7ATER POLLUTION CONTROL RESEARCH SERIES
CLOSED SYSTEM WASTE
MANAGEMENT FOR LIVESTOCK
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollution
in our Nation's waters. They provide a central source of
information on the research, development, and demonstration
activities in the water research program of the Environmental
Protection Agency, through inhouse research and grants and
contracts with Federal, State, and local agencies, research
institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications Branch
(Water), Research Information Division, R&M, Environmental
Protection Agency, Washington, D.C. 20460.
-------
CLOSED SYSTEM WASTE MANAGEMENT FOR LIVESTOCK
by
Dr. Patrick 0. Ngoddy, Assistant Professor
Principal Investigator
Jerome P. Harper, Chemical Analyst
Robert K. Collins, Engineering Technician
Grant D. Wells, Graduate Research Assistant
Farouk A. Heidar, Graduate Research Assistant
Agricultural Pollution Control Laboratory
Department of Agricultural Engineering
Michigan State University
East Lansing, Michigan 48823
for the
OFFICE OF RESEARCH AND MONITORING
ENVIRONMENTAL PROTECTION AGENCY
Project #13040 DKP
June 1971
For sale by the Superintendent ol Documents, U.S. Government Printing Office, Washington, D.C. 20102 - Price $1.25
-------
EPA-Review Notice
This report has been reviewed by the Environmental Protection
Agency and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency nor -does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
ii
-------
ABSTRACT
The vibrating screen separator is examined for liquid-solid separation
of livestock wastewater. A general procedure for performance analysis
and estimation of this type of dewatering apparatus is developed and
verified using methods of dimensional analysis. Tests on swine and
beef cattle wastewaters show that the resistant or slowly biodegradable
solids are effectively removed on vibrating screens ranging in mesh
size from #60 to #120. Although it is measurably less efficient than
conventional dewatering devices such as centrifuges and vacuum filters,
the gravity dewatering screen separator is better suited to the economic
scale of the average livestock operation.
The separated solid fraction from the screen is fibrous, odor free,
stable, storable and does not attract flies. The absence of these
solids from the liquid fraction is shown to stimulate faster rates of
COD and nuisance odor dissipation presumably by enhancing otherwise
rate-limiting transport processes in biological treatment systems.
The salient aspects of this study are integrated into candidate confine-
ment livestock waste management designs featuring
(i) the hydraulic transport of wastes with treated recycled wastewater
(ii) ultimate disposal of separated liquid and solid fractions without
pollution or nuisance odor problem.
This report was submitted in fulfillment of Project Number 13040 DKP
under the(partial)sponsorship of the Office of Research and Monitoring,
Environmental Protection Agency.
iii
-------
CONTENTS
Section Page
I Conclusions 1
II Recommendations 3
III Introduction 5
IV Production Rates and Pollutional Characterization 9
of Hog and Beef Cattle Manures
V Solid-Liquid Separation Studies 21
VI Characteristics of the Component Liquid and 65
Solid Fractions of Beef Cattle and Swine
Wastewaters
VII Comparative Biological Treatability Studies on 77
Swine and Beef Cattle Wastewater Liquid Fractions
VIII Design Alternatives 95
IX Acknowledgements 107
X References 109
-------
FIGURES
PAGE
1 METABOLISM CAGES FOR SWINE WASTE COLLECTION 10
2 METABOLISM CAGES FOR BEEF CATTLE WASTE COLLECTION 11
3 VERTICAL SECTION THROUGH A BASIC SINGLE DECK, 18" 24
SWECO VIBRATING SCREEN SEPARATOR
4 ILLUSTRATION OF AVERAGE MATERIAL TRAVEL ON SCREEN 25
5 OPTIMIZATION PLOT FOR SWINE WASTEWATER 31
6 PERFORMANCE CURVES FOR SWINE WASTEWATER 32
7 AVERAGE PERCENTAGE REMOVAL VS. MESH SIZE OF SCREEN 37
PLOTTED ON LOG-PROBABILITY PAPER
8 SCHEMATIC OF EXPERIMENTAL APPARATUS FOR VIBRATING 40
SCREEN TESTS
9 CLOSE-UP VIEW OF SETUP FOR SCREEN TEST 41
10 ALTERNATIVE PLOTS OF THE SCREEN TEST DATA ON SWINE 51
WASTEWATER
11 PERFORMANCE CURVES FOR SWINE WASTEWATER 56
12 OPTIMIZATION PLOTS FOR BEEF CATTLE WASTEWATER 57
13 SCREEN PERFORMANCE CURVE FOR BEEF CATTLE WASTEWATER 58
14 DIAGRAMATIC SECTION THROUGH EXPERIMENTAL SAND FILTER 60
15 CLOSE-UP VIEW OF SETUP FOR SAND FILTER TEST 61
16 CLOSE-UP VIEW OF EXPERIMENTAL COMPOST SETUP 74
17 TEMPERATURE PROFILE OF COMPOST BEDS 75
18 CLOSE-UP VIEW OF ANAEROBIC DIGESTERS 78
19 DETAIL OF ANAEROBIC DIGESTER 79
20 ANAEROBIC DIGESTER PERFORMANCE PLOTS 84
21 VOLATILE SOLIDS DISSIPATION FOR SWINE WASTEWATER 85
ANAEROBICALLY DIGESTED AT 95 °F
vi
-------
22 VOLATILE SOLIDS DISSIPATION FOR BEEF CATTLE WASTE- 86
WATER ANAEROBICALLY DIGESTED AT 95°F
23 CLOSE-UP VIEW OF BIOXIDATION APPARATUS 87
24 BIOXIDATION OF SWINE WASTEWATER LIQUID FRACTION, COD 90
REMOVAL RATE PLOT
25 BIOXIDATION OF BEEF CATTLE WASTEWATER LIQUID FRACTION, 92
COD REMOVAL RATE PLOT
26 LAYOUT OF PROPOSED WASTE MANAGEMENT SYSTEM 96
27 PROPOSED CLOSED SYSTEM WASTE MANAGEMENT FOR LIVESTOCK 97
28 LAGOON SECTION -- DETAILS 99
vii
-------
TABLES
PAGE
1 ANIMAL WASTE CHARACTERISTICS IN TERMS OF 1000 LB. 13
LIVE WEIGHT
2 SWINE INPUT-OUTPUT SUMMARY 14
3 SWINE FEED MIX 15
4 POLLUTIONAL CHARACTERISTICS OF SWINE WASTE 16
5 MINERAL ANALYSIS -- (FEED, FECES, URINE) SWINE 17
6 FEEDER CATTLE INPUT-OUTPUT SUMMARY 18
7 MINERAL ANALYSIS -- BEEF CATTLE WASTE 19
8 VARIABLES AFFECTING THE PERFORMANCE OF VIBRATING 27
SCREEN SEPARATOR
9 SIEVE ANALYSIS AND THE REMOVAL PERCENTAGE ESTIMATES 35
OF SWINE WASTEWATER SOLIDS
10 SIEVE ANALYSIS OF BEEF CATTLE WASTEWATER SOAKING 35
TIME 24 HOURS, APPROXIMATELY 5% TS
C£ - C D
11 VALUES OF A = (-^r X 100) ^ AS USED IN THIS STUDY 39
Cf Ds
12a RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE 43
WASTEWATER #50 Mesh Screen
12b RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE 44
WASTEWATER #74 Mesh Screen
12c RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE 45
WASTEWATER #120 Mesh Screen
12d RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE 46
WASTEWATER #200 Mesh Screen
12e RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE 47
WASTEWATER #230 Mesh Screen
I3a RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON BEEF 48
CATTLE WASTEWATER #74 Mesh Screen
13b RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON BEEF 49
CATTLE WASTEWATER #74 Mesh Screen
viii
-------
13c RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON BEEF 49
CATTLE WASTEWATER #120 Mesh Screen
13d RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON BEEF 50
CATTLE WASTEWATER #50 Mesh Screen
14 SAND FILTER STUDY -- SWINE WASTEWATER 62
15 SUMMARY SHEET CHARACTERISTICS OF THE SOLID FRACTION 66
(SWINE AND BEEF CATTLE WASTEWATER)
16a SWINE WASTEWATER SCREENED 67
16b NUTRIENT AND MINERAL ANALYSIS OF SWINE WASTEWATER 68
SOLID & LIQUID FRACTIONS
16c BIOLOGICAL OXYGEN DEMAND -- SWINE (80% URINE + 69
20% FECES) (NO DILUTION)
17a BEEF CATTLE WASTEWATER SCREENED 70
17b NUTRIENT AND MINERAL ANALYSIS OF BEEF CATTLE 71
WASTEWATER LIQUID & SOLID FRACTIONS
17c BIOLOGICAL OXYGEN DEMAND -- BEEF CATTLE (60% FECES + 72
40% URINE) (NO DILUTION)
18 ANAEROBIC DIGESTION OF SWINE WASTEWATER LIQUID 81
FRACTION #60 Mesh Screen
19 ANAEROBIC DIGESTION OF BEEF CATTLE WASTEWATER LIQUID 82
FRACTION #60 Mesh Screen
20 AEROBIC DIGESTION OF SWINE WASTEWATER LIQUID FRACTION 89
21 AEROBIC DIGESTION OF BEEF CATTLE WASTEWATER LIQUID 91
FRACTION
IX
-------
SECTION I
CONCLUSIONS
The following conclusions have been derived from this study:
1. The removal of resistant solids from wastewater prior to stabiliza-
tion substantially improves the performance of biological treatment
systems by enhancing the controlling transport and kinetic mechanisms
in such processes.
2. High efficiency removal of the resistant solids can be achieved on
vibrating screen separators at costs which now make the solid separation
step a feasible improvement on present designs of biological treatment
systems for livestock wastewater.
3. The solids from the screen are odorless, stable and storable over
extended periods without an odor-nuisance or pollutional problem.
4. The liquid fraction from the screen can be partially reclaimed by
biological oxidation and recycled as transport water in confined live-
stock operation.
5. A rational procedure for performance evaluation and/or prediction of
vibrating screen equipment for the separation of livestock wastewater
solids has been developed. Since this appears to be the first reported
case in which a systematic analysis of vibratory screen performance has
been attempted, it is hoped that the general formalism developed would
have a wide scope of application in aiding the design, specification and
testing of this type of machinery in the future.
6. As a consequence of this study, a number of candidate livestock
waste management designs integrating the salient features of the study
have been proposed.
7- Sequential anaerobic/aerobic treatment of the liquid waste fraction
shows promise. More complete treatment occurs in such a sequential
process than can be expected in a strictly aerobic process of the same
duration.
-------
SECTION II
RECOMMENDATIONS
As a result of this study we are prepared to recommend:
1. Intensive future investigation concerning the value of the solid
fraction of animal waste as a feed supplement for livestock. This type
of study should include extensive feeding trials in which animal growth
responses and health are closely observed as sole determinants of nutri-
ent availability.
2. A recycle scheme is the most resourceful method of complying with
extended winter storage regulations while minimizing odor nuisance,
pollution potential and expense over the long haul.
3. In the hypothetical situation of a livestock producer operating a
1000-hog confinement unit with recycled flushing water and an aerated
lagoon for treatment of same, an eighteen inch diameter vibrating screen
separator operating thirteen hours per day will suffice.
4. Further intensive research of sand filters is needed before they
can be considered a feasible means of animal waste solids-liquid separa-
tion. Hydraulic studies and management studies directed at plug-up
problems should be given first priority in such studies.
5. In the hypothetical case mentioned above, a technically feasible
system including an aerobic lagoon with diffused aeration, solid-liquid
separator, pumping station and automatic flushing equipment could be
constructed at an estimated 33,650 dollars (1970).
6. Floating aerators can be substituted into the above design to save
approximately 5000 dollars (1970).
7. Since this study was strictly of a laboratory nature, there is a
need to further prove aspects of these designs at the pilot scale before
they can be considered fully operational. Such pilot studies should be
directed toward defining the critical constraints on wastewater recycling,
would provide large quantities of solids for feed trials and should pro-
vide more reliable information on the comparative costs and reliability
of such systems.
-------
SECTION III
INTRODUCTION
General Remarks:
As we ponder the perplexing economic, social and political ramifications
of agricultural operations structured within rapidly broadening bounda-
ries of environmental responsibility, certain very recognizable patterns
of the future begin to emerge. Among such traits can be seen quite dis-
tinctly some of the more obvious constraints within which future waste
management systems in livestock operations must perform. Summarized
into a broad statement of goals, it seems clear that the design of such
systems must integrate:
(a) modern labor-saving methods of materials handling
(b) effective treatment techniques
(c) concepts in waste recycling into conservative closed or almost
closed strategies limiting the total bulk of pollution and minimizing
costs over the long haul.
One significant, relatively recent labor-saving development in animal
waste management is the hydraulic handling and transport of animal
manures in confined livestock operations. As a practice which lends
itself readily to varying degrees of automation, it seems destined to
continue to sustain high levels of acceptance in a circumstance in which
present and projected shortages and high cost of labor have accentuated
the need for even higher levels of automation on the farm.
Although the present use of hydraulic systems is almost exclusively re-
stricted to confinement in-building operations, it could hold the key
for dramatic improvements in the collection and handling of waste de-
posits in large open feed lots. Where such feed lots constitute air
and water pollution hazards -- particularly close to large population cen-
ters it seems likely that these units may have to be paved so that
hydraulic methods can be employed to achieve much higher levels of con-
trol of the waste management aspects.
As presently designed for hog and beef cattle production operations,
hydraulic transport facilities range from modest water-filled pits and
oxidation ditches under slotted floors to fairly sophisticated automated
flushing devices of a variety of designs. Beside their labor-saving
advantages, these systems make for an overall cleaner, safer, odor-free
and more attractive environment for both the producer and the animals.
However, irrespective of the particular design, large quantites of
pollution-laden, odorous water emanate from these systems.
One popular method of-managing this water at the present time is to pro-
vide temporary storage for it in in-building, outside, above-ground or
-------
underground storage tanks. If provided over the winter months when the
ground is frozen and land application is inadvisable, such storage runs
up a considerable expense. For Michigan and a substantial portion of
the midwest region of the United States, a conservative estimate of
minimum winter storage requirement is between 4 to 6 months. Costs of
adequate facilities to provide such storage for a 1,000-hog farm has
recently been estimated to be between $15,000-$25,000 in capital invest-
ment (1). When ultimately spread on the land following extended storage,
untreated wet manure poses serious odor nuisance and water pollution
problems which are responsible for an ever increasing number of legal
suits. In the light of present and growing sensitivities at regional
and national levels concerning environmental issues, a reasonable pro-
jection is that such litigations will occur with even greater frequency.
It seems clear that livestock production must brace to make its opera-
tions compatible with maintaining the integrity of the environment.
An alternative to the store-and-spread practice considered above is the
partial treatment of the pollution-laden wastewater by anaerobic, aero-
bic or combination biological processes intended to minimize the pollu-
tion potential of the wastewater and eliminate the odor nuisance problem
prior to ultimate disposal on the soil. Abatement processes of this
type are exemplified by the oxidation ditch, anaerobic and aerobic
lagoons which are gaining considerable attention of late (2). In general
the design of these treatment systems has been derived from sanitary
engineering practice. It is recognized (2, 3), however, that animal
waste slurries are much too loaded with high oxygen-demanding organic
pollutants to permit the direct transfer of design information from the
relatively dilute domestic or municipal wastewater with which sanitary
engineering is normally concerned. The large number of treatment
studies conducted over the past ten years show conclusively that certain
design modifications are mandatory if biological processes are to meet
with any degree of success in the treatment of highly polluted animal
wastewaters.
As research continues to seek innovative modifications in conventional
designs of biological processes that would make such processes more
effective in the abatement of the unique odor and pollutional potential
of animal wastewater, the essential reduction of the high solid matter
in livestock wastewater prior to conventional biological treatment ap-
pears to be the one requisite task about which universal agreement cur-
rently exists among researchers. A sizable amount of these solids are
non-biodegradable organics and salts whose presence in the treatment
system effectively impedes transfer processes within the biological
"reactor". In hog wastewater for example, such refractory or only slow?
ly-biodegradable fraction amounts to about sixty to seventy percent of
the total solids. The presence of these solids contributes significantly
to the high viscous condition of the mixed liquor and thereby substanti-
ally increases the power demand of oxygenation and mixing equipment of
mechanically aerated systems. It seems logical that the removal of these
resistant solids would considerably improve treatment kinetics of aero-
bic and anaerobic biological processes. These improvements would in turn
-------
enhance the potential to recycle the partially treated wastewater as
wash and transport water for the hydraulic system. Considering the ex-
cessive concentration of oxygen-demanding pollutants in animal waste
slurries, it is essential that treated wastewater rather than fresh
water to recycled as a diluent to bring BOD and COD concentrations down
to the recognized levels of high efficiency treatment.
The study reported here has evolved from the basic rationale that effi-
cient solid-liquid separation of livestock wastewater is a necessary and
integral part of any wastewater management scheme in which water is to
be reclaimed and profitably recycled. The recycling of reclaimed waste-
water offers the promise of minimizing capital investment costs on live-
stock wastewater management systems by substantially reducing the total
bulk of water polluted and subsequently stored and/or processed.
project Objectives:
The broad objective of this study is to develop the technical basis for
a liquid manure processing system which (a) isolates the refractory com-
ponent solids from livestock wastewater, and (b) subsequently handles
the solid and liquid fractions in such a manner that they do not individ-
ually or collectively constitute an aesthetic or pollutional problem.
It is an implied prerequisite that the system remain within the recog-
nized economic scale of the average livestock operator.
The specific tasks to which the study was directed are as follows:
1. To identify and carry out a comprehensive analysis of a technically
and economically feasible solid-liquid separation system for livestock
wastewater.
2. To characterize (physically, chemically and biologically) the solid
and liquid component fractions.
3. To examine where necessary processes by which each component fraction
can be reduced to a stable, storable or reusable end product with mini-
mal odor nuisance or pollutional potential when ultimately disposed of
on the soil.
