mixed sludge and that falls
close to the activated sludge data. This would not be totally unexpected since
the greater quantity of the mixed sludge is the waste activated sludge.
One factor which is not accounted for in the compression roller parameter group-
ing is that of coagulant dosage. This effect is shown separately for each sludge
type in Figure 29. The data shows that for each sludge type even a marginal
addition of coagulant will increase the final cake solids. For example, with the
digested sludge, adding 1 percent (based on dry solids weight) coagulant the
final cake solids was increased from 13.4 (zero coagulant) to 14.8 percent.
However, addition of still more coagulant does not appear to produce a proportion-
ate increase in cake solids. Thus, the effect on cake solids of addition coagulant
appears to be significant only at the lower dosage concentrations.
-------
22
21
20
2 19
to
1»
^ 17
o 16
CO
£15
14
13
12
A. pilot Test Unit, Fig. 3
* Waste Activated Sludge, Polyetectrolyte at 4.0 Ibs/ton
0: Lab Test Unit, Fig. 2
Waste Activated Sludge, Polyelectrolyte at 4.0 Ibs/ton
A: Pilot Test Unit, Mixed
Activated - Primary Sludge, FeCI3 at 172 Ibs/ton
•: Pilot Test Unit, Anaerobically Digested Sludge,
FeCI^ at 170 Ibs/ton
1 1 1 1—
J
F= Com p. Force Ibs/in
_ rO _ Arc Length
eff"" vk Belt Speed
b 2
S.L = Solids Loading (Ibs/fH
J.
_L
-L
±
I 2 3
Roller Compression Parameter
S.L.
xlO
-1
in L sec
i - - j lbmft"2
Fig. 28- Capillary dewatering - final cake solids
(kgs/m2 =0.205lbs/ft2. kgs/ton =2.0 IbsAon)
(kg/m = .036 Ibs/in)
-------
Ln
I
17
16
15
14
13
2 12
o
s
8
I 17
16
15
14
13
12
Sludge: Waste Activated
-Roller Comp. Force: 12.4 Ibs/in
0 5101520304050
C-31 - PolyelectroJyte (Ibs/ton)
Sludge: Waste Activated
Roller Comp. Force: 12.4 Ibs/in
50 100r 200 300 400 500 600
FeCI3- Coagulant (Ibs/ton)
17
16
15
14
13
12
17
16
13
12
Sludge: Anaerobically Digested
Roller Comp. Force: 12.4 Ibs/in
I
50 100 150 200 300 400 500
FeCL - Coagulant (Ibs/ton)
Sludge: Mixed Primary
Roller Comp. Force: 12.4 Ibs/in
_L
l
0 50 100 150 200 250 300 350
FeCL - Coagulant (Ibs/ton)
Fig. 29-Effect of coagulant dosage on final cake solids
(Kgs/m2=0.205lbs/ft2, Kgs/ton(m) =2.0Ibs/ton)
-------
The significance of the compression roller parameter, aside from correlating
the data from both test units, is that it provides a design "tie-in" with the
capillary section. Both the solids loading (S.L.) and belt speed are specified
from design considerations of the capillary section. This leaves only the
roller diameter and compression force to be specified in designing the roller
compression section to achieve the desired final cake solids.
A further consideration in the operation of the roller compression section
involves the preferential adherence of the sludge to the roller. A photograph
depicting this sludge cake detachment is shown in Figure 30. Excessive roller
compression force and sludge over-coagulation with polyelectrolyte were found
to adversely effect sludge cake detachment. High compression forces forced
the sludge solids into the screen decreasing solids capture. With the higher
polyelectrolyte coagulant dosages, the sludge attained a "tacky" (or slippery)
appearance, and the sludge cake would not adhere to the compression roller
surface. Other factors that could effect cake detachment include screen material,
weave, and fiber diameter. These parameters have not been evaluated in this
program.
SLUDGE CAKE SOLIDS CAPTURE
Present practice in sewage sludge dewatering is to normally recycle the process
filtrate back to the head end of the treatment plant. Thus, high filtrate
quality (low BOD, solids, etc.) is not generally demanded. However, sufficiently
good solids capture (^85%) is required to avoid the problems associated with
continued recirculation of fine solids. Ultimately these problems can lead to
deterioration of filter production and inceased chemical conditioning require-
ments .
In the capillary dewatering system, to obtain high solids capture, the sludge
solids must not pass through the screen belt. Solids passage is controlled by
screen mesh size, sludge floe size, and roller normal force. The first two
parameters were studied in separate experiments.
The effect of screen mesh size (screen opening) is illustrated by the data shown
in Figure 31. Four different size screens (100, 200, 325, and 400 mesh) were
evaluated using the swatch test apparatus shown in Figure 31. Each screen was
evaluated by measuring the weight fraction of solids that escaped the respective
screens when the sludge was applied. These tests were conducted with waste
activated sludge applied at a relatively heavy solids loading (0.080 Ibs/ft2)
(0.34 Kgs/m2) and conditioned with 4.0 Ibs/ton (2.0 Kg/metric ton) polyelectrolyte.
As the data shows, increasing screen mesh size (decreasing screen opening)
improves solids capture. This is particularly evident at the higher mesh sizes.
Above 300 mesh, however, there appears to be very little additional improvement
in solids capture with further increases In screen mesh.
