Cincinn.
EPA-6
Processing
Equipment for
Resource Recovery
Systems
Vol. Ill
Field Test
Evaluation of
Shredders
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports ;
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution-sources to meet environmental quality standards.
This document is available to-the pub He through tne National Technical Informa-
tion Service, Springfield, Virginia 22161
-------
EPA-600/2-80-007C
July 1980
PROCESSING EQUIPMENT FOR RESOURCE RECOVERY SYSTEMS
Volume III. Field Test Evaluation of Shredders
by
George M. Savage
Geoffrey R. Shiflett
Cal Recovery Systems, Inc.
Richmond, California 94804
Contract No. 68-03-2589
Project Officer
Donald A. Oberacker
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OH 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
ii
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health
and welfare of the American people. Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of that environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is a necessary first step in problem solution,
and it involves defining the problem, measuring its impact, and searching for
solutions. The Municipal Environmental Research Laboratory developes new and
improved systems technology to minimize the adverse economic, social, health,
and aesthetic effects of pollution. This publication is one of the products
of that research.
This report presents the results of a study of nine shredders used for
size reduction of municipal solid wastes in six processing plants. This is
the third in a series of reports on studies of processing equipment for
resource recovery systems. Volume I - State of the Art including research
needs, and Volume II - Magnetic Separators, Air Classifier and Ambient Air
Emissions Tests were prepared under EPA Contract No. 68-03-2387. Volume III -
Field Evaluation of Shredders, was conducted under EPA Contract No. 68-03-2589.
Francis T. Mayo, Director
Municipal Environmental Research
Laboratory
-------
ABSTRACT
This report presents the results of a program to test and evaluate large-
scale shredders used for the size reduction of solid waste. In all, tests
were conducted on seven horizontal harranermills, one vertical hammermill, and
one vertical ring shredder at six commercial sites (Appleton, Wisconsin;
Ames, Iowa; Cockeysville, Maryland; Great Falls, Montana; Tingon Falls,
New Jersey; and Odessa, Texas). Both two stage size reduction (Ames) and
single stage size reduction were studied as part of this work. Evaluation
and interpretation of the data has resulted in the development of analytical
relationships among the comminution parameters and the establishment of
levels of performance with respect to energy consumption and hammer wear
associated with size reduction of solid waste.
This report was submitted in fulfillment of Contract No. 68-03-2589 by
Midwest Research Institute under the sponsorship of the U.S. Environmental
Protection Agency. This report covers the period from September 1, 1977 to
February 28, 1979.
-------
CONTENTS
Foreword iii
Abstract iv
Figures vii
Tables x
Definitions and Nomenclature xii
Acknowledgment xiv
1. Introduction 1
Overview 1
Site descriptions 2
Hammer maintenance programs. 26
2. Findings and Conclusions 33
Energy and Particle Size 34
Hammer wear 36
3. Recommendations 37
4. Materials and Methods 39
Power monitoring instrumentation 39
Other equipment 42
Size distribution analysis 43
5. Experimental Procedures 44
General procedures for power-flow rate data
collection 44
Appleton procedure 46
Ames procedure 47
Cockeysville procedure 49
Great Falls procedure 50
Tinton Falls procedure 51
Odessa procedure 52
Hammer wear experimental procedure 53
6. Results and Discussion 59
Moisture content of shredded refuse 59
Measured throughputs 59
Size distribution of shredded solid waste 61
Measured power 64
Throughput effects on size 64
Estimation of specific energy requirements 69
Motor sizing 72
-------
CONTENTS (continued)
6. Results and Discussion (continued) 59
Power-flow rate-moisture relationships 72
Discussion of Appleton holdup data 78
Discussion of the results for the Ames primary and
secondary shredders 79
Discussion of bi-directional rotation of
Cockeysville #1 shredder 80
Discussion of Great Falls vertical hammennill
results 81
Discussion of Tinton Falls vertical ring shredder
results 83
Comparison of Cockeysville and Odessa results. ... 83