4. To integrate the essential features of the above findings into least
cost candidate designs at the pilot scale. It is hoped that one or more
of these designs would constitute the basis for possible future large-
scale demonstration studies.
Because of time constraints on the project, these objectives were carried
out only for swine and beef cattle wastewaters.
-------
Report Format:
In accordance with the outlined objectives, the report is presented in
five major sections. The first part gives production and pollutional
characterization data of beef cattle and hog manures determined in the
preliminary stages of this study as compared to similar data presently
available in the literature. The second part presents solid-liquid sepa-
ration studies conducted on hog and beef cattle wastewater. Section
three of the report deals with the characteristics of the component solid
and liquid fractions. Section four presents data and conclusions on
stabilization requirements of the solid and liquid fractions. Section
five consists of a number of proposed pilot designs with cost projections
which integrate aspects of the study.
-------
SECTION IV
PRODUCTION RATES AND POLLUTIONAL CHARACTERIZATION
OF HOG AND BEEF CATTLE MANURES
In summarizing his elaborate survey of existing data on the production
and characteristics of animal manures, Loehr (2) concluded that published
values vary due to differences in:
(i) housing and management practices
(ii) types of rations fed
(iii) analytical techniques employed
(iv) manure handling and collection techniques.
It therefore seems essential that investigators dealing with aspects of
the animal waste problem should report the conditions under which the
animals were fed and housed and the manner in which the waste samples
were collected and analyzed so that a continuity can be maintained in
correlating production and characterization data on these wastes. This
part of the report is presented with this purpose in mind, and also as
a data base for aspects of the study to be taken up in subsequent parts
of the report.
Collection Facilities;
All the manure used in this study was collected from animals confined in
individual metabolism cages. Figures 1 and 2 show the cages used for
swine and beef cattle manure collection. The cage arrangement made it
possible to collect the feces and urine separately. For the hogs, the
feces were held up on the floor screens of the cage while the urine go-
ing through the screens was collected in narrow-neck plastic bottles.
For the beef cattle, the animal was held in place (as shown in Figure 2)
in such a way that the feces were discharged directly into a plastic bag
and the urine was collected in funneled plastic bottles. One unique
aspect of this arrangement is that evaporative losses were held down to
a minimum in the urine.
Collection was done once a day at which time the urine and feces were
weighed separately. In these categories, the manure was stored in its
natural form without any added dilution water. Storage for short periods
before use was done at 4°C. If preserved over extended periods, storage
was done below freezing temperatures.
Feed Rations;
The animals used for collection purposes were in general experimental
animals at the Michigan State University swine and beef cattle farms.
-------
FIGURE 1: METABOLISM CAGES FOR SWINE WASTE COLLECTION
-------
FIGURE 2: METABOLISM CAGES FOR BEEF CATTLE WASTE COLLECTION
-------
For this reason the feeds given to these animals cannot be considered
typical from a commercial standpoint.
The animals had access to as much water as they could possibly drink.
They were fed once a day. A total of 5 hogs in different age groups were
isolated in metabolism cages over periods varying from 3-10 week periods.
The feed was primarily ground shelled corn (79%-86%) with 16%-13% protein
supplement. In Table 3 a complete breakdown of the feed mix for the hogs
is presented. Mineral analysis of the feed, feces and urine are tabu-
lated in Table 5.
The primary feed for the beef cattle animals was corn silage of 35% dry
matter. The feed is tabulated along with the defecation summary in
Table 6. The animals had free access to water. The corn silage re-
ceived some prestorage treatment which was part of an experiment being
conducted by the Animal Husbandry Department.
Analytical Methods;
(i) Biological oxygen demand (BOD), chemical oxygen demand (COD), nitro-
gen, total solids (TS), and total volatile solids (TVS) were determined
using procedures specified in Standard Methods for the Examination of
Water and Wastewater (1965).
(ii) Phosphorus; was determined by first carrying out a perchloric
acid digestion of the sample and then doing the final analysis on a
photometric instrument. Reference: Methods of Soil Analysis, Vol. 9,
Part 2 (13).
(iii) Nutrients and Minerals; (K, Ca, Mg, Zn, Cu, Fe, Mn, Na) were
determined by first digesting the sample sequentially in nitric and per-
chloric acids and then doing the final analysis by flame photometry on
an atomic absorption apparatus. Reference: Methods of Soil Analysis,
Vol. 9, Part 2 (13).
(iv) Samples: In cases where composite samples were analyzed, the
samples were composited as follows:
(a) hog wastes: 20% feces + 80% urine
(b) beef cattle wastes: 60% feces + 40% urine.
These percentage fractions were approximated from the collection data
obtained in this study.
As a rule, nitrogen analysis was conducted as soon as possible following
collection to minimize losses of the volatile forms which are known to
occur.
Results and Discussion of Collection and Characterization Data;
The data obtained from this study on manure production rates and manure
characteristics are presented in as complete a form as was considered
necessary in Tables 2, 3, 4 and 5 for hogs and in Tables 6 and 7 for
12
-------
TABLE 1: ANIMAL WASTE CHARACTERISTICS IN TERMS OF 1000 LB. LIVE WEIGHT
Raw Manure Produccion Rate
(Ibs/day)
Feces/total prod. (7.)
Total Solids (Ibs/day)
Total Solids (7. wb)
Volatile Solids (Ibs/day)
Volatile Solids (7. TS)'
BODj (Ibs/day)
BODj (Ibs/lb of VS)
Reaction Rate
Constant (Io810>
BOD5/COD (I)
Nitrogen (total Kjeldahl)
(1 TS)
(Ibs/day)
P (Z TS)
K (1 TS)
Ca (1 TS)
Mg (Z TS)
Zn (Z TS)
Cu (Z TS)
Fe (% TS)
tti (Z TS)
Na (X TS)
Po
(a)*
50
25
5
10
4.1
83
1.8
0.44
0.8
44
7.0
.35
3.7
3.49
2.47
1.20
0.05
0.05
0.05
0.02
0.63
rk Pigs
(b)
50
50
7.2
14.4
5.9
82
2.1
0.36
33
5.6
0.4
~
1.4
--
--
--
(c)
28-95
8-16
12-28
3.5-8.0
83-87
2.0-5.6
3.30-0.54
19-45
3.3
0.42-0.6
0.5-1.2
1.3-3.2
2.0-4.7
0.3-0.65
0.006
0.002
0.1-0.2
--
E
(a)
35
66
4.9
14
3.8
77
1.7
0.45
0.14
38
6.2
.30
1.7
2.27
1.16
0.47
0.01
0.08
0.01
0.09
eef Catt
(b)
--
80
0.252
9.8
--
--
~
~
--
le
(c)
63
~
9.5
15
3.17
1.02
0.28-0.32
~
31-40
3.1
1.35
3.0
0.8
0.65
0.03
Dai
(b)
88
--
9
10
7.2
80
1.7
0.233
16
4.0
0.36
1.7
~
~
ry Cattle
(c)*
38-74
66
9.5-11.4
13-27
5.7-8.3
0.44-1.53
0.13-0.23
~
8-23
3.5
0.35-0.44
0.3-0.7
1.8-3.8
1.0-2.1
0.4-0.8
0.0015
0.0005
0.01-0.03
--
'Column (a) values obtained by authors of this report (1971)
Column (b) average suggested values by Taiganides (1-971)
Column (c) range of tabulated values by Loehr (1968)
Column (d) Taiganides (1962) Ph.D. Thesis
Column (c)* calculations based on tabulated values by Loehr (1968)
13
-------
TABLE 2: SWINE INPUT-OUTPUT SUMMARY
h
o
a
ii
fl
(4)
initial wt
164 Ibs
final wt
180 Iba
AVERAGE
I
1
S
H
v. o
d ^
fl u
is a
a
v 4-
**.
a " a
-------
beef cattle wastewater. Along with the production and characterization
information are presented complementary data about feed rations, animal
sizes and rate of weight gain which are considered relevant.
In Table 1, a summary of all this data is developed and presented in com-
parison with the most widely accepted values in the current literature.
Overall, there are no differences which cannot be explained on the basis
of differences either in collection practices or in feed rations. In
this connection, it is essential to note that animals confined in cages
such as used in this study do not exercise enough and therefore may eat
and defecate in rather unusual patterns. The methods of analysis for
nutrients and minerals used in this study are believed to be better than
most used previously in animal waste studies. Differences in analytical
techniques may account for some of the differences observed in nutrient
and mineral values.
TABLE 3: SWINE FEED MIX
167. Protein 13% Protein
Ground Shelled Corn 79% 85.5%
Soybean Meal 18 11.5
Dicalcium Phosphate 1.0 1.0
Ground Limestone 1.0 1.0
Trace Mineral Salt 0.5 0.5
MSU Vitamin-Trace Min Mix 0.5 0.5
100.00% 100.00%
15
-------
TABLE 4: POLLUTIONAL CHARACTERISTICS OF SWINE WASTE
Weekly
Composites
1
2
3
4
5
6
7
8
9
10*
Average
COD
(mg/1)
86,000
76,000
61,000
68,000
90,000
80,000
74,000
69,000
81,000
70,000
80,000
BODc
(mg/1)
32,000
30,000
27,000
29,000
47,000
43,000
41,000
29,000
44,000
30,000
35,000
TOTAL N
(mg/1)
8540
7000
5600
5950
9170
6160
6500
6000
8000
5900
6880
NH~ - N
(mg/1)
4611
3780
2856
2915
3668
3449
1950
3240
4240
3186
3390
% NH3 - N
OF TOTAL N
54
54
51
49
40
56
30
54
53
54
49
%
TS*
8.7
8.0
6.3
5.6
9.56
8.00
8.3
8.5
10.5
8.2
8.16
%
VS
80
84
82
83
83.3
83
82
83
84
81
83
Average less than normal because of greater than normal water intake.
-------
TABLE 5: MINERAL ANALYSIS -- (FEED, FECES, URINE) SWINE
Ca
Hg
Zn
Cu
Fe
Mn
Na
K
P
S
N
(ppn)
(ppm)
(PPn)
(ppn)
(ppn)
(ppn)
(ppn)
(ppn)
(ppn)
(ppn)
(ppn)
Expressed as
Ca
Mg
Zn
Cu
Fe
Ma
Na
K
t
S
N
Feed
13% Moisture
7,980
1,690
106
19
140
34.7
2,170
5,930
6,450
3,960
24,700
Grams/Gram Solids
.009172
.0019425
.001218
.00002183
.0001609
.0000398
.0031149
.006816
.0074137
.0045517
.0283908
Feces 65% Moisture
(Trial 1) (Trial 2)
27,780
9,300
432
117
513
213
3,060
11,730
18,110
1,010
34,800
.078474
.026271
.001220
.0003305
.001449
.0006016
.008644
.033135
.0511582
.002853
.098305
22,520
6,740
585
98
398
140
2,210
8,770
15,360
1,070
34,400
.063615
.019039
.0016525
.000276
.001124
.000395
.0062429
.024774
.043389
.0030225
.097175
Urine 96% Moisture
'(Trial 1) (Trial 2)
168
68
3.1
.16
1.1
.32
1,510
2,300
224
980
5,190
.0041379
.001674
.0000763
.0000039
.000027
.000008
.037192
.056650
.005517
.024137
.12783
511
109
1.5
.17
1.2
.31
1,090
2,330
133
1,230
4,770
.01258
.002684
.0000369
.0000041
.0000295
.0000076
.026847
.057389
.003275
.03029
.11748
17
-------
TABLE 6: FEEDER CATTLE INPUT-OUTPUT SUMMARY
(one week trials, 3 animals in each weight division)
Live Weight
(Ibs)
750
750
750
750
1100
1100
1100
1100
750
750
750
750
Feed**
Rate
(Ibs/day)
38
38
33
37
39
40
40
40
35
35
37
34
%
Shelled
Corn
--
21
25
3
55
52
55
52
--
--
%
Mineral
Supplement
5
5
*
-
4
9
4
9
*
*
2
Waste
Elimination
Rate
(Feces+Urine)
(Ibs/day)
22
28
28
51
37
38
36
38
17
27
30
28
Fecal
Material
(as collected)
(Ibs/day)
12
17
19
30
19
17.5
23
17.5
13
20
25
19
**Primary feed was corn silage 35% dry matter in all cases
Free access to water
*Pre-storage treated corn silage
18
-------
TABLE 7: MINERAL ANALYSIS -- BEEF CATTLE WASTE
Composite: 60% feces +40% urine
Ca
Mg
Zn
Cu
Fe
Mn
Na
K
PO,
4
Expressed as PPM
1,700
690
18
27
122
16
130
3,350
7,750
Expressed as % TS
1.16
0.47
0.012
0.035
0.083
0.011
0.088
2.28
5.27
19
-------
SECTION V
SOLID-LIQUID SEPARATION STUDIES
Preliminary Considerations:
Recent studies (4, 5) on solid-liquid separation of livestock wastewater
appear to have established that solid removal efficiencies in excess of
90% can be readily achieved in standard sludge handling equipment such
as centrifuges and vacuum filters. However, a cost survey undertaken at
the preliminary stages of this study showed conclusively that capital
investment costs on centrifuges, presses and vacuum filters far exceed
what can be considered feasible for the average operator.
It was decided early in the project to exert the major research effort
on more realistic solid separation processes. Several chemical floccu-
lants were first examined. These included ferric chloride solutions and
and a whole series of anionic, cationic and nonionic polyelectrolytes
with and without -- pH adjustment. Although a number of these floccu-
lants were effective in concentrating the solids, they invariably pro-
duced an odoriferous sludge which had further to be processed or else
would constitute a major odor nuisance problem when ultimately spread.
It was felt in addition that the close supervision necessary to operate
a chemical process of this nature would be beyond what one can realisti-
cally expect at the level of the individual operator. Besides, the cost
of chemicals would be a continuing expense which can become quite tangi-
ble over time. For these reasons, it was decided that chemical floccu-
lation per se was both insufficient and operationally unfeasible.
Attention was subsequently turned to mechanical gravity-dewatering de-
vices. Of the two recognized gravity units known to be used for sludge
dewatering, the mechanical vibrating screen separator is more easily
accessible, more versatile and less expensive than the centrifuge (6).
In solid liquid separation operations, this type of screen provides
superior quality separated solids since the vibrating action simultane-
ously washes and squeezes the solids to a cleaner and drier end product.
In 1968, a University of California Agricultural Extension Bulletin
authored by Fairbank and Bramhall (7) reported the first documented case
in which a vibrating screen separator was used for the fractionation of
livestock wastewater. In their experience with dairy wastes in Cali-
fornia, Fairbank and Bramhall emphasized the solid washing characteris-
tics, mechanized or semi-automated operation and superior capacity to
handle large volumes of flow as the primary advantages of rotating or
vibro-energy separators. They observed in particular, that the separated
solid fraction exhibited a high degree of apparent stability, was fibrous,
relatively.odor-free, non-fly attracting and can be stored for extended
periods. These observations, though qualitative in nature and scientifi-
cally inconclusive, showed a very high degree of correspondence with the
solid quality criteria considered desirable. A cost survey was-conducted
which showed that capital outlay on this type of equipment lies well
21
-------
within the $1,000-$5,000 margin which we considered comfortable. The
machine is mechanically simple, easy to operate and requires very minimal
maintenance. These attributes were considered very desirable features.
Preliminary tests with laboratory ro-tap reciprocating sieves indicated
that solid removal efficiencies between 50%-70% are well within reach
for swine and beef cattle wastewater on screens ranging in mesh size
from 60 to 120. The resulting solid residue had moisture content be-
tween 70 and 80 percent (wet basis), had a friable non-clumping structure,
was odor-free, and when left exposed over extended periods attracted no
flies. The effluent liquid from the screen contained small particulates
which settled slowly and had a high percentage of matter in the colloidal
range. Small scale digestion tests in anaerobic digesters indicated that
at dilutions of 1:5 and 1:10, over 90% of the volatile suspended solids
were dissipated in 5 to 10 days at 95°F. It was inferred from these
observations that biological processes preceded by solid separation in
a vibrating sieve could hold interesting prospects.
Engineering Analysis of the Vibrating Sieve Separator
In spite of the fact that vibrating screens were used in Germany as far
back as 1954 (8) for sludge dewatering and have been in extensive use as
a multi-purpose separator in a wide variety of industries since then,
there does not exist today any satisfactory method by which this equip-
ment can be selected to do a particular job. The usual method of selec-
tion is on the basis of tests with smaller geometrically similar ma-
chines. However, the installation of such machines for testing purposes
is sometimes not possible, or the supply of sample wastewater may be
limited (as for example from a small pilot plant). Under these circum-
stances it is not possible to predict the performance of large scale
vibratory separators.
The purpose of this section of the study is to attempt to provide a
rational basis for quantitative performance evaluation of vibrating
sieve separators when used for solid separation of livestock wastewater.
The study carried out so far is limited to:
(i) beef cattle and swine wastewaters and
(ii) only one type of vibrating sieve machine -- namely that which
derives its basic vibrational action from the interaction of suspension
drive springs and rotating weights subjected to known angular accelera-
tion.
It is hoped that the general method of attack will have a much wider
scope of application. However, as a note of caution, it is essential to
state at this stage that there can be no real substitute for performance
ratings based on actual tests on geometrically and functionally similar
machines. The effectiveness of such tests, the broad interpretation of
their results and the logical scale-up of such results can be consider-
ably improved by generalized methods of analysis such as will be presented.
22
-------
Basic Operating Principles;
The machine used in this study is the Sweco Vibrating Screen Separator
manufactured by the Southwestern Engineering Company of Los Angeles.
Although the selection of this particular machine was strictly a matter
of convenience, it is very typical of vibrating screen separators which
derive their excitation from rotating weights. It is also one of the two
leading brands in industrial use. The 18-inch diameter model of this
machine was borrowed from a local distributor for the purpose of this
s tudy.
