PB-234 930
SOLID WASTE MILLING AND DISPOSAL ON LAND WITHOUT COVER \
I
VOLUME I, SUMMARY AND MAJOR FINDINGS
CITY OF MADISON, WISCONSIN
PREPARED FOR
ENVIRONMENTAL PROTECTION AGENCY
1974
DISTRIBUTED BY:
KTLT1
National Technical Information Service
U. S. DEPARTMENT OF COMMERCE
-------
BIBLIOGRAPHIC DATA i. !-'
4. Title anUSubt i
PB 234 S3C
SOLI:: Y;A-TV vaLLry? *'<:; DIT^-M ov LAND WITHOUT COVER
7. Authoc(s)
John J.
Robert K. Ham
9. Performing Organization Name ana
City of Madison, "'is cons in
12. Sgonsoring Organization Name and Adoiess
\- U. ". environmental Protection Agency
I Office of Polid 7'aste Management Programs
j- Washington, D.;\ ?04fO
5- Report
6. |
j
8. Performing Organizatior. Rt r; j
No. i
10. Project/Task/Work Un.'. N, .
No.
G06-EC-00004
13. Type of Report & Period
C-«
Final
14.
15, Supplementary Notes
16. Abstracts
;lhe project ber.an as p. pr^cticpl demonstration to investigate the concept of
solid waste for landfill disposal without applying daily cover. The project wasin
to gather datn on the operation and cost of milling equipment, the use of milled
waste in K landfill, and the characteristics of milled solid wafete —all from e
standpoint, ^esrtup 1 ly , howf-ve'", project personnel carried out detailed investig
work on both milled and unprocessed solid waste. To determine whether landfilling
solid waste lived up to the i-.cprovel given by European landfill operators, it bec
necessary to design exper'Jir,er.t.-. invol%n.ng rats, flies, leachate, gas, trees, etc.
hopes of quantifying such f^ot.c,rs.
tenc- d
solid
ative
m.ili
ame
17. Key Words and Document Analysis. 17a, Descriptors '
wente disposal, T1rb9n ereas; Crvishers, Comminution, Grinding mills
17b. Identifiers/Open-Ended Terms
r'olid w^st'; disposal, i'ndison, "'isconsin
17e. COSATI Field/Group
Reproduced by
NATIONAL TECHNICAL
INFORMATION SERVICE
U S Department of Commerce
Springfield VA 22151
18. Availability Statement
Relepse tp public
19. Security Class (This
Report)
UNCLASSIFIED
I 21. No. o
20. Security Class (This
Page
UNCLASSIFIED
I NTIS •>*, ( 10-70)
USCOMM-L'!
-------
SOLID WASTE MILLING AND DISPOSAL ON LAND WITHOUT COVER
Volume I: Summary and Major Findings
This final report (SW-62d.l) on wovk performed
under- Federal solid waste management demonstration grant No. G06-EC-00004
to the City of Madison, Wisoonsin3 was written
by JOHN J. REINHARDT and ROBERT K. HAM
and is reproduced as received from the grantee
, 7 p*
cn^ -
Chicago,
U.S. ENVIRONMENTAL PROTECTION AGENCY
1974
-------
This report has been reviewed by the U.S. Environmental
Protection Agency and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection
Agency, nor does mention of commercial products constitute
endorsement or recommendation for use by the U.S. Government.
The position taken by the U.S. Environmental Protection
Agency on the disposal of uncovered, unmilled solid waste
is published as Appendix B.
An environmental protection publication (SW-62d.l) in the
solid waste management series
II
-------
133110
TABLE OF CONTENTS
Page
INTRODUCTION X
ACKNOWLEDGEMENTS Xl' 1
DEFINITIONS xi 1"f
DEVELOPMENT OF MILLING IN EUROPE 1
MADISON'S MILLED REFUSE OPERATION 3
Historical 3
Original Plant 4
Description of Gondard Equipment 4
Operational Aspects of the Gondard System 9
Gondard Production Rates 11
Plant Expansion 14
Tollemache System 14
Problems with the Tollemache Equipment 25
Tollemache Operation and Production 27
Tollemache Power Consumption 29
Two-mill, Two Shift Operation 29
Mill Accessories 37
Maintenance Programs 37
LANDFILLING MILLED REFUSE 40
CHARACTERISTICS OF MILLED REFUSE 44
Blowing and Particle Size 46
Density 48
Settlement 61
Decomposition 61
Leachate 63
Biotron and Gas Composition Studies 84
iii
-------
Table of Contents (continued)
Vectors
Vegetation
Fires
Odors and Esthetics
USE OF COVER
THE ECONOMICS OF MILLING
Landfilling
Milling Costs
Cost Projections
TRENDS AND DEVELOPMENTS
CONCLUSIONS
APPENDIX A
APPENDIX B
LIST OF TABLES
TABLE 1 Gondard Overall Production Rate Vs. Grate Size-Tons
Per Hour (operating Plus Down Time)
TABLE 2 Gondard Operating Production Rate Vs. Grate Size-Tons
Per Hour (Operating Time Only)
TABLE 3 Operating and Overall Production Rates for Tollemache
Mill
TABLE 4 Operating and Overall Production
TABLE 5 Comparison of Power Data for Tollemache Mill Alone
(1970-71)
TABLE 6 Hourly Breakdown of Average Day (Two-Mill, Two-Shift
Operation)
Page
91
102
105
108
109
110
110
113
128
131
134
137
161
12
13
30
31
32
33
iv
-------
List of Tables (continued)
TABLE 7 Operational and Overall Production Rates (1972)
TABLE 8 Power Consumption - Mills Combined (1972)
TABLE 9 Summary of Test Cells - Olin Avenue Site
TABLE 10 Composition of Madison's Solid Waste (November 1968)
TABLE 11 Olin Avenue Field Density Tests (1967-68)
Results of Truax Field Density Tests (1972)
TABLE 12
TABLE 13
Percent Increase or Decrease of Densities Obtained In
Truax Field Density Tests (1972)
TABLE 14 Water Budget (May 1971 - May 1972)
TABLE 15 Results of Fly Cage Tests (1969)
TABLE 16 Effect of Planting Conditions on Tree Growth (Percent)
TABLE 17 Cost Incurred in Landfilling Milled and Unprocessed
Refuse
TABLE 18 Landfill Costs Using Milled Uncovered Refuse
TABLE 19 Unadjusted Cost Data for Gondard Mill
TABLE 20 Data Used for Computing Amortization of Original
Gondard Installation
TABLE 21 Tollemache Milling Costs
TABLE 22 Stationary Packer and Haul Costs During Tollemache
Evaluation
TABLE 23 Amortization Data for Tollemache Milling System
TABLE 24 Amortization Data for Stationary Compactor and Final
Transportation System
TABLE 25 Milling Costs for Two-Mill, Two-Shift Operation
TABLE 26 Labor Costs for Two-Mill, Two-Shift Operation
TABLE 27 Breakdown of Labor Costs For Two-Mill, Two-Shift
Operation
TABLE 28 Power Costs, Mills For Two-Mill, Two-Shift Operation
TABLE 29 Power Costs, Mill Accessories For Two-Mill, Two-Shift
Operation
Page
35
36
43
50
55
59
60
83
100
103
111
112
114
116
117
119
120
120
121
123
123
124
124
-------
List of Tables (continued) Page
TABLE 30 Lighting Costs For Two-Mill, Two-Shift Operation 125
TABLE 31 Plant Heating Costs For Two-Mill, Two-Shift Operation 125
TABLE 32 Supply Costs, Hammer Maintenance For Two-Mill, Two-
Shift Operation 127
TABLE 33 Stationary Compaction and Haul Costs For Two-Mill,
Two-Shift Operation 127
TABLE 34 Production Estimates 129
TABLE 35 Annual Average Costs Per Ton for Milling, Hauling,
and Landfilling Refuse 130
LIST OF FIGURES
Figure 1 Installation of four Gondard mills at Weisbaden,
Germany, 1966. 2
Figure 2 Refuse Reduction Plant at Madison, Wisconsin, 1967. 5
Figure 3 Gondard mill at Madison, Wisconsin. Housing opened to
show horizontal rotor and hammers 6
Figure 4 Side view of the Gondard mill and rejection tower. 7
Figure 5 Original layout of Madison's Refuse Reduction Plant;
packer truck being loaded for trip to landfill. 8
Figure 6 Gondard Milling System; feed bin in foreground and haul-
away load lugger in background 10
Figure 7 Floor plan of the Madison Refuse Reduction Plant after
completion of expansion in 1971. 15
Figure 8 Vertical-shaft Tollemache mill at Madison, Wisconsin. 16
Figure 9 Cross section of the Tollemache mill. 18
Figure 10 Overhead and Side views for feed and take-away conveyors
of Tollemache System. 19
Figure 11 Tollemache hammers; new (either side), and worn, after
four days use. 20
Figure 12 Tollemache continuous-metal-belt elevating feed conveyor. 21
Figure 13 Tollemache feed conveyor showing tumbling action of
refuse.
22
vi
-------
List of Figures (continued)
Frigure 14 Transfer Packer ,.Sj«sat;iQB, in ToUwaohe System
figure 15 , Stationary, Qjqmpacfcetr w^tfe 35,cu,. yd. storage
Figure 26
Figure 27
Figure 28
Figure 16, Tolleraache hammer pattern in use at Madison at end
of evaluation period (June 1972) .
Figure 17 Maintenance shop for welding and repairs at Madison1 s
milling plant.
Figure 18 Cell locations, Olin Avenue test site.
Figure 19 Test cells of milled and unprocessed refuse at Olin
Avenue Landfill
Figure 20 Tracked Dozer experiencing problems operating on
silty sand cover material during wet weather.
Figure 21 Fully loaded transfer trailer traveling over an
8-foot lift of milled refuse.
Figure 22 Wind-blown film plastic from milled refuse landfill.
Figure 23 Typical particle size distributions for milled refuse
from both mills in 1972.
Figure 24 Compression machine with air hammer in place, used to
compact milled and unprocessed waste during labora-
tory investigations.
Figure 25 Compressibility Tests on Refuse.
One of three 2000-cu-yd. cells constructed at Truax
Landfill to conduct field density tests for milled
and unprocessed wastes.
Pit filled with milled refuse for density tests.
Page
23
24
28
39
41
42
45
46
47
49
51
53
56
57
Pushing milled refuse with a steel-wheeled compactor. 58
Figure 29 Leachate produced by combined refuse in certain
Olin Avenue test cells. 65
Figure 30 Chemical oxygen demand of leachate produced by
combined refuse in certain Olin Avenue Test Cells. 66
Figure 31 Leachate produced by garbage and rubbish in certain
OJ.in Avenue Test Cells. 67
vii
-------
List of Figures (continued) Page
Figure 32 Chemical oxygen demand of leachate produced by
garbage and rubbish in special Olin Avenue
Test Cells. 69
Figure 33 Cross section of lysimeter beds. 71
Figure 34 Lysimeter bed prior to receiving refuse. 72
Figure 35 Diagram of lysimeter bed layout. 73
Figure 36 Lysimeter bed being filled with unprocessed refuse. 74
Figure 37 Lysimeter bed being filled with milled refuse. 75
Figure 38 Cow tanks and barrel arrangement for collecting runoff
from test cells. 76
Figure 39 Leachate production from lysimeters. 77
Figure 40 COD concentration curves for lysimeters. 79
Figure 41 COD production from lysimeters. 80
Figure 42 pH curves for lysimeters. 82
Figure 43 Container of milled refuse used in the Biotron Studies. 85
Figure 44 Cumulative volume of leachate and cumulative
"rainfall" with time.
Figure 48
86
Cumulative COD and TDS vs. cumulative leachate
from Biotron cells.
87
Figure 45
Figure 46 Gas composition data collected by lower probes during
lysimeter studies conducted at the Oscar Mayer site. 90
Figure 47 Bait station used during rat study phase of vector
investigation at Olin Avenue Site. 92
Rodent Test Center for Norway rats at Purdue University,
Lafayette, Indiana.
Figure 49 Typical cage used in studies conducted at Purdue
University.
Figure 50 Scudder grille used to determine numbers of flies
on milled and unprocessed refuse.
Figure 51 Screened cages used during fly survivability tests,
1969.
95
95
98
99
viii
-------
List of Tables (continued) Pu^o
Figure 52 Heavy vegetation growing on a test cell of milled refuse. 106
Figure 53 Fire deliberately started on fresh milled refuse caused
smoldering, but remained on the surface and was easily
extinguished. 107
-------
INTRODUCTION
This report covers the highlights of almost seven years of work at
'ladison, Wisconsin, carried on under a demonstration grant from the
Department of Health, Education and Welfare and the Environmental
Protection Agency. This report is one of two volumes. Volume lisa
summary of the major findings and is intended for those who wish
to gain general information on the project. Volume II contains conden-
sations of data gathered during the project and has been arranged in
sub-reports intended for the researcher who may wish to review the data.
Volume II will also be made available through the National Technical
Information Service (NTIS), Washington, D.C.
The project began as a practical demonstration to investigate
the concept of milling refuse for landfill disposal without daily cover
in what is hereafter referred to as a rnillfill. The project was intended
to gather data on the operation and cost of milling equipment, the use
of milled refuse in landfill, and the characteristics of milled refuse -
all from a practical standpoint. Project personnel ended up doinq
detailed investigation work, on both milled and unprocessed refuse,
which was far removed from the idea of a demonstration project. In order
to show that landfilling milled solid waste without daily cover lives
up to the praise offered by European landfill operators, it was necessary
to design experiments involving rats, flies, leachate, gas, trees, etc.,
which hopefully would quantify such_factors. While some crude techniques
for making measurements on these parameters existed in the solid waste
field, only a few suggestions could be gleaned from the literature or
from similar projects. It was therefore necessary to pioneer in such
areas as to how to measure particle sizes of the milled product and how
to compare the occurrence of flies on a sanitary landfill face and the
surrounding area with their occurrence on a millfill.
The project investigators used not only the numerical evaluations
but also seven years of field experience and observations in drawing the
conclusions presented in this report. It should also be kept in mind
that there is a difference between the idealized, environmental aspects
of a covered unprocessed landfill and what occurred in the field during
years of evaluation. Thus, comparisons betweeen milled and unprocessed
refuse were based on actual day-to-day field evaluations and not on
theoretical considerations.
It v/as interesting to note, during a tour in 1970 of European
milling facilities by the principal investigators in this project, that
the Europeans involved in solid waste management have evolved into
incineration schools, composting schools, landfilling schools, etc.
They seem to be even more opinionated in their approach to processing and
disposal than their American counterparts. But among American solid
waste experts there is also appearing a "school" mentality involving
such processes as baling, sanitary landfill ing, milling, composting and
incineration. In reading this report one should keep an open mind as to
v/here milling of solid waste really fits into the scheme of solid waste
management. Solid waste management operations include storage, collection,
-------
transportation, processing and disposal. Milling is a method of
processing, not the method of processing or a_ method of disposal.
The important thing to consider is not, "What is wrong with milling?"
but "Does milling fit into a particular community's or private
operator's solid waste management system?" Unfortunately, a controversy
rages about milling and running tnillfill sites. As in such controversies
one side is likely to present milling as a magical solution to all
solid v/aste disposal problems and the other to present it as an
expensive, unreliable, impractical approach to a small portion of the
problem. The true value of milling, of course, lies somewhere in the
middle.
There have been many changes in solid v/aste management since this
project was conceived in 1966. Regulations in the areas of air
pollution, water pollution and land disposal of waste have had their
impact on the field. Emphasis in disposal has greatly shifted from
the public health and economic aspects of the problem to the esthetic
aspects. With this change in philosophy, the traditional engineering
economics approach to the solid waste disposal has been altered
somewhat. In the past, a method was chosen for a disposal facility
design based on the acceptable health standards with little regard to
how the process affected the sensibilities of the general public.
Screening of disposal sites, odor abatement control, improved
architectural design of buildings, etc. were usually not included in
the design because they did not turn out to be the most economical
solutions. It has now become necessary to evaluate the esthetic
solution to the disposal problems as well. It is hoped that the
following report will shed some light on how milling can fit into
a locality's solid waste management system from the standpoint of
economics, health,safety, and esthetics.
While the project did not provide black and white answers to
all the questions, it certainly illuminated some of the dusty dark
areas in solid waste management and demonstrated that milling can be
a valuable aid for land disposal, incineration, and some of the future
schemes for resource recovery.
xi
-------
ACKNOWLEDGMENTS
The persons who initiated the refuse reduction project in Madison
included Arnold Meyer, Vice President of the Heil Company; Gerard Rohlich,
Professor of Civil Engineering, University of Wisconsin-Madison; Edwin
Duszynski, Director of Public Works, City of Madison; John Thompson,
former City Engineer, City of Madison; and James Brophy, Superintendent
of Streets, City of Madison.
Overall responsibility for the project has been in the hands of
John J. Reinhardt, Principal Civil Engineer, City of Madison. Robert K.
Ham, Associate Professor of Civil and Environmental Engineering,
University of l.'isconsin-Madison, was in charge of the University
efforts in the project. Evaluations were carried out under direct
supervision of Warren K. Porter and Gerald W. Sevick, Project
Specialists.
Major investigations were conducted by Charles R. Anderson,
Larry Mendrickson, James R. Boyle, Rameschandra Gawalpanchi,
Everett H. Clodfelter, Fred Courtsal, Gier Widgel, Vincent Geier,
and Walter L. Gojmerac.
Assisting in the project have been Gary Doley, Robert Karnauskas,
Kenneth Drunner, Herbert Hanneman, David Potwin, Richard Presney,
Ellsworth Fisher, Earl Ulsrud, Charles Maas, Raymond Dillabough,
Warren Kiraberley, Richard Steinhofer, William Martin, Mary Smits,
Eugene Davenport, and Al Kelly.
Student assistants have included Wendy Arndt, Douglas Lindquist,
John Regal ski, Hilliam iierry, Robert Schmiedlin, and Daniel !'. Benjamin.
Production of the final report for this project has been the
responsibility of John Wolf and Austin Henry.
Special recognition is given to Doris Habich for typing and
retyping the project's numerous reports over the past six years.
Tilts demonstration project fr*SL fieen supported by the U.S. Enviromental
Protection Agency in cooperation with the City of Madison, Wisconsin.
Two of their project officers, David Arella and Roger D. Graham, have
provided valuable assistance.
xii
-------
DEFINITIONS
Several terms used throughout this volume require short defini-
tions. Most of the definitions are adapted from APWA Municipal Refuse
Disposal, Chicago, (1970), and the EPA Publication Solid Waste Manage-
ment Glossary (SW-108ts). 1972.
T)Refuse (also called solid waste). Useless, unwanted, or dis-
carded material. This report is concerned primarily with residential
(also called domestic) and light commercial refuse -- that is, all
solid waste that normally originates in a residential environment.
Industrial, institutional, and large quantities of commercial refuse
are generally collected by private firms in Madison and are not-normally
milled. Large tree cuttings, bulky items, and construction-demolition
debris are also handled separately at Madison.
2) Garbage (also called food waste). Animal and venetable waste
resulting from handlinn, storage, sale, preparation, cooking, and
serving of foods.
3) Rubbish. A general term for solid waste — excluding garbane
and ashes — taken from residences, commercial establishments, and
institutions.
4} Milled Refuse, "efuse that has been mechanicalIv nrnund,
shredded or pulverized.
5) Unprocessed Refuse. To avoid confusion, this term will be
applied uniformly to unmilled, crude, or raw refuse.
6) Cell. A volume of compacted solid waste which may be enclosed
by natural soil or cover material in a sanitary landfill, not covered
as in the case of test volumes of milled refuse, or enclosed in con-
crete beds in other test situations.
7) Sanitary Landfill. A method of disposing of solid waste on
land by utilizing sound engineering and planning principles, by
spreading the waste in layers, compacting it to the least practical
volume, and covering it with soil at the end of each working day.
8) Bulky Items. Madison routinely collects large, bulky items
such as furniture, tires and mlttresses separately. Such material is
not milled but landfilled separately.
9) Rejects. These are items which the mills ballistically
separate because they cannot be pulverized to the desired particle
size.
10) Cover. Soil which is placed over material in a sanitary
landfill at specified intervals, usually daily.
11) Vector. An organism that is capable of transmitting a pathogen.
12) Leachate. Liquid that has percolated through solid waste and
has extracted, suspended, or dissolved material from it.
13) Actual Refuse Density. The figure derived by dividing the
weight of a portion of refuse by the volume of that refuse.
xiii
-------
14) Effective Refuse Density. The figure derived by dividing
the weight of a load of refuse by the volume of landfill space it
takes up (including any cover).
15) Operational Production Rate. Tons processed during the
time in which a mill is actually running. Down time is omitted From
this figure; thus, it is a measure of a mill's efficiency.
16) Down Time. In evaluations of the mills separately, tine when
a mill was shut down due to mechanical problems. In evaluations of the
two-mill operation, all times during which the mills v/ere not grindinq
refuse, between the initial start-up of the mill to the shut down at
completion of the production day (e.g., any time the mill is not run-
ning during normal working hours).
17) Overall Production Rate. Tons of refuse processed divided
by time during which the mill was grinding, plus down tine. Thus, it
is a measure of the plant's efficiency.
18) flillfilling. An engineering method of disposing of solid
v/aste on land without daily cover by milling the waste, spreading it
in layers and compacting it to the smallest volume practical.
xiv
-------
DEVELOPMENT OF MILLING PI EUROPE
Milling was begun in Furope to provide material suitable for composting.
The original, single- or double-rotor harrier mills were designed to handle
homogeneous materials; thus unqrindable items had to be manually sorted and
removed.
A French manufacturer of grain-grinding machines made the crucial
observations that led to the capability of milling heterogeneous material
such as refuse. He v/as trying to learn why his grain mill was unable to
grind high-moisture corn, and during an investigation of this problem he had
a portion of the cover on the machine cut out. He observed that dry grain
particles were readily ground. The softer, spongier, monster grains, however,
remained ungrounJ and were accelerated through the opening in the top of the
machine. The manufacturer of grain grinders thus learned that nongrindablo
objects can be ballistically separated from grindable objects in a mill.
The ballistic rejection feature of the mill was awarded U.S. patent
3,082,963 on 26 March iyr,3.
The idea of ballistically separating nonmillable refuse through a
"chimney" above a hammomi 11 was developed into a test machine which was
installed in Meaux, France in the early 1950's. Since there was no
composting plant in the vicinity, the milled refuse was simply deposited
on land near the machine. Amazingly, the milled refuse did not create a
nuisance.
The original mill at Meaux was fed by hand at the inlet opening.
It soon became apparent that a mechanical feeding machine would be neede-J
to maximize the machine's output. Development of mechanical feeding
machinery took another two years, however.
By 1970 there were over 50 installations employing this type of feeding
and milling equipment in Western Europe (Figure 1). Almost all landfilled
the milled refuse without daily cover.
By the mid-1960's, the first Gondard plant was in operation in North
America — at Montreal, Canada. In this plant, two Gondard mills are used in
a parallel arrangement'for refuse milling by a commercial salvaging company.
About 25 percent of the refuse brought to the mill during its first years of
operation was salvaged and sold. The major item salvaged was paper. The
remainder of the milled refuse was dumped along with unprocessed refuse in an
uncontrolled fashion; thus it was difficult to draw conclusions about many
aspects of landfill ing milled refuse from the Montreal operation.
European experience, however, has indicated that handling milled refuse
at a landfill site is simple and convenient. It is merely dumped and graded
as desired. Since milled refuse is relatively uniform in size and composition,
compaction and settlement are even. The material was reported to provide
excellent support for rubber-tired vehicles, and since glass objects are
reduced to tiny pieces, damage to tires of landfill machinery is greatly
reduced. The nuisance caused by blowing paper and dust produced by packing
machinery is greatly reduced. The danger of fire in the landfill is greatly
lessened. Many vectors, especially rats and flies, are reduced in number or
eliminated from the landfill. Milled refuse is esthetically more pleasing
than unprocessed refuse and therefore is more readily accepted by the public.
Finally, according to European claims, milled refuse that has been landfilled
does not require daily covering.
-------
01
>>
(O
OJ
M
C
to
03
r—
to
to
CD
s_
3
en
-------
MADISON'S MILLED REFUSE OPERATION
Historical
The Heil Company of Milwaukee, Wisconsin mem facturers and marketers
of truck bodies for solid waste collection, learned of i.he Gondard milling
process through sales activities in France, Svn'tze':: v?s Netherlands, and
Germany. Company representatives were impressed c; tie European claims
i out to find a
involving refuse
of the attractive features of milled refuse and thai
city in the United States to undertake a test ,,ro,l3ct
milling.
An incentive to potential users of the new mill ,
Solid Wastes Act of 1965 which provided matching "und
would demonstrate on a production scale new method c
waste. Efforts to interest a community in Milvaukat
new system proved unsuccessful when it appeared v.&t
interfere with development of a county-wide irc-in^-U!'OM system.
Continuing their search, the Heil represanta. iv>, >.ontacted Professor
Gerard Rohlich of the University of Wisconsin-Kadiso.. College of Engineering.
j system was the
for projects which
nawaging solid
•ianty in trying the
project would
igation and encouraged
* Superintendent of
>s and to view the
place in February
Professor Rohlich felt that the system warrante
Madison's Director of Public Works, City Engine
Streets and Sanitation to investigate the po«sii
Montreal operation.
An inspection trip to the Montreal facility
1966, following a prolonged thaw. The haul roads vxo the landfill site
were almost impassable to automobiles. However, the traffic-bearing
quality of the uncovered milled refuse in the Icu.dfvn was remarkable.
The visitors were able to walk on the refuse suvft,cn without overshoes,
and loaded trucks sank only three or four inches Into che material.
The ease of operation at the Montreal landfi'-l even under adverse
conditions, convinced the Madison officials to f^.uim^xl to the Madison
Common Council that they be permitted to apply I'D;' a jrsnt to fund a
demonstration project. The proposed project woi:1U
claims that milled refuse can be landfilled witiicnt
as determine what modifications would be necessary
to refuse generated in this country. The cost or 31
also be carefully documented, and the character's sric
would be studied.
In April 1966, the Common Council approved tie s»;bmittal of a pro-
posal to the Department of Health, Education aad we! fa, e and designated
the Olin Avenue landfill site as the location of n proposed plant.
-rwestigate European
^ly cover as well
adapt, the system
T a system would
of milled refuse
year grant,
t was renewed and
cl pants in the
The City of Madison received a three-and-oue --
number 5-DOI-UI-00004, from HEW in June 1966, The
expanded several times to cover additional work,
demonstration project included:
1) The City of Madison, which provided tha p":ant site, site
improvements, operating personnel, partial matching funds ., and, of
course, the refuse.
2) The Heil Company, which furnished, Insts'Ped, and modified
the equipment and provided partial matching funos to" the original
equipment and for some phases of the evaluation,, Under terms of a
purchase option contract, Madison bought the equipment in 1969 after
it proved successful .
3) The University of Wisconsin-Madison* which collected data
and evaluated the project.
-------
Original Plant
The 01 in Avenue site for the Madison Refuse Reduction Plant is a
GO-acre marshy area which had been operated as an open-burning dunp from
1942 to 19CO. It was operated as a landfill after that tine/ Cells for
evaluation of the milled refuse project had to be constructed above ground
using the so-called area method because the entire 01 in Avenue site had been
levelled with one ten-foot lift of covered refuse prior to the mill inn
project. The water table ranges from 3 to 10 feet below the surface,
depending on the season and location in the landfill.
A minimum floor area for the plant was considered necessary because of
poor subsoil conditions at the site. Soil borings indicated the need for
surcharging the site to compress deep-lying organic silt. This process
delayec the placement of foundations until December T966. The building was
completed a few months later. The original plant was a 64x104 ft. enameled
steel panel building with a concrete floor for refuse storage. A scale was
also installed for weinn inn incoming packer trucks (Figure 2).
The Gondard machinery was delivered completely dismantled, but despite
unfamiliar plans for its assembly, the mill was erected and the plant wired
in only two months. Tno nil! \;as operated for the first tinm on 14 June I'J.'-V
and shakedown runs were made throughout the summer.
Madison began using the mill in September 1967. At the same time, the
city instituted weekly collection of combine"! refuse on the west s.ide of town
to replace separate collections of garbage arid rubbish. Such combined
collection was in accor.; 'n't,h ''adi son's lona-term desires an-! was in line
v.'itn the needs of the demonstration project.
Description of Gondard Equipment
The Gondard hammerm'll installed at the Olin Avenue site in 1967 was
equipped with forty-eight 1-3/16x4x11 in. swinging hammers on four shafts.
The 15-lb. hammers are driven at 1150 rpm by a 150 hp motor (Figure 3).
The patented chimney was included to permit ballistic rejectirn of largo,
unnillablc objects. Ballistic rejection occurs when such objects are strjck
by the liarners sufficiently hard that they pass the ?7-foot length of tho
chute, strike a doflection clato, and cone down a separate chute. Tills
feature allows unqrindabl-" objects to bypass the grates through which nilled
material leaves t'ie mill (Figure 4).
As noted, the reduction plant was designed to minimize floor space
This desiqn led to complex system of three feeding conveyors (Figure 7>).
Tho motors for the three conveyors are controlled by a special unit that
allows the operator to vary the speed of the two rubber-belt conveyors nr
the conveyor in the 7:J-cu. v^. feed bin. The mill also includes a system to
stop all Inren conveyors when nower input to the mill motor exceeds 12r>
percent of rated load for £> consecutive seconds.
The original discharge conveyor was a 3-foot-wide rubber belted
conveyor suspended from a trolley which permitted it to swinn From one to
the other of two I'-cu. >•'•!. portable load lugqer bins.
The Gondard mill and conveyors were imported from France on the basis
of satisfactory performance in turope. Almost immediately, however, it
became apparent th
-------
-------
o
.c:
OJ
c
o>
Q.
o
cr.
c
to
C
o
•1- i-
3 QJ
- E
C fO
"O C
ft3
S-
O
o
-o c
s- o
(0 N
-O T-
c s_
o o
O -c:
CO
OJ
CD
-------
s_
a)
o
c
o
o
O)
cu
-o
V-
(O
-o
c
o
0)
O)
O)
•o
0)
s-
3
CT>
J
-------
(TJ
QL
CO
CL.
c
O
u
3 •
-a r—
GJ t—
a: ••-
c
to
Q.
C T-
O S-
-o $-
(O O
O OJ
-o
3 O
O i—
•
LO
CJ
-------
modification of the Gondard system after installation at Madison. Some of
these modifications included:
1) Addition of 6-in. cleats to supplement the 3-in. cleats in the
feed bin conveyor for hand!inn the large items in Madison's refuse.
2) Installation of spray bars around the area where refuse leaves the
bin conveyor and at the entrance to the mill to help reduce dust problems.
3) Redesign of several components of the feed mechanism to prevent
jams and improve uniformity of feeding.
4) Revision of the servo motor on the bin feeder to allow the operator
to change conveyor speeds rapidly.
These modifications and other improvements in plant operation reduced
down time due to conveyor and mill jams, to well below the 13 minute per day
average recorded between 1 April and 29 November 1968 with the original
equipment.