Specific energy comparison of single versus
multiple stage size reduction of MSW 85
Results of hammer wear studies . 89
Comparison of hammer wear data among sites 96
Site specific comments with regard to hammer
maintenance 99
Summary of shredder performance 99
7. Costs Associated with Refuse Size Reduction 100
Energy costs for shredding 100
Comparison of alternative hammer maintenance
programs 100
Summary of costs 104
References 105
Appendix 106
VI
-------
FIGURES
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Outagamie County Solid Waste Plant; Appleton, Wisconsin
West mill at the Appleton site
View of west mill and motor T
Flow schematic for Appleton facility
Solid waste recovery system; Ames, Iowa
Ames primary shredder shown partially disassembled
Flow schematic for Ames plant
Baltimore County Resource Recovery Facility; Cockeysvi lie,
Mary 1 and
Shredder #1 at Cockeysvi lie
Flow schematic of Cockeysvi lie plant
Great Falls Solid Waste Reduction Facility; Great Falls,
Montana
Exterior view of Great Falls vertical hammermill
Interior view of Great Falls vertical hammermill
Flow schematic of Great Falls facility
Monmouth County Reclamation Center; Tinton Falls, New Jersey
Interior of Monmouth County Reclamation Center, Plant 2
Interior of vertical ring shredder
Flow schematic of Tinton Falls facility
City of Odessa Solid Waste Management Facility; Odessa, Texas
Odessa horizontal hammermill
Paqe
4
5
5
7
8
9
11
13
14
16
16
18
18
19
20
22
22
23
24
25
Vll
-------
FIGURES (continued)
Number
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
Flow schematic of Odessa facility
Hammers undergoing hardfacing at Appleton; weld deposite are
shown on hammer tips
Hardfacing application at the Cockeysville facility
Utilization of pin puller for hammer changes
View of hammers in Ames primary shredder
Inside view of Ames secondary shredder
New hammer for Ames secondary shredder
Worn hammers from Ames secondary shredder
Used and new ring grinders for Tinton Falls vertical shredder
Power monitoring equipment; Appleton
Installation of voltage and current transformers; Appleton ..
Schematic layout of power monitoring equipment
Cleaned hammer being weighed prior to installation in west
mill
Set of newly hardfaced hammers ready for installation at the
Cockeysville facility
Pictoral of vertical hammermill
Internal configuration of the Tinton Falls ring shredder ....
Relationship between nominal and characteristic product sizes
for shredded MSW
Range of size distributions of shredded MSW measured at
Appleton, Ames, and Cockeysville
Average size distribution of raw and shredded solid waste ...
Relationship of product size and throughput for Ames primary
shredder
Relationship of product size and throughput for Ames
secondary shredder
Page
28
29
29
30
30
31
31
32
32
40
40
41
55
55
56
58
62
63
64
67
67
viii
-------
FIGURES (continued)
Number Page
42 Relationship of product size and throughput for Tinton Falls
shredder 68
43 Relationship of product size and throughput for Odessa
shredder 68
44 Specific energy consumption (A.D.B.) as a function of product
size 69
45 Specific energy consumption (W.B.) as a function of product
size 71
46 Net power draw of Appleton East Mill 73
47 Net power draw of Ames primary shredder 74
48 Net power draw of Ames secondary shredder 74
49 Net power draw of Cockeysville #1 shredder 75
50 Net power draw of Great Falls 20 TPH shredder 75
51 Net power draw of Tinton Falls shredder 76
52 Net power draw of Odessa shredder 76
53 Cross-sectional view of discharge under #1 shredder at
Cockeysville 81
54 Relationship between flow rate and specific energy for Great
Falls vertical hammermill 82
55 Representative hammermill and ring grinder power recordings . 84
56 Comparison between Odessa and Cockeysville flow rate-size
relationships 86
57 Hammer wear as a function of allow hardness 90
58 Hammers removed from Great Falls vertical hammermill 93
59 Hammer installation pattern for Great Falls wear experiments 95
60 Hammer wear as a function of alloy hardness and degree of
size reduction 98
61 Cost of energy associated with size reduction 101
IX
-------
TABLES
Number Page
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Summary of Tests Conducted for the Shredder Evaluation
Studies
Appleton Shredder Specifications
Ames Shredder Specifications
Cockeysville Shredder Specifications
Great Falls Shredder Specifications