The essential component parts of the machine are shown diagramatically
in Figure 3. Successful solids separation with the machine depends on
the accomplishment of a number of distinct processes. The first step is
to achieve a pattern of travel over the screen cloth that will give the
efficiency and final product desired. This pattern varies with the char-
acteristics of the material, rotational speed of the drive motor, the
magnitude and setting of the bottom and top weight assemblies. The gross
vibrational force can be attributed to a complex, multi-dimensional gen-
eral acceleration function of sinusoidal origin whose precise representa-
tion is not possible. However, a useful simplification is to consider
the effective components of the vibrational force in the vertical and
horizontal directions. The vertical component force is responsible for
vertical motion of particles on the screen and imparts a compressive
action on the solids tending to squeeze out the liquids. The horizontal
component force is responsible for the horizontal material flow pattern
which moves the dewatered solids out. The pattern of travel on the
screen is regulated by the degree of lead (lead angle) of the adjustable
"bottom motor weight" with respect to the "top motor weight". The rela-
tive size of the weights, and the phase differential in their angular
rotation affect the vibrational force distribution in the vertical and
horizontal planes. These parameters must therefore be expected to vary
with wet and dry materials, particle size, moisture content, percentage
solids, density, opening size of screen cloth, screen analysis of feed,
feed rate and the desired capacity and efficiency.
A distinct step which deserves a special mention is the "roping" or
"balling" phenomenon which plays a very important role in determining
materials travel patterns on the screen. These balls form either as
individual agglomeration of particles or as connected balls. Balling
appears to be very closely related to moisture content and consistency
of the separated solids. Under steady state conditions on the screen,
balls form readily when the material consistency is such that rolling
friction on .the screen is just overcome by the resultant of the vertical
and horizontal component forces. The ball sizes result from the inter-
action between vibrational and sliding frictional forces on the screen
surface. In Figure 4 an illustration of the average material travel
patterns on the screen is presented.
23
-------
Feed
Interchangeable
Sieve
Bucket type
Velocity Reducer
Center tie
down washer
Drive
Springs
Liquids dis-
charge spout
Drive Motor
13"
Top/bottom
weight lead
angle adjuster
-18"-
FIGURE 3: VERTICAL SECTION THROUGH A BASIC SINGLE DECK, 18" SWECO
VIBRATING SCREEN SEPARATOR
24
-------
0* Lead - Average material may
be thrown straight and may give
insufficient separation.
15° Lead - Average material may
begin to spiral.
40° Lead - May give average dry
material maximum efficient
screening pattern.
60° Lead - May give average wet
material maximum efficient
screening pattern.
Maximum Lead may
keep oversize^
material from
being discharged^
and assist in
receiving
maximum thruput
of minus material
which tend to "ball1
Roping Pattern on
the screen
discharge
spout
"Rope" of
material
screen
surface
FIGURE 4: ILLUSTRATION OF AVERAGE MATERIAL TRAVEL ON SCREEN
25
-------
Analysis of Vibratory Screen Performance:
As a solid-liquid separator, the vibrating screen should perform two
functions :
(i) produce a clean centrate, and
(ii) produce a clean low-moisture cake.
The two obvious measures of performance would therefore be (a) centrate
suspended solids concentration and (b) cake solid concentration or mois-
ture content.
It is customary in the estimation of centrifuge and vacuum filter sludge
dewatering performance to define performance in terms of percentage
solids removed. It is convenient to adopt this practice here. We there
fore define performance designated by a symbol R, where:
_ wt. of solids in sludge cake - f. .
wt. of solids in the feed * iuu u;
It is convenient to use a form of equation (1) in which only concentra-
tion terms are employed, so that
R = Cf (Ck . Cc) X 100 (2)
where Cf = feed solids concentration
Cjj = cake solids concentration
C = centrate solids concentration
Equation (2) is approximated by a simpler more convenient form:
Cf * Cc
R = -~ - X 100 (3)
Cf
Variables Affecting Vibrating Screen Performance:
The paramount variable affecting vibrating screen performance is obvious-
ly the design. Even though each design would have certain peculiarities,
there are a number of variables which can be generalized for all machines
which work on the same basic vibrational principle. These can be deline-
ated into two broad categories of parameters -- namely process and machine
variables. From the previous discussion on the operating principles of
this type of machinery a summary table of these variables has been pre-
pared (Table 8) .
Obviously this list is formidable. It does not seem likely that with so
many variables we can intelligently and coherently discuss vibratory
sieve performance. It is therefore necessary to abridge this list to
26
-------
manageable proportions. We need to first assume that we are dealing
with a particular machine separating a particular wastewater. The perti-
nent machine variables now become:
(i) Magnitude of top and bottom weights, (W and W respectively; Ib )
(ii) Lead angle between top and bottom weights («$; dimensionless)
(iii) Percentage opening of screen (§; dimensionless)
(iv) Screen mesh size (D ; ft)
S
2
(v) Effective surface area of screen (A; ft )
The pertinent process variables are:
(i) Average particle size (D ; ft)
3
(ii) Flow rate of feed (Q; ft /sec)
(iii) Residence time (T; seconds)
(iv) Centrate density (p ; j-)
Ib
m
(v) Centrate viscosity (|j, ;
The third process variable, T, is considered to be very important since
it lumps together into one variable all material characteristics and
those machine-material interactions which we cannot easily characterize
individually. One obvious example of this is the material properties
TABLE 8: VARIABLES AFFECTING THE PERFORMANCE OF VIBRATING SCREEN
SEPARATOR
Machine Variables
Process Variables
Screen diameter
Screen mesh size
Screen percentage opening
Rotational speed of motor
Top and bottom weights
Lead angle between top and
bottom weights
Number of springs
Spring constants
Feed point
Degree of artificial feed --
velocity dissipation at
feed point
Residence time on screen
Material characteristics
Particle size of particles
Particle shape
Particle density
Rheology
Solids concentration
Chemical additives and quantities
Liquid viscosity
Liquid density
Feed rate of liquid
27
-------
and their interaction with screen surface frictional forces, and with
vibrational force distributions to bring about balling. It seems quite
clear that any attempt to gain exclusive insight into this complex phenom-
enon would be futile. However, all these forces and a lot more play an
important part in determining the average length of time the individual
particle stays on the screen before it is discharged under steady state
conditions. This quantity T also provides us with a concrete way of se-
lecting optimum performance conditions on the screen. Any combination
of machine variables and process variables which gives the smallest T
value for the same cake moisture content and centrate suspended solids
concentration can be considered optimal.
Development of Model:
If a model is to represent vibrating screen performance accurately, it
must include the essential variables necessary for screen performance.
From the preceding discussions, the general form of any such model can
be represented using methods of dimensional analysis such that:
Recovery = f [(machine variables) , (process variables)]
Using equation (3) and the abridged list of pertinent variables, we can
represent the above relation in the following form:
Cf - Cc
R = C X 100 = f(W W , i, 5, D , A, D , Q,T, p Q ) (4)
1- U D S p C C
We can reduce the number of variables in equation (4) considerably by
defining a modified rate of flow (q) such that
.ft .
Equation (4) now becomes:
C - C
-^ - - X 100 = f (Wfc, Wb, rf, Dg, DpJ q, T, pc, 10,) (6)
The quantity q has the units of velocity and represents the flow rate
per unit surface area of the screen going through the screen openings.
Employing the Buckingham II theorem, we can form the following dimension-
less groups:
C - C
IL = R = -=- - - X 100, (percentage removal)
1 Cf
D
IL = =£ , (ratio of effective particle size and screen size)
28
-------
IL = *- , (dimensionless residence time)
s
q p D
c s
JT _ ^ (dimensionless rate of throughput)
*c
Note that II. is a form of modified Reynolds number defined for the cen-
trate and using screen mesh size as the characteristic dimension. Under
steady state conditions, for feed solids concentration below 5%, the cen-
trate rate of flow measured voluraetrically approximates the feed rate of
flow. The choice of centrate properties in defining IL is therefore a
matter of convenience since a term like viscosity is easier to measure
for suspensions in which the rate of settling of solids is slow enough
so that medium characteristics during measurement remain reasonably un-
changed. Also the density of the centrate closely approximates that of
water at the same temperature.
W
IL = b (dimensionless weight ratio)
wt
Since the magnitude of the bottom weight assembly as compared to the top
weight assembly affects the force distribution in the horizontal and
vertical planes, the ratio IL is a simple measure of the relative effect
of these component vibrational forces in two planes.
IL = $ (lead angle)
fa
It was previously stated that the lead angle regulates travel patterns on
the screen. This parameter must therefore have a strong effect on resi-
dence time.
There is at least one more dimensionless quantity which can be defined
involving the resultant vibrational force, F. This quantity would take
into account the angular acceleration responsible for the vibrational
force and can take the form of a modified Euler Number, F/q D^p . How-
ever, since the angular velocity is usually fixed for a given machine,
this quantity may be quite superfluous. It is particularly undesirable
since F is practically impossible to determine explicitly in a real
machine. Since we have already attained a sufficient representation of
the pertinent parameters in the first six dimensionless groups, we can
meaningfully characterize the system without a 7th group.
Equation (6) can now be written as
IL_ = x (IL,, n 3, n4, n5, n6)
°r Cf - C D . qp D W,
_f c x ,00 _ r_s 21 c s _b ,.
X 100 - (D , D , , , 0) (7)
f - p s 'c t
since the percentage removal is directly related to both effective parti-
cle size and screen mesh size we can extract IL from the parenthesis to
29
-------
obtain:
D C - C qp D W.
5* -V-5 x 10°= x (^' -r* - c - «
s f s ^c t
The left hand side of equation (8) being the product of two dimension-
less groups remains dimensionless. Note, however, that for the same
screen size and the same material, the dimensionless quantity on the
left hand side will remain effectively constant irrespective of the
other variables.
With the number of dimensionless variables specified in equation (8), it
is not possible to proceed directly to set up an explicit functional
predictive relationship between the variables. One simplification which
can make this task less burdensome is to specify a steady state con-
straint on the performance variables. As a continuous flow machine, the
vibrating screen separator relies very much on steady state conditions
for good performance. The pertinent process variables must be adjusted
to achieve a state of equilibrium on the machine.. For a given set of
machine variables represented typically by (^, -.£), the rate of feed
must be adjusted until a condition approximating*^ teady state performance
is observed on the screen. Under this condition, a characteristic flow
pattern maintains a steady stream of cake coming out of the screen. Our
study so far indicates that as long as measurements are taken under a
state of equilibrium, the cake maintains a uniform moisture content for
the same feed material irrespective of the combination of machine and
process variables necessary to generate equilibrium.
With the steady state constraint on the model, there are a number of
alternative paths along which we can proceed. One one considered most
expedient is to approach the problem in two basic steps:
1. The Optimization Step: in which we extract from equation (8) a gen-
eral residence time function in the form:
f ' \ & ' *> (9)
s t
It is now possible to define an optimum residence time within carefully
specified limits by seeking a minimum for equation (9). The experimen-
tal data obtained from vibrating screen performance tests on swine
wastewater have been used to graphically represent the nature of equa-
tion (9) in Figure 5.
This plot shows clearly that the equilibrium values of dimensionless
residence time when plotted as a function of lead angle with the bottom-
top weight ratio as a parameter, has well defined minima. Each minimum
point in Figure 5 represents an optimum residence time obtainable from
the combination of machine and process variables (D ,W^/ Wt, {>, q)
making up the plot. For a given screen, the curve giving the smallest
optimal dimensionless residence time as well as an optimal top-bottom
30
-------
Mesh Size of
Screen #120
Mesh size of
Screen #74
45° 60°
$ (Lead Angle)
FIGURE 5: OPTIMIZATION PLOT FOR SWINE WASTEWATER
31
-------
ixio TT
09
Q
cr
0)
0)
o
B
V
o
1X10
a
ca
4)
r-i
B
O
H
to
1X103
1X10"
D Mesh size screen #50
A Mesh size screen #74
a Mesh size screen #120
_O Mesh size screen #200
Mesh .size screen #230
100=Percentage Removal
A- R X-E
1X10
-3
1X10
-2
1X10
-1
1X10U
[Modified Reynolds Number^"
FIGURE 6: PERFORMANCE CURVES FOR SWINE WASTEWATER
32
-------
weight arrangement using this general procedure.
For swine wastewater, Figure 5 shows that the best residence times for
screen mesh sizes #74 and #120 occur when the top-bottom weight ratios
are respectively 3.16 and 3.46. The experiments on which the plots are
based will be described in detail in a subsequent section. It is appro-
priate to mention here, however, that these experiments were sufficiently
exhaustive to permit some generalization of their results.
2. Design Step: with optimal constraints already specified for «5 and
in step 1, it is now possible to simplify equation (8) to the form:
C£ - C D
X MO)
f s s K
c
Equation (10) is rearranged to the form
D^ = X3 (qPc s A> (11)
s |-l
C - C D
where A = ( X 100)
Experimental data obtained from vibrating screen tests on swine waste-
water are plotted in Figure 6 in accordance with equation (11). The raw
data from which this plot is developed will be discussed in great detail
subsequently. Figure 6 expresses the dimensionless residence time
(qT/Ds) as a function of the modified Reynolds number (qP D /(J. ) with
the dimensionless quantity, A, as a parameter. It is apparent chat very
distinctive patterns of behavior can be discerned from the plot. In
terms of the dimensionless parameters of the plot, the screens appear to
have well defined exponential performance characteristics. Each screen
has a range of Reynolds number in which its performance, defined in
terms of percentage solids removed and throughput capacity, is optimal.
The residence time of the solids on the screen which matches this opti-
mal performance level must be selected from the preceding step in which
the residence time was optimized in terms of machine variables.
While basically incomplete, the system of curves in Figure 6 is believed
to provide a sufficient coverage of the region of practical interest.
Experience with solid separation of both swine and beef cattle waste-
water indicates that mesh screen sizes between #50 and #120 are most
practical. Screen sizes below this range would permit greater capacity
throughput at the expense of solids removal efficiency. On the other
hand, above the range, throughput capacity is drastically reduced with-
out significant improvements in solids removal. Within the recommended
range, we feel that interpolation or extrapolation of the system of
curves in Figure 6 can be carried out with reasonable validity.
33
-------
For each specified wastewater, it will be necessary to develop a chart
similar to that of Figure 6 as a design aid in equipment selection and
ultimately as an operational guide for the estimation of performance
rating of machines. For purposes of design, if a small model of the
machine can be obtained, the procedure presented provides a comprehen-
sive framework within which a test can be most effectively conducted and
the data so obtained can be put to the best use. Where it is not possi-
ble to obtain a small test machine, a first estimate of machine size
necessary to accomplish carefully defined objectives can be made with the
aid of the curves and with the general procedure presented.
The Effect of Sludge Conditioning on Performance Rating:
It is customary in performance evaluation of dewatering centrifuges,
presses and vacuum filters to consider the effect of sludge conditioning
on performance rating. Sludge conditioning refers to the pretreatment
of the sludge.by the addition of chemicals such as coagulants, floccu-
lants, pH control, or predigestion under anaerobic conditions. Limited
tests iii this study indicated that chemical conditioners had little or
no effect on the performance of vibrating screen separators. This obser-
vation is not surprising since vibrating screens do not depend on set-
tling characteristics of the solids for their performance. The bridging
forces which encourage the agglomeration of particles when chemical con-
ditioners are added to sludge appear to be destroyed by vibrational
forces on the screen. Consequently their effect, if there is one, is
negligible.
Sludge age and sludge digestion under anaerobic conditions were found
to change the flow properties of the solids on the screen. In general,
the solids appear to be experiencing greater difficulties in forming
balls. The optimum moisture content appears to change to a value slight-
ly lower than the corresponding balling moisture content of fresh solids.
The throughput capacity under these conditions appears to be measurably
reduced. These observations, though vague, are intended to stimulate
interest in further exploration of this particular feature. Within the
broader context of our overall study, it was not considered advantageous
to predigest the solids prior to separation since this would add an addi-
tional step and cost to the process. With this basic disposition, no
real concerted effort was made to quantify the effect of pre-digesti,on
on screen performance. It is recognized, however, that there may be
circumstances under which sludge conditioning by anaerobic digestion
would be desirable.
Experimental Procedures and Results:
1. Methods of Sieve Analysis; The wet methodj^of sieve analysis adopted
from the 1970 edition of Tyler Handbook #53 was used to estimate the
solids removal potential of vibrating screen separators. The complete
set of results obtained from sieve analysis of swine wastewater is pre-
sented in Table 9. A somewhat less extensive analysis of beef cattle
wastewater is given in Table 10.
34
-------
TABLE 9: SIEVE ANALYSIS AND THE REMOVAL PERCENTAGE ESTIMATES OF SWINE
WASTEWATER SOLIDS
Soaking Time Mesh Size of Screen Percentage Removal (%)
1 Hour 20
48
60
100
200
325
4 Hours 20
48
60
100
200
325
24 Hours 20
48
60
100
200
325
1.2% TS
44.2(%)
51.6
54.9
56.3
-___
43.0
51.8
52.8
57.0
58.4
35.0
40.5
41.1
53.0
55.0
55.8
2.72% TS
55.6(%)
60.7
61.4
63.5
63.9
63.5
53.0
56.3
58.1
59.8
61.2
62.0
51.9
57.0
57.8
58.0
60.0
63.8
5.44% TS
51.2(%)
55.6
57.6
58.9
60.9
61.0
51.2
57.3
60.0
61.8
62.0
42.0
47.1
49.1
50.4
62.5
TABLE 10: SIEVE ANALYSIS OF BEEF CATTLE WASTEWATER
SOAKING TIME 24 HOURS, APPROXIMATELY 5% TS
Mesh Size of Screen
20
40
60
100
% Removal
44
47
50
58
35
-------
Table 9 indicates that both soaking time as well as percent total solids
of the wastewater being fractionated can have measurable effects on
screen performance. These effects can be partially explained by the
recognized relationship between soaking time, total solids concentration
and the soluble component fraction of the solids. It can thus be ratio-
nalized that longer soaking times and low total solids concentrations
are likely to encourage more of the soluble fraction to go into solution.