Operational Aspects of the Gondard System
In the Gondard system, incoming refuse is first weighed and then
enptied in the 75-cu. yd. storage bin or on the floor when the feed bin is
full (Figure 6). A front-end loader is used to push refuse from the floor
into the bin. The combination of refuse storage in a bin and on the floor was
chosen to eliminate the need for an overhead crane and operator, to minimize
handling of the refuse, and to make it easier to clean up the dumping area
after each day's operation.
At the bottom of the storage bin is a conveyor with metal cleats that
move refuse out one end of the bin and onto a rubber belt at right angles
to the bin conveyor. This second conveyor, in turn, dumps the refuse onto a
third belt, again at right angles to the second conveyor. This rubber-belt
conveyor lifts the refuse and drops it into the hammermill through the side of
the chimney. A variable spef?d drive allows the flow of refuse to be controlled
by increasing or decreasing the speed of the conveyors.
In the mill, the refuse is either pulverized to sizes small enough to
pass through a grate or rejected through the chimney.
The original method for transporting the refuse from the mill to the
landfill was'a Load Lugger detachable container system consisting of a
flatbed truck with hydraulic arms to lift a 10-cu. yd. bin onto the bed.
Three bins were provided initially. One bin was placed under the reject
chute and could hold the items rejected during one or two days of milling.
The other two containers were located at the end of the conveyor bringing
milled refuse from the mill. The conveyor v/as mounted on a track so that it
could be switched from one container to the other. When one container was
filled, a man stationed at the truck, levelled the load, switched the
conveyor to the other container, took the full load to the landfill, and
finally replaced the empty container in its original location. By the time
he returned, the second container was usually almost full. This man was thus
continually occupied in emptying containers. Because of the fluffy nature
of the freshly milled material, these 10-cu. yd. containers held only
2,500 to 3,000 Ibs.
-------
10
-------
A change to conventional refuse packer trucks to transport milled
refuse was made in January 1968, when two old packers were pressed into .
emergency service. Since the trucks needed frequent repair and since a
person was needed to actuate the packing mechanism every few minutes, these
trucks were retired in November 1968.
Also during 1968, two models of 20-cu. yd. packers were tested.
Neither was able to keep pace with the volume of refuse discharged from the
mill.
The final collection and transportation system used during the
tests involved two prototypes of a 25-cu. yd. packer truck planned for
production by the Heil Company. These trucks had a packing mechanism that
coiJld bo continuously cycled by mechanical means. This system reduced both
manpower needs and the number of trips to the landfill.
Gondard Production Rates
During the year of experimental trials with the Gondard system, the
grate at the bottom of the mill, through which milled refuse passes, was
changed systematically to determine optimum grate size. Grate size is the
space between bars of the grate. Considerations in optimization included
machine capacity, operating costs, landfill space required, and particle size,
Initially, plans called for using 2-t 3 I/?-, 4-, 5-, and 6 1/4-in. grates;
however, tests with the 2-in» grate were discontinued almost immediately
because it yielded particles finer than required for landfill operating and
was slowing projection.
Tables 1 and 2 show production rates at various seasons and with
different grate sizes. Two production rates were calculated: one based on
machine milling tine alone and the other based on machine milling time plus
down time. Thus, "operating" production rate is the tonnage processed
during the time the machine is actually milling, while "overall" production
rate is the tonnage processed during mill milling tine plus down tiie. 'lot
included in the overall rate is time lost because the plant was out of refuse
or because of the time lapse from arrival of the first load to the start of
milling operations. These factors were not included because they Jo not
reflect machinery limitations in the plant itself. Time lost because of
these nonmachine problems will be taken into account and discussed further
in the section on costs.
The optimum grate size for use in Madison was found to be 5-in. This
yields the'highest "operating" production -- y tons per hour at lowest
cost without producinn blowing paper problems in the landfill. The
"operating" production rates for the 3 1/2- and 6 1/4-in. grates were P..6 and
9.4 tons per hour, respectively, during the tests.
Items rejected from the mill were collected and weighed from
September 1957 through January 1968. During this time it was found that
between 1 and 7 percent, by weight, of the total refuse was ballistically
separated when the reject chute extended 27 feet vertically above the mill.
11
-------
TABLE 2
GONDARD OPERATING PRODUCTION RATE VS. GRATE SIZE - TONS PER HOUR
(Operating Time Only)
Period
Fall 1967
Winter 1967-68
Spring 1968
Summer 1968
Fall 1968
Winter 1968-69
Spring 1969
Average - last full year
Projected Average - based
on improvements in support
equipment
Grate Size
3 1/2
Inches
6.8
7.0
7.1
8.0*
-
8.4*
9.7*
8.7
8.8
(opening
5
Inches
-
-
8.2
8.2
7.9
7.2
-
7.7
9.3
between bars)
6 1/4
Inches
9.3
7.3
8.1
8.4
-
-
-
8.1
9.7
*The opening between the bars was actually four inches.
13
-------
Plant Expansion
The oriainal Gondard project was largely an effort to qain experience.
By late 1968 it appeared that milling refuse aid offer enough advantages for
Madison to warrant enlargement of the project. Also at this time the Heil
Company became interested in evaluating the English-manufactured Tollemache
hammermill. In addition, the City of Madison was interested in revising the
existing facility, in cooperation with the Heil Company, utilizing the
experience gained during the Gondard tests. Under a two-year renewal grant
(2-GOG-EC-00004-04A1) from HEW, Madison, The University of Wisconsin-Madison,
and the Heil Company expanded their objectives to include:
1) Installation and evaluation of the Tollemache mill;
2} Installation and evaluation of a new feed system;
3) Installation and evaluation of a stationary packer with
75-cu. yd., self-unloading transfer trailers;
4) Expansion of the refuse reduction plant to permit
operating two shifts;
5) Evaluation of operating the Gondard and Tollemache
mills simultaneously for two shifts.
Late in 1970 and early in 1971, a 6,500 sq. ft. addition was made to
the original milling plant to accommodate the expanded objectives of the nrojec;
(Figure 7). Of this addition, 600 square feet were devoted to shops, offices,
control room, ami mt.-etinn room. Another 1,500 square feet were assigned to
testing various components of an air-classification, wood-fiber recovery
process being developed by the United States Department of Agriculture
Forest Products Laboratory at Madison. The remaining 4,300 square feet were
devoted to floor storage of incoming refuse and for maneuvering space for
the packer trucks. This increased floor space allows storage of 310 tons
of unprocessed refuse versus 85 tons for the original building. Cost of the
expansion of the facilities was $98,000, with another $32,000 going toward the
research area for experimentation in paper recovery.
The new control room, located on the second floor, is the heart of
the plant operation. It contains all systems controls and gives the operator a
clear view of the dump floor, both mills, their respective feed systems, and
the first section of the takeaway conveyor system. A closed circuit television
set-up allows the operator to see the remainder of the takeaway conveyor as
well as the stationary compactor. Below the control room is the scale room
which was not modified during expansion.
With completion of the expansion, a new, larger front-end loader was
purchased to handle the increased volume of unprocessed refuse. Also, a
Tennant floor sweeper was bought to reduce man hours devoted to cleanup.
Tollemache System
The Tollemache mill installed at the Madison milling plant in 1969 is
a vertical-shaft, ballistic-rejection hammermill (Figure 8). In contrast,
the Gondard mill has a horizontal main shaft. The Tollemache mill has funnoi
shaped outside walls and a rotor which has the shane of a conical surface wnen
spinning. The rotor is driven at 1350 rpm by a 200 hp squirrel cane -rioter now
by a 3-phase, 440-v source.
14
-------
z
o
i £
53 s «
z£ z
~s s°
9 *"^*
d —
~ K>
8
i*
i1
K
i s ^
u.
"v IIIIIH 1111
3 ' ^
11 1
^ 1
i—
a:
c; -~
o
Wl SI,
•i- o
-O •!-
rrj i/>
?c c
«
OJ Q-
-c: x
4-> OJ
O O
C C
ro O
cu
o "Q-
o s
f— O
u_ o
01
i-
15
-------
figure 8. Vertical-shaft Tollemache mill at Madison, Wisconsin.
16
-------
The hammers are connected to the rotor in three distinct layers.
(Figure 9). The hammers in the top layer are mounted to the shortest radius;
the diameter to the hammer tips is 33 inches. This layer is the pre-break
section of the machine. Here incoming refuse is sufficiently ground to
reduce the load to the mill motor. The pre-break section is also the site
of ballistic rejection of some potentially damaging and unmillable material.
From the pre-break section, the refuse falls into the constricted neck
of the mill, which has a diameter of 41 inches. The hammer tip diameter in
this layer is 38 inches. In this section, therefore, there are only 1 1/2
inches of clearance; thus any material that is ungrindable or that has not
been sufficiently reduced in size to pass this restriction will be spun around
the funnel and will exit through the reject chute opposite the feed opening.
The material then enters the grind section, where the hammers have a
43 inch tip diameter. It is here that most of the work is done to produce
the desired particle size. The ground material is discharged centrifuqally
through an opening at the bottom of the machine (Fiaure 10).
Two lavers of metal form the housing t>f the mill. The outer layer
is the "shell" of the unit. The inner layer is a removable protective
lining, upon which are mounted breaker bars. It is the action between the
rapidly moving hanmers and the stationary bars that produces the grinding.
The hammers are 10x4x1-3/T& Inches and weigh 15 IDS. (Figure 11)
Originally 54 hammers were used, but this number has been reduced to 34.
Reasons for this change will be discussed later.
The Tollemache mill is fed by a metal flight conveyor system shown
in Figure 12.The one piece, 45 inch wide conveyor is driven by a 20 hp motor
equipped with a speed reducing mechanism and overload switch. The 18 foot
long horizontal portion of the conveyor is located in the Gondard storage bin
and over approximately one-half the length of the portion of the Gondard
feed conveyor lies within the bin. The inclined portion of the feed
conveyor is about 30 feet lonq and makes an angle of about 45° with the hori-
zontal. The variable drive mechanism allows the operator to adjust the conveyor
speed from 8 feet/minute to 12 feet/minute. An automatic overload stops the
feed conveyor when the mill motor draws 100 percent of its rated capacity.
The conveyor restarts automatically after the mill motor draws less than
75 percent of its rated capacity for 5 consecutive seconds.
The conveyor pit is 5 feet deep. At the point where the conveyor
leaves the horizontal, the walls abruptly pinch down in width; thus, a
constriction is placed on refuse flow. This constriction halts the forward
movement of the refuse mass and creates a tumbling or rolling action which,
in theory, should produce an even feed to the mill (Figure 13).
Soon after installation of the Tollemache mill, an .upgraded method for
collecting and hauling the milled refuse to the landfill became a
necessity. To meet the increased volume produced by the Tollemache mill,
a Heil stationary compactor system was adapted. The stationary compactor
is used to compress the ground refuse into a 75-cu. yd. transfer trailer.
A 35-cu. yd. storage hopper was installed above the compactor to allow the
mill to operate while trailers are being switched (Figure 14,ttem:6 and
Figure 15). The storage hopper is outfitted with an electric eye that auto-
matically turns off the feed conveyor to the mill if the bin becomes full. This
system eliminates both spillage from the bin and backups on the final transfer
conveyor. '>.••
17
-------
-C
o
o
I—
O)
o
c
o
u
a;
o
o
en
en
18
-------
00
o
10
I
o
00
j_
o
(I)
c
o
o
TO
I
Ol
(13
•a
o>
XJ
c
ns
QJ
j=
S-
O)
o
o
O)
CT.
19
-------
20
-------
ri gyre 12. Tollemcfce continuous-metal-belt elevating fettf
21
-------
13. Tollemache feed conveyor showing cum&'nng acil. \ > f rv
-------
0)
.en
u
o
c
c
o
"r*
4->
2
m
i.
cu
o
f-
CO
CJ
3
cn
23
-------
s-
OJ
Q.
fa-
fO
-
o
u
IT5
4-3
I/O
24
-------
Using the stationary compactor, 1t takes 70 to 80 minutes to fill one
trailer with 16 to 20 tons of milled refuse utilizing one Tollemache mill.
One man can switch trailers in 5 minutes. Another 15 minutes is needed to
empty and return the trailers if the round trip is less than half a mile.
The 35-cu. yd. storage hopper provides enough volume to allow milling to
continue during normal switching operations.
Since the plant is located on the landfill site, only two trailers and
one tractor were purchased initially. The trailers have a hydraulically
operated ram to eject the milled refuse. The rear door operates like a
guillotine for filling and swings open for ejection. The Ford tractor is
powered by a 2G6 hp gasoline engine.
In anticipation of hauling milled refuse to outlying landfills
(in effect, using the milling facility as a transfer station), Madison has
acquired two new tractors and three additional trailers. The new trailers
are similar to the original ones except they have swinging gate rear doors
and their hydraulic system will be powered by the tractor itself. The
new tractors are powered by 195 and 225 hp diesel engines for over-the-road
operation.
Problems with the Tollemache Equipment
Experience with the Tollemache conveyor has shown that excessive
loading of the conveyor bin is the prime cause of feed delays. Delays
result when an empty conveyor is moving no refuse into the mill, or when
the conveyor transports unprocessed material in slugs which overburden the
mill motor and trip the automatic overload device. Excessive loading of
the bin causes bridging of refuse near the constriction where the conveyor
leaves the horizontal. Thus, although the bin is full and the conveyor is
moving, an inconsistent flow of material is heading toward the mill. To
overcome this problem the operator limits the amount of refuse placed in
the bin. Although the problem has not been eliminated altogether, the
change has reduced delays and led to a smoother operation,
Other problems with the conveyor included failure of a linkage in the
speed reducer and occasional jamming of the belt when objects became lodged
betv/een the conveyor rollers and track. The linkage problem was traced to
excessive wear of the reducer assembly and was quickly eliminated by an
overhaul of this assembly. Some of the difficulties experienced with the
speed reducer are thought to have been due to the original design. The
manufacturer has redesigned subsequent models of the speed reducer to
eliminate the problems encountered at Madison.
The jamming problem, which occurred three times in 1971-72, caused the
rollers to jump the tracks, resulting in damage to the bin walls, conveyor
flights, and belt tighteners. The problem centered at the point where the
Tollemache conveyor makes a 180° turn at the center of the Gondard bin.
A steel "umbrella" was built over the turn early in 1972, and no further
jams of this kind have been reported.
The stationary compactor has been relatively trouble free. Some metal
fatigue has been noticed on the tracks supporting the ram and also on the
floor of the hopper. The most serious problem v/as a leaky hydraulic cylinder
which operates the hook that holds the trailer to the hopper. On two occasions
the cylinder failed, thus releasing the trailer. The result was that the ram
pushed the trailer forward and eventually through the door of the packer room.
The cylinder i/as replaced and no further difficulties have been experienced.
25
-------
The main problem with the original tractor in the haul-away system
for the Tpllemache mill has been its gasoline engine. Mainly, the motor
does not have sufficient torque at low speeds, thus making it hard to operate
with a full trailer. Some engine overheating problems, caused by the
radiator becoming clogged with dust, have also plagued the tractor. Constant
maintenance of the radiator is needed to keep it open and prevent engine
damage.
The mill itself is the most reliable piece of equipment in the
Tollemache system. Nonetheless, the mill has not been completely trouble-
free. Some problems have included internal jamming, explosions, and trouble
with the radial and thrust bearings on the main shaft.
Internal Jams:
An internal mill jam differs from a feed jam in that it occurs in the
mill proper, usually in the grind section. Such jams have occurred when
heavy wire, bed springs, tires, etc. inadvertently enter the machine. When
this happens the mill motor becomes overloaded and circuits are broken.
The jams, especially when wire was involved, have stopped operations up to
an hour while the machine was being cleared.
During the first few months after installation, these jams occurred
about once a week. Thus, elimination of this problem was of importance.
Prevention has involved close scrutiny of incoming refuse. Any article
deemed damaging or able to cause jams is removed from the incoming stream
of refuse. This material is small in quantity and is landfilled and covered
with the milled refuse. Also, since Madison has a bulky item pickup, refuse
pickers are instructed not to collect potentially damaging articles. The
combination of minor separation at the plant and separation on the collection
routes has minimized internal jams to the point that, on the average, only one
is experienced a month.
Explosions:
As the refuse is being processed a constant array of sparks.is produced
by contact of the hammers and metal in the refuse. Therefore, if a can of
paint thinner, unbroken bottle of alcohol, or a container with any other
flammable liquid happens to enter the mill intact, an explosion is possible.
The explosions themselves do no notable damage to the mill, but they are a
potential hazard to personnel working in the area. During the first 3 years
of operation using the Tollemache mill, five or six such explosions occurred.
After the first two incidents, an explosion chamber was constructed above the
mill. The chamber consists of a heavy steel chute leading from the top of
the mill to the plant roof. Here the chute is capped with light sheet metal
to provide mininum resistance to rapidly expanding gases. The expanding
gases from an explosion are vented through the roof. Experience has shown
that the explosion chamber functions quite well.
Bearing Problems:
By far the biggest concern in respect to the mill itself has been the
bearings that support the rotor. At the extreme upper and low parts of the
rotor are three bearings. Two are radial bearings; the other a thrust
bearing. It has been the unfortunate experience at Madison to have to
replace two lower radial bearings and four thrust bearings.
26
-------
In the original design, the thrust bearing was to draw the cooling and
lubrication oil from a reservoir and throw it up to the two radial bearings.
About a year after installation, the thrust bearing burned out and was
promptly replaced. The replacement bearing was manufactured in the United
States and was not exactly the same as the original manufactured in England.
Less than a week after replacement this new bearing also burned out, as did
a radial bearing. At this time it was decided that the new type bearing was
not drawing enough oil to cool itself and consequently not passing enough
up to the radial bearings. Thus, an oil pump was installed to force-circulate
oil to all bearings. This worked well until about a year later when the
thrust bearing again needed replacement. It was replaced and burned out 3
months later, as did another radial bearing. At this point a careful
reevaluation of what had taken place over the last year and a half revealed
that the problem was not directly related to the circulation of oil but was the
result of the rotor shaft moving up and down. During replacement of the rotor,
it was discovered that a retainer nut holding the rotor assembly together
had worked loose, thus allowing the entire rotor assembly to jump up and
down a fraction of an inch. This resulted in abnormal stress on the
bearings, and they burned out. A key has been placed over the retainer nut
in the new assembly, and no bearing problems have been experienced since that
time.
Tollemache Operation and Production
The Tollemache feed conveyor is a substantial improvement over the
system used in the Gondard operation. While the Tollemasche conveyor was not
originally designed for transporting refuse, it has performed adequately
during evaluations at Madison.
Unlike the Gondard mill with its grates, the Tollemache mill depends
on the number, length, and pattern of hammers to produce the desired
particle size. The original pattern used in England and initially tried
in Madison involved 54 hammers. After a short period of test runs it was
evident that such a pattern did not meet the objectives of milling for
landfill disposal. The product was more suitable for a composting operation,
and production rates were very low. Numerous modifications were then
attempted. A final pattern involving 34 hammers emerged as the most efficient
for Madison's purposes (Figure 16). This pattern yields an average operational
production rate of about 14 tons per hour, or just one ton per hour below the
manufacturer's indicated capacity.
As with the Gondard mill, two production rates were determined for the
Tollemache mill. These are the operational rate (including only time during
which refuse is being milled) and the overall rate (including time during
which mill is milling plus down time due to mechanical problems in the plant).
During test periods, the operational rate of the Tollemache mill ranged
from 11 to 20 tons per hour. The overall production rate has been about
0.5 tons per hour lower than the operational rate, indicating that the
machine operates with a minimum of down time.
A distinct seasonal variation in production rate has been noticed
at Madison which is related to the moisture content of the refuse. During
late spring, summer, and fall, when the average moisture content of the
refuse ranges from 35 to 45 percent on a dry weight basis, the highest
production rates have been experienced. In late fall and winter, when the
moisture content is only 15 to 20 percent, production rates reach a minimum.
27
-------
o
c
•»->
its
c
o
TJ
C i—
L.
O)
r— «3
O >
H- O)
(U
3
cn
28
-------
This fact was brought out during two evaluations of the mill. One was done
during the summer and early fall of 1970, the other during the winter of
1971. Table 3 shows a compilation of production rates on a monthly basis
for the two test periods. Monthly average moisture content is also
indicated.
The table shows that higher production rates-were achieved during the
period of high moisture contents. Thus, when discussing mill production
rates, the figures should be cited in terms of dry tons of refuse per hour.
In this manner rates can be calculated for different applications in other
parts of the country. For this reason, Table 4 is presented using dry
tons milled and computing a production rate 1n units of dry tons/hour.
A comparison of dry tons per hour in*Tabi« 4 shows that the mill
processes refuse at nearly constant rates and that, in fact, the seasonal
variations are due to moisture content changes.
Tollemache Power Consumption
Mill power consumption is broken down Into two distinct categories--
kilowatts (kw) and kilowatt hours (KWH). Kilowatts provide a measure of
peak demands for power during a specific time period.- Kilowatt hours are
a measure of the actual energy usage. As such, kilowatts are not a
function of tons milled, but kilowatt hours are.
Mill demand ranges from T10 to.160 kw/month. The highest values are
experienced from November to March; but since the cost of demand at Radison
is billed on the highest demand experienced in the preceding 12-month period,
the range of demand is not important.at this site. Energy usage measured in
kilowatt hours/ton (KWH/T) follows much the same pattern, peaking in the
winter and reaching lows during late spring, summer, and early fall.
Extensive power data were kept during twor'experimental evaluations of
the Tollemache mill. Table 5 shows the summaries of data collected during
these periods. Notice that energy usage during the summer evaluation was
2.5 KWH/T lower than during the winter tests. Also notice that on a dry-ton
basis the summer evaluation had a power consumption over 2.1 KWH/T lower
than the winter test. Thus, a definite positive relationship between KWH/T
and moisture content is not evident, as was the case when computing production
rates; in fact these data suggest an Inverse relationship v/here power consump-
tion increases as moisture content decreases.
Mill accessories, including feed conveyor;: final transfer conveyor,
stationary compactor, welding machine, etc., all operate from a three-phase,
440-volt source. No equipment was available to secure power data from the
accessories on an individual basis until January 1972, when a single power
meter was installed to record data from the stationary compactor. Prior to
that time, all accessories had been lumped together. Data from both
evaluations has indicated that power consumption for mill accessories as a
whole is about one-fourth that of the mill itself.
Two-Hill, Two-Shift Operation
Plans for expanding operations at the Madison Refuse Reduction Plant
in order to operate on a two-shift basis were begun in the summer of 1969.
The first step toward this goal was taken in the winter of 1970-71 with
completion of the plant expansion. The second step was completed with the
conclusion of the second and final Tollemache evaluation early in 1971.
29
-------
oo
CO
_J
*•••<
^"
UJ
3:
cj
<
UJ
1
o
r
fv^
o
u-
co
UJ
1—
2
z
o
HH
1—
u
=>
o
o
ce
Q-
_l
_l
2
UJ
^>
o
o
•£.
CD
»— i
^_
a:
UJ
a.
c
•M
JL
c .c
V
*S O1
°- C
•'•{3
O J_
o~ Q.
o
,_
r_
0}
t-
x-^ Q)
(/) >
V. O
3
O
* c"
Q) ••_
i?2
OJ
Q-
•o
(A 0)
C i—
O t—
ir
*« S« S« X %Q
co in «a- CM fv.
CO CO ^1" CO CO
*
CM CM CO CM CO
CM vo in r*v f—
• • • * ••
*3" ^1" CO ^1" ^"
r~ t™~ r^ t^ t—
*
CM r— O> CM CO
t^. o oo co in
^f in co *j* ^f~
i— ^~ 9^ r*~ r*~
CD r***. co co co
• • • • •
**• i^ in co in
o o CM co i*»
•— r- i— CO
^ CD in o cc
o «a- CM co in
O C CM CO VC
r— r— r— CO
C CO r— <*• CO
CO l-» O VO r—
^- tn r-» in co
r— r— r— in
x
in
CM
o^
VD
*
CM
f—o
O>
CM
r—
VO
O
•
I-»
CO
o
^t
•
vo
CO
cS
f^
|R— •
*8
CM
CM
O
in
CM
r-»
vo
CO
CM
r— •
r—
VO
*
pw*
CM
fM.
r-
t~~
•
CO
I**
cr>
r~*
tn
P— •
a<
CO
CM
CO
in
t
CM
!•—
CO
CO
CM
f—
J^
\c
•
CO
0
CM
,_
in
•
o
CVJ
CM
VO
CM
•
C
V
o
•°£
i *
i- 4->
0 0
O)
o 53
O r—
•— o
•— i.
f- Q.
E
•o
D- •
to t3
^- OL
JD CO
to
•r- +*
%~ CO
(O QJ
> 5
o
• a.
J- 0
>
w C
4-> O
d.
> en
•— »-
o
o
en •—
*~ en
•— o
CO 4J
•— a> **
JJ fc. O
-(-> E a> vo r-s
(/) »•—
CO O- +•> r-
3 O> O 3
OO
U
0)
C O
•i- O
CO J-
•o o
o >-,
•i- O)
1- >
a> c
a. o
u •
5 O) o'
rl % f— f
4-> O)
cr. -r-
C - «J
•r- 10 T-
S- 34-
3 .c a>
O t— T3
30
-------
TABLE 4
OPERATING AND OVERALL PRODUCTION
Test Period
July 6 to 31, 1970
August 1970
September 1970
October 1 to 9, 1970
July 6 to
October 9, 1970
February 4 to 29, 1971
March 1 to 31 , 1971
Dry Tons
Milled
mo
1165
1180
425
3880
885
1245
Dry Tons
11.05
11.11
9.64*
11.19
10.61
10.25
10.55
per Hour
Overall
10.67
10.82
9.43*
11.10
10.35
10.18
10.25
February 4 to
March 31, 1971 2130 10.40 10.20
* During two periods in September, linkage on the variable speed
drive of the mill conveyor was broken. Thus, the conveyor
could only operate at its slowest speed. This was a feed
problem, not a mill deficiency.
31
-------
TABLE 5
COMPARISON OF POWER DATA FOR TOLLEMACHE MILL ALONE (1970-71)
Period I
July 6-31
August
September
October 1-9
Overall
Period II
February 4-28
March 1-31
KWH
1067?
9856
10288
3152
33968
10272
12864
Tons
Milled
1480
1573
1701
564
5318
1105
1519
KWH/Ton
7.20
6.26
6.04
5.59
6.37
9.30
8.46
Dry Tons
Milled
1110
1165
1180
425
3880
885
1245
KWH/
Dry Ton
9.60
8.46
8.70
7.42
8.75
11.60
10.31
Overall
23136
2624
8.82
2130
10.86
32
-------
The third phase of the plan was attained by December 1971 when a plant
supervisor and three additional men were hired to man the second shift.
Finally, in January 1972, the two-shift operation utilizing both the fiondard
and Tollemache mills was initiated. This section of the report deals with
observations and evaluations of this operation between 1 January and
30 June 1972.
During the initial five years of operation, three men performed all
plant functions. Supervision was provided on a limited basis from the
Madison Street Department. During evaluations of the combined Gondard and
Tollemache systems, it was concluded that three men were sufficient to
handle plant operations. Thus, when preparations for two shifts were made
it was decided that three additional men would be hired to man the night
shift.
In addition, a full-time supervisor was employed. It was hoped that
the increased labor cost in hiring a seventh man would be offset by
increased plant efficiency. This seems to have been a valid assumption.
The supervisor is responsible for personnel and administrative duties
not ordinarily assigned to an employe in this position. In addition, he
performs some of the public relations tasks at the milling plant. Thus,
while the supervisor's salary is assigned to mill operations, about half of
his duties are not directly production related.
Table G shows the prescribed duty hours for each shift and the
supervisor. Also included in the table is the average time span for loaJ
arrivals and average mill operating tine. The average length of the milling
day is 10.5 hours (10:30 AM to 9:00 PM).
TABLE 6
HOURLY BREAKDOWN OF AVERAGE DAY
(Two-Mill, Two-Shift Operation)
6 AM 9 12 PM 3 6 9 11
First Shift ' '
Second Shift i
Supervisor i i 1 •/ vvr
K (on call)
Loai! Arrivals i 1
Milling Time
Packer trucks deposit their loads over a 6.5 hour period each working
day. However, the trucks usually arrive in bunches at three specific tines
during that period--middle morning, noon, and midafternoon. Thus it is not
uncommon for 10 or 12 trucks to be lined up at one time waiting to enter
the plant. By union agreement, packer crews cannot start at staggered times.
Another factor contributing to this problem is a floor plan that will permit
only two trucks to maneuver inside the plant at one time. This problem does
not affect plant operations but does lower packer crew efficiency. One
benefit derived from the situation is that the front-end loader operator
33
-------
does not havo a continuous supply of unprocessed refuse to stack or mov-3,
thus allowinq him free time to perform other duties.
On an average day, 53 loads (190 tons) are handled at the plant,
although as many"as 72 loads (340 tons) per day have been processed. These
figures represent the entire city's collection and that of some private
collectors (3 to 6 loads per day). In 1972, approximately 46,600 tons of
refuse were milled in the two-mill, two-shift operation at Madison. It is
important to note hero that all the residential and light commercial refuse
collected by city crews is now being milled. Bulky items, brush, and
bundled newspapers are being collected separately and are not milled. The
bulky items and brush are landfilled separately, while newspapers are
recycled.
Mill start times must be planned to avoid'production stoppages due to
lack of material, After detailed investigations, the plant supervisor
concluded that a lag of 1.5 hours between first load arrival and mill
start time eliminates almost all production stoppages due to material
shortages.
In the design of the two-shift operation it was estimated that an
average of 280 tons of refuse could be processed per 13 to 14 mi 11-hour
day at a rate of 20 to 22 tons per hour. During the first C months of 1972
an average of nearly 190 tons of refuse was milled per 10.4 mill-hour day
at 18 tons per hour. Although short of design estimates, these figures
are satisfactory for early stages of operation.
Two causes were responsible for the difference between the design and
actual figures. First, during the initial 3 months of operation, refuse
generation was at its seasonal low. Second, mill down time, near 2.8 hours
per 10.4 operating day, was quite high. Efforts are underway to increase
available tonnage by brinning in privately collected refuse and to
decrease down time by improving equipment maintenance and by making more
effective use of plant personnel.
During individual experimental runs of the Gondard and Tollemache
mills, operating and down times were recorded separately. Since this
method would prove difficult in simultaneous operation, simple strip-chart
recorders were installed to measure operating and down time for each mill.
The recorders keep track of mill motor amperage on a continuous time chart.
Thus it is possible to determine v/hen the machines are running as well as
the level of loading. For this operation, all times above "mill idle" are
productive, all other times are considered down time.
Operation and overall production rates for the first 6 months
of 1972 are shown in Table 7. Refuse'moisture content as presented is
computed on a dry weight basis. Also included in this table are production
figures for the second half of 1972, following the evaluation period.
Table 7 shows a general increase in tons'mi lied during the 6 month
evaluation period. This increase was due largely to seasonal variations
in refuse neneration in the city. The table does not show that daily
operating time went from 7.7 hours in January to 13.1 hours in fay. A ratio
of down time to operating time indicates that March had a high of 0.39 hours
of down time for every hour of productive operation. This figure was decreased
to 0.25 by June.