Tinton Falls Shredder Specifications
Odessa Shredder Specifications
Transformer Ratios Used at Each Site
Sample Size
Summary of Average Values of Important Parameters Measured
During the Shredder Performance Evaluation
Throughputs Measured During Shredder Performance Evaluations
Power Measured During Shredder Performance Evaluations
Effect of Variation of Throughput Upon Characteristic Product
Size and Specific Energy
Specific Energy Requirements Versus Product Size Utilizing
Test Results from Vertical and Horizontal Hammermills
Results of Multiple Regression Analyses; Net Power as a
Function of Flow Rate and Moisture Content
Shredding Data Used to Develop Comparisons Between Single and
Multiple Stage Size Reduction
Net Power and Energy Requirements for Single and Multiple
Stage Size Reduction
3
6
10
15
17
21
27
42
46
faU
61
65
66
72
//
37
88
-------
TABLES (continued)
Number Page
18 Calculation of Required Gross Specific Energy to Obtain
Equivalent Product Size for Ames and Cockeysville 88
19 Appleton Hammer Wear Experiments 90
20 Ames Hammer Wear Experiments 91
21 Cockeysville Hammer Wear Experiments 92
22 Great Falls Hammer Wear Experiments 94
23 Odessa Hammer Wear Experiments 94
24 Tinton Falls Wear Experiments 95
25 Normalization of Hammer Wear Measurements 98
26 Typical Costs for Hammer Maintenance 102
27 Cost Summary for Hammer Maintenance Programs 103
XI
-------
DEFINITIONS AND NOMENCLATURE
DEFINITIONS
Characteristic Size
Degree of Size
Reduction (Z0)
Nominal Size
Shredder Holdup
Specific Energy
Tons
ABBREVIATIONS
A.D.B.
Kwh
Kwh/T
MC
Re
Td
TPH
TRY
TW
W.B.
Screen size corresponding to 63.2% cumulative weight
percent passing
Defined by the ratio (F0-X0)/F0. A dimension!ess
parameter used to describe the amount of size reduc-
tion occuring during shredding, e.g. Zp = 0 implies
no size reduction, Z0 = 0.90 imnlies characteristic
product size, X0, is equal to 0.10 F0
Screen size corresponding to 90.0% cumulative weight
percent passing
Material in shredder at any instant in time
Net energy requirement (or consumption) expressed on
the basis of net energy utilized per ton of waste
shredded
In all cases, the word tons refers to metric tons
(1000 kilograms)
Air dry weight basis
Kilowatt hour
Kilowatt-hours per ton
Air dry moisture content
Rockwell "C" hardness scale
Metric tons on an air dry weight basis
Metric tons per hour
Metric tons per year
Metric tons on a wet weight basis
Wet weight basis
xn
-------
DEFINITIONS AND NOMENCLATURE (continued)
VARIABLES AND PARAMETERS
E0 Specific energy consumption on an air dry weight basis
E0 Average specific energy consumption on an air dry
weight basis
EQW Specific energy consumption on a wet weight basis
EQ Average specific energy consumption on a wet weight
w basis
F0 Characteristic feed size
H Shredder holdup
mj Flow rate (throughput) on an air dry weight basis
ri^ Flow rate (throughput) on a wet basis
Freewheeling power
PQ Gross power
PN Net power (PN = PQ - PFW^
w- Average material weight loss per hammer per ton of
1 waste shredded
W0 Hammer wear of a full complement of hammers on the
basis of material loss per ton of waste shredded
X0 Characteristic product size (screen size corresponding
to 63.2% cumulative passing)
XQ Average characteristic product size
Xgn Nominal size (screen size corresponding to 90%
cumulative passing)
Xgg Average nominal product size
ZQ Degree of size reduction, (F0-X0)/F0
err Standard deviation of specific energy on an air dry
weight basis
oi Standard deviation of specific energy on a wet weight
w basis
a0 Standard deviation of characteristic product size
ago Standard deviation of nominal product size
x i i i
-------
ACKNOWLEDGMENTS
The study of shredder operation and performance reported herein, was
developed and conducted by Cal Recovery Systems, Inc. (CRS), Richmond,
California, under Subcontracts No. 4424-D and 4426-D with Midwest Research
Institute (MRI). The principals of CRS have been actively involved in the
area of size reduction of solid waste for a period of 8 years. Their efforts
in this field include pioneering work in the characterization of parameters
governing the process of size reduction, development of analytical relation-
ships governing size reduction of solid waste, and the influence of machine
characteristics, such as rotor rpm and grate opening size upon product size,
energy consumption, and hammer wear.