This in turn would lead to lower percentage recovery of solids on the
screen. These observations indicate a need to establish some type of a
standard specifying time durations of soaking and percentage total solids
when conducting tests and reporting data related to performance ratings
of vibrating screen separators.
No attempt has been made to recommend any particular standard in this
study. The following observations, however, can aid the eventual formu-
lation of such standards. A soaking period of 24 hours would give more
reliable and more reproducible results; it was adopted in our study pri-
marily for this reason even though such a prolonged soaking period is
certainly counter to the need for more rapid laboratory procedures. It
is believed that errors inherent in sampling at low total solids concen-
trations may have contributed significantly to the observed effects of
total solids concentration on removal efficiencies on the screen. In
the low range between 1% and 5% total solids, provided that a sufficient-
ly long soaking time has elapsed, the exclusive effect of percent total
solids on percentage removal of solids should be negligible.
In Figure 7, the percentage removal 100 (Cf - C )/Cf is plotted as a
function of mesh size of screen on log-probability paper. This plot
represents average estimates of screen performance obtained from labora-
tory sieve analysis on Tyler screens for both swine and beef cattle
wastewaters. If a significant size reduction does not take place in
transport as the wastewater is pumped to the screens (assumption con-
firmed by our experience), Figure 7 should provide accurate estimates of
screen removal efficiencies within 5% of the actual value.
From the data on sieve analysis, the fineness moduli of the solids were
determined to be approximately 1.57 for swine wastewater solids and 1.98
for beef cattle wastewater solids. These moduli were then used to calcu-
late the average particle size (Dp) of the wastewater solids using the
following equation given in Chemical Engineering Handbook (9)
D = 4.1 X 10"3 (2) FM (12)
P
where FM = fineness modulus. From these calculations, the following
values of particle size were obtained:
#
D (swine wastewater solids) = 0.012 inches
P
D (beef cattle wastewater solids) = 0.017 inches.
P
36
-------
cfl
O
8P
4J
C
(1)
O
^
0)
p-l
o
o
u
M-l
o
99
9£ -
95
9C
8C
50
4C
3CL
20
10
1
0.5
0.2
0.1
0.05
Iwine Wastewater
"Beef Cattle
Wastewater
20 40 60 80
Mesh Size of Screen
100
120
FIGURE 7: AVERAGE PERCENTAGE REMOVAL VS. MESH SIZE OF SCREEN PLOTTED
ON LOG-PROBABILITY PAPER
37
-------
With the average particle size thus defined, a table of the dimension-
less parameter A [=100 (D /D )(Cf - Cc)/Cf] can be developed for screens
tested in this study. These values are presented in Table 11 for swine
wastewater and beef cattle wastewater.
2. Tests on the Vibrating Screen Separator: As was previously stated,
the 18-inch model of the Sweco Vibrating Screen Separator was used to
obtain all the data presented in this section. Since it is a popular
separator, it is hoped that the data we obtained would be of some direct
value to people who possess this brand of machine. The principle of
operation of most vibrating screen machines is the same, however, irre-
spective of manufacturer. The basic vibrational excitation is provided
by top-bottom weight assemblies mounted out of phase on an electric motor
rotating at a constant angular velocity. The motor is rigidly linked to
the screen assembly. Multiple drive springs (in this case, 6) link the
screen assembly to a stationary base support.
The apparatus used to conduct the vibrating screen tests is shown diagram-
matically in Figure 8. A close-up picture of the general setup is shown
in Figure 9. In developing the test, it was essential to provide and
sustain a continuous operation which would approximate actual farm condi-
tions. The only way to do this in the lab where sample size is usually
limiting is to devise a recycling system in which screened solids and
effluent water from the screen are both returned on a continuous basis
to a completely mixed tank from which the raw wastewater is pumped to the
screen.
A 55-gallon capacity drum, D-^ in Figure 8, was used as the completely
mixed chamber. Vigorous agitation of the contents of the drum was ob-
tained by the action of a high speed impeller type mixer and the action
of a bypass return flow from the main hydraulic line. Both of these
operations established a satisfactory mixing pattern in the drum which
approximated complete mixing. Any overflow from drum D-^ was collected by
a receiving drum I>2'
A high volume constant rate sludge centrifugal pump was used. Flow regu-
lation was achieved by the use of two gate valves V, and V2 in the hy-
draulic line. Vo is a bypass valve which can allow any desired fraction
of the total flow to return to the mixing tank. By adjusting the two
valves, it was possible to obtain a wide range of flow rates during the
test. Gate valves, of course, have the disadvantage of imposing a block-
ing effect on the wastewater solids. For the percentage total solids
level at which all the tests were run, however, the gross effect of the
valves was negligible. Immediately preceding each test, maximum mixing
of the chamber contents was generated by using 100% bypass in addition
to the mixing impeller action. ^
The continuous return of the solid and liquid fractions from the screen
to the mixing chamber permitted each test to be run for any desired time
duration. The return of these solid and liquid fractions was interrupted
only for short periods of time during which cake flow and centrate flow
measurements were made.
38
-------
C - C D
TABLE 11: VALUES OF A =( - X 100) 7^ AS USED IN THIS STUDY
C,. D
f S
(a) Swine Wastewater
Mesh size of screen
Ds (in.)
D (in.)
P
c _ c (Figure 7)
r c
R . , vi f\(\
D
R (/) - A
s
50
.0116
.012
55
62
74
.0074
.012
60
105
100
.0058
.012
64
144
120
.0049
.012
56
175
200
.0029
.012
70*
290
230
.0026
.012
70*
307
(b) Beef Cattle Wastewater
Mesh size of screen
D (in.)
S
D (in.)
C_ - C (Figure 7)
R ^ c. Y i nn
Cf
R <£*> - A
s
50
.0116
.017
45
66
74
.0074
.017
52
120
100
.0058
.017
58
170
120
.0049
.017
62
215
200
.0029
.017
70*
410
230
.0026
.017
70*
457
*70% removal is assumed to be maximum possible.
39
-------
Control Valve
(V
By-pass Valve
(V,)
Velocity
Dissipator
Discharge Spout
for Solids
By-pass flow
Variable Speed
Impeller
Discharge
Spout for Liquids
Screen Separator
Over Flow
Receiver
(D2)
Stand
Centrifugal Pump
FIGURE 8: SCHEMATIC OF EXPERIMENTAL APPARATUS FOR VIBRATING SCREEN
TESTS
40
-------
FIGURE 9: CLOSE-UP VIEW OF SETUP FOR SCREEN TEST
41
-------
In a typical test, the desired machine variables (screen size, top-
bottom weight ratio, lead angle between top and bottom weight assemblies)
were first established. Maximum mixing of the wastewater in the tank was
then established. The flow of wastewater to the screen was regulated
until an approximate steady state condition was observed on the screen.
This equilibrium condition is defined by very characteristic patterns of
motion on the screen. This motion maintains a steady flow of "roped",
dewatered, and "balled" solids via the periphery of the screen to the
discharge spout. For a given set of machine variables, there is only
one flow rate (Q) of feed at which equilibrium can be observed. If the
rate of feed is below this value, the solids tend to crowd at the center.
Their movement becomes so sluggish that inordinately high residence times
are obtained for the average particle on the screen. If the rate of feed
is higher than the equilibrium value on the other hand, a "wash out" of
the screen solids occurs instantly. At equilibrium, a distinctive pat-
tern emerges. Measurements of the cake rate of flow, aClt^/min), and the
centrate rate of flow, Q(gal./min), are taken only under equilibrium
conditions as specified above, a and Q are measured by diverting their
respective flows to appropriate collectors and clocking the accumulation
of material over time intervals of about one minute.
Both the drive motor of the screen and the pump are actuated by a common
electric switch. The instantaneous average mass of cake retained on the
screen under steady state condition has been designated |3 (lbm). This
quantity is determined by simultaneously shutting-off the pump and screen
during steady state operation, blocking the exit spout at the same time
and measuring the residual mass of cake on the screen when it ceases to
vibrate. The mean residence time, T(sec), is then defined as:
_ nb 1 instantaneous average mass of cake re-
P m tamed on screen
Tsec) = = ~~^~~~~~~i~~~~~~~~~~~~~~~^*
O, (Ib /sec) average mass velocity of cake on the
m screen
In Table 12 a complete set of experimental data obtained from vibratory
screen tests on swine wastewater has been tabulated. A similar set of
data on beef cattle wastewater is given in Table 13. In each of these
tables, the raw data are given as the quantities:
Q (gal./min) = total flow rate of feed
a (Ib /min) = mass velocity of cake on screen
m
(3 (Ib ) - instantaneous mass of cake retained on screen.
m
From these basic quantities the following parameters are synthesized:
q = *-£ ( ) = feed rate of flow per unit area of screen open
se° space. It represents the average velocity through
screen opening. It is used to define the modified
Reynolds number.
42
-------
TABLE 12a: RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE WASTEWATER.
Mesh Size of Screen - #50. % Total Solids of Uastewater = 1.4Z-27..
1.02 ft ,
i «= Lead Angle
= 0.0672 lbm/ft-sec, D = .0012 ft, §= .526
0° 15° 30° 45°
60'
75'
90"
wb
wt
C3
2.56
Q (gal/min)
o.=cake flow (Ib/min)
6=cake retained (lbm)
q=Q/A§ (ft/sec) (lO'1)
T=B/aX 60 (sec)
qT/Ds X (103)
\e=<&s Pc/uc X CIO'3)
31.15
2.85
3.8
1.29
80.0
8.624
2.31
31.15
4.2
1.8
1.29
25.7
2.77
2.31
28.75
4.6
1.8
1.19
23.5
2.338
2.13
Wb
Wt
a
2.92
Q
a
P
q X 10"1
T
qT/D8 X 103
NRe X 10-3
25.16
4.8
2.1
1.04
26.2
2.281
1.87
35.94
6.45
1.5
1.49
13.9
1.729
2.67
Wb
Wt
3.46
Q
a
e
q X 10"1
T
qT/D8 x 103
NRe X 10-3
31.15
4.7
2.4
1.29
30.6
3.229
2.31
35.94
5.9
1.7
1.49
17.3
2.152
2.67
35.94
6.9
2.3
1.49
20.0
2.487
2.67
"b
Wt
3.83
Q
a
B
q X 10"1
T
qT/Ds X 103
NRe X 10-3
23.02
5.1
1.9
.956
22.3
1.776
1.71
35.94
5.4
2.1
1.49
23.3
2.898
2.67
43
-------
TABLE 12b: RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE WASTEWATER.
Mesh Size of Screen = #74. % Total Solids of Wastewater = 1.47.-27..
A = 1.02 ft ,uc = 0.0672
4 = Lead Angle
, Dg = 8.17 X 10"4ft, §= .52
0°
15° 30°
45°
60°
75°
90c
wb
Mt
si
2.5
Q teal/min)
a=cake flow (Ib/min)
B=cake retained (lbm)
q=Q/A5 (ft/sec) (lO'S
T=p/aX 60 (sec)
qT/Ds X (103)
"Re^s Pc/uc X (10'3)
7.19
2.0
7.0
.302
210.0
7.763
.367
12.2
2.8
7.4
.512
158.6
9.948
.623
18.7
4.4
7.1
.786
96.8
9.307
.955
22.3
7.7
3.6
.937
28.0
3.21
1.139
19.4
7.6
2.7
.815
21.3
2.125
.991
24.4
5.2
2.7
1.02
31.2
3.914
1.246
24.4
7.2
6.0
1.02
50.0
6.273
1.246
Wb
Wt
2.77
Q
a
P
q X 10"1
T
qT/Dg X 103
NRe X 10-3
16.5
6.0
5.0
.69
50.0
4.242
.843
16.5
5.6
4.6
.693
49.3
4.182
.843
19.4
6.5
2.8
.815
25.8
2.573
.991
20.1
6.9
2.6
.844
22.6
2;336
1.026
21.6
7.5
3.1
.907
24.8
2.754
1.103
21.6
7.0
3.5
.907
30.0
3.332
1.103
27.3
8.5
4.9
1.15
34.6
4.857
1.394
Wb
Wt
3.16
Q
a
8
q X 10"1
T
qT/Ds X 103
NRe X io'3
15.8
5.5
5.6
.66
61.1
4.963
.807
18.7
5.1
6.0
.786
70.6
6.788
.955
18.7
6.2
2.4
.786
23.2
2.231
.955
24.4
7.6
3.0
1.02
23.7
2.973
1.25
31.6
9.4
3.5
1.327
22.3
3.623
1.61
27.3
7.9
4.8
1.147
36.4
5.109
1.39
15.8
8.1
5.6
.66
41.5
3.371
.807
Wb
Wt
3.77
Q
a
P
q X 10"1
T
qT/Ds X 103
NRe X ID"3
15.81
5.5
4.3
.664
46.9
3.812
.807
21.6
7.6
2.3
.907
18.2
2.021
1.103
25.9
9.5
2.8
1.09
17.7
2.357
1.323
30.2
11.0
3.7
1.27
20.2
3.136
LJ542
30.2
11.2
3.3
1.27
17.7
2.748
1.542
30.2
11.1
5.3
1.27
28.6
4.441
1.542
23.0
5.0
5.5
.966
66.0
7.805
1.175
44
-------
TABLE 12c: RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE WASTEWATER.
Mesh Size of Screen = #120. % Total Solids of Wastewater = 1.47.-27..
9 /
A = 1.02 ft , |ic = 0.0672 Ib /ft-sec, D = 4.83 X 10 ft, § = 0.473
Lead Angle
15°
30°
45°
60'
75°
90"
wb
wt
ss
2.56
Q (gal/min)
a=cake flow (Ib/min)
B=cake retained ( m)
q=Q/A| (ft/sec) (lO"1)
T=B/aX 60 (sec)
QT/DS X (103)
NRe=1Ds Pc/uc X (10-3)
20.8
2.4
3.5
.961
87.5
17.401
.69
34.7
5.7
4.0
1.60
42.1
13.967
1.15
34.7
5.5
2.7
1.60
29.4
9.754
1.15
33.1
5.6
4.0
1.53
42.9
13.577
1.09
31.1
3.6
4.0
1.44
66.7
19.833
1.03
31.1
3.6
4.4
1.44
73.3
21.796
1.03
Wb
Wt
2.92
Q
a
P
q X 10"1
T
qT/Ds X 103
NRe X ID'3
22.8
4.5
3.7
1.05
49.3
10.747
.757
34.7
5.4
3.5
1.602
38.9
12.906
1.15
35.9
8.4
2.9
1.658
20.7
7.105
1.19
35.9
7.2
3.5
1.658
29.2
10.023
1.19
26.4
4.0
2.6
1.219
39.0
9.844
.877
26.4
3.6
3.5
1.219
58.3
14.716
.877
Wb
Wt
3.46
Q
a
6
q X 10"1
T
QT/DS x 103
NRe X 10-3
21.56
1.0
2.0
.996
120.0
24.736
.71!
33.54
2.4
2.4
1.55
60.0
19.241
1.113
31.15
7.6
2.8
1.439
22.1
6.582
1.03
29.95
5.2
3.3
1.383
38.1
10.910
.994
31.63
4.8
3.8
1.461
47.5
14.365
1.05
32.35
3.6
5.6
1.49
93.3
28.858
1.07
Wb
Wt
* S3
3.83
Q
a
P
q X 10"1
T
qT/Ds X 103
NRe X 10-3
20.4
3.7
2.6
.942
42.2
8.231
.677
26.4
3.1
2.3
1.219
44.5
11.232
.876
24.0
3.4
2.2
1.108
38.8
8.903
.797
30.0
3.3
2.7
1385
49.1
14.083
.996
31.1
5.8
4.1
1.436
42.4
12.608
1.032
28.8
4.0
5.2
1.330
78.0
21.478
.956
45
-------
TABLE 12d: RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE WASTEWATER.
Mesh Size of Screen = #200. 7. Total Solids of Wastewater = 1.47.-27..
A = 1.02 ft ,
I = Lead Angle
= 0.0672 Ib /ft-sec, D
m - ;
0"
.000242 ft , §= .34
*»
15° 30° 45° 60°
75°
90°
wb
wt
B
2.5
Q (gal/min)
a=cake flow (Ib/min)
8=cake retained (lbra)
q=Q/A5 (ft/secHlO"1^
T=B/aX 60 (sec)
QT/DS X (103)
NRe=q°s pc/uc X CUT3)
5.75
1.9
5.1
.365
161
24.576
.133
7.188
2.5
2.5
.462
60
11.449
.166
6.47
2.6
3.2
.415
73.8
12.676
.149
Wb
W
ss
3.16
Q
a
P
q X 10"1
T
qT/Ds X 103
NRe X 10'3
10.78
5.9
2.2
.692
22.4
6.410
.249
11.5
5.9
3.0
.739
30.5
9.311
.266
10.78
4.9
3.6
.692
44.08
12.615
.249
8.57
2.9
5.2
.551
107.6
24.480
.198
Wb
Wt
e
3.83
Q
a
8
q X 10"1
T
qT/Ds X 103
NRe X 10"3
3.59
2.1
2.4
.231
68.6
6.538
J0803
11.5
5.3
1.7
.739
19.25
5.877
.266
5.75
2.6
2.6
.369
60.0
9.158
.133
6.47
3.2
3.6
.416
67.5
11.594
.149
46
-------
TABLE 12e: RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON SWINE WASTEWATER.
Mesh Size of Screen = #230. % Total Solids of Wastewater = 1.47.-27..