Solids capture data was also obtained while operating the pilot test unit with
activated sludge. This data was taken with machine capacities ranging from 2.0
to 4.6 Ibs/hr ft2 (10 to 22 Kgs/hr m2); coagulant dosages from 0 to over 40 Ibs/
ton (0 to 20 Kgs/metric ton) of a polyelectrolyte; and two different mesh screens
(200 and 300 mesh). The data is given in Figure 32, which shows sludge cake
- 1*6 -
-------
Fig. 30 - Photograph depicting sludge cake detachment,
-------
100
TO
O
o
CO
% 90
o
|
Sludge: Waste Activated
Coagulant: C-31 (Dow) - 4. Dibs/Ton
Solids Loading,- 0.07 (Ibs/ft2)
Data from Swatch Test Apparatus
50 100
150 200 250 300
Screen Mesh Size
350 400 450
Fig. 31 - Effect of screen mesh size on solids capture
(kgs/m2=0,205 Ibs/ft2, kgsAon(m) =2.0 IbsAon)
-------
1
Sludge: Waste Activated
Coagulant: Polyelectrolyte-Cationic
Machine Capacity Screen
•
— a -4.6 Ibs/hr ft* - 300 Mesh
o -2.0 Ibs/hr ft2-200 Mesh
o -3.0 Ibs/hr ft2-200 Mesh'
— • -2.0 Ibs/hr ft2-300Mesh"
10 20 30
Coagulant Concentration (Ibs/ton)
40
Fig. 32 -Capillary dewatering - sludge cake solids capture
(kgs/mz =0,205 Ibs/ft2 , kgs/ton(m) = 2.0 IbsAon)
-------
solids capture as a function of coagulant dosage under various operating
conditions. The data is for the polyelectrolyte coagulant only. The effect
of screen mesh size is as previously predicted in Figure 31.
The Improved cake capture with increased coagulant addition is as expected.
With no coagulant addition, cake capture varied from 50 to 75 percent depending
on the machine feed rate and screen type. Cake solids captures of 95 percent
or higher were achieved at a coagulant dosage of 10 Ibs/ton (5 Kgs/metric ton)
or about $4.00/ton ($4.40/metric ton) dry solids. This is at a machine capacity
of 2.0 Ibs/hr ft/ (10 Kgs/hr m2) . Increasing machine capacity to 4.6 Ibs/hr ft*
(22 Kgs/hr m2) by reducing the solids loading (i.e., maintaining a thinner cake
but increasing the sludge pumping rate and belt speed) diminshed cake solids
capture. This is attributed to the fact that a thinner sludge cake is not as
effective in capturing fines as a thicker cake. Solids capture can be improved
at the higher machine capabilities by increasing coagulant dosage. Thus, a
trade-off between machine capacity, coagulant dosage and cake filtrate quality
can be made.
The data showing cake solids capture for activated sludge with ferric chloride
conditioner is shown in Figure 33. The data was taken with the 300 mesh screen
at a machine capacity of 4.3 Ibs/hr ft (21 Kgs/m2 hr). Comparing this data with
the equivalent data in Figure 32 shows that 90 percent solids capture is obtained
at about a 15 Ibs/ton (7.5 Kgs/metric ton) polyelectrolyte while the equivalent
solids capture for the ferric chloride is obtained at 130 Ibs/ton (65 Kgs/metric
ton). On a cost basis, this means $6/ton ($6.60/metric ton) of dry sludge solids
to condition with polymer and $5.20/ton (5.72/metric ton) to condition with ferric
chloride (FeCl3). On this basis, it must be reasoned that the ferric chloride
is more cost effective than the polyelectrolyte. With the capillary dewatering
system, this difference could be associated with the relative effectiveness of
the two coagulants in effecting sludge cake detachment. The polyelectrolyte has
an inherent tackiness which appears to inhibit cake detachment (cake tends to
adhere to the screen belt) . This consideration is in and above any differences
in the coagulating effectiveness of the two conditioners. When comparing the
two coagulant systems, these factors must be balanced against the larger volume
of sludge produced due to the addition of significant amounts ferric chloride,
the expense of handling the larger volumes of coagulant, and the corrosive nature
of ferric chloride solutions. Lime may be added in the plant to control corrosion
and improve conditioning.
Solids capture data for the anaerobically digested and mixed primary sludges are
shown in Figures 34 and 35. With digested sludge, solids captures over 90 percent
were attained at a coagulant dosage of 300 Ibs/ton (150 Kgs/metric ton). This
data is for a machine operating between 5 and 6 Ibs/hr ft* (24-30 Kgs/m2 hr).
With the mixed primary sludge, the best solids capture achieved was 85 percent.
This is a coagulant dosage of 250 Ibs/ton (125 Kgs/metric ton). Both of these
sets of data were attained using ferric chloride conditioner.
GENERAL MACHINE OPERATION
Operation of the capillary dewatering pilot scale test unit has provided sufficient
experience to anticipate potential problems with a full scale system. The general
operation of the unit on activated sludge was excellent. There were no particular
- 50 -
-------
100
_ 90
#80
to
0>
70
60
30
-—— ""A"" Sludge.- Waste Activated
Coagulant: FeCL
Machine Capacity Screen
•4.3lbs/hrft2 300 Mesh
.2
L 89 Ibs/ hr ft
200 Mesh
I
I
I
50
100 150 200 250
Coagulant Dosage - FeCL (Ibs/ton)
300
350
400
Fig. 33-Capillary dewatering - sludge cake solids capture
(Kgs/m2 =0.205 Ibs/ft2, Kgs/ton (m) =2.0 Ibs/ton)
-------
100
90
^
OJ
CO
;o
"o
l/>
o>
-a*
5
0>
70
60
Sludge: Anaerobically Digested
Coagulant: FeCL
Machine Capacity
• 5.7lbs/hrft2
* 3.2lbs/hrft2
Screen
300 Mesh
200 Mesh
50
1
1
100 200 300 400
Coagulant Dosage - FeCL (Ibs/ton)
Fig. 34-Capillary de water ing - sludge cake solids capture
(Kgs/m2 = 0.205 Ibs/ft2 Kgs/ton (m) » 2.0 Ibs/ton)
500
.52-
-------
100
90
80
VI
u>
I
CO
50
Sludge: Mixed Waste Activated and Primary
Coagulant: FeCU
Machine Capacity: 5.2 Ibs/hr ft2
Screen: 300 Mesh
I
30
100 150 200 250
Coagulant Dosage - FeCI, (Ibs/ton)
300
350
400
Fig. 35- Capillary de water ing - sludge cake solids capture
2 = ).205lbs/ft2, Kgsfton(m) =2.0 Ibs/ton)
-------
problems encountered in pumping the sludge to maintain the uniform and thin
sludge layers required for the higher machine capacity operation. The sludge
cake would detach cleanly (preferentially adhere to the compression roller)
even over a relatively wide range of compression force settings. With this
sludge, the machine could be run unattended for relatively long periods of
time (several hours). The ease of operation with the activated sludge is
attributed to the sludge's homogeneity. The mixed primary sludge exhibited
a similar homogeneity but was heavier (higher solids content). This sludge
has a strong anaerobic odor making it unpleasant to work with.