This study was conducted by Cal Recovery Systems under the direction of
Dr. Louis F. Diaz and Mr. George M. Savage. Actual site testing and data
analysis were performed by Mr. George M. Savage, Dr. Geoffrey R. Shiflett,
and Mr. Stanley M. Boghosian. Mr. Savage and Dr. Shiflett are the principal
authors of this report. The manuscript was typed and edited by Ms. Linda
Eggerth.
The Project Director for MRI was David Bendersky, Principal Engineer.
The EPA Project Officer was Donald A. Oberacker, Senior Mechanical Engineer.
The success of the testing programs at the six sites was to a large
degree the consequence of the excellent cooperation shown to us by those
associated with the shredding facilities. The cooperation of the following
people was instrumental in arranging for testing at each site:
Mr. Robert Brickner, All is Chalmers, Appleton, Wisconsin;
Mr. Eugene Higgins, Outagamie County, Appleton, Wisconsin;
Mr. Arnold Chantland, City of Ames, Iowa;
Mr. Hal Gordy, Teledyne National, Northridge, California;
Mr. Carl Able, City of Great Falls, Montana;
Mr. John Gray, Monmouth County, New Jersey; and
Mr. Dwayne Dobbs, City of Odessa, Texas.
In particular we wish to acknowledge the help and cooperation of the
personnel from each plant, especially that of the plant managers, Mr. Ed
Maloney (Appleton), Mr. Jerry Temple (Ames), Mr. Ken Cramer (Cockeysville),
Mr. Dale Young (Great Falls), Mr. John Gray (Monmouth County), and Mr.
Dwayne Dobbs (Odessa).
xiv
-------
SECTION 1
INTRODUCTION
OVERVIEW
Over the past fifteen years, the use of shredders in the field of solid
waste management has seen steady growth. From first attempts at using size
reduction of solid waste as the initial step in the production of a suitable
material for composting, the use of shredders has grown into the field of
full-scale, integrated resource recovery facilities. In between these
events, shredders have seen service in the areas of ferrous scrap recovery
from solid waste and treatment of refuse for landfill disposal without the
necessity of cover material.
Shredding (or synonymously, grinding, milling, size reduction, or commi-
nution) of solid waste has taken on an added degree of significance since
the emergence of :large, full-scale resource recovery operations. For such
facilities, size reduction often represents the first step in processing the
waste stream. Consequently, the unit process of size reduction affects all
equipment involved in downstream material handling and separation. In addi-
tion, the shredding operation normally accommodates 100 percent of the
waste, whereas other unit processes in resource recovery plants generally
handle only particular fractions of the waste stream. As a result, the
importance of size reduction is generally recognized, albeit poorly under-
stood, by the solid waste industry.
The proliferation of shredders in the solid waste industry has stimula-
ted interest in their operation, evaluation, and performance. For example,
criteria for estimating shredder operation and performance are needed in the
design stage of resource recovery plants. Also, a plant manager may wish.to
know how certain operational changes involving a shredder (such as a change
in size of the grate openings or variation in.shredder throughput) may
affect shredder operation (e.g. energy requirements and size of product).
Until now, only pilot-scale research data has been available for charac-
terizing size reduction of solid waste. In fact, in-depth testing and eval-
uation of shredders has only been conducted on units with a capacity of less
than 15 tons per hour. Consequently, the obvious question remains, as to the
usefulness of the results obtained from pilot-scale shredders when applied
to the large-scale shredding operations of typical plants (in the range of
15 to 100 tons per hour).
The testing and evaluation program described in this report was initia-
ted to extend the knowledge of solid waste size reduction and resolve the,
question concerning the applicability of pilot-scale research results.
1
-------
The underlying motivation was to establish predictive relationships,
design criteria, evaluation techniques, and levels of performance for
large-scale size reduction equipment. The information contained herein has
been developed for the use and consideration of those associated with solid
waste management including shredder manufacturers, plant designers, plant
operators, and researchers.