A = 1.02 ft2, (ic = 0.0672 lbm/ft-sec, Dg = 2.42 X 10"4 ft,
i = Lead Angle
15° 30"
45'
.46
60°
75°
90°
wb
Mt
fS
2.5
Q (gal/min)
a=cake flow (Ib/min)
B=cake retained (lbm)
q=Q/A| (ft/sec) (10'1)
T=3/aX 60 (sec)
JJ/DS x (io3)
NRe=lDs Pc/uc X (ID'3)
7.2
2.0
2.9
.342
87.0
12.291
1.23
5.8
1.0
1.0
.275
144.0
16.388
.991
Wb
Wt
«
2.92
Q
a
B
q X IO"1
T
qT/Ds X IO3
NRe X ID'3
3.6
1.0
.8
.171
48.0
3.391
.6145
10.1
1.5
2.0
.48
80.0
15.85
1.726
11.5
1.8
2.2
.546
73.3
16.54
1.97
Wb
Wt
3.46
Q
a
e
q X 10"1
T
qT/Ds X IO3
NRe X 10-3
7.9
2.4
1.9
.375
47.2
7.363
1.35
8.6
2.6
1.2
.408
27.7
4.674
1.47
11.5
2.4
2.7
.546
67.5
15.231
1.97
Wb
V
3.83
Q
a
8
q X 10"1
T
qT/Ds X IO3
NRe X IO-3
2.9
.8
.9
.138
67.5
3.841
.495
7.2
.9
2.2
.342
L46.7
20.725
1.23
8.6
1.0
2.2
.408
132.0
22.275
1.47
47
-------
TABLE 13a: RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON BEEF CATTLE WASTEWATER
Mesh Size of Screen = #74. 7. Total Solids of Test Wastewater * 27..
1.02 ft , D = 8.17 X 10"
«S = Lead Angle
ft, 5- .52,
0° 15°
6.72 X 10"2 iWft-sec
30'
45"
60°
75'
90°
wb
wt
s
2.56
Q (gal/mtn)
a=cake flow (Ib/min)
6F=cake retained (lbm)
q=Q/A5 (ft/sec) CIO'1)
T=p/aX 60 (sec)
qT/Ds X (103)
NRe=1Ds Pc/uc X (ID'3)
17.25
8.1
5.0
.725
37.0
3.282
.881
18.69
11.1
5.2
.785
28.2
2.710
.954
10.06
5.9
5.2
.423
52.9
2.736
.514
11.50
3.6
7.2
.483
J.20.0
7.095
.587
Wb
W
2.92
Q
a
P
q X 10"1
T
qJ/Ds X 103
NRe X ID'3
11.5
5.8
6.6
.483
68.3
4.038
.587
15.09
8.6
4.0
.634
27.9
2.165
.771
15.81
9.7
4.6
.664
28.5
2.317
.807
15.81
10.1
4.7
.664
27.9
2.268
.807
11.50
5.7
5.5
.483
57.9
2.423
.587
Wb
Wt
ss
3.46
Q
a
0
q X 10"1
T
qT/Ds X 103
NRe X 10-3
8.63
5.6
2.5
.363
26.8
1.189
.441
7.19
6.4
5.3
.302
56.8
2.100
.367
21.56
17.4
6.1
.906
21.0
2.328
1.101
17.25
10.0
6.1
.724
36.6
3.246
.881
10.78
6.2
5.9
.453
57.1
3.165
.550
Wb
Wt
3.83
Q
a
P
q X 10"1
T
qT/Ds X 103
NRe X lo'3
10.06
6.4
3.9
.423
36.6
1.893
.514
19.17
10.1
6.4
.805
38.0
3.745
.979
16.53
8.0
5.9
.694
44.3
3.765
.844
11.50
7.2
6.8
.483
56.7
3.352
.587
48
-------
TABLE 13b: RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON BEEF CATTLE WASTEWATER.
Mesh Size of Screen = #74. % Total Solids of Test Wastewater = 1.3%.
A = 1.02 ft, D = 8.17 X 10
Lead Angle
"
ft, § = .52, k
0° 15e
6.72 X 10~2 Ib /ft-sec
30° 45"
60'
75e
90°
wb
wt
=
2.56
Q (gal/min)
tt=cake flow (Ib/min)
B=cake retained ( m)
a=Q/A 5 (f t/sec) (lO"1)
T=8/a X 60 (sec)
T/DS X (103)
S^sPc/Mc X do'3)
17.97
6.6
4.7
.755
42.7
3.945
.918
17.97
6.7
3.9
.755
34.9
3.225
.918
15.81
5.4
3.6
.664
40.0
3.251
.807
11.5
6.5
11.9
.483
45.2
2.673
.587
TABLE 13c: Mesh Size of Screen = #120. % Total Solids of Test Wastewater = 2%.
A = 1.
.02 ft2, D =.4.83 X 10"4ft, § = .473, (jc = 6.72 X 10"2 Ib /ft-sec.
Lead Angle
15°
30°
45'
60°
75°
90°
?b
w
S3
2.56
Q
a
6
q X 10"1
T
qr/Ds X 103
\e X lO'3
4.31
3.9
2.4
.199
36.9
1.521
.143
49
-------
TABLE 13d: RESULTS OF VIBRATING SCREEN SEPARATOR TEST ON BEEF CATTLE WASTEWATER.
Mesh Size of Screen = #50. 7. Total Solids of Test Wastewater 2%.
A = 1.02 ft , Dg = .00121 ft, I = .526, Uc = .0672 iWft-sec
= Lead Angle
0°
15° 30° 45°
60° 75"
90°
wb
wt
s
3.83
Q
a
P
q X 10"1
T
qT/Ds X 103
^e X 10'3
6.47
4.9
3.5
.268
42.9
.9526
.484
Wb
Wt
3,46
Q
a
P
q X 10"1
T
qT/Ds X 103
NRe X 10'3
5.75
4.3
2.7
.239
37.67
.743
.430
50
-------
30
70
60
50
t(sec)
40
30
20
10
4.0
3.0
2.0
T"40 sec.
0 15° 30° 45° 60° 75° 90
0 15° 30C 45" 60" 75° 90"
T(SSC)
140
120
100
80
60
40
20
0
2.5
i- 3.84
I
23 4 56
q(ft/sec)X 10^
FIGURE 10: ALTERNATIVE PLOTS OF THE SCREEN TEST DATA ON SWINE WASTE-
WATER
51
-------
T - X 60 (sec) = Mean residence time.
qT
r Dimensionless residence time.
s
qpc Ds
NRe = Modified Reynold's number.
_o
The centrate absolute viscosity |J. is taken as 6.72 X 10 (Ib /ft-sec)
in all calculations. This is the best estimate obtained from the recent
data of Kumar, Bartlett and Mohsenin (10). Blanks in the tables imply
that it was not possible to achieve equilibrium performance for the
specified variables. We shall now discuss the results of tests for
swine wastewater and beef cattle wastewater individually.
Test Results for Swine Wastewater:
The most complete data were obtained for mesh sizes #74 and #120. Equi-
librium machine performance was achieved at practically all combinations
of machine variables tested. In Figure 10, the data in Table 12b corre-
sponding to the test on #74 screen, were plotted in a variety of forms.
Each plot shows very distinctive minima irrespective of the variables of
the plot. It is clear that a large number of plots of this type is
possible. The dimensionless form of the same plot can be seen in
Figure 5 which is a graphical representation of the data in Tables 12b
and c corresponding respectively to tests on screens #74 and #120.
The plot in Figure 5 is obviously more concise, inclusive of the primary
factors and thus to a large degree, very comprehensive. Decisions con-
cerning optimum performance conditions made on the basis of this plot
would have taken a good number of the pertinent variables into considera-
tion. Consequently such decisions are to be preferred to those deduced
from a less comprehensive representation on the experimental data.
It is observed that the remaining tables of the swine wastewater screen
test, namely Tables 12a, d and e, do not provide a sufficient number of
data points to facilitate optimization plots similar to Figure 5 for
screen sizes #50, #200,and #230. However, it is still possible to pick
optimum residence time values by inspection of the tables. In Figure 6,
all the data in Table 12 have been combined with those in Table 11 to
obtain dimensionless performance curves for vibrating screen separation
of swine wastewater solids. We shall now illustrate the practical use of
these curves.
Illustrative Example;
A livestock producer operating a 1000-hog confinement unit proposes to
install a vibrating screen separator as a first step in a wastewater
management scheme involving biological treatment in an aerated lagoon
and recycling of the treated wastewater for flushing purposes in the con-
fined building. What size of vibrating screen separator does he need?
52
-------
At what machine variables must he run the unit to obtain optimum perfor-
mance?
Solution;
Assume that each hog produces an average of 1.5 gallons of manure
(urine + feces) daily. For transport and flushing purposes, we assume
that each unit volume of manure is diluted ten times in the flush water.
Each hog has consequently a gross total of 15 gallons of wastewater which
must be screened daily. Since there are 1000 hogs, we must therefore
handle 15,000 gallons of water daily through the machine. If we assume
that this machine operates on the average 10 hours/day, we are then look-
ing for a 25 gpm capacity machine.
We desire a solid removal efficiency of at least 50% on the screen. From
Figure 7 and Table Ha, screen mesh size #74 has an average removal effi-
ciency of 60% (>50% OK). Therefore we select a #74 mesh size of screen.
For #74 mesh size of screen, Figure 5 indicates optimum dii
residence time (qr/D ) optimum = 1.5 X 10 .
_ . dimensionless
. ) ODtimiitn = 1.5 X 10 .
s
This optimum value occurs when the machine variables are:
6 (Lead angle between top and bottom weights) = 30° to 45
W,/W (Bottom- top weight ratio) =3.16
From Table Ha, screen #74 has A = 100 (D /D ) (C - C )/Cf = 105
p s £ c r
From Figure 6:
(qT/D ) optimum = 1500 <=> N_. = qp D /[i =0.06
S K.6 C S C
A #74 = 105
p = 62.4 Ib /ft3
rc m
D ^8.17 X 10"4ft
s
M. ~ 6.72 X 10"2lb /ft-sec
'c m
_2
_ 6 X 10 Mc _ (6 X 10"2)(6.72 X 10"2)
~ p D (62. 4) (8.17 X lO-^)
c s
q = 8 X 10"2 (ft/sec)
for a screen of % opening 52.6%, § = 0.526
q = S-; ^ = q§ = (8 X 10"2)(.526) = 4.5 X 10"2 (f t3/ft2-sec)
£\O A
53
-------
£ _ flow rate
A unit surface area of screen
-2 ft3
= 4.5 X 10 (- 2U ) X 60 (sec/min) = 2.7
ft -sec ft -min
2.7 (-i-) X 7.5 (gal/ft3) =
2 /«./../ ^goi/ j. <- f 2
ft -min (min)(ft )
The 18-inch diameter model of Sweco Vibrating Screen Separator has an
effective surface area,
A ^ 1.02 ft2
Q = 20 gallons . . . 2 ~ 20 gallons
(min) (ft2) ' min
This is less than the 25 gallons/min which we need. However, if we let
the machine run for 13 hours instead of 10 hours a day we have,
_ 15,000 gallons ~ 19 gallons
needed 13 X 60 (min) min
which then makes it possible for this machine to be used.
Alternatively we can specify a 24 inch diameter screen of effective area
A = 2.2 ft", Q = 20 X 2.2 ft2 = 44 Sallon
tc..t- . N mm
(ft -mm)
which is far in excess of the 25 gallons/min required.
Design Summary:
(i) Machine Specifications: Either (a) Sweco Vibrating Screen
Separator Model 24-inch operating for 10 hours per day (estimated cost =
$1,500, 1 hp) or (b) Sweco Vibrating Screen Separator Model 18-inch
operating 13 hours per day (estimated cost - $1,000, 1 hp).
(ii) Machine Variables;
i = 30° to 45°
W./W =3.16
b t
(iii) Process Variables (expected);
R = Cf " Cc =607.
cf
Q = 20 or 44 gallons/min
54
-------
D = 0.012 inches
P
A somewhat different plot of the data is represented in Figure 11. This
plot is basically the same as that of Figure 6 except that in Figure 11,
the performance parameter A is plotted against Reynolds number (Nj^e) with
the dimensionless residence time (qT/D ) as a parameter. It is felt that
there may be circumstances under which the form in Figure 11 would be
found more convenient. The matter of convenience aside, the plot in the
form of Figure 11 shows an interesting feature which was not as well de-
fined in Figure 6. This is the apparent existence of a common point of
convergence. The practical implication of this is that the Reynolds
Number Njje = 0.04 represents a unique solution where the removal effi-
ciency is the same irrespective of screen size and residence time. The
full implication of this from the standpoint of machine operation is
unclear and deserves more attention and study than we have been able to
give it.
Test Results for Beef Cattle Wastewater:
As is apparent from Table 13 we were able to obtain the most complete set
of results from screen size #74. Repeated attempts convinced us that
the fibrous characteristics of the beef cattle solids (cake) appeared to
agree best with this mesh size of screen. Mesh sizes below (represented
by #50 screen, Table 13d) and mesh sizes above (represented by #120
screen, Table 13c) performed comparatively poorly and at best, showed in
each case only one set of variables at which equilibrium was attainable.
This observation permits us to recommend #74 screen categorically as the
most suitable as far as beef cattle wastewater is concerned.
An optimization plot identical to Figure 5 is developed from Table 13a
and represented as Figure 12. In this case the optimum dimensionless
residence time (qT/D ) optimum is 1200 and corresponds to machine
variables:
6 =30° and WV/W = 3.46.
b t
A screen performance curve similar to Figure 6 is developed for beef
cattle wastewater and represented as Figure 13.
Exploratory Investigation of Granular Filters:
A number of different kinds of granular filters have been used for the
purification of water and the treatment of domestic and industrial waste-
waters (11). There does not exist to our knowledge a reported instance
in which this method has been explored for the removal of solids from
livestock wastewater. In the agonizing process of searching for low
cost methods for solids removal from animal wastewater, it was felt that,
the associated plug-up problems aside, filtration of screened livestock
wastewater through shallow beds of silica sand or finely divided anthra-
cite can provide a possible answer to the basic problem of solids re-
moval at low cost. This section of the report will deal briefly with
55
-------
Ul
10'
A
(see
text)
10
10
10
10
-2
qT Dimensionless
D ' Residence Time
10'
X 10'
I l_
-1
10
s (Modified Reynolds Number)
[1C
FIGURE 11: PERFORMANCE CURVES FOR SWINE WASTEWATER
10
-------
o
I-l
X
I
O
g
U
H
CO
Lead Angle
60'
75'
90
FIGURE 12: OPTIMIZATION PLOTS FOR BEEF CATTLE WASTEWATER
57
-------
00
10
4>
U
a
8
to
4)
i-H o
§103
01
l-
er
10
10
-5
10 10
(Modified Reynolds Number) qPcDs
FIGURE 13: SCREEN PERFORMANCE CURVE FOR BEEF CATTLE WASTEWATER
-------
the very preliminary stages of a sand filter study which was undertaken.
Sand Filter Experiments and Results;
In all, three sand filters were constructed and tested. The first two
were made out of plastic tubes, each 5 ft long with 2-inch and 4-inch
inside diameters respectively. The third sand filter was made out of
treated plywood stock. The wooden column was made in 1-ft sections with
a square cross-section of inside dimensions approximating 1 ft. The 1-ft
sections can be bolted together to provide varying depths of sand column.
The flow through a cross-sectional area of one square foot of the wooden
column facilitated loading rate studies.
Figure 14 shows a typical section through the sand filters used. Each
filter was tapped at 1-ft intervals. A close-up picture of the sand
filter setup with sampling accessories is shown in Figure 15. The col-
umns were packed with washed silica sand roughly sieved between screen
mesh sizes #28 and #48.
The first concern of our investigation was to determine potential re-
moval efficiencies obtainable from sand beds as a function of depth.
Swine wastewater diluted 1:10 and 1:20 with tap water and previously
sieved through a #60 mesh size of screen was filtered through 4 ft sand
filters. The feed wastewater and samples collected at depths of 6", I1,
2', 3' and 4' were analyzed for total solids (TS), COD and total nitrogen.
A summary of the results is given in Table 14.
Table 14 shows that removal efficiencies between 90 and 95 percent were
obtained for total solids after the wastewater was filtered through 4 ft
of sand. At the same depth, the COD was depleted 22 to 39%. The corre-
sponding removal efficiencies at various other depths in the bed can be
seen in the table. These removal efficiencies are, to say the least,
impressive. Between the vibrating screen and sand filters, it is quite
apparent that removal efficiencies approaching 99% were well within
reach for the solids. This performance far exceeds what can be expected
from commercial centrifuges, vacuum filters or sludge dewatering presses.
Having been convinced about the tremendous potential of sand filters, the
next logical preoccupation was with the general management of such fil-
ters. Flow rate studies extending over 24-hour periods were conducted
on the one-square-foot filter. A free head of 3 inches was maintained
during the test. Flow through the filter was clocked at approximately
2-hour intervals. These measurements showed that within the first two
hours flow rates approximating 2.5 gallons/min-ft can be achieved. How-
ever, as time progressed the flow rate dropped because of plugging prob-
lems at the -filter surface. At the end of 24 hours the average flow
rate was 0.5 gallons/min-ft . If an average loading rate of 0.5
gallons/min-ft2 were assumed as feasible over 24-hours periods of contin-
uous operation, a livestock operation processing 15,000 gallons of waste-
water per day would require a filter of cross-sectional area
59
-------
Containing Wall
- -: .'.'*
3" gravel laye
2 layers of cheese
perforated plastic plate
FIGURE 14: DIAGRAMATIC SECTION THROUGH EXPERIMENTAL SAND FILTER
Constant head
overflow
tap
1'
I1
60
-------
FIGURE 15: CLOSE-UP VIEW OF SETUP FOR SAND FILTER TEST
61
-------
TABLE 14: SAND FILTER STUDY SWINE WASTEWATER
<^
1:10 Dilution
Origina1
6" Filtered
1' Filtered
2' Filtered
3' Filtered
41 Filtered
1:20 Dilution
Original
6" Filtered
1' Filtered
2' Filtered
3' Filtered
4' Filtered
Total
Solids
.137%
.0538%
.0404%
.0252%
.0160%
.0142%
.0685%
.0082%
.0164%
.0092%
.00613%
.00341%
% Change
T.S. (-)
60.7%
70.5%
81.6%
88.3%
89.6%
76.05%
86.56%
91.02%
95.0%
C.O.D.