With the anaerobically digested sludge, pump clogging was frequently encountered
because of hair and large solids. This problem was circumvented by changing
pumps.
With the heavier sludges i.e., both the mixed and digested sludges, the compression
roller section operation (cake detachment) appeared more sensitive to change in
coagulant dosage, coagulant type and roller force. These are qualitative obser-
vations that need additional investigation.
The most serious mechanical problem encountered in operating the pilot unit was
maintaining the integrity of the screen belt for extended periods. During the
early part of the program, frequent screen failures were encountered (after only
2 or 3 weeks operation). The screen was observed to tear at the seam and near
the edges. These screen failures were not catastrophic but gradual. The failure
at the seam was attributed to the method of fastening the screen together. It
had been sewn. Sewing the screen actually put a non-uniform and point loading on
the screen fibers. This mode of failure was finally corrected by subsequently
bonding the two ends of the screen together. This proved to be an easy and
effective method of screen attachment. The tears at the edge of the screen were
eliminated by edging the screen (doubling over the screen to put a 1/4 inch edge
along both sides) and eliminating screen contact with sharp metal edges on the
unit. For example, the screen guides were removed and a teflon edging was put
on the underside of the feed tray. These changes appeared to signifianntly improve
the screen integrity during the remaining portion of the test program (about 3
months). It is not known whether the fine mesh screens used in this pilot study
would be acceptable in large scale field systems. Heavier (large fiber diameters)
screens may be needed for these systems.
Only one porous belt was used during the course of this investigation. There were
no operating problems encountered with this belt, and the belt, in fact, still
appears in excellent condition.
Several minor machine operating problems were encountered and solved during the
Bourse of this study. This included some spray water replumbing to prevent water
-un-back (into the belt) and retracking the screen belt to take maximum advantage
f the large belt area. The reader is referred to Appendix B, Tables 1 through 4.
ncluded in the tables are some general comments that reflect some of the operating
roblems encountered.
-------
SECTION V
REFERENCES
1. Adrian, D. D., Smith, J. E., Jr., "Dewatering Physical-Chemical Sludges,"
Application of New Concepts of Physical-Chemical Wastewater Treatment,
September 18-22, 1972. Published in Pergamon Press, Inc., pp. 273-289,
2. USEPA, "An Environmental Law—The Federal Water Pollution Control Act
Amendments of 1972—HIGHLIGHTS," published by Office of Public Affairs,
U.S. Environmental Protection Agency, Washington, D.C., December 1972.
3. Fowler, G. J., "Apparatus and Method for Separating Solids From Liquids",
U.S. Patent 2,021,122, November 12, 1935.
4. Zaromb, S., "Apparatus and Method for Separating Slurries Into Solid and
Fluid Components," U.S. Patent 3,207,061, September 21, 1965.
5. Goodman, B. L., et. al., "Sewage Treatment System," U.S. Patent 3,531,404,
September 29, 1970.
6. Kline, S. J., and McClintock, F. A., "Describing Uncertainties in Single
Sample Experiments", Mechanical Engineering, Vol. 75, 1953, pp. 3-8.
- 55 -
-------
SECTION VI
APPENDIX A - Error Analysis
The data from any experiment can only be interpreted within the limits of the
accuracy and reliability of the measurements. These limits are difficult and
often impossible to ascertain in single sample experiments. Nevertheless, it
is necessary for the experimenter to attempt to describe the uncertainties in
the data in order to add credence to the results. Such an analysis would first
be extremely helpful in selecting the apparatus best suited for the experiment,
and secondly, give the experimenter a basis for evaluating his data.
An analysis has been conducted to describe the uncertainties in the data reported
herein. The calculations are based on the equations presented by Kline and
McClintock, (*>) who show that the uncertainty interval Wr in some function R of
n independent variables V.^ is given by
W =
3V, 2
i i.
9V
n
W
n
1/2
where W, is the uncertainty interval in the variable V... In developing the above
equation, it was assumed that the same odds exist for each of the variable
intervals and for the result i.e., if the confidence intervals of the W^ variables
are 90 percent, then the confidence interval on W will also be 90 percent. Using
this equation, the uncertainty in the variables can be estimated. The basic
measured quantities and their respective uncertainty intervals are summarized in
Table A.I. The confidence intervals in the variables were arbitrarily set at
90 percent. The calculations for the error in the derived quantities are
summarized below.