The test program involved detailed measurements of energy and hammer
wear associated with size reduction of solid waste, the two most important
aspects of size reduction due to their effects on operational cost. In
order to ascertain relationships involving energy and hammer wear, a rigid,
scientific protocol was established and implemented. Cooperation of person-
nel at each site was instrumental in maintaining a consistent test procedure
among plants.
The test plans were directed toward assessment of energy consumption and
hammer wear associated with shredding. In order to adequately assess energy
consumption, the test program called for collection of samples to determine
the product size and moisture content of shredded refuse. The samples were
then correlated with energy consumption. The study concentrated on the
evaluation of various hardfacing alloys as well as the base material typi-
cally used at each site for determination of hammer wear. Rates of hammer
wear were determined and interpreted.
The shredders that were tested are installed at facilities located in
Appleton, Wisconsin; Ames, Iowa; Cockeysville, Maryland; Great Falls, Mon-
tana; .Tinton Falls, New Jersey; and Odessa, Texas.
Since the number and types of tests performed on the shredders at each
site were necessarily site specific, Table 1 has been prepared to summarize
the tests that were performed. In all, tests (either power or wear measure-
ments) were conducted on nine shredders. Due to restraints present at some
sites, both power and wear measurements could not be collected for some
shredders, as shown in Table 1.
In order to acquaint the reader with the plants that were visited, the
utilization of each shredder, the general processing sequence, and programs
of hammer maintenance, the following two sections, Site Descriptions and
Hammer Maintenance Programs, give a brief overview of each facility.
SITE DESCRIPTIONS
Appleton
The Outagamie County Solid Waste Shredding Facility (Figure 1) located
in Appleton, Wisconsin, operates two Allis-Chalmers Model KH 12/18 horizon-
tal hammermi 11s (Figures 2 and 3) in parallel to shred'the approximately 160
tons of refuse received daily. Ferrous material is recovered after1shred-
ding using a magnetic belt conveyor, and the remaining^shredded' waste is
compacted into transfer trailers and transported to the landfill. Table 2
provides .a summary of the principal features of the Appleton' shredders, and
a flow schematic for the Appleton facility is presented in Figure 4.
-------
TABLE 1. SUMMARY OF TESTS CONDUCTED FOR THE SHREDDER EVALUATION STUDIES
1.
2.
3.
4.
5.
6.
Site
Appleton, Wi.
East Mill
West Mill
Ames, Iowa
Primary
Secondary
Cockeysville, Md.
Shredder #1
Shredder #2
Great Falls, Mt
20 TPH Mill
Tinton Falls, NJ
Shredder #2
Odessa, Tx
Type of
Shredder°)
HSH
HSH
HSH
HSH
HSH
HSH
VSH
VRS
HSH
Shredder
Manufacturer
All is Chalmers
All is Chalmers
American Pulverizer
American Pulverizer
Tracor-Marksman
Tracor-Marksman
Heil
Carborundum
Newell
Average
Throughput
(TWPH)
24.8
18.1
18.6
22.4
49.5 %
N.D.a)
14.8
60.8
82.0
MFfl
Energy
Consumption
(ll)b)
(3)
(10)
(10)
(12)
(0)
(12)
(12)
(12)
CHRCMCMTC
Particle
Size
di)b!
(3)e)
(10)
(10)
(12)
(0)
(12)
(12)
(12)
Hammer
Wear
(l)c)
(1)
(1)
(0)
(1)
(1)
(1)
(1)
(1)
Notes:
a)Not determined
b)Figures in this column refer to the number of tests conducted on each shredder
c|Figures in this column refer to the number of data sets collected for hammer wear
C')HSH = horizontal swing harnnermill
VSH = vertical swing hamrnermill
VRS = vertical ring shredder
e)This test data was used solely in conjunction with the hammer wear data
-------
Figure 1. Outagamie County Solid Waste Plant; Appleton, Wisconsin.
-------
Figure 2. West mill at the Appleton site.