(PPM)
4,260
3,200
3,100
2,920
2,720
2,590
1,960
1,680
1,560
1,560
1,560
1,520
% Change
C.O.D. (-)
24.8%
27.2%
31.45%
36.14%
39.2%
14.28%
20.4%
20.4%
20.4%
22.4%
Nitrogen
(PPM)
717
658
658
632
624
588
340
322
361
330
330
248
% Change
Nitrogen (-)
8.23%
8.23%
11.85%
12.97%
17.99%
5.29%
2.94%
2.94%
27.05%
-------
A = 15.000 (gal./day) = f
24 (hrs/day) X 60 (min/hr) X 0.5 (gal./min-f t^) square rt
A sand filter 4 ft deep and of dimensions 5' X 5' (if square) or 6 ft
diameter (if circular) would do the job. Since it will be necessary to
scrape the top inch of the filter surface following each 24-hour period
of continuous operation, it will be desirable to have two or more filters
of this type so that daily cleaning would not be necessary.
Recommendations Regarding Sand Filtration of Livestock Wastewater:
From the limited experience we have had with sand filters at this stage,
the following recommendations are in order:
(i) Coarse sand filters have a potential as a high efficiency remover
of solids from livestock wastewater. It is essential that the waste-
water be screened prior to application on the filter.
2
(ii) Flow rates of about 0.5 gpm/ft of filter surface can be achieved
over 24-hour periods of operation. It is essential to point out that
the rate-of-flow measurements in this study were done on small filters
in which wall effects constitute a significant impediment to flow. Higher
flow rates are to be expected from larger filters.
(iii) Plug-up for fresh manure appears to be primarily restricted at the
very top inch of the surface. If the solids are digested prior to fil-
tration through the filter, they tend to penetrate much further into the
bed.
(iv) Further intentive research of sand filters is needed before they
can be considered a feasible proposition. Hydraulic studies and manage-
ment studies directed at plug-up problems should be given first priority
in such studies.
63
-------
SECTION VI
CHARACTERISTICS OF THE COMPONENT LIQUID AND SOLID FRACTIONS
OF BEEF CATTLE AND SWINE WASTEWATERS
Following the separation process on the vibrating screen, the solid and
liquid fractions were analyzed for their physical, chemical and biologi-
cal properties. The data obtained from these analyses are detailed in
Tables 16a, b and c for swine wastewater and in Tables 17a, b and c for
beef cattle wastewater.
The Solid Fraction:
The solid fraction as used here, implies that the waste slurry has been
diluted 1:10 or more times for washing purposes and then screened through
an appropriate mesh size of screen. The solids held back on the screen
constitute the solids fraction.
In general these solids were found to be extremely fibrous. They had a
moisture content ranging between 75-85% (wb). They were odorless and
when exposed over extended periods of time, did not develop any fly
larvae. On the basis of mere visual observations the solids seemed
stable and easily storable. A summary of the essential properties of the
solids are presented in Table 15. A number of the basic characteristics
of these solids appear to be significant at first sight. The most eye-
catching of these is probably the high volatile solids percentage of the
solids in a muffle furnace confirmed the figure of about 91.5% TS shown
in Table 15. The corresponding figure for beef cattle solids is 87.8%TS.
Since these solids are mostly residues of the original feed which passed
through the digestive tracts of the animal without being affected, it
appears strange that they should be so high in volatile solids. Pre-
vious observations in biological treatment systems have shown that these
types of solids are only very slowly biodegradable. The fundamental
question is then why such supposedly biologically inert solids should be
so high in volatile matter. The answer to this question is of course
closely tied to the basic metabolic mechanism by which celluloses and
lignins are hydrolyzed to soluble substrate in biological systems. At
best, such processes, if favored at all, proceed at an extremely slow
pace.
The C:N:P ratio as inferred from COD, nitrogen and phosphorus values in
Table 15 appears to be favorable for some type of microbial action tend-
ing to promote putrefaction. Since the stability of the solids under
conditions of extended storage is of primary interest in the overall
study, an attempt was made to compost the fibrous solids in the meso-
phyllic-thermophilic temperature range so that the intensity of biologi-
cal activity could be observed as an index of stability for the solids
under storage.
65
-------
TABLE 15: SUMMARY SHEET CHARACTERISTICS OF THE SOLID FRACTION (SWINE AND BEEF CATTLE WASTEWATER)
SWINE
Total Solids (TS)
Volatile Solids
COD 249,450
Nitrogen (Total) 7,896
Ca 6,700
Mg 5,500
Zn 300
Cu 5,270
Fe 140
Mn 49
Na 1,150
K % 6,770
P 39,500
17-20%
91.5% TS
1/1 grams
ppm 1.41 a -
** gram TS
.045
.0067
.0055
.0003
.0053
.00014
.000049
.00115
.00677
.0395
BEEF CATTLE
10-15%
87.8% TS
162,350 ppm 1.64 g^_g
3,868 .039
3,000 .003
2,500 .0025
82 .000082
5 .000005
240 .00024
12 .000012
200 .000200
10,000 .01000
16,000 .016000
-------
TABLE 16(a): SWINE WASTEWATER SCREENED
Total Solids
7. Change (-)
Volatile Solids
7. Change
Nitrogen (ppm)
% Change
COD (ppm)
7o Change (-)
BOD (ppm)
7. Change (-)
Expressed as grams/gram
Nitrogen
COD
BOD
Composite
before
Screening
8.3%
82%
10,024
80,000
36,000
of total solids
.1207
.9638
.4337
Effluent
from
60 Mesh
5.367.
337.
73.07.
n o
4,116
58.97.
59,400
19,800
.07679
1.1082
.3694
Effluent
from
100 Mesh
5.087.
397.
73.57.
effective
4,220
57.97.
57,028
18,600
.08307
1.1225
.3661
Effluent
from
200 Mesh
4.997.
407.
73.77.
change
4,140
58.697.
56,210
18,300
.08296
1.1264
.3667
Effluent
from
325 Mesh
4.797.
427.
73.37.
3,940
60.67.
52,195
18,000
.08225
1.08966
.3757
-------
TABLE 16 (b): NUTRIENT AND MINERAL ANALYSIS OF SWINE WASTEWATER SOLID & LIQUID FRACTIONS
00
Ca
Mg
Za
Cu
Fe
Mn
Na
K
P
(ppm)
(ppm)
(ppm)
(ppm)
(ppm)
(ppm)
(ppm)
(ppm)
(ppm)
Expressed
Ca
Mg
Zn
Cu
Fe
Mn
Na
%K
P
Swine Composite
(80% Urine + 207, Feces)
2,050
1,000
40
45
50
14
525
2,900
9,500
as Grams/Gram of total solids
.024698
.012048
.0004819
.0005421
.0006024
.000168
.006325
.034939
. 114457
Screen Dried
(60 Mesh)
6,700
5,500
300
5,270
140
49
1,150
6,770
39,500
.006700
.005500
.000300
.005270
.000140
.000049
.001150
.006770
.039500
60 Mesh Liquid
1:2 Dilution
850
475
18
23
20
6
250
1,440
4,000
.015858
.0088619
.000335
.0004291
.0003731
.0001119
.004664
.0268656
.074627
-------
TABLE 16(c): BIOLOGICAL OXYGEN DEMAND -- SWINE (807. URINE + 20% FECES) (NO DILUTION)
ON
vo
Days
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
*
325 Mesh
(ppm)
0
10,000
16,000
22,000
31,000
32,000
37,000
39,000
41,000
42,000
43,000
43,000
43,000
44,000
43,000
43,000
43,000
43,000
43,000
43,000
43,000
*
60 Mesh
(ppm)
0
18,000
21,000
26,000
31,000
30,000
30,000
28,000
34,000
29,000
27,000
26,000
32,000
36,000
38,000
32,000
28,000
24,000
24,000
**
Composite
(ppm)
0
10,000
15,000
25,000
31,000
37,000
38,000
37,000
37,000
37,000
37,000
37,000
39,000
38,000
38,000
38,000
37,000
37,000
35,000
* Effluent water from screen was analyzed for BOD
** Composite was analyzed for BOD prior to application on the screen
-------
TABLE 17(a): BEEF CATTLE WASTEWATER SCREENED
Composite
before
Screening
Total Solids
7. Change (-)
Volatile Solids
7. Change
Nitrogen (ppm)
% Change (-)
COD (ppm)
% Change (-)
BOD (ppm)
% Change (-)
Expressed as grams/gram
Nitrogen
COD
BOD
14.77.
0
76.87.
8,650
127,560
48,000
of total solids
.05884
.86775
.3265
Effluent
from
60 Mesh
6.387.
56.6%
78.57.
n o
8,425
2.6%
78,670
38.3%
20,000
57.5%
.1320
1.23307
.3197
Effluent
from
100 Mesh
6.38%
56.6%
78.3%
effective
8,340
3.567.
77,088
39.6%
18,600
61.257.
.1307
1.2082
.29153
Effluent
from
200 Mesh
6.072%
58.87.
77.3%
change
8,230
4.85%
74,679
41.5%
15,600
67.25%
.1355
1.2298
.25691
Effluent
from
325 Mesh
5.867.
60.37.
77.6%
7,920
8.439%
71,467
43.97%
15,600
67.50%
.1351
1.2195
.26621
-------
TABLE 17 (b): NUTRIENT AND MINERAL ANALYSIS OF BEEF CATTLE WASTEWATER LIQUID & SOLID FRACTIONS
Ca
Mg
Zn
Cu
Fe
Mn
Na
K
P
Expi
Ca
Mg
Zn
p..
v»U
Fe
Mn
Na
K
P
(ppm)
(ppm)
(ppm)
(ppm)
(ppm)'
(ppm)
(ppm)
(ppm)
(ppm)
ressed as
Beef Cattle Composite
(607. Feces + 407. Urine)
1,700
690
18
122
16
130
3,350
7,750
Grams/Gram of total solids
.011567
.004694
.0001224
.0008299
.0001088
.0008843
.0227S9
.05272
Screened Dried
Solids (60 Mesh)
3,000
2,500
82
5
240
12
200
10,000
16,000
.003000
.002500
.000082
00000 S
. wv/uv/*?
.000240
.000012
.000200
.010000
.016000
60 Mesh Liquid
1:2 Dilution
800
215
14
22
30
6
33
2,100
2,000
.012539
.003369
.0002194
00034ft
. \J\t\J J*TO
.0004702
.000094
.0005172
.032915
.0313479
-------
TABLE 17(c): BIOLOGICAL OXYGEN DEMAND -- BEEF CATTLE (607, FECES + 40% URINE) (NO DILUTION)
Day a
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
*
325 Mesh
(ppm)
0
5,500
12,000
16,000
17,500
19,500
21,000
22,500
25,000
28,000
29,500
29,500
31,000
31,000
31,000
31,000
31,000
31,000
31,000
31,000
31,000
*
60 Mesh
(ppm)
0
12,000
13,000
18,000
23,000
28,000
29,000
28,000
28,000
28,000
27,000
26,000
28,000
27,000
27,000
27,000
26,000
26,000
26,000
**
Composite
(ppm)
0
20,000
28,000
36,000
44,000
48,000
49,000
49,000
54,000
50,000
49,000
48,000
55,000
56,000
58,000
53,000
49,000
43,000
43,000
* Effluent water from screen is analyzed for BOD
** Composite was analyzed for BOD prior to application on the screen
-------
Compost Study on the Screened Solids;
A close-up view of the laboratory cotnposters used in this study is shown
in Figure 16. Approximately 4 cubic feet of each type of waste solids
(swine and beef cattle) were composted. Prior to introduction into the
reactors, the screened solids were squeezed manually to a moisture con-
tent of approximately 65%. The biological reactors were made of glass
cylinders with rubber gasketed metal end plates. Air saturated with
water vapor was introduced through the bottom of each reactor for aera-
tion purposes. By saturating the air in this manner prior to introduc-
tion into the reactors, problems of dehydration associated with artifici-
ally aerated composts were held down to a minimum.
The head space gases of each reactor were exhausted through a tap on the
head plate. The exhaust gases were bubbled through a barium hydroxide
solution for carbon dioxide stripping. Copper-constantan thermocouples
were placed at three locations (close to the bottom, at the middle and
near the top) in each reactor. A multi-channel potentiometer regulated
by a time clock provided hourly read-out of the temperature profile in
each reactor. The whole apparatus was housed in a controlled temperature
chamber which was maintained at 95°F throughout the compost period of 20
days.
In Figure 17, the recorded temperature profiles for each of the composts
are shown. Following initial equilibration at the ambient temperature
approximately 8 hours after the beginning of the experiment, the tempera-
ture at all positions in the reactors showed hardly any deviation from
the room temperature. There was a maximum deviation of 5°F above this
value on the second day, after which the temperature dropped to 95°F and
never rose above it for the rest of the test period. Daily inspection of
the barium hydroxide solution showed no visible precipitate of carbonate
salts. Volatile solids determinations made at weekly intervals showed
that there was practically no measurable dissipation of the volatile
matter over a 20-day period.
The conclusion from these observations is that the solids are indeed
reasonably stable and should store over extended periods without any
odor nuisance or aesthetic problems. Under low temperature winter con-
ditions the screened solids will store well on paved platforms beneath
some type roof. Under warmer climatic conditions, the solids will dry
readily under natural conditions to a fluffy, odorless, storable end
product. It is envisaged at this point that these solids will ultimate-
ly be disposed of on the land. It seems logical that the winter stock-
pile can be spread comfortably during plowing operations the following
spring.
For some time there has been a growing research interest in the recycling
of manure solids as feed for livestock. Nutrient and mineral content of
the screened solids suggest that they would be good candidates for -feed-
ing trials in a future study. An inspection of the comparative data in
73
-------
FIGURE 16: CLOSE-UP VIEW OF EXPERIMENTAL COMPOST SETUP.
Table 17b suggests that a substantial percentage of the minerals and
nutrients in the original feed are still present in the screened waste
solids. Of course, the availability of these nutrients can only be
determined in experiments in which animal growth responses are system-
atically analyzed.
The Liquid Fraction:
The properties of the liquid fraction from the screen are detailed in
Tables 16a, c and c for swine wastewater and in Tables 17a, c and c for
beef cattle wastewater.
A significant reduction in resistant solids has taken place in the screen-
ing process. The solids still present in the liquid fraction are largely
colloidal in nature and are expected to be the most biologically degrad-
able fraction. When further diluted in wash and transport water (a
dilution approximating 1:10 is considered reasonable for this purpose),
74
-------
100-
95
90-
Temp 80
SWINE
5-top
12-middle
2-bottom
BEEF
II- top
4- middle
7- bottom
i 1 1 h
II
H 1 1 h
8
10
days
FIGURE 17: TEMPERATURE PROFILE OF COMPOST BEDS
the total solids level of the wastewater from the screen would be below
0.5%.
A vast majority of the biological oxygen demand, minerals and nutrients
is present in the liquid fraction. The liquids in all cases were odor-
ous and had a pH of about 8.5 when fresh. Under anaerobic storage con-
ditions these liquids will become a serious odor nuisance problem and
would take up a sizable storage space. Treatability studies on the
liquid fraction will be reported in the next section of the report.
Along with these studies, we shall consider management strategies for
beef cattle and swine wastewaters integrating treatment and recycling
concepts to conserve both the gross quantity of water polluted as well
as the size and cost of facilities required to meet mandatory winter
storage specifications.
75
-------
SECTION VII
COMPARATIVE BIOLOGICAL TREATABILITY STUDIES
ON SWINE AND BEEF CATTLE WASTEWATER
LIQUID FRACTIONS
The data reported in Section VI of this report show that in passing
through the vibrating screen separator, the liquid fraction retains:
(a) for swine wastewater:
50-70% of the original volatile solids
50-60% of the original BOD
60-70% of the original COD
50-90% of the original nutrient and mineral contents.
(b) for beef cattle wastewater:
50-70% of the original volatile solids
50-60% of the original BOD
60-70% of the original COD
50-90% of the original nutrient and mineral contents.
The limited data on sand filtration presented in Section V of the report
indicated that although a substantial proportion of the residual solids
has been removed in the sand filter following initial screening, the BOD,
COD, nutrient and mineral contents of the liquid fraction remain substan-
tially potent as a potential pollutant and/or nuisance odor problem.
A large number of previous studies (2, 13, 14) has already established
the fundamental treatability of animal wastewaters in aerobic, anaerobic
or multi-stage biological systems. It is assumed that this basic prop-
erty remains essentially unaltered in the liquid fraction following the
isolation of resistant solids. From the detailed tabulation of the basic
properties of the liquid fraction in Section VI there is no question that
the residual (COD):(nitrogen):(PO.-P) ratio far exceeds the minimum
value of 100:5:1 specified for conventional biological treatment systems.
In view of the recognized adverse effect of excessive solids on transport
processes in such biological systems, it is hypothesized that the reduced
solids level together with the associated reduction in BOD and COD con-
centrations would substantially enhance treatment kinetics of the liquid
fraction. If such improvement can indeed be demonstrated, then a sound
basis would have been established for either relaxing present design
criteria or making such specifications substantially more effective.
In all, three types of treatments were studies, namely: strict anaerobic,
aerobic and combined anaerobic-aerobic stabilization of the liquid frac-
tion.
77
-------
oo
FIGURE 18: CLOSE-UP VIEW OF ANAEROBIC DIGESTERS
-------
ANLROB1C DIGESTER
FIGURE 19: DETAIL OF ANAEROBIC DIGESTER
79
-------
Anaerobic Digestion;
Experimental Apparatus; Six anaerobic digesters (Figure 18) were assem-
bled for the study. Each digester had an operating capacity of 11 liters
and was made of 8-inch diameter, 3/4-inch thick plexi-glass stock. A
close-up view of a typical digester showing its salient features is pre-
sented in Figure 19. The chief components of the digester are:
(a) rubber-stoppered loading port and gas trap on the cover plate
(b) three sampling outlets on the side
(c) a baffled-impeller mixing assembly.
The system was constructed so that it was for all practical purposes air
tight. Gases produced during fermentation could be collected if desired.