Although production for the two 6-month periods of 1972 was virtually
equal, overall TPH went from 17.90 in the first 6 months to 19.47 in the
second G months. Also down tine decreased from 350 hours in the first half
of the year to 213 in the second half.
34
-------
TABLE 7
OPERATIONAL AND OVERALL PRODUCTION RATES (1972)
(Two-Mill, Two-Shift Operation)
January
February
March
April
May
June
Overal1 for
January-
June 1972
Hours
Tons Plant
Milled Time
1
2663
3334
3440
4236
5126
4518
161
205
208
235
262
232
Hours
Down
Time
26
39
74
77
77
57
Overal1'
Rate
Operational
Rate
(TPH)
16.51
16.28
16,51
18.04
19,57
19-48
23317 1303 350 17.90
17.93
17.93
20.00
21.98
22,86
22.24
20.63
Overall
Moisture Dry Rate
Content Tons (Dry TPH)
23
23
25
29
40
38
31
2165
2710
2752
3287
3665
3276
13.44
13.22
13.24
14.00
14.00
14.15
17855 13.71
July*
August
September
October**
November
December
2165
4947
4797
3040
4840
3483
130
232
230
163
243
197
51
37
31
26
29
39
16.64
21.32
20.83
18.61
19,91
17.72
22.49
23.17
22.44
20.16
21.74
19.23
Overall for
July-December
1972 23272
1195 213 19.47
21.72
Overall
1972
for
46589 2498 563
18,65
21,22
* Low production because density tests conducted at landfill diverted
considerable unprocessed refuse.
**Low production because rotor was being changed on Tollemache mill.
1. Elapsed time from first mill start-up to last mill shut-down including
milling plus down time.
2. Summation of hours down time for Gondard mill plus Tollemache mill.
3. Tons milled/hours milling plus down time.
4. Operating rate is defined by the following formula for a two mill
facility, to take into account the different milling capacities
for the two mills:
(22 tons/hr) x Tons Milled
(Tollemache operating hours x 14 tons/hr)+(Gondard operating hours x 8 tons/hr)
35
-------
The data in Table 7 also show an increase in production rates from
January to June. Although this production rate increase reflects improving
plant operation, it is more directly related to an increase in refuse
moisture content. Figures showing the operational rates in terms of
dry tons per hour (PTPH) illustrate this dramatically. The maximum variation
from the overall figure of 13.71 DTPH is only 0.49, while the maximum
variation in the wet basis average of 17.90 TPH is 1.67.
Power usage (KWH) was recorded by a 440-volt watt-hour meter installed
in February 1972 on each mi 11(Table 8).
TABLE 8
POWER CONSUMPTION - MILLS COMBINED (1972)
Demand (kw) Combined Energy KWH/ KWH/
Tollemache Gondard Usage (KWH) Ton Dry Ton
March TTTTC TW ,'~' 3177(1 9.28 11.56
April 125 134 31157 7.38 9.47
May 116 121 31950 6.23 8.40
Juno 120 116 29730 6.60 9.07
OVERALL 7.19 9.61
From the table it is seen that the Gondard mill created slightly more
power demand than the Tollemache mill in all months except June. Previous
evaluations of the separate mills shov/ed that the Gondard mill used nearly
as much energy as the Tollemache mill even though it iis rated at onty
about 60 percent of the capacity of the Tollemache.
Earlier tests indicated that the Gondard mill consumed power at the
rate of 12.5 KWH/ton while the Tollanwhfe m^ used only 7*2 KWH/ton on the
average. The combined overall figure of 7.19 KWH/ton in Table 8 is better
than the average of either machine on previous evaluations. It should be
noted, however, that the 1972 data were collected during a period of the year
which in the past has recorded the lowest power consumption. On a dry-ton
basis, the overall combined figure v/as 9.61 KHW/dry ton. No apparent
relationship exists between moisture content and power consumption, as in
the case of moisture content and production.
On the basis of the first 6 months of the two-shift operation, it
would be unrealistic to assume that an average of 280 tons of refuse could
be milled daily at this stage. Not only would it require 13 to 14 hours of
mill operation daily, but it would also mean that the Gondard mill would be
in operation at an overall production rate of 7 to 8 tons per hour and the
ToTT.emache at 13 to 14 tons per hour -- a combined overall production rate
of 20 to 22 TPM. Table 7 shows that during no month of the evaluation
was an average production rate of 20 TPH achieved. Even if 19.57 TPH, the
highest monthly rate obtained during the period of record, were to be
maintained all year, it v/ould require 14.3 hours of mill operation per day to
maintain a daily output of 280 tons. This v/ould leave only 1.7 hours for
clean-up and preventive maintenance, an insufficient time for such tasks.
It is more realistic to assume that under the present conditions of a
12 to 13 hour mil lino day, at an average of 18 to 19 TPH, a production
of 230 tons per day could be achieved and maintained.
36
-------
The two-shift operation involved new personnel and new operatinq
characteristics during the evaluation period. Thus it is fair to say that.
the system was not operation at its full potential. This is evidenced by
the high incidence of down tine during this evaluation compared to down
time experienced during evaluations of the two mills separately. It is
reasonable to expect, therefore, that the plant will eventually reach a
production rate of 20 TPH over the year. Combined with a 12 to 13 hour
nilling day, this would result in an average daily production of 240 tons
or a yearly average of 60,000 to 65,000 tons.
Mi 11 Accessories
For this evaluation, mill accessories are defined to include feed and
takeaway conveyors as well as welding equipment. This equipment operates
fror: a 440-volt power source, and power consumption for the accessories,
therefore, was determined by taking the difference of the total 440-volt
usage and the 440-volt usage of the two mills. From this information, it
was found that average accessory demand is only 7 percent of the total for
both mills and energy usage is 18 percent of total energy usage for both
mills. Thus, mill accessories required approximately 1.29 KWH/ton or
1.34 KVJH/dry ton of electrical energy.
With simultaneous operation of both mills, an increased demand on the
stationary compactor has occurred. It now takes only 45 to 60 minutes to
fill one 75-cu. yd. trailer. The 35-cu. yd. storage hopper thus does not
provide enough volume to allow milling to continue during normal trailer
si/itching. The mills in combined operation fill the bin in 2 to 5 minutes,
v/hile it normally takes 5 minutes to switch trailers.
A watt-hour meter was installed on the power source to the stationary
compactor in February 1972. Adequate data were not obtained until April.
For the period of April through June, the monthly average power consumption
of the compactor was 0.76 KWH/ton (1.03 KWH/Mry ton), or about 3500 KMH.
Thus, the compactor requires less than 10 percent of the power needed by
the mills and their accessories.
Ma i ntenanee Programs
Preventive maintenance on the conveyor is done once a week. At this
time the pulleys are greased and the tracks oiled. A general inspection to
locate worn parts or potential trouble areas is also conducted to minimize
major breakdowns and to nlan an effective repair proaram.
With the Gondard mill, the hammers, of course, are subject to the
greatest abrasion in the system. The rods on which the hammers swina
require replacement about four times per year when the mill is operating
on a one-shift basis. The spacers between the rotor disks have a si milar
lifespan. Other erodable elements are the grates and the liner or wear
plates. These are maintained by arc-weld applied hard surfacing. The
maintenance progran, when properly executed, can be carried out daily ;!urinq
the clean-up period. Only if grate replacement is required should it be
necessary to perform such \;ork on overtime. The fact that all elements
of the Gondard hammermill are relatively thick and can be welded in the
down-hand position makes it easy to train personnel for thiswork .
Routine maintenance performed on the Tollemache mill mainly involves
hardfacing and/or replacing hammers and mill liners and replacing shafts.
Other areas of the machine, such as reject chute, rotor plates, and cone
liners, need only occasional attention.
37
-------
Hardfacing of harriers and mill liners will be discussed later. The
vertical shafts, K.5 in. in diameter and 23 ft. long, hold the hamners
and therefore also suffer considerable wear. Experience has shown that
approximately four of the ten shafts in the mill need replacement per set
of hammers. Shafts are replaced when daily inspections reveal excessive
wear or that a shaft is bent.
Two-Mill Operation:
There is a 3 1/2 hour period between arrival of first-shift personnel
and mill start time under the current operation at the Madison Refuse
Reduction Plant. This time is used for necessary plant clean-up and to
conduct the preventive maintenance program.
Detailed inspections are performed on Mondays. At this time notes are
made on what repairs or replacements are needed (Figure 17). The necessary work
is then scheduled for completion later in the week. Necessary lubrication is
also applied during Monday inspections. Hammer "tipping" or hardfacing is
performed daily. Portions of the mill interior are hardfaced as necessary.
Hardfacing:
The original recommendation for the Gondard mill was to use plain
carbon steel hammers and to replace them after reversing them to use both
outer faces. Following this procedure only about 300 tons of refuse could
be milled with one set of hammers. Therefore, experiments with hardfacing,
or "tipping", the hammers with arc-deposited metal were begun in February
1968. Results of tipping medium-hard steel (SAE 1060) hammers were
extremely good — hammer life was nearly doubled. Early attempts to tip
harder (SAE 1090) hammers resulted in frequent fractures, but recent
experience has been more successful. Nearly 1500 tons can now be milled
with one set of properly tipped hammers in the Gondard mill.
Because of the success of the hammer tipping program with the Gondaru
mill, a similar but more intensive hammer maintenance regimen has been in
force with the Tollemache system since its installation. This program
involves application of Amsco Super 20 hardfacing alloy to each hammer
before mounting it in the mill. The hardfacing alloy is applied to the
wearing edges of the hammer. The wearing edges are in constant contact
with the refuse and consequently deteriorate more rapidly than other surfaces
of the hammer. In addition to the initial tipping, 18 hammers in the lower
section of the mill are tipped daily. The remaining 16 hammers in the
pre-break and neck sections of the mill are inspected daily and tipped when
needed.
In full-scale operation, a set of 34 Tollemache hammers maintained by
tipping will process 1,000 to 1,200 tons of refuse. A set of untippeJ
hammers will process 400 to 500 tons. From this comparison, it is obvious
that tipping greatly increases hammer life; but it was not evident that tinning
is economical. To stU'Jy the economics of tipping, a test was nerforme:: usinn
tipped and untipped hammers in the nil! under normal operating conditions.
The comparison was made on a cost-per-ton basis, and the results showed that
a tipping program, rigorously applied, will save approximately 23 cents per
ton of refuse milled. The entire savings is due to the lower number of
hammers required. Labor and welding materials for the tipping program were
about equal to labor for hammer changing alone in the control program using
untipped hammers.
38
-------
"O
i.
Q.
O
1-
c
re
CD
s.
o
Q-
O
JC
cc -
c
(O
O!
-M
C
QJ
i-
3
39
-------
The hull of the Tollemache mill is protected by replaceable manganese
steel liners or wear plates. During the first 2 years of operation, no
attempt was made to reduce wear of these liners because they did not show
significant loss of metal. A set of untipped liners was found to last
through approximately 12,000 tons of refuse. When the plant went to a
two-shift basis in January 1972, the prospect was for the Tollemache mill
to process 45,000 tons of refuse annually. At that rate of production,
four sets of liners would be needed each year. At over $1,300 per set of
liners and with 32 man-hours of labor needed per change, the outlook
suddenly became expensive. Consequently, methods of increasing liner life
were investigated. The first method tried was hardfacing. Some difficulties
were experienced, however, since an adequate welding rod that would deposit.
weld vertically was needed. After trying many types of rods, a Stoody 3/8 in.
#57159 proved most efficient. Liners are now extensively hardfaced prior to
installation and are touched up in the machine when extensive wear is noticed.
Early results indicate that this procedure has increased liner life to
20,000 tons per set. Other methods to increase liner life are now being
studied, such as rotating or inverting them periodically to distribute wear
evenly arid welding manganese rods to the liners to prevent wear of the base
metal itself. Also, the manufacturer has redesigned the liner plates to
give extra thickness on the lower half of the plate.
LANDFILLING MILLED REFUSE
To provide a direct comparison between landfill milled refuse without
daily cover and the sanitary landfill technique using the unprocessed
refuse, refuse was placed in piles — called cells — above the level surface
of the 01 in Avenue landfill site during 1967 and 1968.(Figure 18). The cells
were 5 to 6 feet in height and were level!ed:>«nd sloped -to provide'surface
drainage. Lengths and widths varied, but the smallest cell was at least
40 feet in its shortest dimension (Figure 19). Covered unprocessed cells
and uncovered milled cells were constructed simultaneously. Records were
kept on the season during which the more than 20 cells were constructed
and (for milled refuse) grate size used in the mill (Table 9). Both
cell types were compacted with a D-7 Caterpillar tractor. In the case of
covered cells, the cover material was a sandy silt obtained five miles fron
the site.
Strictly speaking, those cells constructed with unprocessed refuse and
covered v/ere not sanitary landfills. Insufficient refuse was avai'labls to
construct an entire cell, or even a major portion of a cell, in a sing1'?
day. A choice hat1 to be made, therefore, whether to cover the small
amount of refuse placed each day, cover all exposed refuse daily except for
the working face, or cover each cell upon its completion. It was decided
to avoid having the cells consist of small pockets of refuse bounded by
soil, because this would lead to difficulties in tracing and under starK'v/K;
moisture and gas movement. But it would have been poor practice to leave
an entire unprocessed cell uncovered until its completion. Therefore, only
the working face at the close of each day's operation was left exposed.
In addition to the Olin Avenue experimental cells, special cells were
constructed at other sites for specific studies. These will be described
and discussed later. After the Olin Avenue cells had been completed in
September 1953, milled refuse was landfilled in a specified arna at t!">?
40
-------
-------
-------
TABLE 9
SUMMARY OF TEST CELLS - OLIN AVENUE SITE
Cell No.
And Type
MILLED CELLS
1
2
3
9
10
11
13
15
16
17
19
20
21
23
5
7
Type of Waste Grate Size
ComposItityn Tons Used Jn.Milling Period of Construction
Refuse
Refuse
Refuse
Refuse
Refuse
Refuse
Refuse
Refuse
Refuse
Refuse
Refuse
Refuse
Refuse
Refuse
Garbage
Rubbish
534
450
703
474
689
844
319
617
739
733
499
430
709
420
103
310
3 1/2
2
6 1/4
6 1/4
3 1/2
6 1/4
3 1/2
3 1/2
6 1/4
5
5
4
6 1/4
5
6 1/4
6 1/4
Sept. 18 - Oct. 6, 1967
Oct. 9 - 27, 1967
Oct. 30 - Nov. 17, 1967
Dec. 11 - 29, 1967
Jan. 2 - Feb. 1, 1968
Feb. 2 - March 18, 1968
March 18 - 29, 1968
April 1 - 19, 1968
April 22 - May 13, 1968
May 13 - 31 ,1968
June 3 - 14, 1968
July 8 - 20, 1968
July 29 - Aug. 20, 1968
Oct. 10 - 22, 1968
Nov. 20 - Dec. 1, 1967
Nov. 27 - Dec. 8, 1967
UNPROCESSED CELLS
4A Refuse 1045
4B Refuse 683
12 Refuse 1456
14 Refuse 197
18 Refuse 1268
22 Refuse 548
6 Garbage 38
8 Rubbish 400
Sept. 18
Oct. 23 -
Jan. 8 -
March 18
April 1 •
July 29 -
Dec. 4 -
Feb. 12 -
- Oct. 18, 1967
Nov. 17, 1967
March 15, 1968
- 29, 1968
May 31, 1968
Aug. 20, 1968
8, 1967
March 5, 1968
43
-------
01 in Avenue si to without daily cover. During the entire demonstration
period, conventional sanitary landfill operations for unprocessed refuse
were carried on at the 01 in Avenue site (and at another major city-operated
site).
After Christinas 1967, many Christmas trees from Madison's West side
were deposited at the project site. To determine if milled refuse could be
used as a substitute for cover soil'.to cover the trees and fill the voids,
the trees were bulldozed, into as small a pile as possible, and milled
refuse was dumped on top of them. A bulldozer then worked the refuse
into the voids. The results with milled refuse cover were as satisfactory
as with regular cover material. Only minor pockets of settlement occurred after
the cover operations with milled refuse.
In another test, during the spring of 1968, milled refuse was used to
cover the working face of a pile of unprocessed refuse. This was also
concluded to be satisfactory, and resulted in a neat and smooth workinq
face.
The landfill operation with milled refuse quickly provided a test for
claims that milled refuse has traffic bearing characteristics superior to
unprocessed refuse. In the past, Madison normally had difficult traffic
operating conditions in its landfills during spring and fall (Figure 20).
The problems were associated with wet conditions during these seasons and
with silty-sand cover material used at the landfills. During the demonstra-
tion project, however, nilled refuse was used to construct access roads to
desired dumping areas, and experience showed that a 2-foot depth of milled
refuse provides adequate access.
Trucks carrying refuse from the plant to the landfill are now routed
over gravel roads built to the milled refuse area and then over the top of
the milled refuse pile at the site. Although some trucks weigh nearly
73,000 Its., they have experienced little difficulty in maneuvering.
Tests with both empty and loaded tanker trucks, loaded to and in some cases
exceeding legal road limits, also have shown that milled refuse has
satisfactory supporting capacity for truck traffic (Figure 21).
Tire 'damaqe to equipment at the dumping face of sanitary landfills
is commoi. because of larne pieces of glass and other sharp objects.
Because of the lack of such objects in milled refuse, tire damage has not
been a problem for trucks or rubber-tired loaders used on the milled
material in the landfill. The only tire damage experienced during this
project has been with worn tires on the end loader used in the mi 11 inn
plant.
CHAHACTLRISTICS OF [1ILLEO REFUSE
The most noticeable feature of milled refuse is its homogeneous
character. Milled refuse has the general appearance of oversized confetti.
llass is virtually disintegrated, being ground to particles less the 3/3 in.
in size. Many of the cans arc completely crumbled.
In early 1972, a magnetic separator was installed in the plant to
remove ferrous motals after milMnq. Currently, this metal, mostly cans,
is being hauled by semi-trailer to the Wisconsin Chemical Corporation in
Milwaukee whore it is de-tinne-1 and then sln'pprd to copper mines in the
western United States for USP in extraction of copper. This will be
discussed in more detail in the section on Trends and Developments.
44
-------
"O
c
ai en
•r- C
QJ r
a. 3
x -a
i- to
ai -i-
N $-
O 0)
Q +J
IT3
-O E
a>
-^ t-
u a>
i- o
I— u
o
CM
ai
s-
45
-------
Blowing and Particle Size
Blowing Titter has been one of the major objections to some of
Madison's sanitary landfill operations for unprocessed refuse. The extent
of the problem depends on wind direction and speed as well as exposure of
the working face to the wind. For example, unprocessed refuse dumped at
one of Madison's other landfills may be caught by the wind and blown over
the boundary and site fences. On occasion, special work details have been
sent out to pick up litter from the lawns of homes near landfills. In 1969
alone, some $22,000 was spent for manpower to control and pick up blowing
paper from city sanitary landfills. Even 15-foot-high movable fences
placed downwind from the working face have failed to solve the blowing
problem associated with landfilling unprocessed refuse.
As noted earlier in this report, various grate sizes and hammer
patterns were used to obtain a particle size which would reduce blowing
problems at the landfill. With properly ground milled refuse, the blowing
problem was found to be minimal. Madison's Director of Public Works has
stated that this feature alone justifies a milled refuse operation.
Landfilling has been carried out with milled refuse during winds
up to 60 mph on a flat site with only minor problems. Those blowing prob-
lems that are experienced are usually due to sheets of plastic, which are
not thorouqhly shredded in the mill and therefore tend to roll across the
fill surface. Such items do not become airborne, and are readily caught
by low fences (Finnre 22).
Three factors may be given to explain the lack of blowing of milled
refuse. First, particles ot milled refuse tend to become entangled in
each other so that they are discharged in clumps rather than as individual
particles which can be blown away. Second, the snail surface area of
individual particles of milled refuse provide a small target for the wind.
This is in contrast to a page of newsprint which, caught in a high wind,
Figure 21. Fully loaded transfer trailer traveling over an
8-foot lift of milled refuse.
46
-------
Figure 22. Wind-blown film plastic from milled refuse landfill
47
-------
acts like a sail and blows for long distances. Observations at the landfill
site have confirmed that bits of freshly milled refuse blow only a few
feet before corning to a rest. Finally, in a landfill, a crust similar to
papier mache forms at the surface of the exposed milled refuse in a few
weeks.
A study completed in 1972 was designed to analyze the particle size
of milled refuse produced by both the Gondard and Tollemache mi 11s.(Figure 23),
It v/as found that the Tollemache mill with a new set of 34 hammers producad
nearly the same distribution of particle sizes as the Gondard mill with a
5-in. grate. Between 00 and 90 percent of the particles from both mills
passed through a 2-in. screen, and 15 to 30 percent passed through a 0.2-in.
screen. All figures are on a dry-weight basis. Particles larger than
1 inch were mostly paper, rags, and plastics, while particles between 1.2
and 1 inch were largely paper, rock, glass, wood, garden trimmings, leaves,
and metal pieces. Particles smaller than 0.2 inch were finely nround glass,
sand, and ash. Typical particle size distributions for the milled product
when both mills are in operation are shown in Figure 23. The variations
in grind are due to changing refuse composition, moisture content, and ham-
mer wear.
During this investigation it was found that increased moisture content
of refuse results in a more finely ground product, while an increasing
degree of hammer wear in the Tollemache mill results in a more coarsely
ground product.
Milled refuse appears to be bulkier after it comes out of the mill
than in its unprocessed state. It is thought that this is due to "fluffing"
of paper arid paper products. The fluffing, or bulking, is the reason that
the original container system for hauling milled refuse was inadequate for
its task (see page 29).
Density
It is important in estimating the life of landfill sites to know
whether there is a significant difference in density between milled and
unprocessed refuse and whether heavy compaction equipment can produce the
same or higher densities as achieved by milling.
Although it is bulkier than unprocessed refuse immediately after
milling, milled refuse is reported from European experience to become more
dense than similar but unprocessed refuse after it has been compacted in
the landfill. To determine relative densities of milled and unprocessed
refuse, several field and laboratory tests were conducted in conjunction
with the fladison demonstration project.
In 1967 and 1968 field density tests were run at the 01 in Avenue
landfill. An evaluation of the test results indicated that additional
information was required. The major objections-to these first tests were:
first, the tests did not take into account refuse moisture content; second,
due to building the cells above ground in the form of mounds, more cover
was required tlian would ordinarily be used. Finally, the amount of
compactive effort was not held constant or controlled during tost cell
construction.
In 1971 a laboratory testing program was begun to investigate in more
detail the density of milled and unprocessed refuse under identical
conditions, and to examine the effect of vibration on densities. The tost
results led to the desinn of a large-scale field experiment in which as
48
-------
CM
r-.
en
to
O
CU
1/1
CU
s_
O)
S-
o
to
o
+J
3
l/l
-o
(U
N
to
CD
O
•r-
-(->
S-
o.
ro
o
n
csi
O)
3
cn
-------
nuch control ari^1 precision as could be reasonably provi>'p.J u!Hor fi>!,'
conditions \;».s jse'. Thi? final tost on density w.is run in 1 •>"".'.
All three testinp programs will be described in more dotai! in th<<
remainder of tin's soccion.
P.efuse Composition:
Table 10 gives the composition of refuse at Madison as determined by
the Bureau of Solid Mas to "ananemerit in November 196B. llhile other separation
analyses have been run at various times throughout the project, the results
have not been found to vary greatly from the composition indicated in the table
TABLE 10
COMPOSITION OF MADISON'S SOLID WASTE*(November I960)
(Percent Wet Weight Basis)
Category
Minimum
Maximum
Average
Food Wastes
Garden Hastes
Paper Products
Plastics, Rubber, Leather
Textiles
Wood
Metals
Glass Ceramics
PxOck, Ash, etc.
4.4 28.9
0.0 31.1
35.1 53.2
0.3 3.7
0.1 7.3
0.0 2.6
5.0 1/1.5
4.4 17.6
0.6 17.6
15. .1
42.4
i.e
1.1
f.7
10.1
7 ''
*Determined by Federal Solid Waste Management Personnel.
Laboratory Study:
To provide detailed laboratory-scale data on densities, comparative
tests utilizing botii milled and unprocessed refuse were conducted vn'tn a
compression machine, able to apply loads up to 1,000,000 Ibs., in the
Engineering Mechanics laboratory of the University of Hisconsin-"'adison
(Figure 24). To more accurately simulate a landfill compaction operation,
50
-------
Figure 24. Compression machine with air hammer in place, used to compact
milled and unprocessed waste during laboratory investigations.
51
-------
an air hammer was used to produce vibrations .:uf;n>; the laboratory tests.
To determine what vibrations were actually producer! in the field, a
vibration meter was brought to the landfill site and set up near a working
D-7 Caterpillar tractor. The vibration pick-up was placed on a steel plate
located on the refuse surface approximately 4 i'ect from the tracks. Results
indicated that the tractor transfers vibrations of greater displacement when
moving in reverse than when moving forward. The average displacement was
found to be 14.02 mils (1 mil = 1/1000 in.), and the frequency was crudely
measured at 6 cycles per second. The air hammer was able to simulate these
vibrations quite well.
All refuse for the laboratory density test was collected on August 10,
1971. A single truckload of residential solid waste was mixed and 1,000
Ibs. were milieu with the Tollemache mill; another 1,000 Ibs. were removed
for testing but left unprocessed. The moisture content was about 45
percent on a dry weight basis. Samples of refuse of each kind were placed
separately in a specially built container and compressed under various loads
and vibrations. The pressure was increased in increments of 1,0^9 to 2,000
Ibs. (acting over a 4-square-foot area of refuse), and the volume of the
refuse was recorded after each increase in pressure. This was continued
until there was little or no further compaction. In all cases there was a
rapid increase in density to a pressure of about 15 psi. After then,
increased pressure resultei, in relatively snail density increases. The
results are shown in Figure 25.
The density of milled refuse was higher than that of unprocessed refuse
under all pressures tested. Initially, average actual refuse densities
in the containers were 386 Ibs. per cu. yd. for milled refuse and 313 Ibs.
per cu. yd. for unprocessed refuse all on a wet weight basis. Since it
was noted earlier that freshly milled refuse is bulkier than unprocessed
refuse it is apparent that some compaction of the milled refuse occurred
between milling and initial density tests. This compaction is due to
the weight of the compression plates in this test and to the tendency of
milled refuse to become more dense with time under its own weight. At 5
psi and with no vibrations, the actual refuse density for the milled
material was 755 Ibs. per cu. yd., while the unprocessed refuse density
was 500 Ibs. per cu. yd. With medium vibrations (displacement of approx-
imately 20 mils), the actual refuse densities at 5 psi were 028 Ibs. per
cu. yd. for the milled refuse and 560 Ibs. per cu. yd. for the unprocessed
refuse, all on a wet weight basis.
The average pressure exerted by a 35,000-1b. D-7 Caterpillar tractor
is 7.44 Ibs. per sq. in. An extrapolation from these test results at the
level of vibration closest to that measured in the landfill indicates that
the expected density of refuse corppacted^ijbh a D-7 in a landfill would be
WO Ibs per cu. ydT for milled refuse and 660 Ibs. per cu. ytf. Tor
unprocessed refuse on a wet weight basis.
Field Tests:
Field determinations of actual and effective refuse densities were
conducted which may be compared with the laboratory results. The first field
tests were completed in 19(17 and 1968 at the 01 in Avenue site. All cells,
both milled and unprocessed, were compacted in 2 ft. lifts to a 6-foot depth by
a D-7 Caterpillar tractor. These tests indicated that milled refuse in a
landfill situation has an average actual density of 930 Ibs. ner cu. yd.,
while unprocessed refuse has an average actual density of 31" Ibs. -i^.r cu.
yd. and an average effective density of 570 Ihs. per cu. vd., ::1<
-------
Q«VA 01803 M3d
- A1ISN3Q
c
o
5
0>
>>
CO
o>
o.
in
CM
o>
en
53
-------
on a wet weight !),•• is,
in formula Form ir Table
savings, the effective re-
they include the volume •:;•!'
Results of t:ie 01 •' >
because of the lac*, of cr>,
material usage, mo;:, tun
compa ri s on. T he - >•> for c1,
densities of milled urn.1.
effort and to compare ';•<•
placed under "a-Jeni-aLe" :•
1 andf i 11 mac!, i nc ooera v. '
adjacent to the city's r
filled with approximate-
with a similar weiqht tv.
refuse used in tin's t-:-"
the separation slujio? .
cells v/as col l-icto by ,.: •
entire city, '.'•:
and August, mor.t
normally exn-i
between 37 an..! ,'/• p^r
was pi ace-i on ! i." <• -i
ted to place os;ly ,:io
that, wit.h equal .-•-•n
density than t'i.i>rr, • s
these tes Ls n«y ':c «. h
careful considerai '< n
under field cori'ii t i
qreater effcctiv ' xn
this differs,ic' :n "
more commonly 1,0; t ;• - r»
by "bridqinn" . ' '<
Ti - ; i ;c rnfi. • o. Allies are c'ofined
•' "/' i'"ir jiO'v'1 of ;.'if iiiarinq i andf ill space
••rr, • «,y t'iqurrs arr-.1 uore '"saninqful because
- ':;<.; "trial used,
^':-'i. -"oni-ily ',.•:.; • .;ore in uuestion, however,
. ••• • -,,rh r/.-prciis a"- C'J'"iMcti(jn tine,, cover
' • ;- rrh.'ie, -.'i.i :-: u.e (•( ttie cells used for
••• ; : •jf::. n noJ c;- c\n i; a*-£ the in-place
-'-• rr -.,:;;: i'-lai:eo v," fn onual compactive
"... i •. : s o! mi'i u-:J rit.c1 unprocessed refuse
• '.--• '. ' i-\ pyr-'.t'. ! '.'•/ : oxn^i'iVncnd sanitary
"o ,.' • ' •.:>. ;/•'. .,x'!s ''.'.Tf ..-xcavated
;"•"'! si lc/'"! ;',"• ..T{ "i" ) ,. Mne cell v;as
je',, r.ne other two
Vie composition of
,,s detemined by
.' in fil'linq the
i-ir: rurs over the
vore filled in July
C'.'P"" ,.".. it. ion is
MU refuse varied
;! ccver of soil
of.rrator v/as instruc-
d under ordinary
e -..
,ve'"; <>rn; ;crnpacted
!!-•., !'o ,,1' !nu'i ate a
>v. • a.'I} ., or a total
• |"".trorrHkic tion
fuse densi ty o"" the
•-il rf'n^e density
,'iroh ("• ? i'et weight
••, hr.(,i an actual
{•••'• biisi? after it had
•if; to ai-r.'jt ^ percent
11 I?}.
.'••*ct.ory estimates
••h if ioontly higher
' '€' > ')!'ildined in
:qures do merit
"",,, "illled refuse
.i-o-'us"' density
;'i-rc(Mil. qreater
'" 1 ,IP .s'-'tfin ce when
,;: ; -• •'.'' percent
•5 •'."• reason for
'.i'VOv ,. ;sed rofuse
, '',.ru'.:«i"J cans or
-------
TABLE 11
OLIN AVENUE FIELD DENSITY TESTS (1967-68)
Actual* Effective**
Cell No. Refuse Cover Refuse Refuse Refuse
Volume
(cu. yd.)