Table A-l. EXPECTED UNCERTAINTY IN MEASURED VARIABLES
Quantity
1. Belt active length
2. Belt active width
3. Sludge total suspended solids
a. Activated
b. Mixed-activated primary
c. Anaerobically digested
4. Sludge sample weights (gram balance)
a. Beaker wt.
b. Beaker wt. + wet solids
- 56 -
Typical value and
confidence interval
50+1.0 inch
(cm = .39 in)
10.5 + 0.25 inch
19,000 + 100 mg/£
40,000 + 400 mg/fi,
45,000 + 450 mg/£
173 +0.1 gm
360 +0.1 gm
-------
Table A-l (continued). EXPECTED UNCERTAINTY IN MEASURED VARIABLES
Quantity
Typical value and
confidence interval
c. Beaker wt. + dry solids
5. Time measurements (stop watch)
a. Solids capture measurement
b. Sludge feed measurement (low)
c. Sludge feed measurement (high)
6. Sludge volume measurement (graduate
cylin.)
a. Activated sludge (low feed rates)
b. Activated sludge (high feed rates)
c. Mixed-primary-activated
d. Anaerobically digested
7. Chemical feed rate
200 + 0.1 gm
60 + 1.0 sec.
15 + 0.25 sec.
10 + 0.25 sec.
600 + 10 m&
1000 + 25 mi
1000 + 50 m£
1000 + 50 mi
10+1 m£/min
Experimental Uncertainty in Sludge Flow Rate (W.A.S.)
F.R. = Flow Rate - Sludge Volume/time = S.V./t
From Equation A-l, the uncertainty in sludge flow rate becomes;
1/2
WF.R.
F.R.
Now, from Table A-l,
M
ws.v _±io
S.V. ~ 600 -
1.7%
_
t
60
+ 1.7%
(A-2)
- 57 -
-------
Therefore, from Equation (A-2),
WF.R.
F.R.
+2.4% (low sludge feed rates)
Similarly, for the high feed rates,
W,
F.R.
F.R.
= +3.5%
b. Experimental Uncertainty in Machine Capacity (W.A.S.)
, , „ Sludge Solids Wt x Sludge Feed Rate x Conversion Factory
Machine Capacity = B R2n- AT.-,
or M.C. =
Belt Area
S.W. x F.R.
B.A.
Thus, the uncertainty in machine capacity is;
WM.C.
M.C.
S.W.
F.R.
B.A.
1/2
W.
B.A.
The uncertainty in belt area, ' ', is given by; B.A.
width) thus, from Table A-l,
= length x
W.
B.A.
B.A.
i.or + 1.25
50
10.5
1/2
= +3.3%
Thus; from Table A-l and above, (low capacities);
1/2
M. L.
[(.005r + (.024) + (.033r] » +4.1%
Similarly, for the higher capacities,
WM.C.
M.C.
= +4.8%
- 58-
-------
c. Experimental Uncertainty in Cake Solids Capture (W.A.S.)
„ , _,, _ _ „ Dry Solids From Roller D.S.R.
Solids Capture = S.C. = _. ^ g . . ,— —3— = _ _ _
* Dry Solids From Feed D.S.F.
Therefore,
WS.C.
S.C.
w.
2 rW .2
DSR[ | DSF
DSR J [DSF
1/2
(A-3)
The uncertainty in the measurement in the dry cake solids from the compression
roller is a result of three factors:
1. weighings +0.5%
2. sampling time +1.7% (60 second sample with a +1.0
second accuracy)
3. edge losses +2% (estimate)
On this basis, then, the uncertainty in the dry cake solids wt. from the compression
roller is,
W.
DSR -2.2%
DSR +4.2%
The uncertainty in feed dry solids wt. is basically in measurement of this sludge
feed rate.
W
WDSF
DSF
+2.4%
Thus, from Equation (A-3), the uncertainty in cake solids capture is;
ws.c.
S.C.
(.022)
, t
(.042)2 + (-
1/2
Similarly,
W
S.C.
S.C.
+4.9%
-3.2%
(low feed rates)
b • L* » "TO * J« /. , , - , . v
^7~ = -4.1% (high feed rates)
Similar calculation follows for the mixed activated-primary sludge and for the
anaerobically digested sludge using the uncertainties given in Table A-l . The
results of the calculations are given in Table 3 of the text.
-59-
-------
SECTION VII
APPENDIX B - A Chronological Summary Listing of
Test Data—Pilot Test Unit
A summary of test data from the pilot unit is given in Tables IB through 5B.
The data is arranged in chronological order. Included In each table are CD Run
number which merely corresponds to the date of testing i.e., month and day (521 ~
May 22) and (2) Type of sludge - ACT - waste activated sludge, DIGST - anaero-
bically digested sludge and MIXPRIM - mixed primary activated sludges. The
remaining columns are self explanatory.
Table IB summarizes data obtained during the initial start-up and system shake-
down series. During this time, most of the machine operating problems were
solved and improved experimental methods for obtaining solids recovery values
were developed. The data is limited to basic machine performance on activated
sludge using one screen type.
Table 2B shows data that reflects a continuation of developing base line system
performance. Methods for obtaining solids recovery data were Incorporated during
this time along with attention to measuring roller compression forces. Initial
data on the digested and mixed primary sludges was also obtained.
Tables 3B and 4B summarize data obtained with activated sludge during that portion
of the program that involved developing a more basic understanding of various
system parameter interactions. This included varying coagulant dosage rates,
belt speeds, sludge feed rates, screen type, and roller compression forces.
These sets of data constitute the most reliable and informative compilation of
tests conducted on waste activated sludge during this program.
Table SB summarizes data taken in the anaerobically digested and mixed primary
sludges.