The mixing impellers in all six digesters were operated by a common
drive system consisting of a variable speed electric motor and six bevel
gear arrangements driven by a main horizontal shaft. The rotational
speed of the impellers could be varied between 0 and 500 rpm. The elec-
tric motor was operated on a percentage timer which allowed a 10-minute
mixing interval every hour. The entire assembly was housed in an envi-
ronmental walk-in chamber capable of controlling temperatures to within
+ 1°F of the desired value. All digestion tests were performed at 95°F.
All tests were run in duplicate.
Acclimation Procedure; An acclimation period of 15 days preceded the
actual experimental runs. Each digester was seeded with a filtered mixed
culture obtained from the bottom sludge of a functioning oxidation ditch
at the Michigan State University swine farm. Each digester was operated
until visible floes began to form. Thereafter, the unit was settled
daily for a period of 30 minutes, the supernatant liquid drawn off and
the system refed with the prospective test sample. After the 15th day
each digester had reached an impressive mixed liquor suspended solids
level of between 2,000-4,000 mg/1. At this point the unit was considered
ready to receive its first load of the test sample.
Experimental Procedure; This study was strictly carried out to determine
the comparative merit of solids removal as it affects kinetic parameters
in the digester. The digesters were operated for limited detention peri-
ods of between 10 and 15 days -- just long enough to see established
kinetic patterns. The digesters were operated on a batch-fed, draw-and-
fill basis. Mixed liquor volatile solids and COD were measured either
daily or on alternate days as indices of digester performance. The pH
was maintained in the range between 7 and 8 throughout the test period.
Experimental Results and Discussion; The data obtained fr$m the anaero-
bic studies undertaken are summarized in Table 18 for swine wastewater
liquid fraction and in Table 19 for beef cattle wastewater liquid frac-
tion. The loading rates of the digesters are reported either as a dilu-
tion factor or in the more conventional manner of giving them as pounds
80
-------
TABLE 18: ANAEROBIC DIGESTION OF SWINE WASTEWATER LIQUID FRACTION
Mesh Size of Screen = #60. Digestion Temperature = 95°F.
Dilution
or
Loading Days
0
1
2
3
1:5 - 4
0.34 5
,lb VS. 6
1 cfd ^
/
8
9
10
0
1
2
3
4
1:10 - 5
0.16 6
,lb VS. 7
cfd «
o
9
10
11
12
13
14
15
0
2
3
1:20 5
0.115 6
.lb VS. 7
cfd 8
9
10
(7.)
Total Volatile
Solids (TVS)
0.55
0.486
0.374
0.355
0.327
0.343
0.304
OQA.7
J*r/
0.343
0.341
0.341
0.256
0.232
0.210
0.205
0.209
0.201
0.194
0.203
0.207
0.214
0.198
0.218
0.205
0.208
0.201
0.185
01 ftl
.LQJL
0.147
0.139
01 95
-L£J
01 *4A
. 1JO
0.111
0.124
0.131
0.111
0.132
(7.)
Reduction
in TVS
0.00
11.70
32.00
35.50
40.60
37.70
44.80
37
37.70
38.00
38.00
0.00
9.40
18.00
20.00
18.40
.21.50
24.30
20.80
10 nn
iOf UU
19.20
16.50
22.80
14.90
20.00
18.80
21.50
0.00
2 on
. Au
20.60
24.90
&D 3U
40.00
33.00
29.20
40.00
28.70
(ppra)
COD
Remaining
5,490
5,480
5,241
4,800
4,409
4,300
4,050
3,600
3,100
4,200
4,000
4,200
3,950
3,800
3,900
3,700
3,700
3,600
3,400
3,000
2,600
1,497
1,331
1,200
950
850
K (base 10) Predicted Deten-
COD Removal tion Time for
Rate Constant 507. COD Removal
(0.145) 7 days
day
(.071) 15 days
day
t 1 fil "i A 5 HJIVQ
I- J.OA / o j nays
day
81
-------
TABLE 19: ANAEROBIC DIGESTION OF BEEF CATTLE WASTEWATER LIQUID FRACTION
Mesh Size of Screen = #60. Digestion Temperature = 95°F.
Dilution
or
Loading Days
0
2
3
4
1:5 - 5
0.61 6
*lb VS. 7
( cfd >
eta g
9
10
0
1
2
3
4
5
1:10 - ,
o
0.3 ?
rlb VS.
( cfd > 8
9
10
11
12
13
14
15
0
1
2
3
4
1:20 - 5
.165 6
1 v trC 7
.lb VS. /
Cfd j
9
10
tt)
Total Volatile
Solids (TVS)
0.99
0.88
0.76
0.72
0.65
Ofifi
QW
a. 615
0.625
0.629
0632
. VJf.
0.678
0.470
0.440
0.417
0.403
0.378
0.371
0.339
03A2
. J*vA
OA/ »
.345
0.326
0.344
0.308
0.315
0.309
0.310
0.305
0.264
0.260
0.238
0.222
0.209
0.210
0.185
OOAC
** J
0.200
0.199
0.216
(%)
Reduction
in TVS
0.00
U2O
. f.\i
23.30
27.30
34.40
MAn
. *»v
37.90
36.90
36.50
in 9n
31.60
0.00
6.40
11.30
14.30
19.60
21.10
27.90
27.30
OC £A
26.60
30.70
26.90
34.50
33.00
34.30
34.10
35.20
0.00
1.60
9.90
16.00
18.90
18.50
30.00
24.30
24.70
18.20
(ppra)
COD
Remaining
12,900
11,148
9,817
10,040
8,560
8,200
7,800
7.750
7,800
7,400
7,200
7,000
6 Ann
,400
6,600
6,400
6,000
5,000
2,912
2,682
2,590
? 175
A, LI J
2,270
K (base 10) Predicted Deten-
COD Removal tion Time for
Rate Constant 50% COD Removal
ft\ nft? % 19 A A
day
(0.076) 14 days
day
(0.087) 12 days
day #
82
-------
of volatile solids per cubic foot of digester per day. Within the broad
context of this study, the dilution method is convenient since we are
ultimately interested in the management of waste slurries subjected to a
mandatory dilution in wash and transport water.
Semi-log plots of the COD removal data in Figure 20 furnishes us with
approximate COD removal rate constants for the anaerobic tests. These
constants are tabulated as K (base 10) in Tables 18 and 19. Since the
plots of Figure 20 approximate a first order kinetic pattern, we can
assume the following metabolic equation to apply:
E = KT/(1 + KT) (4-1)
where E = COD removal efficiency
K = COD removal rate constant (I/day)
T = detention time (in days)
Equation 4-1 is easily rearranged to the form:
T = E/K(1 - E) (4-2)
Using equation 4-2 we can now estimate expected detention periods neces-
sary to achieve 50% COD removal from the various digester operating
conditions. The calculated values are recorded in Tables 18 and 19. For
swine wastewater liquid fraction diluted 1:20 or else loaded into the
digester at the rate of 0.115 (Ib.VS/cfd), we need approximately 7 days
to dissipate 50% of the initial COD. The characterization data in seq-
tions IV, V and VI suggest a COD/BOD ratio of between 2 and 4 for swine
wastewater liquid fractions. It is therefore safe to conclude that a
50% COD reduction is equivalent to BOD reduction far in excess of the
90% level. The implication of this now becomes that 7 days of anaerobic
digestion of the liquid fraction under the operating conditions specified
would eliminate over 90% of biological oxygen demand. Under similar
loading and operating conditions (but without solids separation),
Taiganides (14) reports that a minimum detention period of 10 days would
be required for comparable treatment of hog wastes in an anaerobic di-
gester. It appears therefore that considerable improvement in treatment
kinetics can be achieved by solid separation processes preceding anaero-
bic digestion.
Although the experiment was set up to simulate high rate anaerobic di-
gesters, the results are expected to be no less true for anaerobic la-
goons subject to varying ambient temperature conditions. The performance
of such lagoons can be considerably improved by sensible removal of re-
sistant solids prior to charging the lagoons. Present design specifica-
tions of such lagoons have not met the nuisance odor problem adequately,,
Solids separation could help to improve lagoon performance to a point
where odors can be further suppressed.
83
-------
10
10
l:5-Dilution
l:20-Dilution
(a) Anaerobic Test
Beef Cattle Wastewater Liquid Fraction
Following Screening through #60 Mesh Screen
7 ^ 8
Days
10 11 12
13 14
15
10
10
l:20-Dilution
(b) Anaerobic Test
Swine Wastewater Liquid Fraction
Following Screening through #60 Mesh Screen
123456789
Days
10 11 12 13 14 15
FIGURE 20: ANAEROBIC DIGESTER PERFORMANCE PLOTS
84
-------
00
co
T3
rl
iH
O
0)
iH
O
cd
4J
O
H
C
O
O
TJ
(1)
1:10
Dilution
5 6
Days
10 11 12 13 14
15
FIGURE 21: VOLATILE SOLIDS DISSIPATION FOR SWINE WASTEWATER ANAEROBIC ALLY DIGESTED AT 95°F
-------
00
rl
rH
O
CO
Q)
cd
rH
O
rH
CD
4-1
O
H
C
H
(3
O
H
4J
O
12 13 14 15
FIGURE 22: VOLATILE SOLIDS DISSIPATION FOR BEEF CATTLE WASTEWATER ANAEROBICALLY DIGESTED AT 95°F
-------
In the short interval of the test, considerable dissipation of volatile
solids was obtained. Figures 21 and 22 show the percentage reduction in
volatile solids as a function of time. The general unsteady behavior of
the solids is typical of the mixed liquor and reflects the complex dynam-
ics of growth, synthesis and dissipation of volatile solids occurring
simultaneously in these systems. This characteristic pattern serves as
additional evidence that the digesters were operating satisfactorily.
In the absence of more generally accepted behavior indices such as total
volatile organic acids and alkalinity, the total volatile solids per-
formed a very essential service.
Aerobic Digestion Tests:
^Experimental Apparatus: The bioxidation system together with some of
the instrumentation used is shown in Figure 23. The basic system con-
sists of four batch-fed, fil1-and-draw type units equipped with porous
stone diffusers for aeration purposes. Each unit had an operating
capacity of 8 liters. Enough air was forced through the diffusers to
provide adequate mixing while maintaining dissolved oxygen levels no
less than 2 ppm throughout the test period.
FIGURE 23: CLOSE-UP VIEW OF BIOXIDATION APPARATUS
87
-------
Acclimation Procedure; A procedure similar to the one previously des-
cribed for the anaerobic tests was used. Each bioxidation unit was
seeded with a filtered mixed culture taken from the effluent of an oper-
ating oxidation ditch on the Michigan State University swine farm. The
unit was then aerated until visible floes were formed. On each subse-
quent day thereafter, the system was settled for 30 minutes, decanted
carefully, refed and aerated. This process continued until a mixed
liquor suspended solids level of about 6000 mg/1 was reached. At this
point the unit was considered ready to receive the first load of test
sample. Again the acclimation period was 15 days.
Experimental Procedure; The bioxidation units were fed once a day on a
draw-and-fill basis. The detention time on the average test was 10 days.
However, the test was discontinued whenever sufficient data had been
obtained to estimate the removal rate constant. Most of this type of
data was obtained between 6 and 8 days.
Two basic types of raw sample were used, namely:
(a) screened only or screened and sand filtered prior to bioxidation and
(b) screened and anaerobically pretreated prior to bioxidation.
In each test the pH was maintained between 7 and 8. Daily COD measure-
ments were made as the primary performance index. The operating temper-
ature of the digesters was about 70-75°F.
Experimental Results and Discussion; The data obtained from bioxidation
tests on swine wastewater liquid fraction are summarized in Table 20.
Kinetic plots of this data are presented in Figure 24. Similar represen-
tations of the experimental data on beef cattle wastewater can be seen in
Table 21 and Figure 25.
The removal rate constants for COD are particularly impressive. In the
treatment of municipal wastewaters by extended aeration in aerated lagoons
it is customary to use estimates of K in the range 0.55 to 1.0 per day
for cases in which the removal rate constant has not been explicitly
determined. Compared with the above value, K values ranging between 0.4
and 0.8 per day obtained in the various tests reported here are indeed
impressive. The removal of resistant solids from the mixed liquor sub-
stantially improves transport processes such as oxygen transfer and sub-
strate diffusion. The absence of these solids also effectively reduces
the viscosity of the liquor which is directly related to mixing opera-
tions in the biological reactor.
It is interesting to note that sand filtration following the initial
screening step does not seem to improve the removal rate constant in both
cases tested. This would suggest that additional solids removal in a
sand filter may be superfluous.
88
-------
TABLE 20: AEROBIC DIGESTION OF SWINE WASTEWATER LIQUID FRACTION
oo
vo
Dilution
1:10
1:20
Days
0
1
J.
2
3
4
5
6
7
8
9
10
0
1
2
3
4
5
6
7
8
9
10
(1)
Screened on #60 Mesh
Prior to Bioxidation
C.O.D. (ppm)
1,081
Qf\r\
yuu
750
600
600
400
395
398
"*"
(2)
Screened on #60 Mesh
4- Sand Filtered (4ft)
Prior to Bioxidation
C.O.D. (ppm)
1,000
700
500
340
300
__.__
(3)
Screened on #60 Mesh
Anaerobically Digested for
10-days Prior to Bioxidation
C.O.D. (ppm)
871
620
442
340
240
150
169
225
-------
Prior to
Bioxidation
(1) »> Screened through 060 mesh
(2) > Screened + sand filtered
(3) »> Screened + anaerobically digested
for 10 days.
3456789 10
FIGURE 24: BIOXIDATION OF SWINE WASTEWATER LIQUID FRACTION, OD
REMOVAL RATE PLOT
90
-------
TABLE 21: AEROBIC DIGESTION OF BEEF CATTLE WASTEWATER LIQUID FRACTION
Dilution
1:20
Days
0
1
2
3
4
5
6
7
8
9
10
(1)
Screened on #60 Mesh
Prior to' Bioxidation
C.O.D. (ppm)
1,415
1,100
866
600
540
480
320
320
(2)
Screened on #60 Mesh + Sand Filtered
(4 ft) Prior to Bioxidation
C.O.D. (ppm)
1,303
1,000
808
650
500
400
320
320
-------
e
a
o
o
10
(1) «> Screened on #60 Mesh Screen j Prior to
(2) «> Screened + Sand Filtered Bioxidation
'(1) K base (10) - 0.56/day
(2) K(base 10) - 0.56
day
I
123456789 10
Days
FIGURE 25: BIOXIDATION OF BEEF CATTLE WASTEWATER LIQUID FRACTION,
COD REMOVAL RATE PLOT
92
-------
Anaerobic digestion prior to bioxidation of the screened swine waste-
water liquid fraction shows a very definite improvement in the oxidation
kinetics. The removal rate constant was effectively doubled (see
Figure 24). This is not at all surprising. An increasing amount of
experimental evidence is being brought to bear on the fact that combined
anaerobic-aerobic treatment systems are both functionally and economically
best suited for managing highly polluted wastewaters such as concern us in
this study. Biochemical theory of anaerobic metabolism appears to favor
the combined process since the organic intermediates of anaerobiosis make
easily available and usable energy sources for the aerobic culture. A
synergism develops in which insoluble, otherwise unavailable waste or-
ganics are first hydrolyzed in the anaerobic step and then totally
oxidized in the terminal bioxidation step.
93
-------
SECTION VIII
DESIGN ALTERNATIVES
The primary object in this section is an attempt to provide design alter-
natives derived from integrating the salient features of this study into
existing design concepts.
As a problem, we specify a Michigan hog producer raising an average of
1,000 market weight (150 Ibs) animals at any time. This producer would
process between 2,000 and 3,000 (say 2,500) hogs through his operation
annually. It is our purpose to provide alternative designs to meet the
following criteria.
1. In-the-building operation with automatic hydraulic collection and
transport of wastes.
2. Provide odor-free storage for a minimum of 6 months over the winter
period. When ultimately spread on the land following storage, there
should be no odor-nuisance or pollution problems.
3. Employ concepts of treatment and recycle to minimize costs, con-
serve space and limit the gross bulk of water polluted.
4. Recommended waste characteristics as summarized from this and pre-
ceding studies are as follows:
(i) BOD =2.1 lb/1,000 Ib of hogs
(ii) Manure production rate = 1.5 gallons per animal per day
(iii) Assume that in the process of collection and transport the manure is
diluted 1:20 times.
Design #1; Aerated Lagoon with Diffused Aeration System
The basic system layout proposed can be seen in Figure 26. Its primary
features are:
(i) In-building automatic flushing system-
(ii) Vibrating screen separator which takes out refractory solids prior
to lagooning of the liquid fraction.
(iii) A control (inlet-outlet) structure. This structure should be
housed in a metal shed large enough to contain an air compressor and the
vibrating screen separator. Heat discharge from the compressor would
keep the structure warm enough during winter to prevent freezing in sub-
zero weather.
95
-------
VO
SUBMERGED
TUBING.
(DIFFUSED
AERATOR)
ntiun I
SPRAY IRRIGATION
ON FARM LAND
VIBRATING SCREEN
SEPARATOR
I INFLUENT
I «
WASTE
CONTROL
([INLET-OUTL]
|STRUCTURE
HOG
HOUSE
(1,000
T) ANIMALS)
RECYCLE WASTEWATER
28,500 gpd
(>1,500 gpd)
FIGURE 26: %AYOUT OF PROPOSED WASTE MANAGEMENT SYSTEM
-------
VO
FIG. 27 =
PROPOSED CLOSED SYSTEM
WASTE MANAGEMENT FOR LIVESTOCK
-------
(iv) An aerated lagoon equipped with a diffused aeration system consist-
ing of submerged perforated plastic laterals branching from metallic main
pipes connected to an air compressor. A unique feature of the lagoon is
that it is baffled to prevent short circuiting.
(v) Pumps and connecting pipe lines.
(vi) A solid storage space under snow cover draining into the lagoon and
equipped for self loading and unloading.
(vii) Equipment (portable spray irrigation) for soil disposal of the la-
goon effluent under favorable weather conditions.