MILLED CELLS
1
3
9
10
11
15
17
21
UNPROCESSED CELLS
4A
4B
12
18
22
1120
1500
1090
1820
2090
1370
1340
1250
3010
1440
4390
2865
1245
Volume Weight
(cu. yd.) (wet tons)
534
703
474
6B()
844
617
733
709
AVE.
670 1045
550 683
2190 1456
1455 1268
635 548
AVE.
Density
(Ib./cu. yd.)
950
940
870
760
810
900
1090
1130
930
700
950
660
880
880
810
Density
(Ib./cu. yd.)
950
940
870
760
810
900
1090
1130
930
570
690
440
590
570
570
* Actual Refuse Density = Wt. Refuse/Vol. Refuse
** Effective Refuse Density = Wt. Refuse/(Vol. Refuse + Vol. Cover)
55
-------
fe»WSf« '&*>•*$•*]$.^ "-t"'&;'!'.'',|$4*'if li-'l^'-'rf!";* y""1",
**^,•**$?* ^ ^>-%'j^ ^^if*- 4|Ji" ^i$ ' l£f ^ ^IHk A ^\'i-'' Mr't^i/ '•'nfe, -
' '!&*' -' "'! ;' tf^S*feb-| , v' ,%**' 'cJ>'\"". J'-'W'i*' W^'' 2|%«'"""'•
> '-t .- }. . v, ji'a, ~yt' 'JSfo:':-'.i^'< •• ^,,4*, <¥* r ™:-"*-v-yPl* K i i§H*<% :
'*>, •«? "•'-.,, o4H!> ' .lia^C*' * ;• y. . ?•-'<• ASK-1S ^*is *lfe **!*,,.
u
c
o
o
c
re
(U
X -M
X)
•M O
U i.
3 Q.
t- c:
4-» 3
U1
c -a
o c
u > o
( H~
3
O VI
I 4J
O Vi
cr> CD
O 4->
CXI
>>
OJ -*->
OJ -r-
S-. in
jc. c
+-> CD
"O
<4-
O -O
cy a;
c: -r-
o s-
to
CXI
cu
s~
3
CJ>
-------
"v *•, -W-
'>?.', .. '"<*
I? •• . ,J *
* • • * * '* »
t *w
S» ,' • - o
'-"'' ' :' '•'''••'-••>titv f '•
:^.H^V-v; V- -*•
t* * »
- T :- »•"'-"<
VI
4J
01
0>
c
•r—
x
"O
O)
IX
CVJ
OJ
3
O)
57
-------
s_
o
Q.
O
U
-o
O)
'oi
-------
cw
r—
UJ
_l
CO
-
1—
i— i
CO
•y*
UJ
o
o
_J
UJ
h-l
1 1
X
•
>rtl -t 1
w t-'
f- «/) H-
+> Z3 W
O <4- C
0)0)0)
«4_ o; o
<+_
UJ
$ >,
r— T-
3 3 to
4-> «t- G
O O) O!
«t ct: o
O)
to
«*-
0)
o:
(U
to
M-
0)
o;
i.
o
0
0)
CO
3
«*-
03
C£
C
C
4J
0
«*— *»
•
•o >-
,>,g
LJ
•
3
O
25
•
TD >-
"7°
3
0
«^|—
• UJ
J3 3
* —
*~-H
to
-!-> C
-C O
cn-M
•r-
>
3 S-
TJ
to
4-> C
J= O
C7HJ
•r™
(U •*->
3 tt)
S
•K
i— O)
re E
•»-> 3
O r—
1— 0
•
01 -o
E >>
f— •
O 3
> O
^-.
•
0) T3
E >^
f— •
O 3
> 0
QJ *
E w
UJ 1- i-
Q.I- IT
E
O
O
^wX
o>
O.
>>
H-
^-
f ' '
0) 0
CJZ
«•
CM
|^^^
I^V
O
o
o
0
i—
CO
cn
r^
FH—
fHK
UD
O
CO
•!*•
r—
r-«*
<•
CM
CM
CM
ID
CO
CM
CM
O>
cn
r*"~
un
•
^~
CM
"C5
qj
in
to
OJ
u
o
J_
o.
C
Z3
r_
co
0§
VJ
CD
^f
CO
«sj-
co
cn
in
CM
•«a-
m
<*
r^
CD
CO
r—
r_
CD
CO
to
O
CO
CO
CD
in
pw
r>.
•
co
CM
•o
OJ
r—
^«
•r—
s:
CM
co
CD
^*
0
CM
«3-
0
CO
CO
r-.
CM
^**
CD
r*.
rx
to
CO
CM
r—
CO
CM
O
CM
«5f
in
r—
CO
CO
cn
r—
f _
•
sf
2
^^
•fW
z
CO
0)
to
3
•4—
O)
O)
U
(O
O.
1
C
*r~
O)
^~
-L_
•*->
s~
o
o
T3
C
«o
0)
"o
>
(U
to
3
U-
O)
S-
Ol
(—•
_A__
•!->
^_
(O
•^
^j
CT
OJ
•M
0
c:
U)
0)
o
TJ
CJ
E
3
"o
>
"TO
4-*
O
H*
*
•
TJ
O)
4->
U
4->
IT)
5-
O)
>
O
o
to
fO
O)
to
3
*4—
O
s.
(11
J=
4->
O
•M
C
•r-
S-
C
•r~
•P
<4-
«^-
to
-o
(O
-o
O) •
4^ f*~
O r—
to
Q. (U
Er-
O -Q
U
l|— r—
•r*
(/) (O
«O 4J
3 0)
en
t
59
-------
CD
>
CJ> I- O •!-
c o o>
fO M- C
J= C «»- 0)
O >— i UJ O
O
O
•
CM
CM
a:
Q
UJ
CU
•r- >>
CU +J +J +J
cn —i UJ Q
O
o>
V
•a-
co
CD
C
i— i CM
i- £:
i— < en
CT> i- (O «r-
C C 3 t/1
re 4-> c
^: c u cu
O 1-1
UJ "Z.
Q UJ
Q
°3
LU UJ
££
UJ
CJ
CJ -4-> i— -M
CT »-
»
CSJ
CM
co
CSJ
V)
«o
JD
ex
•o
OJ
to
I/I
a>
u
o
CL
c
o
o
a.
c
«— o
CM
CO
•p
c
o
o
OJ
o
*
60
-------
Settlement
Plates to measure settlement of milled and unprocessed refuse were
installed in the oldest Olin Avenue test cells and the lysimeter beds,
but accurate readings over long periods have proven difficult and the
results inconclusive. Factors v/hich have made it impossible to draw
conclusions from settlement data include changing refuse composition and
noncomparable degrees of compaction.
Observations by project personnel have been that miUed -refuse settles
less than unprocessed refuse, undoubtedly due to the higher initial density
and less "bridging" of milled refuse. Attempts were made to recompact a
milled refuse cell at theOlin Avenue site after 1 1/2 years, but very
little additional compaction was achieved. This again indicates the relative
stability of milled refuse in a landfill situation.
Observations will be continued on existing refuse cells at Madison in
an attempt to make quantitative determinations of settlement of milled
and unprocessed refuse.
Decomposition
The top foot or two of a milled refuse cell that has been in a landfill
for several months begins to decompose to a browh, mulchlike material,
except for stable items such as plastic and glass, which remain intact.
The rate and extent of degradation, as well as the rate and extent of
removal of matter from milled and unprocessed refuse in a landfill, are
dependent on a variety of interrelated factors. The two basic mechanisms
resulting in a change of appearance of refuse and in removal of matter
from refuse are: 1) physical or chemical leaching and 2) biological
decomposition.
Physical and chemical leaching is brought about by the flow of water
which rinses matter from the refuse. Biological decomposition refers to
the degradation of refuse to Teachable matter, gas, or more stable decom-
position products by biological activity. These two basic mechanisms are
dependent on the following:
1. Presence of water
Water falling on a cell of milled or unprocessed refuse, covered or
uncovered, will either run off the surface, evaporate back into the
atmosphere, or infiltrate downward into the cell. Water which infiltrates
will increase the moisture content of the surface layer until that layer
can hold no more moisture. After that, the addition of more water to the top
layer will cause some water to flow to the next lower layer. As the process
continues, more layers of refuse become saturated (or reach what is defined
as field capacity) until the cell can hold no more water. At this point,
any additional water added to the cell will displace a like amount of water
into the soil or bedrock beneath the cell as leacha'te. Water flowing
through a refuse cell in this manner, plus the water originally landfilled
with the refuse, brings about physical and chemical leaching and is a
prerequisite for biological activity. The amount of water present in the
refuse at the tine of landfillinq is generally well below field capacity and
so will not by itself result in leachate production; however, it does
commonly support initial decomposition processes.
61
-------
It is important to observe that the rate of increase of moisture
content to field capacity, and the amount of leachate neneratod, are functions
of the surface characteristics of a refuse cell or landfill. The field
capacity of refuse v/ould be expected to be basically the same whether it is
unprocessed or milled. Therefore it follows that volume production of leachate
is dependent more on the surface characteristics of the landfill than
whether or not the refuse is processed.
While it is possible to increase runoff from a refuse cell by using
clay and silt cover material and by controlling slope to avoid ponding, it
is almost impossible in humid climates to prevent infiltration short of
employing artificial barriers such as plastic sheeting. With milled refuse
left uncovered, therefore, it may be predicted that there will be little
runoff and considerable infiltration. It may also be predicted that the
presence of paper particles on the surface of such a cell will act to
promote evaporation, offsetting in part the increased infiltration and
reducing the amount of leachate.
2. Temperature
The greater the temperature within a refuse cell, the more quickly
biological activity proceeds. The ambient temperature is important as
it modifies the refuse temperature.
3. Presence or absence of air
Aerobic decomposition,, which takes place in the presence of air, is the
characterized by rapid activity which produces sufficient heat to raise
refuse temperature as much as 30° to 40° F above ambient at a depth of 6 feet.
If the rate of oxygen use exceeds the rate of replenishment, the refuse
cell becomes anaerobic and a new group of organisms predominates. In the
first stage of anaerobic decomposition, organisms which can tolerate the
presence of some oxygen begin the decomposition of organic matter. Thus,
partially decomposed organic matter is made available to the leachate,
resulting in hinh levels of chemical oxygen demand. Since some of these
organics are acidic, pll drops, and some inorganic matter is solubilized.
As decomposition proceeds further, all oxygen is depleted and methane-
forming bacteria predominate. These organisms decompose organic matter
more completely to methane and carbon dioxide. At this point the chemical
oxygen demand of the leachate decreases and the pH rises. This second stage
of anaerobic decomposition is commonly associated with little or no
temperature rise.
4. Effect of milling and mixing
Milling is thought to enhance the rate of physical-chemical leaching
and biological decomposition by increasing the surface area of the refuse.
Also, the mixing produced by milling reduces pockets of relative
inactivity that are common in piles of unprocessed refuse. Further, water
flows more evenly through the entire volume or refuse if it is milled,
raiher than flowinn through channels as it often does in unprocessed refuse.
These factors lead to more uniform and rapid decomposition of milled refuse.
With these basic mechanisms of decomposition in mind., studies were
undertaken to determine the effect pt decomposing mil led refuse on the
environment as well as the effect milling has on the rate of stabilization
of refuse. Three major studies wore conducted — two under field conditions
62
-------
and one under controlled conditions at the University of Wisconsin-Madison
Biotron (a controlled-environment test facility). The results of these studies
are discussed in the following two sections.
Leacliate
01i n Avenue Tests:
The first set of tests to evaluate the comparative effects of milled
and unprocessed refuse with respect to leachate production was performed usino
the test cells at the nlin Avenue site. These cells were constructed between
October lf;67 and October 1968. Unprocessed refuse cells were covered with at
least 6 inches of soil; the milled refuse cells were not covered. Tn addition,
four special cells v/ere built to represent the extremes in landfill situations.
Two of these cells were composed of specially collected garbage -- one
unprocessed and covered, the other milled and not covered except for the sides.
The other two cells were composed of rubbish -- again, one unprocessed and
covered and the other milled and not covered.
Fifteen cells, including the four special cells, were equipped to
collect leachate. A plastic sheet was placed under portions of each of
these cells. Leachate was channeled into collection reservoirs by proper
contouring of the plastic sheets. Leachate was sampled by pumping from the
reservoirs. In obtaining a sample, the first 2 liters were discarded, and
the sample was obtained from the second 2 liters. The reservoirs were then
Dumped dry and the total volume noted.
Data obtained in the .field included the amount of leachate pumped
(in milliliters) and the temperature of the leachate (in degrees centigrade).
Leachate commonly contains a wide varietyiof contaminants and .other
substances, often in large concentrations. It was decided to categorize the
most important substances and to determine their presence as a group
wherever possible. Certain other parameters were analyzed because of their
individual importance. These other parameters included chlorides, total
and calcium hardness, alkalinity, iron, nitrogen, phosphates, and pH.
The two basic analyses performed on the leachate, however, were chemical
oxygen demand (COD) and specific conductance. COD is a widely used
evaluation of the amount of oxygen needed to oxidize chemically the matter
in a v/ater sample. As used in these studies, the COD test results were
primarily related to the amount of organic matter in the leachate, although
inorganics can add to the COD in anaerobic waters. Biological oxvgen demand
(POD) was not measured because of problems of precision at the very high
dilutions require,, because BOD is not as good a measure of organic
content as is COD, and because of reservations about the usefulness of B^D
results in the leachate system.
Specific conductance, also called conductance, is a gross indicator
of the total concentration of dissolved inorganic matter, or ions. Thus,
conductivity anJ COD roughly measure the inorganic and organic content,
respectively, of a water sample reasonably free of particulate matter.
The presence of particulates was observed to vary widely depending on how
a sample was taken. For example, the exact location of the sampling hose
markedly influenced the amount of sediment or particulates pumped out.
Thus, all samples were allowed to settle for 4 hours before the supernatant
was drawn for analysis. This procedure minimized the effect of changes in
particulates resulting from sampling. The settled sample probably represents
63
-------
more closely the quality of leachate leaving a landfill site since
undoubtedly only the finest participates are not removed from leachate
during passage through the first layer of soil.
the Olin Avenue cells differed in size, thus making comparisons of
leachate data from the various cells difficult. Different amounts of
leachate were collected from cell to cell, and the actual amount
collected from a single cell may not have been representative of that
cell's production. Because of this, neither a determination of the water
budget nor a presentation of data in terms of volume of leachate per
volume or weight of refuse could be made from the Olin Avenue studies.
Some of these difficulties were overcome when similar tests were run under
more controlled conditions. These studies will be discussed later.
Four covered cells of unprocessed refuse were instrumented for these
studies, but only two provided useful information on leachate. Similarly,
of seven milled uncovered refuse cells originally intended for the study,
results from only two could be used. The remaining cells had leaks in
the plastic sheet, were damaged during subsequent landfill operations, etc.
Figure 29 summarizes the leachate production for the four useful cells.
For reasons discussed previously, the rates of leachate production cannot
be strictly compared from cell to cell, but the shapes of the curves for
each are thought to be significant. The production rates for all four cells
seem to have stabilized early in 1969, approximately 1 to 1 1/2 years after
placement. The two unprocessed cells produced leachate at about the same
rate as did the two milled cells until cell 3 was covered. The rates of
production increased in all cases during wot periods of the year and
decreased during the winter. The effect of cover is seen in the curve for
cell 3, where the rate of production was similar to its sister cell, cell 2,
until cover was applied to cell 3, 2 1/2 years after construction. All that
time the rate of leachate production dropped in cell 3.
COD concentration curves are shown in Figure 30. With the unproces-
sed refuse cells, COD concentration was cyclic, with peaks in the summer
months and valleys in the winter. Also, COD levels remained fairly high
throughout the test period, showing no consistent rise or fall over the 2 years
of data. Mote that data collection stopped when landfill operations
encroached upon and finally covered these two test cells. With the
milled refuse cells, the COD concentration began at relatively high values
but dropped ram'dly, with ninor seasonal fluctuations during the period,
to consistently low levols. The rise in COD for cell 2 in the summer of
1970 should be discounted, because large amounts of water were applied
inadvertently to this cell, rinsing out unusally large amounts of COD
over this period. Conductivity levels closely mirrored the COD concentrations
in both types of cells.
Figure 31 shows the leachate production from the special garbage
and rubbish cells. Tho garbage cells quickly reached a stable level of
Icaciiate production, while it took nearly a year longer for the rublish
cells to produce loachate steadily. Comparing first the milled uncovered
cells of each type, the paper fraction is evidently important in nromoting
evaporation and retarding the downward flow of leachate, because the milled
rubbish cell produced considerably less leachate than did the milled
garbage coll. The paper content may explain to some extent the differences
in production rate for the two special unprocessed covered cells, since
64
-------
V)
»J
_J
I*J
o
s
p
il
g
UJ
2
H
O
O
o
Id
|
O
<
til
*J
1 1 j
ir
1 8 *
ul
tli
to
x:
o
re
01
a;
i-
65
-------
66
-------
o
•M
(/(
Q>
I—
3
C
OJ
o
c
«J
4J
QJ
O
10
•r~
JD
3
J-
OJ
en
to
S-
(O
-a
-------
paper has the nullity to hold noisture closer to the cover-snl -j,1 uaste
interface for eventual evaporation. Another explanation, perhaps of some
importance, for tho difference in leachate production rates anonn the
special cells is the differences (such as size and presence of leaks) in
the plastic sheeting underlying the cells.
The COD concentration data from the four special cells are summarized
in Figure 32. The garbage cells showed very high COD concentrations compared
to the rubbish cells (note the shift in scale). The concentrations for
both garbage cells began very high, but rapidly dropped off to low values.
In contrast, the rubbish cells produced leachate considerably lower in COD, and
did not exhibit as sharp a dropoff in COD concentrations as did the
garbage cells. These results are in line with the putrescible character of
garbage and vn'th the high paper content of rubbish.
As was the case with the refuse cells discussed earlier, the special
garbage arid refuse cells show that milling promotes decomposition. Thus,
the concentration of COD was stabilized at a low level more quickly in the
milled, uncovered cells than in the unprocessed, covered cells. A period
of only 1 1/2 years was sufficient to stabilize the milled garbage cell to
the point that consistently low COD concentrations were produced in the
leachate. In contrast, the milled rubbish cell produced relatively low
levels of COfj from the beginning. It is doubtful that either the
unprocessed garbage- cell or the unprocessed rubbish cell has stabilized
to any great extent. They were still producing significant COD levels
nearly as high as the earlier peak levels in the last summer of testing,
and probably voulJ have continued to exhibit summer rises for several years
had monitoring continued, ''ote that data collection on cell 3 was
terminate.', a year early because of the encroachment of landfill operations.
Thus, the Olin Avenue tests strongly suggested that organic material
is removed from milled refuse by leachate at a faster rate and over a shorter
period than from unprocessed refuse. The unprocessed cells showed mon-
striking rises and falls in organic removal with seasonal changes. While
the milled cells generally produced more COD the first year, comparable
unprocessed c^-lls prouucod larger amounts of COD in subsequent ycarc.
Decomoosition appears to proceed more rapidly in milled cells,
quickly reaching a steady state in which few Teachable substances are
released. During the initial high period of COD production in milled
refuse, a corresponding low pH sugnests that aerobic and first-stage
anaerobic decomposition Is resulting in partially decomposed and Teachable
organic substances, including orqanic acids. A few months later, the pH
rises to near neutrality and the COD drops, suggesting that anaerobic
decomposition is predominant, producing more methane and more completely
dcgra.'ino t.'ie orqanic matter. CDC and pH curves for unprocessed cells
indicate that tiiose colls did not reach second-stage anaerobic decomposi tio:
to any groat extent durinn the period of the Olin Avenue tests.
Conductivity levels showed that inorganics are also removed more
quickly from milled than from unprocessed refuse, although the difference
is no!, as pronounced as in the case of COD. Except in the case of the
special colls, a.nductivity and other specific ion curves closely followed
the; shapes of the related COD concentration curves.
Finally, lechate production rates seemed to be higher for the milled
refuse cells t;ian for tho unprocessed refuse colls. This difference in
leachate production was nrobably more a function of whether the refuse1
68
-------
CD
QNVWJfl
69
-------
was covered tvl!.i>r th,',<; whether i' './-is Mil;.'',. ."'K i\.)it. I T. h-n , )', f.'i/ru
leachate production rervni' tentative, !iii\?> \"T,, LOC.-UIV ,;n >c>v;v .' wtrr
balance coul'l not b.° d''4,e;-Tnned for the 'I'lri ,">vemie tost cells. •.uLsjeu.'nt
studies wepv.1 directed in part to provide nore accurate water b.-i.mce in for-
mat ion.
Lysimeter Studies:
To overcome some of the limitations imposed by the conditions of the
Olin Avenue test, another study was initiated. The object of this project
was to isolate various refuse cells from outside influences such as those
that limited the usefulness of some of the 01 in Avenue cells. ", secord
objective was to obtain a direct comparison of pollutiinal loads from
milled and unprocessed refuse. Third, the effect of cover soil on
degradation aed pollution loads fron refuse was to be letermined. finally,
a water budqt.t was to fo kepi, to determine the percont^nes o^ precipitatio
v/hich qo to runoff, evaporation, and infiltration (leach.'t'"1).
Four cells of equal size were constructed in two obsolete sludge dryinn
beds at the Oscar '"aver Opinany's waste treatnonb plant i»- Madison, (Ticures
33, 34 and 35). The colV, '.vre" filled io vjiJ-Seoterlicr l(-7 i as
follows :
#1 unprocessed refuse covered with approximately 6 inches of
soil (Figure 36};
#2 milled refuse covered with approximately 6 inches of soil;
#3 milled refuse covereci after approximately 6 months;
#4 milled refuse left uncovered (Figure 37},
Each cell contained about 100 tons of residential and light
corcuercifvl rr-f'ise colV-cted "itliin ;\ one-ve.ek ;i:ri'-ii:?i?e any !iri"orences in
refuse c^innsi tic:, fror c°l 1 to c^l'i. !'',r>'Nroy.imately 7r. tons of silty-
sai.d cover '.".'re "i.t..HJ;i''n vis "'i'1 to ir'c.'.r"e t! :,t ri-rrrf .--"id all lea civ to •,•••? re
collect;}1 (fin-ire 3B). Uas r;rc-Juction (riiscusso;! in tlse '"ol lo',.'1n^ socticv),
'cei.ii.c.ratun., ind Moist'-^e \'-T'': ?"'sc -ioni t,nrcv;.
Data on leachate production are summarized in Fifjure 39. Production
(Yon all (;t "'.'•(• C'lls, but -7rr.n C^i urr.ro^csscd cover';' roll i r, ^articular,
varied '-.'itii thi seasons, j^a1'!;'" in s'irinr, ,-ii.d fall "i Hi ',!;"• onset u!" tiiavs
an! liO'ivy rai^s ,ind dT'i'-oin" o^f i,; s'!"ii!;r -"uv" winter, "-'ter reicin'rn
field cana.il1', tuce niili^r! uncovered C'-M! p.roou'ce- 1^-tcha.;'1 at ^. In'^iier
rate than the unprocessed covered call. Drastu: month-to-month fluctuations
in I'-.ae.firUG p-ro,' 'C'.inp ^r-^i his c'll •p'ii'-.nte ~chat le>ic '•.:"!:c "reduction
fro.: UMCOV TO--' "liVls:1"! ri!r'is<' ii 'nr^ dp'-.i-.i ls?nt, on sh^rt t< rn p"iNCtuatie. s
in incident rai nfa i1.
The other milled refuse cell that was linf/iany left uncovered
Mro.'.uceH loach it-, at a "'-•''( -on^istont wi Vi t.:-- fir-sf; uncovered fail! ed cell.
However, this cell was cove^e-' '.ii/. C '. rchos uT .;e Serial in "-. e-r-'ered i^-iodi.\roly ^Iso
ttxhi bite.' I'vc'ial.-- :v.-o.' ict,'i"r '•-P ^ i 1 .-•••- , r- ,^ ,_ ^..-ocerser' -over;1 ' cell.
It may be concluded fror; t.1 ose a.. .: L^J'; soi . cover has a major
impact OK lenchai '.roe-tc ;.i •'... ar.fl I'1'-. dn.. >M ": 1 be -Ms^vss^d farther
in later portirnr, ^f P' i'- r"1--r\ .
vn
-------
.75
'A-
o
-------
1
.
z o o
uj 2 t
O Q O
u.
in
Q
DQ
CVJ
Q
§
1
CO
^co
\
u
o
o
®\
I )?>
tn
o
o
t£
UJ
©
o
UJ
O
UJ
OD
(T
UJ
1
o
>>
(O
o>
.n
s_
CD
+->
-------
OJ
I/)
3
M-
OJ
s_
-a
OJ
QJ
U
o
S-
D-
-a
ai
en
c:
OJ
QJ
0)
-t-J
ai
ro
Oi
-------
O)
I/)
3
s_
-O
ID
s;
-X3
O)
c
•r-
01
T3
Ol
fc-
Ol
4-J
3
O)
-------
0)
o
-------
200
IOO
zoo
too
U
<
o
o
a: N*
CL g 200
III
LJ n
h- J
IOO-
X
o
LJ
-J
200
100
CELL 1 - UNPROCESSED .COVERED
i M t » i ? T i i M r i i i i i t r i M-
CELL £-l*»LLI0f COVt«IO
i n I »i r 11 Ti I T i 111 i 11 iT-1
-------
The COD concentration levels for the cells are shown in Figure 40.
Note first that the general shapes of the curves for cells 1 and 4 match
very well v/ith the corresponding curves for the 01 in Avenue leachate
studies. The unprocessed covered cell at the Oscar Mayer site showed a
slight initial COO rise, possibly indicatinn the release of some readily
leachable^orgam'cs. The COD concentration then dropped to lower levels
until spring when a large volume of water and rising temperature combined
to bring about a substantial removal of organics. thereafter, the
general trend of the curves is slowly downward, interrupted occasionally
by especially uet periods.
Data from the Olin Avenue study indicated that the typical milled
cell had a high initial peak of COO concentration followed by a decline
to low levels, and that a smaller secondary peak usually occurred the
following summer. In the lysimeter studies at the Oscar :1ayer site, the
milled uncovered cell also showed such a curve. The same characteristics
were observed in the milled cell which v.-as covered after G months, except
the initial and secondary COD peaks lasted longer than in the milled
uncovered cell. Hven more apparent was the heightening and prolonging
of the pea!, levels of COO concentration in the milled cell covered
immediately. Again, tli^ presence or absence of soil cover has a pronounced
effect on the degradation and pollutant production from milled refuse,
an effect which '..'ill be analyzed later.
Obviously, COD"co'nc^ntration is not in itself of great importance in
its effect on the environment, since a small amount of leachate with a high
COD concentration may be no more damaging than a large volume of leachate
at a low COD concentration. Thus, to obtain the actual amounts of COD
substances produced from a cell, the COD concentration value is multiplied
by the average volume of leachate produced per day between samplings.
COD production is shown in Figure 41.
Generally, the COD production curve followed the COD concentration
curve in the case of the unprocessed covered cell. With all the milled
cells, COD production curves rose to a peak very quickly after the cells
reached field capacity. The peak for the covered milled cell was not as
high as the other two milled cells, but later COD production rates for
this cell were higher than for any of the other cells. The excessive
peak in April 1971 for the milled cell covered after 6 months was un-
doubtedly due to a large extent to squeezing of leachate out of the cell
by heavy equipment during covering operations in Maneh. This cell and
the milled uncovered cell had initial peaks which indicate that COD-
producing organics were being removed from these cells quite rapidly.
Note that for the unprocessed covered cell, little trend upward or
downward is observable in the COD production curve. In contrast, the
general trend of all three milled cells is toward lower COD production
with time; the weakest such trend is for the milled cell covered
immediately and the strongest for the cell left uncovered. There is a
second summer rise for the milled uncovered cell; the second summer
rise will probably by very weak or not present in subsequent summers if
the Oscar Mayer lysimeter results continue to correspond to the Olin
Avenue findings.
78
-------
Q
12
CELL I - U*P*OCt*$t 0 COVEftCi)
i i i i i n ii i HI » ? i r i in r i
CELLlHMLtED,
22*4-
O 30
<
UJ
24
18
12
6
III I Ml M it iff fill I i I 1 I I I I I
CCLL 3-lr,>.LED, COVE RED -
3/22/71 _
I I I I T I! T T IT
T I T I I I I I I I
NOT
Figure 40.
ITT FT TTT I I TTTi MM T rTTTl TT
NDJFMAMJJASONDJFMAMJJASOWD
I Itfl I ft ft
COD concentration curves for lyslmeters.
79
-------
CELL 1 - UNPROCESSED, COVERED
MM?!
CELL a-
P I I I M M M M if II If M
GILL S-
M M 1
M M M M 1
CELL 4
, WOT ammo-
TIT! I! i rrri M n
NDJFMAMJJASONDJi-MAMJJ A £ 0 N I)
f
Figure 41. COD production frcr.i iy:,1".Deters
-------
The COD production observations can be explained by the nature of
milled refuse and the effect of cover. Initially in milled refuse cells with-
out cover, more precipitation infiltrates the refuse and becomes leachate.
The finer particles in the milled cell present more surface area to the
infiltrating water. Together these two factors result in more grams of
COD per day being produced at first in the milled uncovered cells than in
the other cells. At the same time, the increased amount of leachate tends
to lower the COD concentration. As time goes on, the greater flow of
leachate in the uncovered milled cell has less effect on grams of COD
produced because the readily degradable and Teachable matter has already
been removed.
With unprocessed refuse, organic matter particle size is much larger
and organic matter is not as well mixed as in the milled cells. In
addition, cover material reduces infiltration in the unprocessed cell,
with the result that less organic matter is initially leached from this
type of cell.
Curves for conductance and specific ions closely followed the COD
concentration curves and thus will not be discussed further.
The pH measurement is of interest because it can be used as an
indication of what type of degradation process is occurring. Once
second-stage anaerobic degradation occurs, the pH should rise to near-
neutral levels, the COD production should decrease, and methane should
be produced. Acidic pH levels do not promote degradation because of
their adverse effects on many microorganisms. Acid pH values are found
during the transition between aerobic and anaerobic conditions.
The pH in the unprocessed covered cell has remained acidic
throughout the period of this report (Figure 42). This indicates that
organic acids are being produced and that organics are not being
reduced to their highest state of degradation. Thus, either the refuse
itself or the decomposition processes have not reached stable, relatively
harmless conditions.
The striking effect of soil cover on the degradation process is
observed in the pH data from the milled cells. The covered cell main-
tained an acidic pH, whereas the other two cells more closely approached
neutrality (pH = 7). The leachate from the uncovered milled cell has
been less acidic than that from any of the other cells. This, together
with the fact that methane production has begun within this study period
in the two cells originally uncovered, indicates that the degradation
process has become relatively stable in these two cells. This is further
indicated by the low COD levels being produced in these cells during the
last year of monitoring.