- 60 -
-------
Table 1-B Pilot Test Unit - Summary of Data From May 22 through Aug. 10,
1972 - System Installation and Shakedown
Run
522
524
602
606
607
608
609
612
614
627
628
713
718-A
718
719-A
719
720
720-A
726
728
731
801
802
803
808
809
810
Sludge
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Coagulant
Type
None
C-31
FeCl3
C-31
C-31
C-31
C-31
C-31
C-31
Fed 3
FeCl 3
Fed 3
FeCl3
FeCl3
FeCl3
Fed 3
FeCl3
FeCl3
None
None
None
None
None
None
None
C-31
C-31
(Ibs/
ton)
0
17
103
5
5
7
14
1.3
8
28
27
300
95
140
52
270
230
21
0
0
0
0
0
0
0
5
5
Cake
Solids
(%)
13.2
13.5
16.3
16.2
15.9
15.7
15.5
16.4
15.8
17.2
18.4
15.4
17.8
18.1
17.3
20.3
17.4
18.9
17.1
17.0
18.0
16.2
16.6
16.8
16.5
17.5
16.6
Comp .
Force
(Ibs/
in)
_
-
-
-
-
-
-
-
-
-
-
-
_
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Solids
Loading
(Ibs/
ft2)
0.0097
0.0135
0.0121
0.0084
0.0083
0.0091
0.0063
0.0098
0.0110
0.0095
0.0093
0,0066
0.0088
0.0088
0.0081
0.0071
0.0071
0.0082
0.0078
0.0068
0.0081
0.0081
0.0077
0.0077
0.0080
0.0070
0.0070
Machine
Capacity
(Ibs/
hr ft2)
2.32
2,04
2.18
2.52
2.40
2.72
1.89
2.93
1.33
2.83
2.78
1.98
2.65
2.65
2.41
2.13
2.20
2.46
2.44
2,12
2.53
2.62
2.30
2.29
2.60
2.50
2.50
Solids
Recovery
(%)
__
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
*
-
_
-
-
-
_
-
-
-
Run
Time
(hrs)
3.0
3.0
2.5
4.0
4.0
4.0
3.5
4.5
4.5
4.0
2.5
3.0
4.0
3.5
3.0
4.0
4.0
4.0
3.5
4.0
4.5
4.0
4.5
2.0
4.0
3.0
2.5
i
Screen
Type
(Mesh)
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
*
200
200
200
200
200
200
200
200
Comments
Shakedown - comp . force not
measured
Belt tore at seam — resewn
Changed compression roller -
machined stainless steel.
Chera. feed pump not functioning.
Screen repaired prior to run.
Repaired feed pump.
Comp. roller hanging up.
Screen wash water spilling into
belt effluent stream - All prior
solids recovery data invalid.
Screen tore.
Changed screens - new one 200.
Bonding screen together.
Belt dirty—wash w/ soap & water.
Comp. roller stalling
Changed position of comp. roller.
Re tracked belt.
H
i
(Kgs/ton(m) =2.0 Ibs/ton, Kg/m = 0.056 lbs/in., Kgs/m2 = 0.205 Ibs/ft2)
-------
Table 2-B Pilot Test Unit - Summary of Data From Aug. 16 through Oct. 4,
1972 - Base Line System Performance
1
ON
IVk
IV
1
Run
816
817
818
821
823
824
825
828
829
8 30- A
830-B
830-C
830-D
830-E
908
911
913
913-A
915
915-A
918
919
920-25
927
1003
1004
Sludge
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Digst
Digst
Act
Act
Act
Act
Act
Act
Digst
Act
Mix-
Prim
Mix-
Prim
Coagulant
Type
C-31
C-31
C-31
C-31
C-31
C-31
C-31
C-31
C-31
Fed 3
FeCl3
FeCl3
FeCla
FeCl3
Fed 3
ST-260
None
C-31
C-31
C-31
C-31
C-31
ST-260
C-31
C-31
Ibs/
ton)
5
4
3
7
10
19
9
2
0.5
190
95
48
19
9
9
2
0
9
5
6
50
34
2
26
-
5
Cake
Solids
CO
17.6
17.9
17.8
18.6
17.6
19.8
18.3
18.7
19.4
19.3
20.0
19.1
20.2
20.3
20.0
19.9
18.9
18.8
-
-
18.6
19.2
21.2
19.7
-
20.8
Comp.
Force
(Ibs/
in)
18.5
18.5
16.4
18.5
16.4
20.4
18.5
20.4
20.4
20.4
20.4
20.4
20.4
20.4
16.4
-
17.4
16.4
-
-
20.4
20.4
-
20.4
-
-
Solids
Joading
(Ibs/
ft2>
0.0079
0.0088
0.0078
0.0081
0.0078
0.0082
0.0086
0.0078
0.0077
0.0079
0.0079
0.0079
0.0079
0.0079
0.0135
0.0165
0.0085
0.0085
0.0071
0.0071
0.0047
0.0045
0.0011
0.0061
-
0.0084
Machine
Capacity
(Ibs/
hr ft2)
1.89
2.10
1.87
1.95
1.89
1.98
2.06
1.88
1.86
1.89
1.89
1.89
1.89
1.89
3.21
3.96
2.05
2.05
1.70
1.70
0.88
1.07
3.10
1.48
-
2.32
Solids
Recovery
(%)
70.3
66.2
74.2
68.1
74.8
84.8
67.5
73.1
62.8
80.8
75.2
67.8
75.2
57.3
66.0
75.3
59.0
87.8
84.0
89.0
90.0
84.0
65.6
92.5
-
-
Run
Time
(hrs)
2.5
2.5
4.0
4.0
4.0
4.0
4.0
4.0
5.0
1.0
1,0
1.0
1.0
1.0
2.0
3.5
1.0
1.0
0.5
0.5
1.5
2.0
8.0
3.0
-
4.0
Screen
Type
(Mesh)
200
200
*
200
200
200
200
200
200
200
200
200
200
*
200
200
200
200
200
200
200
200
200
200
200
*
Comments
Solids recovery data obtained —
comp. force recorded.
Holes noticed in screen. New
screen 200.