The best-known manufacturer of the diffused aeration system proposed
above is Hinde Engineering Company, 654 Deerfield Road," Highland Park,
Illinois 60035. The system has the advantage that aeration continues
undisturbed even in -sub-zero weather when the lagoon surface is frozen.
It is thus possible to operate the system on a recycling scheme all the
year around.
Design Outline:
1. Estimated Organic Load: We assume a 50% reduction in BOD of raw
waste following separation on the screen (see Section VI of the report
for supporting data).
T j mnr> u v 150 Ibs .. (1.05 Ibs BODs)
Organic Load = 1000 hogs X r X .Kn-n 'iv «u~
" 6 hog 1000 Ibs of hog
= 158 Ib BOD5 per day
2. Hydraulic Load; 1000 hogs at 1.5 gpd/hog = 1500 gpd of waste
slurry. This will be diluted 20:1 by recycled water during transport.
Assume no spray irrigation during the six winter months. Then
storage for 6 months = 6 (months) X 30 a^- X 1500 gpd
= 270,000 gallons = 36,200 cubic feet
3. Pond Specifications: Estimated organic load = 158 Ib of BOD5/day.
Assume required removal = 90% => 142 Ib of BOD_/day.
Allow 150 Ib of BOD_/acre 10 ft water depth (from Hinde Company Design
Handbook).
1 *^Q
Area required = rrr- = 1.0 acre or 43,560 square feet. *
J.-3U
Volume of pond = 150' X 300' X 10'
= 450,000 cubic feet
= 3,380,000 gallons
98
-------
Baffle
75'
<=-
in
75'
out
(a) Top View of Lagoon
300'
75'
L 75>
2'-freeboard Baffle
10' to 11' water clay sealing
150 ' ^- 4, J< 4
65'
(b) Vertical Cross-section
FIGURE 28: LAGOON SECTION -- DETAILS
99
-------
Detention at 30,000 gpd = 3|o?°A??° =" HO days
5U f \J\J\J
4. Pond (Construction; See Figure 28 for construction details.
3
Storage depth required = 36?200 £t = i ft
43,560 ft
i.e., over 6 months, pond depth increases from 10 ft to 11 ft.
5. Operation of Lagoon System; The start-up should be planned to coin-
cide with good temperature conditions -- preferably summer or spring. The
start-up procedure should include microbial seeding of the lagoon with
about 5% of its volume of active return sludge from a properly function-
ing activated sludge system or extended aeration facility.
The lagoon should be filled with tap water to approximately the 10-ft
depth mark as part of the start-up operation. Since the lagoon is de-
signed for a 100-day detention period, discharge should be withheld for
the first 100 days following initial input of wastes into the lagoon.
Subsequently, spring, summer and 'fall discharges from the lagoon through
a portable sprinkler irrigation system should average no less than 1,500
gpd to prepare the lagoon for storage of influent wastewater during the
following winter.
It has been assumed that in Michigan and parts of the midwest region
evaporative losses from the lagoon closely approximate rain water gains.
This is generally true over a one-year period. For certain parts of the
year, however, evaporative losses from the lagoon would be so high as to
create a need for supplementary fresh water.
The solids stored over the winter months should be spread on the land
the following spring. During favorable weather, accumulation of the
solids over extended periods should depend on the availability of storage
space. Our laboratory determinations show that the solids have a bulk
density of 60 Ib per cubic foot following screening. If 100% of the
solids were taken out on the screen, assuming that the fresh manure con-
tains 10% total solids then, over a 6-month period, a unit such as the
one under consideration should provide space for solids storage of
approximately :
l'5 f **lof X 1000 hogs X 22*25. X 6 months X *£*"» x (0.i) X
(hog) (day) e month . gallon
o
* £* ^4000 cubic feet
60 Ib
m
A storage house 20' X 20' X 10' high would do the job. Since the actual
percentage of solids removed by the screen on the average is only half of
our design value, the storage space provided would be adequate for year
around storage and should therefore allow for considerable flexibility
in the management of solids disposal.
100
-------
Under warm summer and spring conditions the solids would dry quite readi-
ly if exposed. Solids dried down to 10% moisture content (wet basis)
were found to have a bulk density of about 20 Ib per cubic foot. If it
is intended to store this type of solid, a space three times that pre-
viously recommended would be necessary.
Since the handling of this amount of solids would require considerable
manpower, it is advisable to mechanize the transport of the solids from
the screen to storage and also the unloading of the solids from storage.
Construction Cost:
Assume good soil on the site and clay to seal the lagoon bottom on site.
1. Lagoon
(a) Strip site -- $200.00
(b) Dikes -- 18 (^ ) X (1,232 ft) (^) = $8,900
yd
(c) Baffle -- 300' at $6/ft = $1,800
(d) Grass seeding = $1,000
(e) Clay seal = $1,000
(f) Piping (in and out) = $1,000
(g) Aeration equipment:
Compressor, Tubings, = 10,000
etc.
Labor to install = 2,000
Subtotal $25,900
2. Pumping Station
(a) Sump (sludge) pump $ 500
(b) Recycle and Spray irriga-
tion pump 250
(c) Control structure 1,000
(d) Metal building over
control structure 200
(e) Electrical supply 300
101
-------
Subtotal 2,250
3. Solids Separation System
(a) Vibrating screen separator 1,500
(b) Solids storage 500
(c) Solids handling equipment 500
Subtotal 2,500
4. Automatic Flushing Equipment 3,000
Subtotal 3,000
Gross total $33,650
If we assume a five year amortization period, with 2500 animals being
processed through the facility each year, then the investment costs in
the system would be equivalent to $3 per hog excluding intere.st rate if
money were borrowed to build the system. This amount of money may seem
prohibitive at first sight, but compared with a winter underground stor-
age tank costing $20,000, this would be a wise investment over the long
run.
Design #2; Aerated Lagoon with Floating Surface Aerator
By using a floating aerator instead of the diffused system of Design #1,
we can substantially reduce the cost of the aeration component. The
major disadvantage of this alternative is that winter operation of the
aerator is uncertain. To be safe, the design of a system using floating
aerators should consider the lagoon as an inactive storage system during
the six winter months. A six-month period of inactivity in the lagoon
would have serious implications in the basic system design. It would
mean that:
(a) recycling of lagoon water cannot be practiced during winter or in
early spring just following lagoon thaw.
(b) the aeration equipment would be substantially overdesigned to handle
the heavy oxygen requirement which must be satisfied immediately follow-
ing the spring thaw to prevent the persistence of nuisance odors from
the lagoon over the first few weeks of spring. A modified design is
presented as follows:
1. Estimated Organic Load: from previous calculation this was found
to be 158 Ib of BOD_ per day.
102
-------
2. Hydraulic Load; 1000 hogs at 1.5 gpd/hog = 1500 gpd of slurry.
This will be diluted 20:1 with fresh water during 6 months of winter
storage when recycling of wastewater is impracticable.
Storage for 6 months of winter = 6 (months) X 3° - X 30,000 gpd
month
= 540,000 gallons = 72,400 ft3
3. Pond Specifications; Assume mean BOD removal rate constant (K)
of 0.1 per day. (From our bioxidation tests on the liquid fraction, this
is quite conservative.) The required lagoon detention time is calculated
from the pseudo-first-order equation:
E = KT/(1 + KT) K = removal rate constant (per day)
T = detention time
or T = E/(1-E)K
E = efficiency of removal
For 90% BOD removal
- 90
At the peak period just following the spring thaw we must satisfy approxi-
mately 2 X (158 Ib of BOD per day) = 316 Ib of BOD per day.
Surface area of pond = 1 acre or 43,560 square feet (assumed)
Water depth = 10 ft (assumed)
Volume of pond =150' X 300' X 10' = 450,000 cubic feet
= 3,380,000 gallons
Actual detention at
30,000 gpd = 110 days.
4. Aeration Equipment; Let D = Applied 5-day BOD in Ibs/hr
L = Ratio of ultimate BOD to 5-day BOD
(assume 1.47)
P = % BOD reduction
NOR = Normal (biological) oxygen requirement
in Ibs of oxygen per hour
O-i f
then: NOR = LXDXP = 1.47X ^ X 0.9 = 17.5 Ibs of 02 per hour for
our system. Two 5-hp surface aerators with standard oxygen transfer rate
of 3 Ibs of 02/hp-hr will be satisfactory for oxygenation and mixing re-
quirements .
103
-------
5. Lagoon Construction and Operation: This item would be identical to
that for the previous lagoon, except that in winter no recycle from the
lagoon is planned.
Construction Costs: These would be identical to the previous lagoon ex-
cept that the floating aerators would cost approximately $3000 per unit
and installation costs would be considerably less. In all, our estimate
is that floating aerators would lead to savings of at least $5000 from
the original estimate.
Operating costs for the floating aerators would be considerably less
since they would be off all winter. Also in cases of malfunction, the
surface aerators are more easily accessible and do not require any extra-
neous maintenance. The diffused aeration system in contrast, must be
flushed twice a year with dilute hydrochloric acid and would be difficult
to get at in cases of malfunction.
Other Designs; There are a number of other possible systems which can
be developed from some of the completed or on-going research studies in
livestock waste disposal, which would in effect, combine the essential
features of the systems proposed.
Taiganides and White (15) of the Ohio State University are currently per-
forming a demonstration study on a system which separates the refractory
solids on a stationary screen. The liquid fraction is reclaimed in an
oxidation ditch and then recycled for the collection and transport of
influent hog wastes. The solid fraction is stabilized in a wet oxidation
tank. The tank is integrated into the system in such a way that heat
recovered from the wet oxidation process is used to maintain the tempera-
ture of the oxidation ditch under winter temperature conditions. This
system has been in operation for less than two months (at the time this
report was written). It is therefore too early to evaluate its perfor-
mance. However, it is the type of bold and resourceful deviation from
conventional practice which is likely to provide new directions in live-
stock waste management.
At Iowa State University, Smith, Hazen and Miner (16) are looking at two
approaches to recycling of renovated wastewater for hogs involving:
(a) direct bioxidation of fresh hog wastewater in an oxidation ditch
and
(b) sequential treatment of the hog wastewater in an anaerobic lagoon
followed by an oxidation ditch prior to reuse.
t*
This study together with an earlier study by Smith and Jenkins (17) indi-
cates that excessive salt build-up would not be an acute problem in the
operation of treatment systems designed on a recycling concept.
Experimental studies of an aerated lagoon system with spray irrigated
effluent has been carried out by Dale et al. (18) at Purdue University.
104
-------
The results of this study show that aerated lagoon effluents are odorless
and can be resourcefully and conveniently disposed-of by sprinkler irri-
gation on farm land.
All of these studies show a very definite promise of radical new designs
which would match the social and economic challenges of livestock waste
disposal.
105
-------
SECTION IX
ACKNOWLEDGEMENTS
Dr. B. A. Stout, Professor and Chairman, Department of Agricultural
Engineering is gratefully acknowledged for carrying a major portion of
the administrative burden and providing inspiration to the project to
the very end.
The invaluable support of Mr. J. B. Barrows, President-Manager of Indus-
trial Equipment Company of Detroit, Michigan, is acknowledged with sin-
cere gratitude. Mr. Barrows furnished a Sweco Vibrating Screen Separator
at no expense to the project for as long as was necessary to complete
the test program.
In the early stages of its development, the following people provided
administrative guidance to the project: Dr. M. L. Esmay, Professor,
Department of Agricultural Engineering, and Dr. C. W. Hall, formerly
Professor and Chairman, Department of Agricultural Engineering, Michigan
State University and presently Dean of the College of Engineering, Wash-
ington State University, Pullman, Washington. The assistance of these
individuals is gratefully acknowledged.
The Animal Husbandry Department of Michigan State University provided
valuable assistance in various ways. We acknowledge in particular the
unfailing help of the following members of that department: Professor
Edward C. Miller, Dr. Elwyn R. Miller and Dr. Duane E. Ullrey.
The grant support by the Office of Research and Monitoring, Environmental
Protection Agency, and the help of Mr. Marion R. Scalf, the grant Project
Officer, is acknowledged with sincere thanks.
107
-------
SECTION X
REFERENCES
1. Boyd, J. S. (1971) Professor of Agricultural Engineering, Michigan
State University, Private Communication.
2. Loehr, R. C. (1968) Pollution Implications of Animal Wastes -- A
Forward Oriented Review. Report of Federal Water Pollution Control
Administration, Robert S. Kerr Water Research Center.
3. McKinney, R. E. and R. Bella (1967) Water Quality Changes in Con-
fined Hog Waste Treatment. Project Completion Report, Kansas Water
Resources Research Institute, University of Kansas.
4. Cassell, E. A., A. F. Warner and G. B. Jacobs (1966) Dewatering
Chicken Manures by Vacuum Filtration. In Proceedings of National
Symposium on Management of Farm Animal Wastes, Michigan State
University, ASAE Publication #SP-0366, pp 85-91.
5. Holmes, L. W., D. L. Day and J. L. Pfeffer (1971) Dewatering of
Oxidation-Ditch Mixed-Liquor by Centrifugation. Paper presented at
the 2nd International Symposium on Livestock Wastes, The Ohio State
University Center for Tomorrow.
6. Gilbert, J. J. (1959) Mechanical Screens for Industry. Proceedings
of the Eighth SMIWC, pp 21-31.
7. Fairbank, W. C. and E. L. Bramhall (1968) Dairy Manure Liquid-
Solids Separation. Bulletin #AXT-271 of the University of Cali-
fornia Agricultural Extension Service.
8. "Sludge Dewatering", WPCF Manual of Practice #20. (1969)
9. Perry, J. H. (1963) Chemical Engineers' Handbook. McGraw-Hill
Book Company, New York.
10. Kumar, M., H. D. Bartlett and N. N. Mohsenin (1970) Flow Properties
of Animal Waste Slurries. Paper #70-911 presented at ASAE Winter
Meeting, Sherman House, Chicago, Illinois, December 8-11.
11. Fair, G. M., J. C. Geyer and D. A. Okun (1969) Water and Waste-
Water Engineering. Vol. 2, John Wiley and Sons, Inc., New York.
12. Methods of Soil Analysis. Part 2, Chemical and Microbiological Prop-
erties. Editor in Chief: C. A. Black, published by the American
Society of Agronomy, Inc., Madison, Wisconsin (1965).
13. Jones, D. D., D. L. Day and A. C. Dale (1970) Anaerobic Treatment
of Livestock Wastes. Bulletin 737 of the University of Illinois
College of Agriculture and Agricultural Experiment Station.
109
-------
14. Taiganides, E. P. (1963) Characteristics and Treatment of Wastes
from a Confinement Hog Production Unit. Unpublished Ph.D. Thesis,
Iowa State University.
15. Taiganides, E. P. and R. K. White (1971) Automated Handling, Treat-
ment and Recycling of Wastewater from an Animal Confinement Produc-
tion Unit. Presented at the 2nd International Symposium on Livestock
Wastes, The Ohio State University, Columbus, Ohio.
16. Smith, R. J., T. E. Hazen and J. R. Miner (1971) Manure Management
in a 700-head Swine Finishing Building; Two Approaches Using Reno-
vated Wastewater. Paper presented at the 2nd International
Symposium of Livestock Wastes, The Ohio State University, Columbus,
Ohio.
17. Smith, R. E. and J. D. Jenkins (1969) Salt Concentrations in a
Recycling Aerobic Waste Disposal System. Paper #69-927 presented
at the Winter Meeting of ASAE,'Sherman House, Chicago, Illinois,
December 9-12.
18. Dale, A. C., J. C. Ogilvie, M. P. Douglas and A. Cheng (1969)
Disposal of Dairy Cattle Wastes by Aerated Lagoons and Irrigation.
In Proceedings of Animal Waste Management, pp 150-159, Cornell
University Conference.
19. Farm Animal Wastes -- 1970. Bulletin of the North Central Research
Committee, NCR-67 (NC-69) Farm Animal Waste Disposal (in prepara-
tion) .
. GOVERNMENT PRINTING OFFICE : 1972484-484/159
-------
1
Accession Number
w
5
« Subject Field & Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
ACTTI nil tirral F.ncnnefiri ncr Denart.ment,
Michigan State University
East. T,arising. Michigan 48823
Title
CLOSED SYSTEM WASTE MANAGEMENT FOR LIVESTOCK,
10
Authorfs)
Ngoddy, Patrick 0.
Harper, Jerome P-
Collins, Robert K.
Wells, Grant D.
Heidar, Farouk A.
16
Project Designation
EPA/OEM Project Report No. 13040 DKP 06/71
21
Note
22
Citation
23
Descriptors (Starred First)
*Vibrating Screen, *Hydraulic Transport, *Ultimate Disposal, *Aerobic Lagoon
Treatment,*Recycled Washwater, *Dimensional Analysis, *Livestock Waste
25
Identifiers (Starred First)
*Physical Treatment, *Biological Treatment, *Recycle, *Livestock Waste
27
Abstract
The vibrating screen separator is examined for liquid-solid separation of livestock
wastewater. A general procedure for performance analysis and estimation of this type
of dewatering apparatus is developed and verified using methods of dimensional
analysis. Tests on swine and beef cattle wastewaters show that the resistant or
slowly biodegradable solids are effectively removed on vibrating screens ranging
in mesh size from #60 to #120. Although it is measurably less efficient than con-
ventional dewatering devices such as centrifuges and vacuum filters, the gravity
dewatering screen separator is better suited to the economic scale of the average
livestock operation.
The separated solid fraction from the screen is fibrous, odor free, stable,
storable and does not attract flies. The absence of these solids from the liquid
fraction is shown to stimulate faster rates of COD and nuisance odor dissipation
presumably by enhancing otherwise rate-limiting transport processes in biological
treatment systems.
The salient aspects of this study are integrated into candidate confinement
livestock waste management designs.
Abstractor
', D. Anderson
Institution
EPA/ORM
Wt?:I02 (REV. JULY 1989)
WRSIC
SEND. WITH COPY OF DOCUMENT. TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 20240
* GPO: 1970-389-930
------- |