Water Budget:
The water budget for the four cells was determined for the period
from May 1971 to May 1972 (Table 14). This period is used because by this
time the cells were behaving consistently — that is, all the cells had
reached field capacity and were producing leachate regularly.
It is interesting that the evaporation percentages are nearly equal
for all four cells, indicating that approximately 32 percent of incident
precipitation becomes either runoff or leachate after field capacity has
been reached. The effect of cover is to divide this 32 percent into
approximately equal percentages of leachate and runoff. Without soil cover
a larger amount of precipitation infiltrates and becomes leachate, while
the evaporation rate increases slightly. As the surface attains more of
a paper mache, and eventually a soil-like texture, the runoff percentage
for this cell should increase somewhat.
81
-------
CILL 1 - UNPROCESSED, COVERED
o
<-> 8
O 7
6
^ *
O 5
O
o:
£.£
fc *
§1
5
P
HI 8
f» *
7
6
5
Tl II 1 1 i I I ffI f I I ll 1 M I I 1 "I ! 1
CELL 2 - MLLCD , COVERED
TIT T I I TT I IT T I I 1 1T I! I T I I M I
CELL 3 - M ILLEDt CO\CftC0 3/22/TI
T ITTT I 1
CELL 4
I I I I I I I i
-MILLED,
1 I I 1 1 I I I
COVERED
/Tytfl
TTTT I
NO J F M
I I I I 1 I I I I ! II i I ! I I ! 1 I
W4J ASONDJ FMAI^UJ ASOND
I 1972
Figure 42. pH curves for r/s:mpters.
80
-------
Direct comparison of the unprocessed covered cell and the milled
uncovered cell corresponds quite well with the results from the 01 in Avenue
study. In addition, some further results of the present study include the
fact that leachate production occurs at a faster rate in uncovered
cells than in covered cells whether they be milled or unprocessed. While
milling and lack of cover do promote a rapid achievement of a mature
degradation system, both factors also allow larger quantities of organic
matter to be leached out before formation of such a mature system has a
chance to develop.
TABLE 14
WATER BUDGET
(MAY 1971 - MAY 1972}
Precipitation
(measured)
Liters
Runoff
(measured)
Liters %*
Leachate
(measured)
Liters %*
Evaporation
(by difference)
Liters %*
Cell 1
Cell 2
Cell 3
Cell 4
104200
104200
104200
104200
15070 14.5 19018 18.2
15800 15.1 17551 16.8
13670 13.1 19249 18.4
288 0.3 28922 27.7
70112 67.3
70849 68.1
71281 68.5
74990 72.0
*Percent of precipitation
83
-------
Biotron and Gas Composition Studies
The Olin Avenue landfill and Oscar Mayer lysimeter studies produced
valuable results, but they both had one drawback -- unpredictable and
uncontrollable weather conditions. Thus a study of decomposition of
milled and unprocessed refuse without soil cover and under identical
conditions was undertaken in the University of Wisconsin-Madison Biotron.
The Biotron is a controlled-environment facility in which a computer
regulates and records parameters such as rainfall, light, humidity, and
temperature.
Approximately 1,200 Ibs. of each type of refuse were compacted under
10,000 Ibs. pressure into two specially designed containers which were
then placed in two separate but identical test chambers programmed to
simulated a hot, humid climate with high rainfall (Figure 43). The test
ran for 270 days, during which time data were collected from the refuse
itself and from the leachate and gas produced. It is important to stress
that the Biotron study differed significantly from the field studies in
that the unprocessed refuse was not covered.
Both cells exhibited peak temperatures in the upper layer of refuse
after one week. A rise in temperature deeper in the refuse beds
occurred later and was closely related to the depth that moisture had
penetrated into the cell. The highest temperature recorded was 102.0
degrees F for the milled and 102.5 degrees F for the unprocessed cell.
Of special interest in the Biotron study was the ability to closely
monitor the movement of water through the refuse cells. Figure 44
presents the cumulative production of leachate from each cell and the
cumulative amount of "rainfall" with time. The unprocessed cell began
producing small amounts of leachate well before the entire volume of
refuse reached field capacity and before steady leachate production was
achieved. This is due to channeling, or relatively rapid downward move-
ment of leachate, through the unprocessed refuse.
Figure 44 also shows that the unprocessed cell produced about 40
percent more leachate than the milled cell. Since no runoff was allowed
from either cell, this difference attests to the increased capability of
milled refuse to evaporate water in comparison with the unprocessed cell.
Evaporation was about 30 percent of incident precipitation for the milled
cell and 10 percent for the unprocessed cell. Note again that neither cell
was covered.
Moisture sensing probes installed at various levels within each cell
allowed direct monitoring of the movement of the moisture front through
the refuse. Average penetration of the moisture front for the milled and
unprocessed cells was calculated to be 3.03 and 3.02 inches per inch of
percolated water, respectively. The moisture content change was from the
original 14.9 and 16.1 percent water on a dry-weight basis to 138 and 116
percent for the milled and unprocessed cells, respectively.
Figure 45 shows the cumulative COD and Total Dissolved Solids (TDS)
versus cumulative leachate volume. The Total Dissolved Solids is commonly
considered equivalent in concept to specific conductance, because both tests
are primarily a measure of the dissolved inorganic matter, or ions.
84
-------
3
+J
I/)
C
O
•»-»
o
CQ
OJ
O)
t/>
3
en
O)
O)
S-
OJ
C
o
o
CO
cu
•3
CD
-------
E
-Jj
(O
rO
i_
O)
4->
(0
3
E
O
T3
C
OJ
(O
o
fO
(O
r™~
rs
86
-------
d>
o
c
o
CQ
E
O
S-
0)
o
>
(U
U
QJ
0)
3
3
U
(fl
>
-o
C
Q
O
E
o
LD
l-
3
CD
87
-------
It is observed that the milled cell greatly outproduced the unprocessed
cell during the initial stages of leachate production with respect to both
COD and IDS. This is in keeping with the results of the field studies.
After the initial accelerated rate of production, the milled cell tapered
off to a lower rate of production of COD, such that for equal volumes of
leachate production, the cumulative amount of COD produced was the same as
for the unprocessed cell. In other words, the unprocessed cell never
produced COD at as high a rate as the milled cell did initially; however,
the unprocessed cell produced COD at its highest rate during the production
of considerable volumes of leachate, such that in total effect the cells
become equal.
The fact that the unprocessed cell curve extended to more total
leachate, and that the unprocessed cell had produced more COD at the
conclusions of the experiment, is related to the decreased ability of
unprocessed refuse to evaporate water, since unprocessed refuse is not
left uncovered in a landfill situation. No general conclusions will be
drawn from this observation.
Similar analysis of the TDS curves shows that the milled cell produced
TDS at an accelerated pace intially and then quickly dropped to a more
moderate rate. The unprocessed cell, in contrast, produced TDS slowly,
gradually increasing the amount produced to a rate equal to that of the
milled cell (i.e., the two curves became parallel). Thus it is concluded
that the milled cell produced more cumulative TDS at all stages of
cumulative leachate production, and that the relative closeness of the
total TDS production of the two cells at the conclusion of the experiment
is largely a function of the increased leachate production from the
unprocessed cell.
Except for some early pH values between 6 and 7, the pH remained very
low (5.0 to 4.6) for both cells throughout the study. Since the Biotron
project lasted a relatively short time, it would not be expected that pH
would rise toward neutrality as it had in later stages of the field studies.
Gas:
The Olin Avenue, Oscar Meyer (lysimeter), and Biotron refuse cells were
also outfitted to collect gas being given off by the decomposing refuse. The
major gases sampled were oxygen, carbon dioxide, and methane.
Problems arose with the Olin Avenue studies when 9 of the 13 test
cells proved unable to yield long-term gas data. Some useful results
were salvaged, however. In general, it was noted that the concentrations
of oxygen at deep levels in all cells were higher in early stages of the
study than later. This indicates a change from aerobic to anaerobic
conditions within the cells.
The levels of carbon dioxide did not vary much over the test period.
This is probably due to some carbon dioxide being solubilized into the
leachate at times of peak carbon dioxide production. Since the alkalinity
of the leachate did increase during periods of substantial degradation, it
seems likely that the increased carbon dioxide production was being solu-
bilized by the leachate.
-------
The methane concentrations remained low during tho period or the OHn
Avenue study, although they were higher in the milled uncovered cells than in
the unprocessed covered cells. This finding corresponds with the
leachate results which indicated that milled refuse undergoes decompo-
sition more rapidly, and therefore enters a more stable, methane-producing
stage more rapidly, than unprocessed refuse.
It was apparent at times throughout the test period that several of
the milled uncovered cells had higher oxygen concentrations and lower methane
concentrations than the unprocessed covered cells. These observations
were difficult to relate to the leachate results, which always indicated
that milled refuse decomposes more quickly than unprocessed refuse and
should therefore be producing methane in larger quantities. It was
therefore theorized that cover may be of considerable importance,
limiting the passage of air (and oxygen) into, and methane out of, the
unprocessed cells. Conversely, the absence of cover allows oxygen to
enter more readily the milled uncovered cells and methane to leave. For
this reason some doubt rose "about the validity of using gas concentration
data to compare cells with different amounts of cover, and thus only
concentration changes, rather than absolute levels, were used to draw
conclusions. Further studies were performed to provide more information
on this matter.
The three milled cells in the lysimeter studies at the Oscar Meter site
provided an opportunity to observe the effects of cover on gas composition in
the cells under a wide variety of cover conditions. The gas composition data
from the lower set of probes are given in Figure 46. Methane production
occurred first in the two then-uncovered milled cells and since that time
the methane concentration has been higher in these two cells than in the
milled cell covered immediately. This indicates again that degradation
to a stable state occurs more quickly in uncovered refuse. The fact that
methane production remained more consistent in the milled cell which was
covered after 6 months than in the uncovered milled cell and that it is
present in larger concentrations seems to contradict the premise that
degradation occurs at a faster rate in an uncovered milled cell.
However, the presence of oxygen in larger quantities in the uncovered
milled cell than in the milled cell covered later (shown better in the
results from the upper probes, which are not given in this report)
suggests that air and other gases can circulate more readily in and out
of an uncovered cell. Thus methane can more readily escape, and oxygen
more readily enter the uncovered milled cell than the covered cells.
Measured gas compositions, therefore, must be viewed as products of
gas transfer as well as gas production, especially in uncovered cells.
Note that the unprocessed cell still had not begun to produce methane as
of the close of the reporting period. This is another indication of the
relatively lengthy period required before unprocessed covered refuse
reaches a state of stable anaerobic decomposition.
The rate of gas production by the refuse, rather than simply the gas
composition in the refuse, was analyzed in the Biotron studies. The
inlet and exhaust air from the sealed test chambers was tested for
89
-------
50
40
30
20
10
50
40
30
20
10
50
40
30
20
10
50
40
30
20
CELL UNPROCESSED COVERED
«• % C02
••• % 02
* WATEK COLLECTED WITH SAMPLE
CELL 1 MILLED, COVERED
$ WATER COU.ICTED
CELL 3 MftXH),
4, MNLUEE^ HBT
Figure 46.
Gas composition data collected by lower probes during lysimeter
studies conducted at the Oscar Meyer site,
90
-------
carbon dioxide and methane and, knowing the rate of air flow, the
production of these gases could be calculated. Concentrations of carbon
dioxide peaked for both cells a few days after the start of the experiment.
After the peak, carbon dioxide concentrations dropped to levels just
slightly above the carbon dioxide levels in the supply air. The recorded
concentrations were so low that it was suspected that most of the carbon
dioxide produced was not being measured in the exhaust air. A calculation
based on pH and alkalinity data was made to estimate the amount of carbon
dioxide dissolved in the leachate. It was found that less than 0.2 percent
of the total carbon dioxide production escaped as gas, the rest being
dissolved in the leachate.
The same problem was not as likely with methane, which has a very low
solubility in water. Methane production was negligible for the first 80
days, as would be expected during aerobic decomposition. After 176 days
an apparent rapid increase in methane production occurred in both cells.
During the remainder of the test period, methane concentrations remained
relatively high with an average of 23 ppm for both cells (compared with
an average methane concentration of 2.78 ppm in the supply air). Despite
this significant rise at 176 days, the total 270-day production of methane
was just 0.052 cu. ft. Thus, methane production during the experiment was
virtually negligible, as would be expected in cells that had not reached
neutral pH levels in leachate, and the attendant stable anaerobic stage
of decomposition.
Vectors
Studies were undertaken to compare the relative attraction of milled
and unprocessed refuse to two major vectors — rats and flies. Efforts
were also made to determine whether rats and flies could survive in milled
refuse.
Rats:
Field studies of rats were conducted at the 01 in Avenue test cells in
1968. Because the two basic cell types — milled uncovered and unprocessed
covered -- were intermixed through the 01 in Avenue site, they could be
compared with respect to rat infestation.
Thirty bait stations using bait without poison were placed as
uniformly as possible over the site (Figure 47). The bait containers
were weighed regularly to determine the amount of bait loss. The
consumption of bait was considered to be proportional to the amount of
rat activity in the vicinity of the bait station. Other evidence, such
as droppings, presence of tracks, and new burrows, was also recorded.
In the test's first phase, the bait stations were set up to reveal
initial areas of rat activity. Results did not show a preference between
cell types. The highest take occurred in a station located on the special
garbage cells. Since garbage is an obvious attractant, the rats were
probably living on this cell before the bait stations were placed. This
conclusion was supported by burrow counts and rat sightings on and adjacent
to the garbage cells.
91
-------
>'. ,>
,;x
10
c
o
OJ
o
O)
o
l/l
_c
o.
-a
3
CD
c
•f—
S-
3
-o
-o •
QJ a>
w -i->
3 1-
l/>
O O)
•r- 3
-4-> C
tO >
i/i «s:
*j c
•r- 'r~
-------
With the remainder of the stations, the highest takes occurred on the
edge of the test site. This led to the conclusion that rats preferred
the peripheral area of the site more than they preferred either milled or
unprocessed refuse cells.
In the next phase, the bait stations were moved to new locations. If
the stations did not draw rat activity with them, it would indicate that
the previous areas of activity were so desirable that rats would not leave
them to seek out a bait station. If the new stations did draw activity,
it would be important to note which cell type — milled or unprocessed ~
had the greater increase. Since the stations were identical, the rats
would base their choice of a new site on considerations other than the
presence of a bait station.
Nearly all of the original stations were removed and nine stations
were established in the central area, where previously there had been
negligible activity.
The results showed that rat activity was drawn readily for distances
up to 100 feet. This migration led to much test drilling, resulting
eventually in 18 new burrows. Of these, 12 were on the two unprocessed
covered cells while six were on one of the four uncovered cells. Thus,
it appears that the unprocessed covered cells provided better drilling
and living conditions than did the milled cells (minimum of 6 inches
compacted cover).
The fact that rats were drawn by the bait in the second phase does
not negate results from the first phase, since the rats were undoubtedly
dependent on the bait after 70 days of the first phase.
The final phase of the rat field tests involved adding poison
to the bait and observing the rate of kill. To assure a thorough
kill, the bait stations were returned to their original locations
and were replenished with nonpoisoned bait. This was continued
for 3 weeks prior to poisoning to foster dependency by the rats
on the bait stations. Three to nine days after the addition of
5-percent-by-weight anticoagulant rodenticide to the bait, the
kill was essentially complete. The rate of decrease of bait take
was higher on the milled cells than on the unprocessed cells. This
may or may not properly suggest that rats frequenting the milled cells
were more dependent on the bait for food than were rats associated
with unprocessed refuse.
It was noted during these tests that any irregularity in the
surface of a cell, whether milled or unprocessed, was likely to lead
to test drilling or a burrow. Erosion of cover material, for example,
produces irregularities which may lead to test drilling. In milled
uncovered cells a break in the surface is not as likely to occur, and
if it does occur it is not as likely to result in burrows. This is
because the interior of milled cells offers only more of the same material
as is found on the surface. Many signs of test-drilling without burrow
development were found on milled cells.
93
-------
The field studies suggested that milled refuse without cover
was less attractive to rats, especially for burrow development, than
was unprocessed refuse covered with soil. It remained to be shown,
however, whether milled refuse by itself would attract rodents..
To determine this, several tons of milled refuse were placed in a
remote location within a Madison residential area. A snow fence
was set up to enclose the refuse and discourage spreading by children.
The site was checked periodically for signs of rodent activity.
No activity was observed at any time and after 12 months, the test
was terminated.
The above test could not be considered conclusive, since there
was no assurance that rats were living in the area, nor was there
reason to expect that any rats nearby would leave their previous
surroundings in favor of milled refuse. Consequently, a similar
pile of milled refuse was placed in a remote location at the 01 in
Avenue landfill where rats were known to be present. After several
months, no signs of activity were noticed, and the test was ended.
It is felt that the combination of field tests on rats and
refuse provides rather conclusive evidence that milled refuse as
processed at Madison will not result in rat infestation at a land-
fill. The fact that no rats have been sighted in the landfill in
over four years since the rats were poisoned, and that a mother duck
has felt sufficiently secure to develop a nest, lay, and hatch eggs
near the center of the site, supports the findings of the study.
Since the conclusions of the field test were strictly applicable
only to Madison, cage tests were performed in which rats were forced
to depend solely on milled refuse for sustenance. It was felt that
the results of these supplementary tests would have wide applicability
since they would show that rats either can or cannot survive on milled
refuse.
The tests were conducted at the Purdue University Rodent Test Center
at Lafayette, Indiania (Figure 48). The facility is maintained jointly
by the school's Rodent Control Fund and the United States Department of
Interior for testing baits and poisons. Some 750 Norway rats (Rattus
norvegicus) are kept in two large areas where they can live under nearly
normal conditions until they are trapped for test purposes.
Prime Norway rats, each weighing over 200 grams, were used for
the tests. Five males and five females were placed in each test cage.
The metal cages were 3 x 7 x 2 ft. high (Figure 49).
Three test series were run. For the first series, the test
cages were kept inside at 70 degrees F and in total darkness. For the
other two series, the tanks were kept outdoors under mild summer
conditions; these tanks were covered to maintain darkness. All
tests were conducted to a logical conclusion or over a 15-day
period, whichever came first.
-------
Figure 48. Rodent Test Center for Norway rats at Purdue University,
Lafayette, Indiana.
Figure 49. Typical cage used in studies conducted at Purdue University,
95
-------
The first series was run in 1969, when fresh milled refuse
with about 15 percent (wet-weight basis) garbage content and two-
year-old milled refuse of approximately similar garbage content
were used. All refuse samples were sent to Purdue from Madison.
Some 30 Ibs. of each type of refuse were placed in separate test
tanks, replenished as necessary to assure the availability of over
three times the minimum daily nutritional requirements of the rats
at all times. For purposes of calculation, only the food wastes or
garbage content of the refuse was considered to be edible by and
nutritious for the rats.
In the tank with aged"mi lied refuse, the rats dug holes and
scattered the heavy, compact material in search of food. All
animals showed weight loss by the end of the fourth day. During
the fifth night, one animal was cannibalized. All but one animal
had been eaten by the end of the test, and the survivor died the
next day. The results were virtually identical when the test was
repeated.
Although there appeared to be more available food in the
freshly milled refuse than in the aged, it was not sufficient to
sustain the rats. During the first test, two animals were cannibalized
on the eleventh night. Only three animals survived the 15-day test,
and they were so weak that they had to be killed. Again, similar
results were obtained during a replication of this test.
The conclusion of this series of tests is that freshly milled or
aged milled refuse, as received from Madison, cannot sustain a rat
population.
In the next test series conducted in 1972, freshly milled refuse
samples with various fractions of garbage were used. The four levels
of garbage content tested were 17, 32, 47, and 59 percent wet garbage
on a wet-weight basis. No animals died as a result of starvation
or cannibalism during this series of tests. It was quite obvious,
however, that the samples sent to Purdue had been poorly milled, or
perhaps were not milled as a result of procedural problems during
sample preparation at Madison. Thus, large chunks of putrescrible
matter which could be used as food were available to the rats, and they
survived. Obviously, then, particle size is of special importance in
determining the suitability of milled refuse to sustain rats.
The third series, conducted later in 1972, was designed to
correct the problems encountered in the previous test series.
Special care was taken to obtain a representative milled product
for the two samples, which contained approximately 10 and 20 per-
cent wet garbage on a wet-weight basis.
In the 20 percent cage, all the rats died within 10 days, in
the 10 percent cage, five animals survived the full 15 days and one
dominant male seemed to thrive quite well, although the others were
very weak. Although these results are not quite as conclusive as
other tests had been, the results of this final series still leads
to the conclusion that rats cannot survive indefinitely on a diet
96
-------
consisting only of milled refuse containing up to 20 percent garbage
on a wet-weight basis. Based on this fact, there is very little
possibility of rat infestation and survival in a milled refuse
landfill.
Since the fall of 1966, when the rats at the 01 in Avenue test
site were poisoned, only two or three rats have been observed on the
landfill or the vicinity. Many City and University personnel regu-
larly inspect the Olin Avenue site but since 1968 have observed no
burrows that can be attributed to rats.
It has been necessary to institute a rodent-control program at
the milling plant, however. Rats and mice are continually transported
to the plant in refuse packer trucks and must be poisoned on a regular
basis. No migration from the mill building to the landfill has been
observed.
Flies:
Field studies at the Olin Avenue test site as well as laboratory
studies were conducted in the summers of 1968 and 1969 to evaluate fly
problems which might arise from not covering milled refuse in a landfill.
A comparison of the relative numbers of flies on or near each of
the two cell types at the Olin Avenue landfill was made by direct count.
The testing was done using a Scudder Grille to attract the flies for
counting (Figure 50). This is a standard test procedure to evaluate fly
populations in barns, etc. The results indicted no marked differences
in the number of flies on milled uncovered or unprocessed covered refuse
cells.
Another study was designed to determine the relative numbers of
flies emerging from comparable amounts of milled and unprocessed refuse
(Figure 51). Screened cages were placed over similar piles of each kind
of refuse, about 1000 pounds of refuse in each case, and the files in
each cage were counted periodically over the 1 month duration of the test
(Table 15). The cage over the unprocessed refuse (without cover) reached
a population of some 4000 flies while the greatest number observed in the
cage over milled refuse was 15.
To determine whether the lack of flies emerging from milled refuse
was due to lack of viable maggots or to the inability of milled refuse
to support flies, 1200 flies and 2000 maggots were introduced into the
cage over a second pile of milled refuse. In this case, the flies
survived for about a week, but the maggots were unable to complete their
life cycle and thus did not produce more flies.
97
-------
O)
in
3
4-
OJ
s_
I/I
o;
o
o
Q.
-a
T3
-o
rs
u
oo
o
LD
QJ
S_
3
cn
la-
98
-------
f.
CTi
CO
CTi
to
O)
4J
.0
(O
s_
3
CO
>>
«*I
cn
•r-
S-
3
T3
•O
Ol
CO
3
CO
O
O
o>
c
OJ
CD
S-
o
CO
Lf)
O
i-
3
cn
99
-------
TABLE 15
RESULTS OF FLY CAGE TESTS (1969)
Days
0
1
6
12
15
20
21-30
Cage 1
Unprocessed, Compacted
More flies than cages 2 or 3
Approximately 1000 flies
Approximately 1000 flies
Approximately 4000 flies
Remaining at large number
Remaining at large number
Number beginning to decline
Cage 2
Milled
-
10 flies
none
1200 adult flies and
2000 maggots introduced
Approximately 100 flies
2 flies**
none
Cage 3
Milled
-
15 flies
none
none
* none
none
none
* Indicates most adult flies survived.
**Indicates maggots were unable to complete life cycle and that initial
1200 adult flies had died.
100
-------
Continuing this line of investigation, samples of fresh and
6-month-old milled refuse were used for laboratory studies to determine
whether this material can support flies. Approximately 1000 fly eggs
were introduced into separate cartons containing fresh and aged refuse.
Similar cartons had no eggs added. The refuse was kept moist and humid
(40 to 70 percent relative humidity) and warm (80 degrees F) for 3 weeks.
These conditions are commonly cited by entomologists as "optimal" for
growing flies.
With the freshly milled refuse to which no eggs were added, no flies
emerged throughout the test. In the carton of fresh refuse to which 1000
eggs had been added, approximately 1000 flies emerged at the end of 3 weeks.
Thus, when fresh milled refuse is subjected to optimal environmental
conditions, it is capable of supporting the growth of flies.
With the 6-month-old refuse to which no eggs were added, a few flies
did emerge, but these were not houseflies. They probably arose from eggs
or maggots picked up by the refuse while it was in the landfill. The
aged refuse was not able to support the life cycle of the added eggs, for
approximately the same number of flies emerged as compared to the carton
of aged refuse to which no eggs were added.
Thus, under "optimal" conditions, including controlled refuse moisture
content and proper temperature and humidity, fresh milled refuse can
support the fly life cycle. Aged milled refuse, however, is a poor medium
for housefly development even under optimal laboratory conditions.
It remained to show whether maggots are killed during passage through
a hammermill. On two occasions, the Gondard mill was cleared by stopping
the feed conveyor. In the first trial, 6000 mature housefly maggots were
scattered on about 100 Ibs. of refuse on the conveyor. This refuse was
then run through the mill. The second trial was identical, except 12,000
maggots were used. The emerging refuse was examined for living maggots.
The milled refuse was then exposed to ideal environmental conditions in
the laboratory to insure that any viable maggots which were overlooked
would emerge as flies. In refuse from the first trial, no flies emerged;
in the second 84 flies were counted.
It is possible that some maggots were lost in the mill, although care
was taken to avoid this. The most likely explanation for the large
decrease in viable maggots is that most of them were macerated during the
milling process.
The fly studies showed that there are several mechanisms which would
lead to reduced fly populations at landfills with milled refuse without
cover. First, the milling process itself destroys the great majority of
maggots. Second, freshly millled refuse can support the fly life cycle
only under optimal environmental conditions that are not normally found
in a landfill. Finally, when refuse has aged for several months, even this
ability under optimal conditions is destroyed.
N«t one of the tests described above provides absolute proof that no
fly problems will ever exist with milled refuse. Taken togetner, however,
the evidence from these tests becomes quite conclusive. Five years of
experience at the landfill support the conclusions of this study, for
there have been few flies reported on the milled refuse at the site.
101
-------
Vegetation
Trees:
A study was initiated at the 01 in Avenue site in 1969 to determine
the ability of milled refuse to support tree and shrub growth. The type
and thickness of cover soil as well as tree and shrub species were
varied to determine which combinations gave the best growth. Experimental
plots were established on both milled and unprocessed refuse to determine
if there were any differences in growth due to the composition of the
underlying refuse. Additional plots were established to determine the
effects of fertilizer on the growth of tress on the landfill. Ten species
of trees and shrubs were planted.
Growth measurements were made on all ten species planted, but only
white ash, jack pine, red pine and buffaloberry were selected for
representative analyses (Table 16).
Soil type was found to be a significant factor in producing growth
differences in a single year. Most of the species other than jack pine
increased in height and diameter more on topsoil than on subsoil. This
is reasonable because the low fertility of the subsoil was probably
adequate for the low-demanding jack pine.
Many trees (19 percent) died the first year of the study. The much
greater initial mortality on the unprocessed site as compared to the
milled site is attributed to differences in the condition of the
planting stock and soil at the time of planting. Greater initial
mortality occurred on the unprocessed topsoil block than on the unprocessed
subsoil block; this difference can be partially explained by the greater
weed competition on topsoil than subsoil.
Studies of plant growth on spoil banks have indicated that most tree
mortality occurs the first year after planting, with very small increases
in mortality occurring in later years. The results of this study disagree
with this finding. Little additional mortality occurred on either site
until the fall and winter of 1971-72 when a rapid upsurge in mortality
(65 percent in total) occurred; this later mortality was attributed to
a lack of adequate soil aeration. One reason for this hypothesis is that
all species survivied better as of 1972 on topsoil on the milled site
while all species did better on subsoil on the unprocessed site. The
better survival on topsoil on the milled site is probably because the
topsoil had 15 cm. more soil cover than the subsoil. This extra soil
acted as a buffer between the tree roots and gases produced by the
decomposing refuse. Since the trees also had better survival on the
unprocessed than the milled site, it was suspected that gas production
by milled refuse was greater than for unprocessed refuse, as discussed
earlier in this report.
102
-------
TABLE 16
EFFECT OF PLANTING CONDITIONS ON TREE GROWTH (PERCENT)
(November 1970 to September 1971)
Plot*
Milled (A)
(B)
(C)
(D)
(E)
(F)
Unprocessed (A)
(B)
(C)
(D)
(E)
(F)
White Ash
Hts. Diam.
(D
43.0
23.5
53.7
48.4
50.5
27.6
61.7
93.8
71.1
41.7
19.7
(D
51.3
28.4
61.5
47.6
70.4
15.1
38.2
54.9
69.2
71.4
24.6
Red Pine
Ht. Diam.
2.8
2.8
1.6
11.1
14.8
18.9
9.9
10.1
11.2
13.0
(2)
(2)
8.0
3.5
10.6
19.6
18.6
20.6
5.5
13.0
8.8
27.2
38.0
9.0
Jack
Ht.
29.3
41.6
12.8
(3)
19.0
25.8
6.6
11.1
12.914
29.1
24.6
17.3
Pine
Diam.
19.7
34.2
9.4
(3)
4.7
17.8
7.6
18.8
.6
10.6
16.7
15.1
Buff
Diam.
32.6
27.5
6.7
30.2
33.5
35.7
57.6
55.6
67.8
39.2
87.7
12.3
* Plot A - 15 cm. subsoil
B - 30 cm. subsoil
C - 45 cm. subsoil
D - 45 cm. subsoil
E - 30 cm. subsoil beneath 15 cm. topsoil
F - 15 cm. subsoil beneath 15 cm. topsoil
Notes 1. No measurements taken
2. Tops of trees were damaged, limiting growth,
3. All trees had died by September 1971.
103
-------
Compaction caused by grading operations on the study sites greatly
influenced the moisture, aeration, and strength characteristics of the
soils, and thereby reduced growth and efficiency of the root system.
Greater tap root and overall root penetration occurred on the unprocessed
than on the milled site. However, the amount of top growth did not
always correspond to the size and vigor of the respective root system.
Because of compaction, the soils had both low total water
capacity and low available water capacity. The compaction also
caused the formation of a crust which intensified dry soil conditions
by causing much rainfall to run off. Large amounts of rainfall were
sufficient to completely saturate the soil pores, but the low hydraulic
conductivity and high moisture retention then caused the pores to
drain very slowly, greatly intensifying soil aeration problems.
Soil aeration on the site is very poor. The production of
carbon dioxide and methane by the refuse, in conjunction with the
low amounts of gas-filled pore space, provides a great obstacle for
growing trees. Thus, it can be seen that the high mortality in the
fall and winter of 1971-72 was probably due in part to insufficient
oxygen present in the rooting zone.
Measurements indicated that the refuse had no detectable effect
on the soil temperature and that the fertility status of the landfill
soils is generally adequate for tree growth. It appears that the factors
limiting tree growth are more likely physical than chemical.
Because root systems were limited in extent and function by
deficient moisture, deficient oxygen5 or high soil strength, fertilizers
were apparently not utilized by the trees in sufficient quantities to
cause measurable growth changes.