Chem. feed pump malfunction —
comp. roller stalling.
Comp. roller stalling
Comp. roller again stalling.
Replaced chem. feed pump.
Changed screen after test runs-200.
Changed over to digested sludge —
using clips to fasten belt - pump
clogged.
Comp. roller stalling - poor cake
pick up.
Changed back to activated sludge.
Cake solids not measured.
Corap. roller stalling.
Sludge diluted by heavy rainfall.
Sludge diluted by heavy rainfall.
Changed to digested sludge — wait-
ing for normal RAS conditions.
Changed back to activated sludge.
Tried mixed primary - no data
Screen tore on cleaning — replaced
with 300 mesh.
-------
Table 3-B Pilot Test Unit - Summary of Data From Oct. 11 through Nov. 16,
1972 - Initial Study of Machine Parameters on Waste Activated Sludge.
Run
1011
1012
1012-A
1016
L017-A
101 7-B
1017-C
1018
1027
1031
1031-A
1102
1106-A
1106-B
1106-C
1107-A
1107-B
1107-C
1110-A
1110-B
1113-A
1113-B
1114-A
1114-B
1116-A
1116-B
Sludge
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Coagulant
Type
C-31
C-31
C-31
C-31
C-31
C-31
C-31
C-31
None
C-31
C-31
C-31
C-31
C-31
C-31
None
C-31
C-31
C-31
C-31
None
C-31
C-31
C-31
None
C-31
(Ibs/
ton)
2.8
4.5
6.7
-
4.5
2.0
3.9
4.5
0
94.0
70.0
18.7
29.0
44.5
62.5
0
3.0
11.6
5.4
13.4
0
3.0
5.1
7.6
0
3.2
Cake
Solids
(%)
18.2
18.0
17.6
-
-
18.2
18.7
-
18.0
19.1
18.3
17.0
17.7
17.7
16.7
16.0
16.5
16.7
15.5
16.5
17.3
17.3
17.2
17.0
16.0
16.6
Comp.
Force
(Ibs/
in)
18.5
18.5
18.5
-
-
20.4
20.4
-
20.4
16.4
16.4
20.6
18.5
18.5
18.5
16.4
16.4
16.4
18.5
18.5
16.4
16.4
16.4
16.4
16.4
16.4
Solids
Loading
(Ibs/
ft2)
0.0071
0.0069
0.0069
-
0.0067
0.0067
0.0067
0.0213
0.0057
0.0054
0.0054
0.0042
0.0043
0.0043
0.0043
0.0031
0.0031
0.0031
0.0028
0.0028
0.0075
0.0075
0.0067
0.0067
0.0042
0.0042
Machine
Capacity
(Ibs/
hr ft2)
1.98
2.05
2.05
-
1.85
1.85
1.85
-
1.36
1.29
1.29
3.02
3.06
3.06
3.06
2.79
2.79
2.79
3.54
3.54
1.81
1.81
1.73
1.73
3.02
3.02
Solids
Recovery
(*)
88.0
82.0
83.0
-
-
83.0
90.0
-
74.0
99.5
99.1
91.1
88.0
85.5
91.0
59.0
75.9
94.0
77.0
84.0
69.4
68.0
75.0
85.0
51.1
60.3
Run
Time
(hrs)
4.0
4.0
1.0
3.5
1.0
1.0
1.0
1.0
1.5
2.0
1.5
2.0
1.0
1.0
1.0
1.0
1.5
0.5
2.5
2.5
1.0
1.0
1.0
1.0
1.0
1.0
Screen
Type
(Mesh)
300
300
300
300
300
300
300
300
300
*
200
300
300
300
300
300
300
300
300
300
300
300
300
300
300
300
*
Comments
Now using 300 mesh screen.
Lost data.
Capillary profile study.
Capillary profile study.
Capillary profile study.
Capillary profile study. Screen
developing holes.
Screen tore. Changed screens for
next run .
New screen.
Problem w/sludge cake detachment.
Capillary profile study.
Capillary profile study.
Screen tore. Changed screens
after run.
I
OS
-------
Table 4-B Pilot Test Unit - Summary of Data From Dec. 1 through Jan. 1,
1973 - Continue Study of Machine Performance on Waste Activated
Sludge.
Run
1201-A
1201-B
1201-C
1201-D
1204-A
1204-B
1218-A
1218-B
1218-C
1219-A
1219-B
1219-C
1220-A
1220-B
1220-C
109 (73
117 (73
ludge
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Act
Coagulant
Type
None
C-31
C-31
C-31
C-31
C-31
C-31
C-31
C-31
FeCl3
FeCl3
FeC 13
FeCl3
FeC IT
FeC IT
FeC 1^
FeCli
(Ibs/
ton)
0
8.2
5.6
2.6
18.5
11.7
55.0
41.0
12.0
57.6
34.6
23.0
542.0
401.0
293.0
209.0
127.0
Cake
Solids
(%)
14.5
14.1
14.5
15 ."5
14.5
16.1
15.5
15.8
-
13.1
15.0
13.7
16.3
18.3
14,3
15.0
15.2
Comp.
Force
(Ibs/
in)
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
-
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
Solids
Loading
(Ibs/
hr ft2)
0.0044
0.0044
0.0044
0.0044
0.0045
0.0045
0.0043
0.0043
0.0043
0.0042
0.0042
0.0042
0.0042
0.0042
0.0042
0.0042
0.0051
Machine
Capacity
(Ibs/
hr ftz)
4.6
4.6
4.6
4.6
4.6
4.6
4.5
4.5
4.5
4.3
4.3
4.3
4.3
4.3
4.3
4.3
5.3
Solids
Recovery
(%)
53.0
80.0
75.0
64.0
92.2
68.0
87.0
87.0
85.0
77.0
65.0
60.5
93.0
93.5
94.0
90.5
90.8
Run
Time
(hrs)
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.5
1.5
1.5
1.5
1.5
1.7
2.2
2.1
1.5
1.6
Screen
Type
(Mesh)
300
300
300
300
300
300
300*
300
300
300
300
300
300
300
300
325*
300
Comments
Using new screen. Running belt
at high speeds.
Capillary profile study.
Capillary profile study.
Capillary profile study.
Capillary profile study.
Shaft on drive roller broke.
Question solids recovery number.
New screen - same mesh
Capillary profile study.
Capillary profile study.
Capillary profile study.
Capillary profile study.
Capillary profile study.
Capillary profile study.
Capillary profile study.
Capillary profile study.
Screen tore-replaced w/325 mesh.
Capillary profile study.
-------
Table 5-B
Pilot Test Unit - Summary of Data From Jan. 26 through Feb. 14,
1973 - Continue of Study of Machine Performance with Digested
and Mixed Primary Sludges.
Run
126 (73)
129-A(73)
129-B
129-C
130-A(73)
130-B
131-A(73)
131-B
213-A(73)
213-B
213-C
21 3-D
214-A(73)
214-B
214-C
Sludge
Digst
Digst
Digst
Digst
Digst
Digst
Digst
Digst
Mix-Prim
Mix-Prim
Mix-Prim
Mix-Prim
Mix-Prim
Mix-Prim
Mix-Prim
Coagulant
Type
None
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
None
FeCl3
Fed 3
FeCl3
FeCl3
Fed 3
FeCl3
(Ibs/
ton)
0
43
73
14.2
170
242
345
460
0
17.5
52.6
89.4
172
246
369
Cake
Solids
a)
13.4
15.3
15.9
14.8
15.8
18.4
16.4
16.0
14.0
14.5
14.7
15.0
15.il
15.0
15.3
Comp.
Force
(Ibs/
in)
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
12.4
Solids
Loading
(Ibs/
ft*)
0.0056
0.0060
0.0060
0.0060
0.0050
0.0050
0.0053
0.0053
0.0056
0.0056
0.0056
0.0056
0.0060
0.0060
0.0060
Machine
Capacity
(Ibs/
hr ft2)
5.74
6.18
6.18
6.18
5.17
5.17
5.45
5.45
5.05
5.05
5.05
5.05
5.05
5.42
5.42
Solids
Recovery
(%)
55.3
65.2
71.7
61.3
81.0
88.4
92.4
97.3
56.8
59.5
67.0
74.6
80.5
84.3
84.9
Run
Time
(hrs)
2.0
1.5
1.5
1.5
2.0
2.0
2.0
2.0
1.5
1.5
0.5
1.5
1.5
1.5
2.0
S crccn
Type
(Mesh)
325*
325
325
325
325
325
325
325
325
325
325
325
325
325
325
Comments
New screen.
-------
SECTION VIII
APPENDIX C - Economics of Capillary Dewatering on
Waste Activated Sludge
Any product or process which is to be used by customers and for which there
are substitute products available must be economically justifiable in order
to be useful. To this end, an estimate of the cost of dewatering waste activated
sludge by the capillary suction method must ultimately be considered.
The task of providing a detailed cost prediction for this process under all
possible operating conditions is extremely difficult because of the number of
parameters of choice which the customer can vary. Since machine capacity and
performance is the ultimate design parameter which will set the equipment size,
the foremost question is under what conditions will the capacity be measured.
The parameters of choice are independent of each other over small ranges but
interdependent over gross ranges. These include:
• Exit solids concentration
• Solids capture percentage
• Coagulant type (inorganic or polyelectrolyte)
• Feed solids concentration
In addition, dependent variables such as belt contact times and solids loading
rates have an effect on capacity also. Earlier in this paper, curves were
presented showing the capacity as a function of polymer dosage and solids loading
rate. If similar curves were available for feed solids concentration and solids
capture as functions of the other parameters, then an economic analysis could be
made that would be very inclusive.
To eliminate these problems and still provide a justifiable economic analysis,
cost estimates were made for three selected operating conditions which were
demonstrated during the course of the program. The calculations assume a constant
machine size and an activated sludge of 2% as feed to facilitate run to run
comparisons.
Using these assumptions as a basis, sewage plant size will be a variable in the
comparisons. Extension of the assumptions to larger scale plants such as 10 to
100 mgd would require substantiating the data at plant scale.
Costs were developed (Table C-l) at two levels of machine capacity i.e., 2.0 and
4.5 Ibs/hr ft (10 and 22 Kg/hr m2) and for two coagulants at the higher capacity;
ferric chloride and a cationic polyelectrolyte. The only costs considered are:
machine capital
maintenance and repair
electrical
coagulant
labor
- 66 -
-------
Costs such as building, land, feed systems, removal systems, and ultimate
disposal were not considered due to the wide variations possible in these
systems and the fact that many of these facilities are available in existing
plants. The costs shown in Table C-l are for the dewatering function only.
There are several interesting items to note in these costs. As usual, the labor
costs are a significant portion (50% or more) of the cost of dewatering the
sludge. The 1/2 man day estimate may be high since once the unit is started,
it should run automatically. Because of the preparation and clean up time,
however, a half man day is probably realistic.
The coagulant costs for both ferric chloride and polyelectrolyte are similar
with a higher cost for the polymer. The ferric seemed to perform better but
has corrosion problems. A cost of $3/ton ($3.30/metric ton) is not unreasonable
for this type of treatment however.
The power costs in this system are of note because they are so low. At $130/year,
they are only 3% to 5% of comparable costs for centrifuges or vacuum filters.
With the cost of power increasing very rapidly, and present emphasis on energy
conservation, the low power cost makes this method attractive.