Recommendations concerning site preparation, tree planting, and
cultrual practices to maintain trees on landfill sites, will depend upon
the proposed use of the landfill, the composition and preparation of the
refuse, the kinds of soil materials available, and the characteristics of
the site. Choice of a species depends upon use since trees grown for
aesthetic reasons should be chosen primarily on the basis of survival
and appearance while overall growth is more important for economic pur-
poses. The refuse composition and preparation will also influence
these choices since the type and amount of gas production is determined
by the refuse and may contribute significantly to deterioration and
death of trees.
A medium texture, we'll-structured soil material should be applied
to the refuse using a method which limits compaction as much as possible
around the base of the planted tree. The depth of soil material should
be as great as is economically feasible, providing adequate soil volume
for root expansion and a buffer between root system, and gases produced
by the refuse. Species to be planted should have the ability to develop
lateral root systems with diffuse branching; white ash and crab developed
such roots in this study. Relatively small seedlings should be planted
in planting holes which have mulch material cinci fertilizer packets added.
104
-------
Small seedlings can more easily adapt their root systems to the
environment in which they are planted. In cases where the planting
of larger trees is more desirable, planting holes should be back-
filled with topsoil and a mulch material to provide a good initial
root environment.
Other Vegetation:
Unless an uncovered cell of milled refuse is continually worked,
volunteer vegetation will develop within one or two years after placement.
During the summer of 1968, a diverse plant community, ranging from
weeds to garden vegetables to trees, became established spontaneously
on all the milled uncovered cells at the 01 in Avenue site. This growth
may have been due in part to seeds in the refuse itself. Heavy plant
growth has continued on the milled cells in subsequent years. In fact,
the vegetation of the milled refuse cell is so dense that it is difficult
for an observer standing on an old milled refuse cell to see that he is
indeed on refuse and not on soil (Figure 52).
Also in 1968, a slimy growth was noted on many of the milled uncovered
cells. It persisted throughout the summer and reappeared the next year.
It was identified as a slime mold, Fuligo septica, which is commonly found
in heavily wooded areas. It grows on material of high cellulose content.
This slime mold is not a threat to public health.
Fires
In August 1969, the Madison Fire Department evaluated fire hazards
on uncovered milled refuse. A freshly constructed cell and one that was
over a year old were used for the tests at the 01 in Avenue site. The
older cell had a cover of vegetation which was bulldozed off prior to
the tests. Moisture levels in the cells were lower than average as a
result of prolonged dry weather.
Attempts were made to ignite the cells by several methods which
simulated potential fire sources in actual landfill situations. Surface
fires were started by igniting oil which had been poured on the cells
and by igniting dry hay placed over the refuse. In all cases the
refuse smoldered but did not support flames once the oil or hay had
burned completely. Even though fans were used to create a 8-mph
wind during the hay tests on the aged refuse cell, combustion spread
only 25 feet after one hour. With fresh refuse, surface propagation
did take place, although no flames were evident. The smoldering re-
mained on the surface in all cases and was easily extinguished by water
spray or soil cover (Figure 53).
A fire starting from flying embers was simulated by placing hot
charcoal briquettes on the surface of the two cells. The charcoal had
virtually no effect on the aged cell. On the freshly milled cell,
however, combustion began slowly and spread, eventually encompassing the
entire cell surface. In this case the combustion was also limited to the
cell surface where it could easily be controlled.
105
-------
1 "* Hi "•«•'• •''--sdy%
rfe* "' Rr ~ *T; -&; £• -J » -r r.'^SfeiW "• **<'
"Jttr "!(*" '* *w ''' "
,-**r^A ^ -v|^* **' " '^ '» - . > ^
f»? " '" ' f •' J" • ^
Figure 52. Heavy vegetation growing on a test cell of milled refuse.
106
-------
01
o
3
CO
~, X
i— >>
rtS i—
I- •<-
O> to
JD (O
•r- QJ
0) to
-o to
Ol
J- -a
•r- C
ro
tn
cu
S-
cr>
107
-------
To simulate spontaneous internal combustion, a 1,200-watt electric
heating element was buried 3 feet deep in each cell. The element pro-
duced a temperature of 1,500 degrees F for 20 hours. In both cases,
refuse within an inch or two of the element was charred, but no com-
bustion occurred.
In summary, flame!ess combustion was supported by aged milled
refuse, but the combustion did not spread. A slowly spreading fire
was produced on the surface of the freshly milled cell. In both
cases the combustion could be arrested with water or soil.
The results of the fire tests were supported by an experience during
a dry spell in the spring of 1970, when a fire, apparently begun by
a lighted cigarette, began on a milled refuse cell. The fire burned on
the surface but did not penetrate to the interior of the cell. It was
extinguished by surface compaction with a front-end loader.
It is believed that the combination of a lack of voids and the
ready venting of flammable methane to the atmosphere is primarily
responsible for the lack of fire potential in piles of uncovered milled
refuse.
Odor and Esthetics
The 01 in Avenue landfill is bounded by a playfield on one side,
residential areas on two sides, and the Dane County Coliseum on the
other The Coliseum is a 10,000-seat facility. Thus, there was a
large "audience" available to make known their complaints if any odor
problems developed. Fortunately, no such problems have occurred.
The lack of unpleasant smells is one of the most notable features
mentioned by visitors to the milled refuse landfill areas. Project
personnel theorize that ready access to air and the accompanying drying
of the surface of the milled refuse cells produce an aerobic buffer zone
which treats or modifies odors produced deeper in the cells. In
support of this theory, it is noted that by digging 3 to 6 inches into
a cell, one begins to detect odor typical of decaying refuse. Upon
digging a foot or more, a most disagreeable odor is produced.
Some minor odor problems have developed during unusually wet
periods when, due to improper drainage of depressions between the
test cells, ponds of water formed. These problems have been readily
solved by filling the low areas or by providing drainage channels.
As noted earlier, milled refuse is relatively homogeneous and
looks like oversized confetti. Viewed from a distance, milled refuse
is nondescript and unobnoxious since it contains no large recognizable
items. Of the thousands of lay people who have viewed the 011n Avenue
landfill, no one has objected to the sight of uncovered milled refuse.
An independent evaluation of the public acceptability of the 01 in
Avenue reduction plant and landfill was published in Compost Science
in the January-February 1973 issue.
108
-------
Use of Cover
Cover soil 1s generally prescribed for a sanitary landfill to h1d«
the refuse, to reduce odors, to control blowing paper, to lessen the
danger of fire, to discourage vectors, and to limit leachate and gas
production. As we have seen in preceding sections of this report, most
of the problems solved by covering unprocessed refuse are similarly
reduced or eliminated by landfill1ng milled refuse without cover. In
other words, the Initial European claims for milled refuse have been
substantially borne out during the Madison demonstration project.
Except for possible considerations of groundwater contamination
due to rapid initial pollution loads 1n leachate from milled cells,
it appears that no daily cover 1s necessary. And, in fact, the question
of groundwater contamination is more dependent on local hydrogeological
considerations at the landfill site than on differences between milled
and unprocessed refuse. Thus, 1t 1s possible that in a multilayer milled
refuse landfill, the exposed layer may remain uncovered until the next
layer is placed on top of 1t. This sequence could be followed until
the landfill reaches final grade. (Some Intermediate cover may be re-
quired 1f final grade is not reached in a reasonable period, such as 6
to 12 months, or if local conditions dictate such procedure.) Once
final grade 1s achieved, the milled refuse should be covered to pre-
pare and reclaim the landfill site for other uses.
During construction of the 01 in Avenue test cells, 1t was the
practice to cover the top and sides of the unprocessed refuse cells but
not the daily working face. This amounted to what 1s commonly called
intermediate cover. The corresponding milled refuse cells were normally
not covered; however, special tests were conducted to estimate cover
soil requirements after a milled refuse landfill has been brought to
final grade. Based on these tests, 1t 1s felt that a practical depth of
dirt to cover milled refuse completely and smoothly is 6 Inches. In
comparison, a practical depth of cover dirt on unprocessed refuse was
found to be 14 inches on the top and 18 inches on the side of a cell
using the same equipment as with the milled cells. This amount of
cover provides a uniformly smooth and refuse-free surface of the same
quality as did the 6 inches of cover on the milled refuse. Although
some landfill operators will disagree with these figures, the ratio of
the cover required for equal quality of the finished surface for milled
and unprocessed refuse 1s felt to be valid.
This difference in cover requirements for the two cell types is
due to the more even surface that can be obtained with milled refuse.
Spreading, compacting, and filling unprocessed refuse 1s more difficult
than for milled refuse due to the extremely variable compactabll1ty of
the hetrogeneous unprocessed refuse which results in problems in obtain-
ing a level surface. The differences 1n compactabllity result in the
bulldozer leaving an uneven surface for unprocessed refuse which is
more difficult to cover completely. Milled refuse, on the other hand,
is easy to spread evenly without local depressions or rises because of
its smaller and more homogeneous particles. Also, the smaller particles
of milled refuse are not as readily pulled up by billdozer tracks during
compaction or pulled up through the soil during covering operations.
109
-------
To give an Indication of the monetary savings in cover soil between
milled and unprocessed refuse, two Madison landfills were compared.
The first was the 01 in Avenue site which was completely converted to a
milled refuse landfill in mid-1971 and thus did not use daily cover.
The other site was the Truax landfill for unprocessed refuse. The
Truax site operates under strict sanitary landfill procedures and
therefore uses considerable amounts of cover material. For this com-
parison it was assumed that the cover material used at Truax is obtained
at no cost for the material, although this is not always the case. Costs
for excavating and hauling the cover are included, however. The truax
operation handled about twice the volume of material as the Olin Avenue
site during the period of record, January through June 1972 (Table 17).
The table reveals that Madison expended three times the amount of
money, on a per-ton baiis, to operate an unprocessed refuse landfill as
it spends to operate a milled refuse landfill. This substantially higher
cost of a conventional landfill operation as experienced in Madison is
partly due to the handling of cover materials.
THE ECONOMICS OF MILLING
Landfill ing
In the preceding section, it was stated that the operating cost of
landfilling milled refuse amounted to $0.988 per ton for the period of
January through June 1972. During this period, Madison spent $23,043
while landfilling 23,317 tons of milled refuse at the Olin Avenue site.
This does not Include charges fdr land, site preparation or final covering.
By far the greatest cost of the 01 in Avenue landfill operation was
labor. The site is manned by one full-time compactor operator who works
a 7:30 a.m. to 4:00 p.m. shift. The operator averages 6 hours on the
fill and 2 hours on maintenance of landfill equipment.
During the 6 month evaluation period, no difficulties were encountered
handling the average daily load of 180 tons. It is believed, 1n fact,
that an average of 500 tons per day could be handled routinely with no
increase in men or machinery at the landfill.
Another labor expenditure was for cleanup of the area. Cleanup
consists of picking up paper, cutting grass, etc. This chore averaged
184 man-hours per month, with another 2 hours per day for supervision.
An auxiliary operator spent 32 hours on the site during the entire
6 months on read repair.
Permanent equipment at the Olin Avenue site consisted of a Steel
Wheel Compactor, which was operated between 5 and 7 hours per day. The
city garage sets an hourly rate of $6.25 per hour for use of the equip-
ment including amortization.
Stone and oil are used on access roads to the site. The stone
allows good wet weather operation. The road oil keeps dust down in
the summer dry periods.
Tabi* 18 presents the actual costs of operating the Olin Avenue
site during the first six months of 1972.
110
-------
Table 17
Cost Incurred in Landfill ing Milled and Unprocessed Refuse
(January 1 - June 30, 1972)
Cost Item
01 in $/ton**
Milled Refuse
Truax $/ton***
Unprocessed Refuse
WAGES*
Compaction
Supervision
Caretakers
Paper pickup
Loading (cover)
Hauling (cover)
Road repair
Scale operator
SUBTOTAL
EQUIPMENT
Compaction
Loading (cover)
Hauling (cover)
Paper pickup
Road repair
Misc. maintenance
Amortization
SUBTOTAL
MATERIALS - AREA IMPROVEMENT
Stone
Road oil
SUBTOTAL
TOTAL
0.310
0.090
0.270
none
none
none
0.010
none
0.680
0.206
none
none
0.020
0.010
none
0.035
0.271
0.019
0.018
0.037
0.988
0.449
0.229
0.374
0.222
0.182
0.335
0.010
0.121
1.922
0.449
0.209
0.211
0.015
0.008
0.076
0.105
1.073
none
none
0.000
2.995
*Includes all fringe benefits
**23,317 tons
***48,660 tons landfill
111
-------
TABLE 18
Landfill Costs Using Milled Uncovered Refuse
(January 1 - June 30, 1972)
Cost
WAGES*
Compaction $ 7,112
Supervision 2,413
Cleanup 6,518
Auxiliary Operator 221
SUBTOTAL $15,994
EQUIPMENT
Operation (Landfill Compactor Only) $ 3,859
Auxiliary (Road Repair) 310
Amortization (Landfill Compactor Only) 2,025
SUBTOTAL $ 6,194
MATERIALS - AREA IMPROVEMENT
Stone $ 435
Road Oil 429
SUBTOTAL $ 855
TOTAL $23,043
*Includes all fringe benefits
112
-------
'i 11 lug Costs
At Madison, there has been a considerable savings in landfill costs
by using milled refuse without daily cover; however, the additional costs
of the milling process itself must obviously be added to the landfilling
cost to indicate the total cost of refuse disposal by the milling method.
The following sections will investigate these costs and conclude with
cost-per-ton projections which are reasonable for operations like the
one at Madison.
Gondard System:
Cost data for the Gondard system are for the third year of the
demonstration project, from June 1968 through May 1969, by which time
the Gondard operations were refined. Although it is proper to report
the costs incurred in the Madison plant, one must be cautioned about
applying these costs to other installations because this project
began as a pilot plant demonstration whose operation is probably more
sxpensive than that of future plants. More importantly, one must recognize
cho regional variations in labor, power costs, heating costs, and
depreciation methods. Inflation during the years since this study was
undertaken must also be considered.
Furthermore, the unit costs in Table 19 is higher than would be the
case for a larger and differently designed plant because:
1) Refuse was not conveyed to the mill as fast as the mill could
grind;
2) A similar plant without extensive foundations and extra
conveyors would be less costly;
3) Adaptation and improvements can still be made in haul-away
operation.
As in other sections of this report, land costs are not included,
because in most instances they are a negligible part of the total cost
and vary too greatly from area to area to be meaningful. As an example,
the Olin Avenue site was purchased at $1,000 per acre; similar land in
the area is now selling for $5,000 to $10,000 per acre.
The costs per ton are calculated by dividing the annual cost by the
annual tonnage. The annual tonnage figures are projected by using the
overall Gondard production rate, the average number of working hours
per day, and the number of working hours per week.
Table 19 lists the cost per ton for each of the major cost categories
and for three grate sizes.
113
-------
TABLE 19
UNADJUSTED COST DATA FOR GONDARD MILL
(JUNE 1968 THROUGH MAY 1969)
Cost Item
Labor
Amortization
Power
Lighting
Water
Gas-heat
Hammer Wear
Mill Maintenance
Small Equipment
General Supplies
Front End Loader
Operation
Transportation
to Landfill*
Other
TOTAL
Annual Cost
$39,800
$32,200
variable
$ 2,300
$ 200
$ 1,200
$1,600-1,710
$850-950
$ 800
$ 1,100
$ 500
$ 3,250
$ 1,700
3-1/2 Inch
10,750
$3.70
$2.99
.34
.21
.02
.11
.16
.08
.07
.10
.05
.30
.16
$8.29
ton
Grate Size
5 Inch
Annual Tonnage
11,500
$3.46
$2.80
.30
.20
.02
.10
.15
.08
.07
.10
.04
.28
.15
$7.75
ton
6-1/4 Inch
12,050
$3.30
$2.67
.30
.19
.02
.10
.14
.00
.07
.09
.04
.27
.14
$7.41
ton
*Based on round trip of less than 1/2 mile
114
-------
The average hourly wage for the three plant workers together during
the evaluation period was $19.15, including all fringe benefits and over-
time. Yearly labor cost is obtained by multiplying $19.15 by the
number of working hours in a year (2080). This yields $39,830 per year for
labor costs. To get labor cost per ton, this figure was divided by
projected annual tonnage of 11,500 tons.
The annual cost of amortization was calculated by assuming machinery
life and interest rates depending on source of funds. Table 20 lists data
used to arrive at the annual amortization figure of $32,100.
The experimental periods did not exactly coincide with the utility
companies' billing periods. It was therefore necessary to project
power consumption to a monthly basis in order to determine equivalent
monthly costs. The costs of power ranging from $0.34 to $0.30 per ton
are the weighted averages during the times that the three grate sizes
were used.
Computation of other costs was straightforward and needs no further
explanation.
Tollemache System:
Two experimental runs using only the Tollemache mill were undertaken
in the summer of 1970 and winter of 1971. During both of these runs,
extensive cost data were collected on labor, power, repairs, replacements,
etc. Thus, a comparison is possible between summer and winter operations
of the same plant and equipment. To make the comparison equitable, changes
in wages and other increases in cost between the two runs must be
considered. Therefore, actual and adjusted costs will be presented for
Lhe data obtained during 1970, with adjusted cost being computed using
the wages and prices in effect during the 1971 test.
Cost of the milling operation (excluding landfilling) is best
presented in two general categories: that of milling, including
conveyance of the unprocessed and milled material; and that of
transportation to the final disposal site, including the stationary
packer and hauling equipment.
Table 21 contains an extensive breakdown of the costs incurred
during the two test periods on a per-ton basis. Affecting these figures
are such uncontrollable factors as the average moisture content of the
refuse being processed. Therefore the costs stated here can be viewed
only as an indicator of what was experienced at one installation and
as such they cannot be expected to be generally applicable to other
locations.
An analysis of Table 21 indicates that on an adjusted basis the
cost per ton for period 1, summer 1970, is about $0.43 less than period
2, winter 1971. There are two prime factors contributing to this.
First is the fact that on the average fewer tons were milled per day
during period 2; thus, relatively stable or fixed costs such as amortiza-
tion increased on a per-ton basis. The decrease in average milled tons
per day is due to seasonal variations in refuse quantities and moisture
content. Also during period 2 gas and lighting costs increased due to
changed weather conditions.
115
-------
TABLE 20
DATA USED FOR COMPUTING AMORTIZATION OF ORIGINAL GONDARD INSTALLATION
Cost Item
Building
Grinder and
Conveyors
Scale
Front End
Loader
Packer
Trucks (2)
Original
Cost
$133,100
126,700
6,900
15,400
38,000
Total
Estimated
Life-Years
20
15
20
8
10
Annual Cost - Amorti
Interest
Rate
5.8
5.8
7.0
7.0
7.0
zation
Salvage
Value
$4,000
4,000
1,000
3,000
3,000
Annual
Cost
$11,300
12,700
600
2,300
5,200
$32,100
116
-------
TABLE 21
TOLLEMACHE MILLING COSTS
Labor***
Amortization
Hammers & Shafts
Power
Welding Rod
Plant Supplies
Front End Loader Maintenance
Gas - Heat
Lighting
Water & Sewer
Contracted Repairs
Replacement Parts
TOTAL
Test
Period
$/Ton
$2.395
1.134
0.137
0.245
0.033
0.029
0.055
0.000
0.052
0.001
0.011
0.038
$4.130
#1*
Adjusted
$/Ton
$2.758
1.134
0.137
0.284
0.033
0.029
0.068
0.000
0.059
0.002
0.013
0.045
$4.562
Test
Period #2*
$/Ton
$2.577
1.327
0.206
0.351
0.031
0.042
0.079
0.154
0.105
0.006
0.040
0.074
$4.992
* 14 wks. - July 6 - October 9, 1970, 5318 tons milled, average - 77
tons/day and 5.3 hrs. machine time/day
** B wks. - February 4 - March 31, 1971, 2624 tons milled, average -
66 tons/day and 5.1 hrs. machine time/day
*** Includes all fringe benefits
117
-------
The average overall adjusted cost per ton of approximately $4.80
can be misleading. It must be remembered that this figure has been
derived from experimental runs. At the time of the tests the plant
was only operating at three-fourths of its rated capacity. At full
capacity the plant could average 90 to 100 tons/day as compared to the
65 to 75 experienced during the tests. This would mean that fixed
costs, labor and amortization, would be reduced by nearly 27 percent
over the tabulated cost per ton, and the overall cost would be reduced
by 23 percent to an average of $3.70/ton.
Table 22 contains the detailed breakdown of costs for the sta-
tionary packer and transportation to the landfill on a per-ton basis for
test periods 1 and 2. The actual cost as well as the adjusted cost per
ton are presented for period 1 as in Table 21. Again, it is felt
that transportation costs experienced at Madison during this evaluation
are not indicative of transportation costs that would be expected at
other localities under different conditions. Thus, these costs are
presented separately to stress tht they apply to Madison's operation
only.
The computations of costs during the Tollemache evaluation were
made similarly to those in the Gondard economics section. However,
several differences should be noted. First, the average total hourly
wage, including fringe benefits and overtime, for the three plant
personnel had risen from $19.15 to $20.39. Also, amortization figures
were somewhat different, as shown in Tables 23 and 24.
Two-Mill, Two-Shift Operation:
Extensive cost data on all plant functions were collected during
the first 6 months of 1972. As with the Tollemache system, these data
will be presented in two parts - costs of milling and costs of compaction
and final transportation.
Again, it is important to stress that the figures presented here are
strictly applicable only to Madison's operation. This qualification is
especially important in this section, in which two vastly different mills
are being discussed.
The total cost of the 6-month run will be reported first. Table 25
gives a detailed breakdown of expenditures and costs per ton for the period
of record. Total tons milled during that period were 23,317.
The total 6-month expenditure for mill operation was $91,000, or
$3.90 per ton. Ilotice that labor and depreciation account for $2.75
per ton, or 71 percent of that total. Thus it is imperative for ef-
ficient operation that the maximum tonnage of refuse be milled during
the 16-hour working day. If plant production could be increased from
the current rate of 46,000 tons per year to a feasible 60,000 tons per
year, the per-ton cost of milling could be reduced 15 percent to approxi-
mately $3.29 per ton.
118
-------
TABLE 22
STATIONARY PACKER AND HAUL COSTS
DURING TOLLEMACHE EVALUATION
Test
Period
$/Ton
Amortization $0.448
Labor (Driver)*** 0.120
Power 0.016
Tractor & Trailer Maintenance 0.018
Packer Maintenance 0.010
TOTAL $0.612
#1*
Adjusted
$/Ton
$0.448
0.137
0.019
0.027
0.012
$0.643
Test
Period #2**
$/Ton
$0.524
0.147
0.023
0.017
OiPJL2.
$0.723
* 14 wks. in length - July 6 - October 9, 1970, 5318 tons milled,
average - 77 tons/day and 5.3 hrs. machine
time/day
** 8 wks. in length - February 4 - March 31, 1971, 2624 tons milled,
average - 66 tons/day and 5.1 hrs. machine
time/day
*** Includes all fringe benefits
-------
87,600
5,916
15,400
15
20
12
5.9%
6.5%
6.5%
4,000
1,000
3,000
8,640
660
1,700
TABLE 23
AMORTIZATION DATA FOR TOLLEMACHE MILLING SYSTEM
Original Effective Interest Salvage Annual
Item Cost Life (Yrs.) Rate Value Cost
Building and
Foundation $133,188 20 5.9% $4,000 $11,390
Tollemache Mill
And Conveyors
Scale
Front-End Loader
TOTAL $22,390*
* The total annual amortization cost of $22,390 can be proportioned to the
14 week and 8 week evaluation periods on a straight line basis. Amortiza-
tion for the 14 week evaluation is 14/52 ($22,390) = $6,034 and for the 8
week period 8/52 ($22,390) = $3,448
TABLE 24
AMORTIZATION DATA FOR STATIONARY COMPACTOR AND
FINAL TRANSPORTATION SYSTEM
Item
Stationary Compactor
and Hopper
Two Trailers
One Tractor
Building Addition
TOTAL $8,850
Original
Cost
$ 19,150
33,000
13,625
16,801
Effective
Life (Yrs.)
15
12
15
20
Interest
Rate
6.5%
6.5%
6.5%
6.5%
Salvage
Value
$1 ,000
1,500
1,500
1,000
Annual
Cost
$2,000
3,960
1,390
1,500
120
-------
TABLE 25
MILLING COSTS FOR TWO-MILL,
TWO-SHIFT OPERATION (JANUARY THROUGH JUNE 1972)
Total Cost Cost/Ton*
Labor $44,684 $1.912
Amortization 19,570 0.838
Replacement Parts 8,097 0.346
Power 4,382 0.188
Hammers and Shafts 3,786 0.162
Heat - Gas 2,391 0.102
Supplies 2,196 0.094
Lighting 2,023 0.087
Front-End Loader Maintenance 1,653 0.071
Welding Rod 1,128 0.048
Contracted Repairs 1,047 0.045
Water and Sewer 44 0.002
TOTALS $91,001 $3.895
* Based on 23,317 tons milled
121
-------
The final figure of $3.90 per ton as presented in Table 25
represents a sizable reduction from the earlier stated costs for
experimental runs of both the Gondard and Tollemache Systems separately.
The figure is 48 percent lower than that actually experienced during
the Gondard evaluations of 1968 and 1969. It is also 19 percent lower
than the average figure obtained during the Tollemache evaluations of
1970 and 1971. The reduction is the result of better supervision and
the 125 percent average daily increase in tonnages milled over that
of the single-shift operations.
Total milling labor costs as charged to the plant during the 6
month of record are shown in Table 26.
The average total hourly wage for the three regular plant personnel
needed for each shift is now $20.98. The plant supervisor's hourly wage
is $6.25. Labor rates are based on job classification and length of
service with the city.
A breakdown of labor costs into three main categories is given in
Table 27. No differentiation is made between times devoted to the
Gondard or Tollemache system individually.
The data indicate that nearly 73 percent of the total labor cost,
not including supervision, is a result of mill operations, while only
15 percent and 12 percent is the result of repair and hammer maintenance,
respectively.
Based on data from the Gondard and Tollemache systems, separately,
the total amortization for the 6-month period was $19,570. On a per-ton
basis the figure is $0.838.
Power costs for the two mills combined are presented in Table 28.
Table 29 contains power cost data for the mill accessories.
In both tables the demand cost is constant each month, and in
the case of the mills themselves, the demand charge is more than the
energy used. This is an important factor in the overall power cost,
as an increase in tonnage milled will decrease the total cost per
ton. For example, during the Tollemache runs, power averaged $0.24
per ton at an average rate of 70 tons of refuse milled per day.
Power costs for the Gondard runs averaged about $0.28 per ton back
in 1968, at much lower rates, for an average of 46 tons of refuse milled
per day. During 1972 the combined mill operation consumed power at
the cost of $0.168 per ton, which represents a 53 percent reduction
over the single-mill operation. The reason for the decrease is that
187 tons of refuse were processed per day in 1972.
Lighting and other small services are supplied by 220-volt service.
Table 30 contains monthly 220-volt service costs and the amounts of
electricity used.
The plant is heated by radiant natural gas heaters. Table 31
contains a monthly breakdown of heating costs. The total expenditure
reflects 3 months of winter heating bills and 3 months of much lower
spring bills. Past experience has shown that heating costs have gone
almost to zero from June through September.
122
-------
TABLE 26
LABOR COSTS - TWO-MILL, TWO-SHIFT OPERATION
(JANUARY THROUGH JUNE 1972)
Plant Personnel
Supervision
TOTAL
* Includes all fringe benefits
** Based on 23,317 tons milled
Actual Man Hours
5456
1040
6500
Cost*
$38,180
6,504
$44,684
Cost/Ton**
$1.634
0.278
$1.912
TABLE 27
BREAKDOWN OF LABOR COSTS - TWO-MILL, TWO-SHIFT OPERATION
(JANUARY THROUGH JUNE 1972)
Han-Hours
Hill Operation
Repair Maintenance
Hammer Maintenance
SUBTOTAL
Supervision
TOTAL
* Includes all fringe benefits
** Based on 23,317 tons milled
Total
3969
850
641
5460
1040
6500
Per Week
152.8
32.7
24.7
210.0
40.0
250.0
Cost*
$27,754
5,944
4,482
$38,180
6,504
$44,684
Cost/Ton**
$1.188
0.254
0.192
$1 ,634
0.278
$1.912
123
-------
TABLE 28
POWER COSTS, MILLS, TWO-MILL, TWO-SHIFT OPERATION
(JANUARY THROUGH JUNE 1972)
January
February
March
April
May
June
OVERALL
* Based on 23
POWER COSTS,
January
February
March
April
May
June
OVERALL
Demand
Cost
$341
341
341
341
341
341
$2046
Energy
Cost
$257
311
332
329
330
321
$1880
Total
Cost
$598
652
673
670
671
662
$3926
Cost/Ton*
$0.225
0.195
0.196
0.158
0.131
0.146
$0.168
,317 tons milled
TABLE 29
MILL ACCESSORIES, TWO-MILL, TWO-SHIFT
(JANUARY THROUGH JUNE 1972)
Demand
Cost
$ 19
19
19
19
19
19
$114
Energy
Cost
$ 47
56
60
60
61
58
$342
Total
Cost
$ 66
75
79
79
80
77
$456
OPERATION
Cost/Ton*
$0.025
0.023
0.023
0.019
0.015
0.017
$0.020
* Based on 23,317 tons milled
124
-------
TABLE 30
LIGHTING COSTS, TWO-MILL, TWO-SHIFT OPERATION
(JANUARY THROUGH JUNE 1972)
January
February
March
April
May
June
OVERALL
Demand
(kw)
38.6
40.0
40.0
40.0
40.0
40.0
Energy
(KWH)
18,788
17,088
19,286
14,734
12,772
13,294
Demand
Cost
$ 58
61
61
61
61
61
$363
Energy
Cost
$ 319
293
326
258
228
236
$1660
Total
Cost Cost/Ton*
$ 377 $0.14
354 0.106
387 0.113
319 0.075
289 0.056
297 0.066
$2023 $0.087
* Based on 23,317 tons milled
TABLE 31
PLANT HEATING COSTS, TWO-MILL, TWO-SHIFT OPERATION
(JANUARY THROUGH JUNE 1972)
Usage
January
February
March
April
May
June
OVERALL
100 Cu. Ft. Gas
8,679
8,297
5,124
1,983
618
540
Cost
$ 823
788
468
194
63
55
$2391
Cost/Ton*
$0.309
0.262
0.136
0.046
0.012
0.012
$0.102
* Based on 23,317 tons milled
125
-------
Total water usage equaled 6510 cu. ft. The cost of this water
equaled $21.00. Sewer charges are 140 percent of the total water cost.
Thus the water and sewer bill for 6 months equalled $44.45, or $0.002
per ton.
Items such as hammers, hammer shafts, and welding rods constitute
supplies used in the hammer maintenance program. Table 32 contains
all pertinent data in respect to the numbers of each item used and the
resultant expenditure. Hammer maintenance supplies constitute the
fourth largest expense in plant operations.
As seen in Table 25, the cost of replacement parts was over $8,000
and is therefore the third most expensive item on the list of plant
expenses. Parts replaced during the period of record were mill grates
(Gondard) and wear plates (Tollemache) as well as conveyor belting - all
of which are very expensive items. A set of Tollemache liners, which lasts
approximately 6 months, costs nearly $1,300. Gondard grates, also lasting
6 months, cost nearly $800 per set. Other parts such as small motors,
conveyor slats, and bearings, make up the remainder of the expenditure in
this area.
Other expenses included miscellaneous supplies, contracted repairs,
and front-end loader maintenance. Supplies consisted of janitorial
requirements, office materials, grease and oils, etc. The total expendi-
tures for sunplies, $2,196, is almost $0.10 per ton. Contracted repairs
include all labor and material charges for repairs made by outside agen-
cies. The total cost, $1,047, is less than $0.05 per ton. Front-end
loader maintenance is dependent on the hours of vehicle use. The city
garage charges $4.27 per hour of use to cover vehicle maintenance such
as oil, minor repairs, grease, etc.
Expenses for compaction and hauling include labor, depreciation,
power, and equipment maintenance. Not included are minor expenses
due to heat, lighting, and water which are grouped under milling
costs. Table 33 gives a complete listing of all expenses attributed
to final handling of the milled material, excluding landfilling costs.
Labor expenses, as was the case with milling costs, constitute
the largest expenditure for compaction and hauling. A total of 780
hours, or 30 man hours per week, was spent in transporting milled
material to the landfill (round trip is less than 1/2 mile). Vehicle
maintenance is computed on a per-mile charge. Tractors are charged at
a rate of $0.20 per mile and transfer trailers at the rate of $0.25
per mile. The charges cover all fuel, oils, grease, and minor repairs.
Table 33 indicates that the total power cost is almost insigni-
ficant at $0.014 per ton. The stationary compactor maintenance at
$0.032 per ton consists of labor and parts for all repairs to the com-
pactor. As the figures show, the compactor is not prone to breakdowns.
126
-------
Used
1000
43
600
Unit
Cost
$ 3.27
12.00
1.88
Total
Cost
$3,270
516
1,128
Cost/Ton*
$0.140
0.022
0.048
TABLE 32
SUPPLY COSTS, HAMMER MAINTENANCE, TWO-MILL, TWO-SHIFT OPERATION
(JANUARY THROUGH JUNE 1972)
Hammers
Shafts
Welding Rod (Ibs.)
OVERALL $4,914 $0.210
*Based on 23,317 tons milled
TABLE 33
STATIONARY COMPACTION AND HAUL COSTS, TWO-MILL, TWO-SHIFT OPERATION
(JANUARY THROUGH JUNE 1972)
Cost Cost/Ton*
Labor $ 5,216 $0.223
Amortization 4,380 0.188
Compactor Maintenance 753 0.032
Haul-Vehicle Maintenance 597 0.025
Power 334 0.014
OVERALL $11,280 $0.482
*Based on 23,317 tons milled
127
-------
Cost Projections^
Based on the extensive cost data on operations at the 'Edison
Refuse Reduction plant, several projections have been meule on tht>
costs of various milling systems employing the Toller.tache mill.
The estimates take into account one through four mills operated
either one or two milling shifts per day and an additional shift
for maintenance.
It is important to realize that, while the projections are thought
to be accurate, they apply strictly only to Madison. Wage rates,
power and utility rates, depreciation, etc. are all based on mid-1972
costs as experienced in Madison, Wisconsin. To arrive at projected
costs for similar operations at other localities, it will be necessary
to make an economic study using appropriate base rates.
It was assumed in these projections that each Tollemache mill will
operate an average of 7 hours per shift at a rate of 14 tons per hour.
It was also assumed that the plant will operate 245 days (or 49 weeks)
per year, with 3 weeks for repairs. Each plant will include a minimum
of one mill, one feed conveyor, two transfer trailers, and one tractor.
Two mill plants will have two mills and feed conveyors, but will
otherwise be similar to a one mill installation. Plants with three or
four mills will have two discharge conveyors, two stationary compactors,
four or five trailers, and two tractors. In addition to the one or two
milling shifts is an additional maintenance shift. The maintenance
shift duties involve mill maintenance, plant clean up and maintenance,
and machine repair work.
Under these conditions, each mill will produce approximately 100
tons per day, or 24,500 tons per year. Daily and annual tonnages for
combinations of mills and shifts are tabulated in Table 34.
A cost analysis of the various plant sizes on a one and two shift
milling operation resulted in projected milling, transfer, and land-
filling costs per ton as shown in Table 35. A detailed cost analysis
is provided in Appendix A of this report.
Thus, if a municipality generates 196,000 tons of mi 11 able
refuse per year, this refuse could be milled and landfilled for a cost
of $2.75 per ton using four Tollemache mills for two shifts per day.
This corresponds to Madison's present costs of $4.88 per ton for milling
and landfilling refuse and $3.00 per ton for landfilling unprocessed
refuse.
It is to be stressed that milling is but an alternative method
which can be used in conjunction with landfill. It is not meant as a
replacement for landfill. Instead, its characteristics may enable a
higher set of operational standards to be followed at similar or slightly
higher costs than standard sanitary landfilling.
128
-------
TABLE 34
Production Estimates
(For one and two shifts, and one to four mills)
(Number of Mills
_J 2 3 4
One Shift
Daily tons 100 200 300 400
Annual tons 24,500 49,000 73,500 98,000
Two Shifts
Daily tons 200 400 600 800
Annual tons 49,000 98,000 147,000 196,000
129
-------
TABLE; 35
Annual Average Costs per Ton for [-1111 ing, Hauling, and Landfill ing Refuse
Number of Hills
1234
One Shift
Annual tonnage
'Tilling, stationary
compactor, and haul
costs/ton*
'"illfilling costs/ton
TOTAL costs/ton
Two Shifts
Annual tonnage
ili 11 ing, stationary
compactor, and haul
24,500 49,000
5.27
1.43
C.34
3.74
0.75
4.59
49,000 98,000
73,500
3.48
0.73
4.21
93,000
3.33
0.57
3.90
147,000 19C,090
costs/ton*
Mill fill ing costs/ton
TOTAL costs/ton
3.75
0.75
4.50
2.70
0.50
3.20
2.52
0.57
3.09
2.30
0.45
2.75
*Cost includes amortization, labor, operation, and milled refuse haul to
landfill less than ^ mile av;ay. Land cost excluded.
130
-------
TRENDS AND DEVELOPMENTS
Before the Madison milling project began in 1967, the only milling
of solid waste in the United States was for the production of compost.
Since Madison's project was initiated, most of these composting
facilities have been closed down for one reason or another, but a number
of other milling facilities have been constructed and many more are in
the planning or construction stages. The facilities that have been
constructed or that are being planned are for the purposes of:
1. Hilling refuse for baling in a continuous baler - San Diego,
California.
2. Hilling refuse for energy recovery by burning it in a conven-
tional coal or gas fired power plant - St. Louis, flissouri.
3. Milling refuse for energy recovery in new types of municipal
incinerators in which most of the burning takes place in air
suspension - Hamilton, Ontario, Canada.
4. Milling refuse for landfill disposal - Pompano Beach, Florida;
Mil ford, Connecticut; Vancouver, Washington; etc.
b. Milling refuse for resource recovery - at the oresent time,
March 1973, the City of Madison is one of the few facilities
known to be magnetically separating the ferrous metal from
milled refuse. A number of future plants will magnetically
separate ferrous metal and will also attempt more advanced
schemes of resource recovery.
The original application for the fladison demonstration project was
entitled, "Solid '.laste Reduction/Salvage Plant". It was envisioned
that narketable material could be picked from the solid waste feed bolt
by hand and marketed through local salvage dealers. While time has
shown that this idea was rather naive in terms of recycling, it does
illustrate that the concept of material recovery from solid waste v/as
one of the motivating factors for the Madison project.
Since then, the City of Madison has entered into a separation-at-
the-source newsprint recovery project; a cooperative project with the
Continental Can Company for magnetic separation and marketing of ferrous
rnetal from the milled refuse; and a log recycling project in cooperation
with Urban Hood Fiber, Inc. A cooperative agreement with the Forest
Products Laboratory, U.S. Department of Agriculture, for recycling of
v/ood fiber has resulted in work on wood fiber separation. Following is
a more detailed description of these projects.
131
-------
In 1967, the 'lational Committee for Paper Stock Conservation
approached the City of Madison with a proposal that the Committee and the
City of Madison work together on a project to collect newsprint separated
at the source. Since early contacts with local secondary material dealers
had indicated that any paper that had been in a packer truck would not
meet existing paper specifications, the City of Madison expressed interest
in the recycling project proposed by the Committee, In the fall of 1968,
the City of Madison entered into a pilot project on the east side of the
community which in 1970 was extended to the west side of Madison. The
newspapers are bundled separately by cooperating citizens, placed at
curbside along with the remainder of the solid wastes, and placed by the
collection crews on specially built racks under the packer bodies. In
1972 approximately 2,800 tons of newsprint vie re collected; this amounted
to about 1.5 percent of the total amount of solid waste collected and
brought to City of Madison disposal sites.
A study of the realities of recycling by the City of Madison led
to the conclusion that probably only newsprint meets the criteria for a
successful separation-at-the-source program. Newsprint is easily
identified by the home owner, available in large quantities, and fairly
easily marketable. When the newsprint project was beginning, a new
de-inking mill was going on line at Alsip, Illinois and this created a
new demand in the Midwest for newsprint.
A study of further recovery of wood fiber has led to the conclusion
that mechanical methods are necessary for large-scale recovery. Dialogue
with the Forest Products Laboratory in Madison led to a cooperative
agreement in 1970 between that institution and the City of Madison to
cooperate on fiber recovery. Madison's main role in this project was to
provide working space at the Refuse Reduction Plant and to supply
milled refuse for the Forest Products Laboratory projects.
The pilot separation facility established by the Forest Products
Laboratory at the 01 in Avenue P^efuse Reduction Plant is a flexible
dry-separation system. The system is designed to use air currents to
separate light materials (paper and sheet plastics) from heavy materials
and to separate the paper into several types. Equipment includes two
fans, two cyclone units, an air classifier, a dry screen, and conveyor
and collector bins. All separation equipment in the pilot facility
is available commercially. Specific units were acquired solely on
the basis of availability and their application to the pilot scale.
Other commercial equipment may be equally suitable for the same
purposes.
The first fiber recovery experiments began in the fall of 1971.
Since then, the Forest Products Laboratory has used wood fiber to
make paper and building products out of the material in sufficient
quantity and size to run standard tests on the quality of materials
from urban v/aste. Lfforts are currently underway to obtain large enough
samples of the wood fiber to run production-scale tests of various
products.
132
-------
j:
In September 1971 the City of Madison entered into an agreement
with the Continental Can Company, Inc., which granted that company
5 years of salvage rights of the ferrous metals and alloys removed from
;-iilied refuse. Under terms of the agreement, the Continental Can
Company installed a magnetic separator, provided the trucks to haul
the material away, and developed the markets. They also paid the
City of "ladison 10 percent of any profits. Perfecting the mechanics of
magnetic separation has taken approximately a year and has required
some developmental work. A new magnetic separation unit has been
installed on the basis of the experience gathered with the first unit.
The new unit is obtaining approximately 95 percent of the ferrous metals,
3nd during a run in February 1973, 10 percent by weight was removed as
ferrous metals from approximately 20,000 Ibs. of solid waste.
After lengthy negotiations v/ith Urban Wood Fiber, Inc., the City
of Madison has obtained an outlet for market-sized logs from municipal
and private tree clearing operations in the City of Madison. This
operation was a direct spin-off of attempts to develop a milling system
for trees and logs generated in the City of Madison.
Another study undertaken through the City of Madison milling
.voject has been work by Professor Norman Bra ton of the University of
Hisconsin-Madison Mechanical Engineering Department on developing
techniques for shredding frozen tires. The tires are first frozen in
liquid nitrogen and then dropped into a hammermill. The frozen
rubber is quickly and easily separated from the remainder of the tire
carcass. Professor Braton proposes to develop equipment on a railroad
car which could be shipped from place to place to freeze and fragment
tires, thus producing marketable rubber.
There are over one hundred plants in the United Kingdom and on the
European continent which are milling refuse for landfill disposal without
daily cover. Some of these facilities have been in operation for 20
years. A result of the Madison project has been the construction and
operation of a number of facilities in the continental United States for
milling refuse for landfill disposal. In addition, there are also on the
drawing boards or under construction many more of such facilities.
Uhile only time will determine the ultimate success or failure of the
milling approach, it is felt that milling is definitely assuming a
significant role in the solid waste management field in the United
States.
133
-------
COIiCLUSIONS
Among the conclusions of the demonstration milling oroject in
Madison between 1967 and 1972 are the following:
1. As operated at Madison, the Gondard hammermill has a capacity
of 9 tons per hour with a 5-inch grate, and the Tollemacho hammermill
has a capacity of 14 tons per hour with a 34-hammer pattern. The grate
size and hammer pattern were chosen to grind refuse as coarsely as
possible without producing problems of blowing litter at the landfill
and without leaving food wastes accessible to vectors. (On a dry-weight
basis, between 30 and 90 percent of the particles produced by both mills
pass through a 2-inch screen.) Evaluations of the two mills separately
showed that the Gondard mill uses nearly as much electrical energy as the
Tollemache mill while producing only about 60 percent as much milled
product as the Tollemache.
2. Aside from some minor problems with the mills themselves, most
of the early operational problems were associated with conveying refuse
to the mills and carrying milled refuse to the landfill. The steeply
inclined feed conveyor and the stationary compactor with a 75-cu. yd.
transfer vehicle used with the Tollemache mill have greatly increased
the ability of the Madison plant to handle unprocessed and milled refuse
on a production basis.
3. Residential and light commercial refuse as collected at Madison
can be milled in either type of hammermill without extensive presorting,
with minimal hand-picking of unmillables, and with negligible downtime
due to mill stoppage.
4. Milled refuse has been left in a landfill without cover for
up to 6 years, and no complaints have been received about odors,
unsightliness, blowing litter, rodents, or insects. Public acceptance
of the milling plant and the landfill has been unusually good.
5. experience with milled refuse without daily cover indicates that
the quality of operation at this type of landfill is superior to
sanitary landfill operations at Madison with respect to travel over the
fill and at the face of the fill, dust, tracking of trucks on highways,
appearance during operating hours and maintaining a uniformly high
level of operation during cold and wet weather. Fully loaded trucks
weighing nearly 73,000 Ibs. can drive on milled refuse in inclement
weather. Also, tire problems have not been caused by travel on uncovered
milled refuse.
G. experience and specific testing have shown that there is less fire
hazard with milled than with unprocessed uncovered waste in a fill.
7. Rats are not able to survive on properly milled refuse
containing up to 20 percent wet garbage on a wet-weight basis. Cased
on this finding and on observations of the landfill site, there is very
little likelihood of rat infestation and survival in a properly operated
milled refuse landfill.
8. Under optimum weather and moisture conditions, flies probably
can breed in freshly milled refuse; however, once such refuse has aged
several months, this ability is evidently lost. Tests with the Gondard
mill showed that nearly all fly maggots passing through the mill during
134
-------
normal operation were killed. Fly counts and operating experience
at Madison indicate that there is no fly nuisance problem associated
with mil led refuse.
9. Compaction of cover soil, as v/ell as production of methane
and carbon dioxide by underlying refuse, created poor aeration
conditions for tree roots and thus led to a high mortality of trees
planted on milled and unprocessed refuse cells after 2. years. White
ash and crab were the most successful of the tree varieties planted
in that they developed effective lateral root systems in the densely
compacted cover soil.
10. Actual refuse density of milled refuse on a wet-weiqht basis
v;as found to be approximately 27 percent greater than the actual refuse
density of unprocessed refuse given equal compaction. Under the same
conditions, the effective refuse density of milled refuse was
calculated to be nearly 35 percent greater than that of unprocessed
refuse.
11. Leachate production occurs at a faster rate in milled uncovered
cells than in covered cells, milled or unprocessed. In the absence of
cover, milled refuse develops a relatively mature degradation pattern
and thus lowers the organic pollution load leaving the refuse in
leachate. Before a mature degradation condition develops in milled
refuse, large quantities of organics in particular are leached from
milled refuse.
12. The covered unprocessed refuse cells never produced organics
at as high a rate as did the milled cells during initial stages of
decomposition; however, the unprocessed cells continued to produce
organics at a fairly consistent rate throughout the duration of the
project. Thus, the milled refuse cells could be characterized as
producing more leachate contaminants during initial stages of decompo-
sition but less during later stages of decomposition than the
unprocessed refuse cells.
13. A water budget analysis shows that in Madison, Wisconsin
about 68 percent of incident precipitation on covered refuse in a
landfill evaporates, while the remaining 32 percent is divided almost
equally between runoff and infiltration (leachate). Virtually no
runoff and slightly more evaporation occurs with uncovered milled refuse.
14. In the first evaluation of the Gondard mill with a 5-inch
grate in mid-1968, a per-ton cost of $7.75 - including process and
hauling costs but excluding landfill ing costs - was experienced.
When this figure was adjusted to exclude factors related solely to
the experimental aspects of the operation, a comparable cost of $5.33
per ton emerged as a reasonable estimate for a production facility.
Juring an evaluation of the Tollemache mill in the summer and fall of
1970, a cost of $4.13 per ton for milling and hauling (but not
landfill ing) was calculated. A winter evaluation early in 1971
yielded a cost per ton of $1.99. These two Tollemache figures are based
135
-------
on the weight of refuse as received at the plant and include increased
labor costs incurred during the second evaluation.
15. For a two-mill, two-shift operation during the first 6 months
of 1972, a milling cost of $3.90 per ton was determined, based on 23,317
tons milled; another $0.48 per ton must be added to this figure to cover
compaction and hauling less than 1/2 mile. The operating costs of land-
filling for the first 6 months of 1972 in Madison were $0.99 per ton for
milled refuse and $3.00 per ton for unprocessed refuse, excluding any
land and development costs. During this period, regular sanitary land-
fills in Madison handled about twice as much material as did the milled
refuse landfill.
16. Based on the findings of this study, a cost of $3.11 per ton
of milled refuse was projected for a two-shift, two-Tollemache-mill
operation. This figure includes milling, hauling less than 1/2 mile,
and landfilling and assumes a continuous supply of millable refuse.
The City of Madison will continue milling refuse at the 01 in?Avenue
Milling Plant. In the Summer of 1973, the 01 in Avenue Mil Ifill operation
will be completed. The City has recently acquired another landfill
approximately nine miles from the milling plant. The milling plant will
serve as a central processing and transfer station from which the milled
refuse will be hauled to the new millfill facility.
136
-------
APPENDIX A
COST PROJECTIONS FOR NEW MILLING PLANTS AND LANDFILLS
This segment of the report includes detailed data in respect to
operating requirements and cost projections for a combination of one
through four mills operated either one or two operating shifts with
one maintenance shift. The estimates given in this portion of the
report are based on data collected utilizing the Tollemache vertical
shaft hammer-mill in Madison, Wisconsin. It is important to realize
that the cost projections have been developed on the basis of costs
at Madison. The wage rates, power and utility rates, depreciation, etc.
are all based on 1972 Madison cost data. To arrive at projected costs
for similar operations an economic study following the lines of that pre-
sented below may be made using the appropriate base rates as applied to
the area being studied.
Basic Design Criteria
The following design criteria were used in plant design and eventual
cost projections:
(1) Each mill will operate an average of 7 hours per shift at
a rate of 14 tons per hour.
(2) The plant will operate 245 days or 49 weeks per year. Three
weeks are allowed for repairs and breakdowns.
(3) The milling production day (one or two milling shifts) will
be followed by an eight-hour maintenance shift.
(4) Each plant will have as many feed conveyors as mills. One
and two mill plants will have one discharge conveyor and one
stationary compactor.
(5) Plants containing 3 or 4 mills will have two discharge con-
veyors and two stationary compactors.
(6) Cost projections are primarily based on pilot plant and
two shift studies conducted with the equipment mentioned
in the text.
(7) All pertinent data as to wage rates, utility rates,
depreciation, etc. are based on the evaluation data
from Madison.
Annual Tonnage
Each Tollemache mill operating one shift will have a daily capacity
of 14 tons/hour x 7 hours/day, or approximately 100 tons per day. The
corresponding annual tonnage for one mill shift will be 100 tons/day x
137
-------
245 days/year = 24,500 tons per year. Dally and annual tonnages for
combinations of mills and shifts are listed in Table A-l.
TABLE A-l
Daily and Annual Tonnage Processed for
Combinations of Mills and Shifts
Number of Mills
1234
One Shift
Daily Tonnage 100 200 300 400
Annual Tonnage 24,500 49,000 73,500 98,000
Two Shifts
Daily Tonnage 200 400 600 800
Annual Tonnage 49,000 98,000 147,000 196,000
Size of Plant
Experience has shown refuse quantities to vary throughout the year.
Peak tonnage rates are about 1.5 times the average daily tonnage. Good
plant design involves increasing available storage to allow for mill
breakdowns, etc. A factor about 1.5 times the average daily tonnage
through the plant is used to determine storage space. The maximum storage
requirement then becomes 150 percent of the average daily tonnage processed
on a one shift basis and is reduced to 125 percent of the daily tonnage
processed on a two shift basis. The reduction in excess capacity is due
to an increase in scale.
The maximum storage which must be provided is listed in Table A-2.
The values in Table A-2 were determined by multiplying the daily tonnage
in Table A-l by 150 percent for plants operating one shift, and by 125 per-
cent for plants operating two shifts.
TABLE A-2
Maximum Storage Requirements for
Combinations of Mills and Shifts - Tons
Number of Mills
1234
One Shift 150 300 450 600
Two Shifts 250 500 750 1,000
Assuming a density of 400 pounds per cubic yard on the dumping floor,
and an average stacked storage height of 8 feet, 0.53 ton can be stored per
square yard of floor space, or 0.059 tons per square foot of floor space.
The square footage of floor storage is computed by dividing the tonnages
in Table A-2 by 0.059 ton per square foot.
138
-------
TABLE A-3
Square Footage of Floor Storage Space Required
For Combinations of Mills and Shifts
Number of Mills
1234
One Shift 2,500 5,000 7,500 10,000
Two Shifts • 4,300 8,500 12,700 17,000
The total plant size including refuse storage space on the floor, con-
veyors and mills, and office is tabulated below. (A plant with three or
four mills will be considered to be equipped with two discharge conveyors
and two stationary compactors.)
TABLE A-4
Total Plant Size - Square Feet
Number of Mills
1 2 3 4
One Shift
Office & Workshop 1,000 1,000 1,500 1,500
Employee Facilities 300 300 500 500
Conveyor(s), Mill(s)
and Compactor(s) 3,500 5,000 9,000 10,500
Floor Storage-
Refuse 2,500 5.000 7.500 10.000
TOTAL 7,300 11,300 18,500 22,500
Two Shifts
Office & Workshop 1,000 1,000 1,500 1,500
Employee Facilities 500 500 700 700
Conveyor(s), Mill(s)
and Compactor(s) 3,500 5,000 9,000 10,500
Floor Storage-
Refuse 4.300 8.500 12.700 17,000
TOTAL 8,800 15,000 23,900 29,700
139
-------
The cost of foundations and building is computed by multiplying the
total space requirements in Table A-4 by $20 per square foot. The entrance
road and site grading is computed on the basis of 20 percent of the founda-
tions and building construction costs.
TABLE A-5
Cost of Foundations, Buildings, Entrance, Roads, and Grounds
For Combinations of Mills and Shifts
One Shift - Foundations and
Buildings $146,000 $226,000 $370,000 $450,000
- Entrance Roads
and Grounds 29,200 45,200 74,200 90,000
TOTAL $175,200 $271,200 $444,200 $540,000
Two Shift - Foundations and
Buildings $176,000 $300,000 $478,000 $594,000
- Entrance, Roads
and Grounds 35,200 60,000 95,600 11,900
TOTAL $211,200 $360,000 $573,600 $605,900
Labor Requirements and Costs
The number of men needed to man the combination of mills and shifts
under consideration are shown in Table A-6 as are the annual costs of labor.
Past experience at Madison has shown the need for proper supervision at all
levels of plant operation; thus, one supervisor is required for each case
shown. Also included are two maintenance men for one and two mill installa-
tions and three maintenance men for three and four mill installations. These
men will work an 8-hour maintenance shift following each day's milling
operation, whether one or two milling shifts are used.
Mill Maintenance - Annual Cost
(1) Hammers - 1100 tons/set at $106 per set
(2) Shafts - 500 tons/shaft at $l2.00/shaft
(3) Wear Plates - 20,000 tons/set at $1600 per set
(4) General Maintenance:
Mill - replace bearings and rotor once every three years for
one shift operation, $2,500 for each mill operating one
shift, and 80 percent additional for second shift.
Conveyors - $600 for the feed conveyor, and $200 for each dis-
charge conveyor for one mill operating one shift. The
cost for the second shift will be an additional 80 percent.
140
-------
(5) Welding Rods - 34 rods/set at $.50/rod; or $17.00 per set of hammers
50 Ibs./liner set at $1.60/lb.; or $80.00 per set of liners.
(6) Plant Supplies - $1,200 per one mill shift; $500 additional for each
mill; 50 percent additional for second shift.
TABLE A-6
Annual Labor Requirements and Costs
Number of Mills
Hourly Annual
Wage* Wage
One Shift
Supervisor $6.26** $13,000
Reduction Plant
Foreman-Operator $8.02 $16,660
Reduction Plant
Operator $7.21 $14,990
Public Works
Maintenance Man $6.20 $12,890
Scale Man $4.73 $ 9,840
Total
Two Shifts
Supervisor $6.26 $13,000 1111
Reduction Plant
Foreman $8.02 $16,660 2224
Reduction Plant
Operator $7.21 $14,990 2344
Public Works
Maintenance Man $6.20 $12,890 2233
Scale Man $4.73 $ 9,840 - 122
70,430
$ 85,420
$108,150
$139,800
Total $102,080 $126,910 $164,630 $197,950
*lncluding 30 percent fringe benefits and estimated overtime.
**0ther employees salary higher due to longevity pay program in City of Madison.
141
-------
TABLE A-7
Annual Mill Maintenance Costs
Number of Mills
One Shift
(1) Hammers
(2) Shafts
(3) Wear Plates
(4) General Maintenance
Mill(s)
Conveyor(s)
(5) Welding Rods
(6) Plant Supplies
Total
580
370
1,160
1,600 3,860
2,500 5,000
800 1,600
740
1.200 1,700
1,740
5,760
2,800
1,110
2.200
$ 2,330 $ 4,660 $ 6,990 $ 9,320
2,320
7,680
7,500 10,000
4,000
1,480
2,700
$ 8,580 $18,720 $28,100 $37,500
Two Shifts
(1) Hammers
(2) Shafts
(3) Liners
(4) General Maintenance
Mill(s)
Conveyor(s)
(5) Welding Rods
(6) Plant Supplies
Total
$ 4,660 $ 9,320 $13,980 $18,640
1,160 2,320
1,440 2,800
740 1,480
1,800
2,550
5,480
3,868 7,680 11,520
5,040
2,220
3,300
4,640
15,360
4,500 9,000 13,500 18,000
7,200
2,960
4,050
$18,160 $35,150 $53,040 $70,850
142
-------
Power for Mills, Conveyors, and Stationary Compactor
(1) The maximum demand for each mill and conveyor system is 160 kw.
(2) The maximum demand for the compactor is 20 kw.
(3) The power consumption for each mill and conveyor system averages
8.8 KWH/ton.
(4) The power consumption for each compactor is 16 KWH/hour.
The monthly demand charge for one mill and conveyor system is:
$2.00 for the first 10 kw, $2.00 per kw for each of
the next 90 kw, and $1.00 per kw up to the 160 kw
required for a total of $242
The monthly demand charge for each additional mill and conveyor system is:
$1.00 x 160 kw = $160
The monthly demand charge for each compactor is:
$1.00/kw x 20 kw = $20
The power consumption for one mill and conveyor system operating one
shift is 8.8 KWH/ton x 14 tons/hour = 123.2 KWH/hr. The monthly con-
sumption then is 123.2 KWH/hour x 7 hours/day x 5 days/week x 4.0
weeks/month = 17,250 KWH/month. The monthly consumption cost for one
mill operating one shift is: 0.026 x 500 KWH + $0.015 x 1000 KWH
+ $0.011 x 8500 KWH + $0.010 x 7,250 = $194. The monthly consumption
for all additional mill-shifts is $0.010 x 17,250 KWH = $172. The
monthly power consumption for each compactor working each shift is
16 KWH/hr x 7 hrs./shift day x 5 days/week x 4.0 weeks/month = 2,240
KWH/month. The corresponding monthly consumption cost for each com-
pactor shift is $0.080 x 2,240 KWH « $22. The above listed power con
sumption costs are summarized as follows:
The monthly power consumption cost for the first mill-shift
is $194.
The monthly power consumption for all additional mill-shifts
is $172.
The monthly power consumption cost for all compactor-shifts
is $22.
143
-------
TABLE A-8
Annual Power Costs for Mill(s)
Conveyor(s), and Compactors
Number of Mills
Lighting
One Shift
Demand
First Mill $242 $ 242 $ 242 $ 242 $ 242
Each Addnl. Mill 160 - 160 320 480
Each Compactor 20 20 20 40 40
Power Consumption
First Mill-Shift 194 194 194 194 194
Each Addnl. Mill-Shift 172 - 172 344 516
Each Compactor-Shift 22 22 22 44 44
Total Monthly Charge $ 478 $ 810 $ 1,184 $ 1,516
Annual Charge $5,736 $9,720 $14,208 $18,192
Two Shifts
Demand $242 $ 242 $ 242 $ 242 $ 242
Each Addnl. Mill 160 - 160 320 480
Each Compactor 20 20 20 40 40
Power Consumption
First Mill-Shift 194 194 194 194 194
Each Addnl, Mill-Shift 172 172 516 860 1,204
Each Compactor-Shift 22 44 44 88 88
Total Monthly Charge $ 672 $ 1,176 $ 1,744 $ 2,248
Annual Charge $8,064 $14,112 $20,928 $26,976
The consumption is to be proportioned on the basis of:
(1) The actual consumption in the reduction plant for one shift
from July 1970 through December 1970, and for two shifts from
January through June of 1972.
(2) The relative sizes of the buildings.
The original building used for a one-shift operation was 60 ft.
x 100 ft. = 6,000 sq. ft. The expanded building used for the two-shift
operation is approximately 13,000 sq. ft. The ratio of floor space for
144
-------
the projected plants, as listed in Table A-9 to 6,000 sq. ft. or 13,000
sq. ft. where applicable, is the relative size factor. The relative
sizes are:
TABLE A-9
Relative Sizes of Buildings
Number of Mills
One Shift 7,300 = 1.22 11,300 = 1.88 18,500 = 3.08 22,500 = 3.75
6,000 6,000 6,000 6,000
Two Shifts 8,800 - .68 15,000 = 1.15 23,900 - 1.84 29,700 « 2.28
13,000 13,000 13,000 13,000
Table A-10 shows the actual consumption for the existing plant per month
studied, and the proportional usage based on relative building sizes, for one
shift. Table A-11 contains the same information as related to a two-shift operation.