22 2
Decreasing machine capacity from 4.5 pounds/hr ft (22 Kg/hr m ) to 2 Ibs/hr ft
(10 Kg/hr m^) essentially doubles the total cost. The machine capacity markedly
affects the system economics. The exit dry solids in the lower capacity case
are significantly higher than the high capacity case which will offset some of
the cost. These figures point out the trade off capability of the system.
The costs detailed here are competitive with existing systems such as vacuum drum
filters, pressure filters, and centrifuges. In fact, on waste activated sludge,
achievement of 15% to 17% solids is extremely difficult with other systems.
- 67 -
-------
Table 1-C
Cost Estimates For Capillary Dewatertng
Basis for Cost Calculation
Assumptions
1. Capillary device capital cost is $300 per active square foot of capacity and
is linear with capacity.
2
2. Machine size is 110 ft of active areas.
3. Operation schedule is 8 hour/day, 7 days/week, 360 days/year.
4. Machine performance as in run 1218A data sheet.
Performance data: 55 Ibs/ton ferric chloride as coagulant
4.5 Ibs/hr ft2 machine capacity
87% solids capture
15.5% exit solids cake
Waste activated sludge feed of ^2% solids
5. No costs for overhead, peripheral feed equipment, or building.
Annual Cost Calculations
1. Capital
110 ft2 machine at $300/ft2 = $33,000
Amortization over 20 years at 6% $2,877
(m2 = .093 ft2, Kgs/ton(m) = 2.0 Ibs/ton, Kgs/m2 = 0.205 lbs/ft2,
Kg = 2.2 Ibs, ton(m) = 1.1 ton, (HP- M = .986 HP)
2. Coagulant cost at 5c/lb ferric chloride
4.5 Ibs , in f2 8 hours 360 days 1 ton _. „ rt
- — x 110 ft x — r - x - *— x -nnA = 712.8 tons
.. 2 day year 2000 Ibs
hr r t
3
3
(Approximately 4 mgd plant (16 m /day))
712.8 tons x _ x
ton lb
3. Power: assume 5 hp 230 v 30 motor
230v x 13a 3Kw
Shrs
Kw-nr day year
- 68 -
-------
Table 1-C (continued)
4. Maintenance @ 5.5% of capital/year
$33,000 x .055 $1,815
5. Labor 1/2 man day/day at $5.QO/man hour
4 man hours $5 0,n , , onr.
x Ir x 360 days 7,200
day man hour
Total annual cost for above items $13,804
Cost/ton = $13804/712.8 $19.36/T
Run 12Q1B same conditions but polyelectrolyte
Performance data: 8.2 Ibs/ton C-31 polyelectrolyte as coagulant
4.6 Ibs/hr ft^ machine capacity
80% solids capture
14.1% exit solids cake
Waste activated sludge feed of ^2% solids
1. Capital $2,877
2. Coagulant cost at 40c/lb C-31
4.6 Ibs 1in ft_2 8 hrs 360 days 1 ton
n x 110 ft x ——— x J— x n--- ..,— = 729 tons
hr ft* day year 2000 Ibs
(approximately 4 mgd plant (16 nH/day))
-no 8.2 Ibs $.40 ^ .
729 tons x x y = $2,391
ton Ib
3, Power $ 130
4. Maintenance $1,815
5. Labor $7,200
$14,413
Cost/ton = $14,413/729 $19.77/ton
Run 1012A
Performance data: 6.7 Ibs/ton C-31 polyelectrolyte coagulant
2.05 Ibs/hr ft2 machine capacity
83% solids capture
17.6% exit solids cake
Waste activated sludge feed at ^2% solids
-69-
-------
Table 1-C (continued)
1. Capital $2jg77
2. Coagulant cost at 40 870
3. Power $
4. Maintenance $1
5. Labor $7,200
$12,892
Cost/ton = 12892/325 $39.67/ton
- 70 -
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA 670/2-74-017
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
EVALUATION AND DEMONSTRATION OF THE CAPILLARY
SUCTION SLUDGE DEWATERING DEVICE
5. REPORT DATE
March
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
T. E. Lippert and M. C. Skriba
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGMMIZATION NAME AND ADDRESS
Westinghouse Electric Corporation
Research and Development Center
Pittsburgh, Pennsylvania 15235
10. PROGRAM ELEMENT NO.
132043 -RQAP 21-ASP Task 09
11. CONTRACT/GRANT NO.
68-01-009^
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Research and Development
National Environmental Research Center
Environmental Protection Agency
Cincinnati. Ohio 4*5268
13. TYPE OF REPORT AND PERIOD COVERED
Final report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A device for dewatering waste activated sludge that uses the capillary suction in
a porous belt has been demonstrated on a pilot scale test unit. The system uses
capillary action to initially dewater the sludge and form a thin sludge cake which
is then mechanically compressed in a final step to obtain a still dryer cake. Both
dewatering steps are incorporated into a single system endless belt design. Sludge
feed capacities from 2 to 4.5 Ibs/hr ft2 (10 to 22 Kgs/hr m*) have been demonstrated
with the cake solids at discharge ranging from 15 to 18 percent. These results were
obtained with a conditioned waste activated sludge at a coagulant cost of approxi-
mately $4.00/ton ($4.4o/metric ton). The device has also been tested on an
anaerobically digested and a mixed primary-activated sludge.
KEY WORDS AND DOCUMENT ANALYSIS
Jc. COSATI Field/Group
DESCRIPTORS
b.lD!
*Sludge, «Sewage, Sludge Drying, Cakes,
Filtration, *Sludge treatment. Deposition,
Force, Conditioning (treating)
8. DISTRIBUTION STA I bMbN I
Release to public
i
EPA Form 2220-1 (9-73)
*5iu
------- |