TABLE A-10
Actual and Proportioned Power Consumption (KWH)
For Lighting-One Shift
Actual KWH Number of Mills
Month Consumption 123 4
July 1970 4,414 5,385 8,298 13,595 16,552
August 1970 4,104 5,007 7,715 12,640 15,390
September 1970 4,018 4,902 7,554 12,375 15,068
October 1970 5,860 7,149 11,017 18,048 21,975
November 1970* 5,900 7,198 9,109 18,172 22,125
December 1970* 5,400 6,588 10,152 16,632 20,250
Average 6,038 8,974 15,243 18,560
*Figures adjusted - due to construction
145
-------
TABLE A-11
Actual and Proportional Power Consumption (KWH)
For Lighting-Two Shifts
Actual KWH Number of Mills
Month Consumption 1 23
January 1972 18,788 12,775 21,606 34,570 42,836
February 1972 17,088 11,620 19,651 31,442 38,960
March 1972 19,286 13,114 22,179 35,486 43,972
April 1972 14,734 10,019 16,944 27,110 33,593
May 1972 12,772 8,650 14,687 23,500 29,006
June 1972 13,294 9,040 15,288 24,460 30,310
Average 10,868 18,393 29,590 36,446
Based on the data given in Tables A-10 and A-11 a fairly good monthly average
of lighting power consumption for a one and two-shift operation can be obtained.
Table A-12 contains this average in relation to number of mills.
TABLE A-12
Average Monthly Power Consumption (KWH)
For Lighting One and Two Shifts
Number of Mills
123
One Shift 6,038 8,974 15,243 18,560
Two Shift 10,868 18,393 29,590 36,446
The demand charge would also be proportional to the size of the
building, but would be nearly constant throughout the year because the
charge is based on the high during the preceding 12 months. The maximum
demand for the original one-shift plant was 20 kw and for the two-shift
plant 40 kw. The maximum demand to be expected in any of the projected
plants is calculated by multiplying 20 kw or 40 kw whichever is applicable
by the relative building sizes as listed in Table A-9. The maximum demand
is tabulated below.
146
-------
TABLr. ^
Maximum Demand For Lighting - kw
Number of Mills
1/34
One Shift 24 38 62 75
Two Shifts 27 46 74 91
The annual cost of lighting can be computed using the monthly average
consumption (KWH) and maximum demand as shown in Tables A-12 and A-13 and th
current utility rates. Table A- 14 contains the summary of lighting costs.
TABLE A- 14
Annual Cost of Lighting for Combinations
Of Mills and Shifts
Number of Mills
1 2 3 4
One Shift
Demand $ 120 $ 696 $1,272 $1,584
Consumption 936 1,320 2,148 _ 2,592
Total $1,056 $2,016 $3,420 $4,176
Two Shi
Demand $ 432 $ 888 $1,560 $1,968
Consumption 1,572 2,568 4,044 _ 4,944
Total $2,004 $3,456 $5,604 $6,912
•'^•^PWABMMH^V^^VBM^^Bff'^^M'^^i^M^i^^M^^i^^^^^BIV^'VHlMfeMi^^B^BPVlVHMBBV^BB^HW^VI^^^^b'^VB^hlM'^^^^HmilVfl^B^V^I^^^*
Gas Heat
The gas costs are computed in a similar manner as for lighting costs
because consumption is dependent on building size. The actual consumption
is to be proportioned on the basis of:
(1) the actual consumption in the reduction plant for one shift
from July 1970 through December 1970, and for two shifts from
January through June of 1972, and
(2) the relative sizes of buildings.
The relative building sizes have been previously computed and are
shown in Table A-9 . The actual gas consumption for the existing plant,
and the proportioned usage based on relative building size are shown in
the following two tables.
147
-------
TABLE A-15
Actual and Proportioned
Gas Consumption (Heat) - One Shift
Month
July 1970
August 1970
September 1970
October 1970
November 1970
December 1970
Average
Actual
Consumption
(Cu.ft.xlOO)
0
0
30
780
3,420
3,600
0
0
37
952
4,173
4,392
1,590
Number of Mills
2 3
0
0
56
1,466
6,430
6,768
2,450
0
0
92
2,402
10,533
11,088
4,020
0
0
112
2,925
12,825
13,500
4,890
The consumption for two shifts is listed in the following table. The
proportioned consumption was obtained by multiplying the relative building
size by actual consumption used in the expanded plant under two-shift operation.
TABLE A-16
Actual and Proportioned
Gas Consumption (Heat) - Two Shifts
Month
January 1972
February 1972
March 1972
April 1972
May 1972
June 1972
Average
Actual
Consumption
(Cu.ft.xlOO)
8,679
8,297
5,124
1,983
618
540
5,900
5,640
3,480
1,350
420
370
2,860
Number of Mills
2 3
9,980
9,540
5,890
2,280
710
620
4,840
15,970
15,270
9,430
3,650
1,140
990
7,740
19,790
18,920
11,680
4,520
1,410
1,230
9,590
148
-------
Since only 6 months of good gas consumption data for a one and two
shift operation at Madison is available, only a rough estimate of gas costs
can be calculated. This estimate is computed by doubling the total cost
figure obtained from the proportioned consumption as presented in Tables A-15
and A-16. The estimate is valid because each six-month's term studied con-
tains equal periods of warm and cold weather in relation to that period not
studied. Table A-17 contains the projected yearly costs for gas as heat, based
on natural gas rates in effect at the time of evaluation*
TABLE A-17
Annual Cost of Gas (Heat)
One and Two Shifts
Number of Mills
1234
One Shift $1,746 $2,580 $4,104 $4,950
Two Shifts $2,976 $4,902 $7,722 $9,516
Water and Sewer
The usage is proportional to the tonnage milled, and has been found
to cost $0.002 per ton. Rather than going through the tedious procedure
used for estimating lighting and gas costs, the water and sewer costs are
estimated by multiplying the annual tonnage by $0.002 per ton. This method
is felt to be valid because the cost per ton was nearly constant throughout
the periods tested, and because the volume consumed is low enough that the
same utility rate applies.
TABLE A-18
Annual Water Costs for Combinations
Of Mills and Shifts
(Based on Cost of $0.002 Per Ton)
Number of Mills
1234
One Shift $49 $ 98 $147 $196
Two Shifts $98 $196 $294 $392
Tractor and Transfer Trailer Requirements
The number of trailers and tractors required is computed below. Based
on actual data a 70-yard trailer loaded with 15 tons of milled refuse would
have a density of 430 Ibs. per cubic yard. Each trailer is limited to
15 tons because of State highway regulations. Past experience has shown
that switching and unloading of each trailer averages 30 minutes when the
plant is located on the fill site. At a mill production of 14 tons per hour
one mill will fill one trailer in 64 minutes. Two mills operating at 14 tons
per hour each will fill one trailer in 32 minutes. Thus a one mill plant
will need a minimum of two trailers. A two mill plant could also function
with only two trailers; but would be advised to have three to minimize pro-
duction down time because of trailer breakdowns or delays in the switching-
unloading process. Both a one mill and two mill plant would require only
one tractor to pull the trailers. The above data would be applicable to a
one shift or two shift operation.
149
-------
As mentioned previously any three or four mill plant should contain
two stationary compactors. Taking this and the above data into considera-
tion > a three mill plant would require four trailers and a four mill plant
five trailers. Each three and four mill plant should be equipped with a
minimum of two tractors. Table A-19 summarizes the above data.
TABLE A-19
»
Tractor and Trailer Requirements
Number of Mills
123
One Shift
Tractors 1123
Trailers 2345
Two Shift
Tractors 1123
Trailers 2345
Annual Tractor and Transfer Trailer Operation and Maintenance Costs
Maintenance on the tractors and trailers is charged on a mileage basis;
$0,20/mi. for tractors and $0.25/mi. for trailers. To compute maintenance
costs it is first necessary to determine the number of loads taken to the
fill site; based on the information given above.
TABLE A-20
Number of Loads Per Day
Number of Mills
1234
One Shift 7 14 21 28
Two Shifts 14 28 42 56
The annual operating cost for the tractors can be computed as follows :
multiply the number of daily trips by round trip distance traveled, 0.5 mile
in Madison, x $0.20/mi. x 245 operating days/year. The computation for the
trailers is the same except $0.25/mi. is used instead of $0.20/mi. The ap-
propriate annual cost for tractor and trailer maintenance is shown in Table A-21.
150
-------
TA8LK A-2\
Annual Operation and Maintenance Cost
For Tractor and Transfer Trailer Operation Per Unit
Number of Mills
1 JL 3 4
One Shift
Tractors
Trailers
Total
Two Shifts
Tractors $344 $ 688 $1,032 $1,376
Trailers 430 860 1,290 1,720
Total $774 $1,548 $2,322 $3,096
$172
215
$387
$
$
•••V^^^v*
344
430
774
$
$1
^^^WPB«
516
645
,161
$
$1,
688
860
548
Annual Stationary Compactor Maintenance Costs
Past studies conducted at the Madison plant have revealed that on the
average, stationary compactor maintenance has cost approximately $0.02/ton,
based on a one shift operation. It would be reasonable to assume that a
50 percent increase in costs would be experienced when the machinery is
operated on a two shift basis. Using these figures as a guide the annual
operating costs for the combination of shifts and mills being studied is
shown in Table A-22. It should be reemphasized that all three and four mill
plants are designed for two compactors.
TABLE A-22
Annual Stationary Compactor Maintenance Costs
Number of Mills
1 2 3 4
One Shift
First Compactor $ 480 $ 960 $ 960 $ 960
Second Compactor _- - 4-80 960
Total $ 480 $ 960 $1,440 $1,920
Second Shift
First Compactor $1,440 $2,880 $2,880 $2,880
Second Compactor _^ - 1,440 2,880
Total $1,440 $2,880 $4,320 $5,760
151
-------
Annual Front End Loader Operation and Maintenance
For one and two mill operations a small end loader would be sufficient
to handle all tonnage processed. For a three and four mill plant, a medium
range end loader would be required. The operation and maintenance of both
pieces of equipment is assumed to cost $4.25/hr. The end loader will operate
5 hours and 10 hours per day for a one mill, one shift and one mill, two shift
operation, respectively. The same piece of machinery will operate 6 hours and
12 hours per day for a two mill, one shift and two mill, two shift operation,
respectively. The end loader will operate 7 and 14 hours for a three mill,
one and two shift operation, and 7 and 14 hours for a four mill, one and two
shift operation, respectively.
TABLE A-23
Annual Operation and Maintenance Costs
For Front End Loader
Number of Mills
1234
One Shift $ 5,205 $ 6,246 $ 7,287 $ 7,287
Two Shifts $10,410 $12,492 $14,575 $14,575
Amortization*
Amortization data and annual costs are listed below for all depreciable
items .
Building
Amortize over 20 years at 4.0 percent interest, salvage estimated at
3.0 percent of original cost as contained in Table A-5.
^Amortization or annual cost is calculated on the basis of the following
equation: **•
A.C. = (P-L) (CRF) - Li
where A.C. = annual cost
P = initial investment
L = salvage value
CRF = capital recovery factor
i% = interest rate
N - rated full life
The interest rate, i%, used in the computations > is a function of the
expenditure involved; i.e., short term notes or long term bonds. The
City of Madison usually funds large expenditures by long term bonds.
The interest rates in effect at the time of writing are 4.0 percent on
long term bonds.
Grant, Eugene L. and Ireson, W. Grant. Principles of Engineering
Economy, Fourth Edition. New York: The Roland Press, 1964.
152
-------
TABLE A- 2 4
Annual Cost o' Voum'..-i( UM\ .itu'
One Shift
Two Shifts
Number of Mills
23
$12,720 $19,690 $32,250 $39,200
$15,330 $26,130 $41,640 $43,980
TABLE A-25
Amortization* Data and Annual Costs
Cost Item
Depre- Int.
Original ciation Rate, Salvage Annual
Cost Rate-Yrs. % Value Cost
Scale
$15,000
20
4.0 $ 1,000 $ 1,076
Front Endloader
Michigan
Case
Trailers (ea)
Tractor (ea)
$32,000
$22,000
Grinder and Conveyor (ea) $140,000
Stationary Compactor (ea) $ 22,000
$ 19,000
$ 15,000
12
12
15
15
12
15
4.0 $ 5,000 $ 3,089
4.0 $ 3,000 $ 2,153
4.0 $ 4,000 $12,400
4.0 $ 1,000 $ 1,930
4.0 $ 750 $ 1,982
4.0 $ 1,500 $ 1,275
Annual Projected Operating and Amortization Costs
For Combinations of Mills and Shifts
Tables A-26 and A-27 summarize all costs of operating the various
sized plants and the amortization rates of each.
*NOTE: This is not a surplus fund for replacement of equipment, but
reflects municipal approaches for amortization.
153
-------
TABLE A-26
Annual Cost for One Through Four Mills - One Shift
Number of Mills
123
Labor
Mill Maintenance
Power - Mills, Conveyors,
and Compactors
Lighting
Gas (Heat)
Water
$ 70,430 $ 85,420 $108,150 $139,800
8,580 18,720 28,100 37,500
5,740
1,060
1,750
9,720 14,210 18,190
50
2,020
2,580
100
3,420
4,100
4,180
4,950
150
200
Tractor and Trailer Operations
and Maintenance 390
Front End Loader Operation
and Maintenance
Compactor Maintenance
5,210
480
770
6,250
960
1,160
7,290
1,440
1,550
7,290
1,920
Subtotal, Operating Costs $ 93.690 $126.540 $167,920 $215.580
Amortization
Building
Scale
Front End Loader
Mill(s) and Conveyors
Stationary Compactor(s)
Trailers
Tractor(s)
$ 12,720 $ 19,690 $ 32,250 $ 39,200
1,080
2,150
1,930
3,960
1,280
1,080
2,150
1,080
4,090
1,930
5,940
1.280
3,860
7,960
2,560
1,080
3,090
12,400 24,800 37,200 49,600
3,860
9.940
3,840
Subtotal, Amortization $ 35.520 $ 56,870 $ 88,000 $110,610
Total Annual Cost
$129.210 $183.410 $255,920 $326,190
154
-------
TABLE A-27
Annual Costs for One Through Four Mills - Two Shifts
Number of Mills
Labor
$102,080 $126,910 $164,630 $197,950
Mill Maintenance
Power - Mills, Conveyors,
and Compactors
Lighting
Gas (Heat)
Water
Tractor and Trailer Operation
and Maintenance
18,160 35,150 53,040
70,850
8,060 14,110 20,930
2,000
2,980
100
3,460
4,900
200
5,610
7,720
290
26,980
6,910
9,520
390
780
1,550
2,320
3,100
Front End Loader Operation
and Maintenance 10,410 12,490 14,580 14,580
Compactor Maintenance
1,440
2,880
4,320
5,760
Subtotal Operating Costs $146,010 $201.650 $273.440 $336.040
Amortization
Building
Scale
Front End Loader
Mill(s) and Conveyors
Compactor(s)
Trailers
Tractor(s)
Subtotal Amortization
Total Annual Cost
$ 15,330 $ 26,130 $ 41,640 $ 43,980
1,080
2,150
1,930
3,960
1,280
1,080
2,150
1,930
5,940
1,280
1,080
3,090
12,400 24,800 37,200
3,860
7,960
2,560
1,080
3,090
49,600
3,860
9,940
3,840
$ 38.130 $ 63.310 $ 97.390 $115.390
$184,140 $264,960 $370,830 $451,430
155
-------
Landfilling Cost Projections
Landfill cost projections will be based on the tonnages processed by
each combination of mills and shifts discussed in the previous section.
Costs will include labor for compaction, machine operation and maintenance,
and equipment amortization. Land costs are excluded. No charges are pro-
jected for cover material since no daily cover is used when landfilling milled
refuse at Madison.
Labor and Equipment Requirements and Costs:
Projections based on data obtained at Madison reveal that one operator
utilizing a steel wheeled compactor can efficiently handle 300 to 360 tons of
milled refuse per day. Based on this projection and the assumption that
landfills adjacent to the projected plant operations will be operated only
one 8-hour shift per day, the following labor and equipment requirements
are made,
TABLE A-28
Daily Landfill Labor and
Equipment Requirements
Number of Mills
1234
One Shift
Tons per day
Man Shifts
Trash Pak(s)
Two Shifts
Tons per day
Man Shifts
Trash Pak(s)
100
1
200
200
400
300
600
400
800
156
-------
TABLE A-2 9
Annual Labor Costs - Landfill Operations
Number of Mills
Annual*
Employee Wage
Operator $13,000 $13,000 $13,000 $13,000 $13,000
Site Control 11,300 5,650 5,650 11,300 11,300
One Shift Supervisor 15,100 3.800 3,800 7,500 7,500
TOTAL $22,450 $22,450 $31,800 $31,800
Operator $13,000 $13,000 $13,000 $26,000 $26,000
Site Control 11,300 5,650 11,300 11,300 11,300
Two Shift Supervisor 15,100 3,800 7,500 15,100 15.100
TOTAL $22,450 $31,800 $52,400 $52,400
^Includes 30% fringe benefits excluding overtime.
To compute equipment operating and maintenance costs it is assumed
that the landfill compactor will operate a minimum of 2 hours per 8 hour
day and a maximum of 6 hours per 8 hour day at the rate of spreading and
compacting 45 to 65 tons of milled refuse per hour. Operation and maintenance
for landfill equipment are charged at the rate of $6.25/hour of operation.
Compactor amortization is charged at the rate of $5,960 per year for an
initial investment of $45,000 and an 8-year equipment life.
TABLE A-30
Annual Operating and Maintenance Costs -
Landfill Equipment
Number of Mills
1 2 3 4
One Shift
Hours 990 1,225 1,470 1,715
Compactor Cost $6,130 $ 7,660 $ 9,190 $10,720
Misc. Equipment Cost 610 770 920 1,070
Total Cost $6,740 $ 8,430 $10,110 $11,790
Two Shifts
Hours 1,225 1,715 2,940 3,430
Compactor Cost $7,660 $10,720 $18,380 $21,440
Misc. Equipment Cost 770 1,070 1,840 2,140
Total Cost $8,430 $11,790 $20,220 $23,580
157
-------
Total Annual Projected Landfilling Costs:
TABLE A-31
Annual Projected Landfilling Costs
Number of Mills
Landfill 123
One Shift
Labor $22,450 $22,450 $31,800 $31,800
Compaction Equipment -
Operation and
Maint enance 6,740 8,430 10,110 11,790
Subtotal Operating Costs $29,190 $30,880 $41,910 $43,590
Depreciation -
Compaction Equipment $5,960 $ 5,960 $11,920 $11,920
Total Operating Costs $35,150 $36,840 $53,830 $55,510
Two Shifts
Labor $22,450 $31,800 $52,400 $52,400
Compaction Equipment -
Operation and
Maint enance 8,430 11,790 20,220 23,580
Subtotal Operating Costs $30,880 $43,590 $72,620 $75,980
Depreciation -
Compaction Equipment 5,960 5,960 11,920 11,920
Total Operating Costs $36,840 $49,550 $84.540 $87,900
158
-------
Summary of Annual Milling and Landfilline Costs
TABLE A-32
Annual Cost of Milling, Milled Refuse Transfer System and Landfilling
One Shift
Number of Mills
234
Annual Tonnage
24,500 49,000 73,500 98,000
Reduction Plant & Transfer
Operating Costs $ 93,690 $126,540 $167,920 $215,580
Amortization 35,520 56,870 88,000 110,610
Total Reduction Plant
$129,210 $183,410 $255,920 $326,190
Cost Per Ton*
Millfill
Operation Costs
Amortization
Total Landfill
Cost Per Ton**
TOTAL ALL OPERATIONS
COST PER TON
$5.27
$3.74
$3.48
$3,33
$ 29,190 $ 30,880 $ 41,910 $ 43,590
5,960 5,960 11,920 11,920
$ 35,150 $ 36,840 $ 53,830 $ 55,510
$1.43
$0.75
$0.73
$0.57
$164,360 $220,250 $309,750 $381,700
$6.70
$4.49
$4.21
$3.90
*Cost includes amortization, labor, operating, and milled refuse haul to
landfill less than one-half mile round-trip distance. Land cost excluded*
**Millfill cost includes labor and equipment costs with amortization. Land
cost and site preparation costs are excluded.
1
159
-------
TABLE A-33
Annual Cost of Milling, Milled Refuse Transfer System and Landfilling
Two Shifts
Annual Tonnage
Reduction Plant & Transfer
Operating Costs
Amortization
Total Reduction Plant
Cost Per Ton*
Mlllfill
Operating Costs
Amortization
Total Landfill
Cost Per Ton
TOTAL ALL OPERATIONS
COST PER TON**
Number of Mills
234
49,000 98,000 147,000 196,000
$146,010 $201,650 $273,440 $336,040
38.130 63.310 97.390 115,390
$184,140 $264,960 $370,830 $451,430
$3.75
$2.70
$2.52
$2.30
$ 30,880 $ 43,640 $ 72,620 $ 75,980
5,960 5.960 11,920 11,920
$ 36,840 $ 49,600 $ 84,540 $ 87,900
$0.75
$0.50
$0.57
$0.45
$220,980 $314,560 $455,370 $539,330
$4.50
$3.20
$3.09
$2.75
*Cost includes amortization, labor, operating, and milled refuse haul to
landfill less than one-half mile round-trip distance. Land cost excluded
**Millfill cost includes labor and equipment costs with amortization. Land
cost and site preparation costs are excluded.
160
-------
APPENDIX B
POSITION ON LANDFILLING OF MILLED SOLID WASTE *
A. BACKGROUND
The landfilling of milled solid waste without daily soil cover began
in Europe with claims that it was an environmentally acceptable and economic
method of final disposal. In June of 1966, a solid waste demonstration
grant was awarded to Madison, Wisconsin to evaluate the European experience
and to determine the feasibility of landfilling milled solid waste without
daily cover in this country.
In January 1971, the Madison project personnel met with OSWMP personnel,
a consulting engineer, and entomologists from the Bureau of Community
Environmental Management .(USDHEW), to review the progress and findings to
date from the Madison project. OSWMP concluded that the policy governing
soil cover for milled solid wastes should be as stated in Sanitary Landfill
Facts:
"The compacted solid wastes must be covered at
the conclusion of each day, or more frequently
if necessary, with a minimum of six inches of
compacted earth.11
It was also concluded that further investigation at Madison and in
other geographic and climatic areas was needed to fully resolve the
policy issue. In a February 2, 1971 memorandum, Mr. Richard Vaughan
expressed these findings to OSWMP Senior Staff and Regional Representatives.
Additionally, environmental evaluations of landfilling milled solid
waste made at the Madison demonstration site have been augumented by
information from site visits to other facilities. An increased interest
in the procedure is evidenced by the knowledge of six new sites being
planned, ten new sites under construction, and five sites operational.
Some of these sites are constructed and operated with provisional
approvals, some are operated in opposition to local regulations but in
all cases the operations do not adhere to the position stated by the
OSWMP on February 2, 1971.
Recent articles, based on European experience, findings from the
Madison project, and other new sites within the United States, have
appeared in engineering and public work journals. This information,
combined with equipment promotional activities has generated an increased
interest in the process particularly where problems exist in achieving
satisfactory sanitary landfill operations or where milling may compliment
resource recovery.
*This appendix, not a part of the grantee's original report, was prepared
and added by the Office of Solid Waste Management Programs, U.S. Environmental
Protection Agency.
161
-------
B, CURRENT POSITION
Landfilling milled solid wastes can be an environmentally accep-
table method of final disposal. The same sound engineering principles
involved in sanitary landfill sites, including a properly located,
designed, financed, and operated milling facility must be provided to
insure successful operations and to minimize adverse environmental
impacts. Since environmental, economical, and operational conditions
vary from existing sites, the need for cautious planning to meet local
conditions and to determine the feasibility of each new site must be
emphasized.
It must be recognized that this position is based on detailed
investigations at the Madison site augumented by general knowledge
from a few additional sites. The ability to mill, grind, or shred
wastes such that it is environmentally acceptable to landfill them
without daily cover is dependent on the process, its operation, and
local conditions such as the environment and the waste content.
It is, therefore, recommended that conditional approvals be given
by regulatory agencies contingent upon verification that the quality
of operation necessary to minimize environmental hazards is maintained.
Such verification should be supported by operational controls and
monitoring.
Except as modified below, the position statement on sanitary
landfill applies to milled solid waste disposal operation. Comments
relating milled solid waste to sanitary landfill requirements are
listed below in the order presented in the pending "Guidelines for
the Land Disposal of Solid Wastes.!!
1. As an alternative to sanitary landfill, landfilling milled
solid waste without daily soil cover can result in increased surface
water infiltration and accelerated decomposition which in turn can
result in earlier leachate production and temporarily increased
pollutional concentrations. Under the usual situation of landfill
construction over a period of years, peak leachate production and
concentrations occur only in a small part of the fill at any one
time. In areas where rainfall infiltration exceeds evaportrans-
piration and field capacity is reached, the total production of
leachate constituents has been shown to be equivalent to a sanitary
landfill which reaches field capacity and produces leachate.
Therefore, in accordance with the sanitary landfill position, it is
necessary to prevent leachate from entering surface or underground
sources of water supply. This can be accomplished by preventing
leachate production and/or by collecting and treating leachate
should it occur.
162
-------
2. As with sanitary landfill operations, design and operation must
conform to applicable air quality standards; specifically, open burning
of solid waste must be prohibited.
3. As with sanitary landfill cover, compacted, milled, uncovered
landfill surfaces must be left undisturbed to prevent odor. This does
not preclude vehicular traffic but precludes excavation of a finished
surface.
4. Although milling solid waste reduces the tendency for paper
to blow during placement, satisfactory control requires that the waste
be spread to a smooth contour and compacted promptly after placement.
5. A milled, uncovered solid waste landfill is much less
obnoxious than an open dump and to many observers is no more
obnoxious than bare earth,
6. Free venting or loss of gases from milled solid waste,
experienced in test cells, indicates that milled solid waste without
cover is less likely to trap gases in pockets or cause horizontal
gas migration. However, the addition of cover or possible migration
through fissures or broken pipe lines, etc. requires the same
attention to gas control as a sanitary landfill.
7. European experience, verified by tests at Madison, Wisconsin
and Purdue University indicates that:
Rats cannot extract sufficient food to sustain life;
from properly milled combined residential, commercial
solid waste (7-1/2 % organics wet weight in test) nor are
they attracted more readily to an uncovered milled solid
waste landfill than to a sanitary landfill (baiting studies);
the milling process kills nearly 100% of the maggots present
in incoming solid waste virtually eliminating fly emergence
(sampling studies); and flies are not attracted more readily
to an uncovered milled solid waste landfill (Scudder Grill
Study).
8. Undetected hazardous materials in incoming wastes have
been known to explode or ignite during the milling process. Protection
against explosions such as blow-off stacks and personnel shields
must be provided. Equipment to extinguish fires which may exist in
incoming solid waste or which may be ignited during the milling
process, during transport or on the landfill must be provided. No
operation should be located where birds might be a hazard to
aircraft flight operations.
163
-------
9. Site selection on an engineering basis is similar to that
for a sanitary landfill operation except the availability of daily
cover material is not required. The availability of emergency cover
is required (see operational plan requirements below). Final cover
and final use criteria should be the same as for a standard sanitary
landfill.
10. Only properly milled residential and commercial solid wastes
should be accepted in an uncovered milled solid waste landfill.
Items not accepted in a conventional sanitary landfill and volatile,
flammable, explosive or sludge wastes accepted in small quantities
at a conventional sanitary landfill, should not be accepted for
milling. Final disposal of all wastes not suitable for milling
must be in accordance with pending "Guidelines for the Land Disposal
of Solid Wastes."
11. All operations and aspects including lighting, dust control,
and noise levels must meet the requirements of the Occupational
Safety and Health Act of 1970. All solid waste storage areas must
be maintained and cleaned at the end of each day's operations, or
during continuous operation, as necessary, to prevent fly, rodent,
or other vector problems. All equipment must be maintained to
control spillage and to achieve a milled product quality necessary
to prevent environmental hazard.
12. All operational personnel must be specially trained and
instructed on the proper operation, maintenance, and safety
aspects of the facilities and equipment.
13. The operational plan must include provision for removal
and proper disposal of wastes within 24 hours should the mill
facility cease to meet the above conditions because of either a
temporary equipment breakdown or a loss of quality operation.
The operational plan must include provision of a stock pile of
emergency soil cover material and provision to convert the
operation to a sanitary landfill.
Preliminary project planning must include a detailed cost
analysis including means of establishing a sound financing and
revenue system, in order to guarantee that the quality of operation
necessary for environmental acceptability can be sustained. Milling
and landfilling residential and commercial solid wastes is usually
not cost competitive with conventional sanitary landfill disposal.
Cost comparisons to justify milling as an alternative to more
extensive disposal systems including transfer stations or cover
material transport must be evaluated on a local basis. Each
community or private operator must make their own thorough economic
evaluation of the alternative disposal systems. Milling costs
including labor, amortization, utilities, maintenance, and supplies
164
-------
recorded at and relevant only to the Madison project were as high
as $7.07/ton for a single 9 ton/hr. Gondard mill operating 5 to
6 hours a day. Costs for a single 15 ton/hr. Tollemache mill operating
about 5 hours a day have been recorded at $5.10/ton while costs for
a similar operation with "hard to mill" wastes ran as high as
$6.44/ton. Transportation to the adjacent landfill averaged about
$0,40/ton additional. Spreading and compacting costs averaged an
additional $0.50 ton. Cost projections for the. combined operation
of one Gondard mill at 9 ton/hr. and one Tollemache mill at 15 ton/hr.,
milling 280 tons/day or a two shift operation is approximately
$3.50/ton excluding transport and disposal. These costs reflect
local labor rates, union contracts, construction costs, and electrical
costs, etc.
C. REFERENCES
1. Ham, R. K., W. K. Porter, and J. J. Reinhardt. Refuse milling
for landfill disposal. JEn Solid Waste Demonstration Projects;
Proceedings of a Symposium, Cincinnati, May 4-6, 1971. Washington,
U.S. Government Printing Office, 1972. p.37-72.
2. Solid waste reduction/salvage plant; an interim report; City of
Madison pilot plant demonstration project, June 14 to December 31,
1967. Washington, U.S. Government Printing Office, 1968. 25 p.
3. Sanitary landfill guidelines. U.S. Environmental Protection Agency.
(in press.)
\
4. Brunner, D. R., and D. J. Keller. Sanitary landfill design and
operation. Washington, U.S. Government Printing Office, 1972. 59 p.
5. Stirrup, F. L. Public cleansing: refuse disposal. Oxford, Pergamon
Press, 1965. 144 p.
6. [Great Britain]. Bepartment of the Environment. Refuse disposal;
fceport of the Working Party on Refuse Disposal. London, Her Majesty's
Stationary Office, 1971. 199 p.
7. [Great Britain]. Department of the Environment. Report of the
Working Party on Refuse Disposal. Circular 26/71. Apr. 1971. 7 p.
ya555
165
------- |