(feed conveyor and discharge conveyor shown at top
and bottom of photograph, respectively)
Figure 3. View of west mill and motor (Appleton)
-------
TABLE 2. APPLETON SHREDDER SPECIFICATIONS
Shredder Summary
Item
Manufacturer
Model
Grate opening (cm)
Number of hammers
Hammer mass (kg)
Hammer material
Tip to tip diameter (m)
RPM
Drive
Tip velocity (m/sec)
Freewheeling power (kw)
East
Allis-Chalmers
KH 12/18
12.7 x 30.5
24 (4 rows of 6)
27
US Steel T-l Type B
.12-.21% carbon
22-31 as cast
Rockwell C hardness
1.23
1014
V-belt
65.3
47.0
West
Allis-Chalmers
KH 12/18
12.7 x 30.5
48 (4 rows of 12)
9
Carbon steel
.4-.5% carbon
1.23
869
V-belt
56.0
41.0
Motor Summary
Item
Manufacturer
Model
RPM
Rating (kw)
Voltage (volts)
Current (amps)
East
All is-Chalmers
114, Type G
1775
298
480, 3 phase
462
West
Allis-Chalmers
114, Type G
1775
298
480, 3 phase
462
-------
Con-
veyor
Pit
Truck
Floor
Control
Room
Con-
veyor
Pit
Non-Fe to landfill
Non-Fe to landfill
Figure 4. Flow Schematic for Appleton Facility.
Refuse trucks are weighed at the scale house as they enter and leave the
site, thus maintaining a record of the total tonnage processed. After ini-
tial weighing, trucks enter the building and dump refuse into either one of
two receiving pits, the floors of which are steel piano-hinged conveyors.
Each receiving conveyor transports refuse to an inclined steel piano-hinged
shredder input conveyor. Refuse is visually inspected while it is on the
inclined conveyors by plant personnel, and any items which have proven dif-
ficult or dangerous to shred are removed. After shredding, the output from
both shredders drops onto a horizontal rubber belt discharge conveyor which
transports the shredded product to an inclined rubber belt conveyor for
removal to the ferrous recovery station. After removal of the ferrous
material, the remaining refuse is loaded into transfer trailers for trans-
port to the landfill.
Ames
The Ames Resource Recovery System (Figure 5) in Ames, Iowa, shreds
roughly 180 tons of refuse per day through two American Pulverizer Model
6090 horizontal hammermills (Figure 6) operating in series. Both material
and energy recovery are practiced at the Ames plant. Most ferrous material
is magnetically separated from the refuse after primary shredding, and the
remaining material is subjected to secondary shredding and air classifica-
tion so as to recover a refuse derived fuel. The refuse derived fuel is
pneumatically conveyed to the nearby city-owned power plant where it is
cofired with coal in the utility's boilers. The plant has the capability of
recovering aluminum from the non-ferrous heavy fraction. A summary of the
specifications for the shredders at Ames is provided in Table 3, and Figure
7 shows a flow schematic for the Ames facility.
-------
00
Figure 5. Solid waste recovery system; Ames, Iowa,
-------
Figure 6. Ames primary shredder shown partially disassembled.
-------
TABLE 3. AMES SHREDDER SPECIFICATIONS
Shredder Summary
Item
Manufacturer
Model
Grate opening (cm)
Number of hammers
Hammer mass (kg)
Hammer type
Tip to tip diameter (m)
RPM
Drive
Tip velocity (m/sec)
Freewheeling power (kw)
Primary
American Pulverizer
6090
22.9 x 25.4
48 (4 rows of 12)
57
Manganese steel
14% manganese
1.83
691
direct
55.0
53.2
Secondary
American Pulverizer
6090
8.9 x 12.7
48 (4 rows of 12)
29
Carbon steel
.2% carbon
20 Rockwell C hardness
1.83
691
direct
55.0
40.5
Item
Manufacturer
Model
RPM
Rating (kw)
Voltage (volts)
Current (amps)
Motor Summary
Primary
Allis-Chalmers
46791-1
691
746
4160, 3 phase
151
Secondary
Allis-Chalmers
46791-2
691
746
4160, 3 phase
151
10
-------
— Air
Solids
Figure 7. Flow schematic for Ames plant.
Incoming refuse trucks are weighed at the entrance to the plant and sub-
sequently dump their contents onto a large tipping floor. The refuse is
scanned by plant personnel, and any potentially hazardous materials (i.e.
gasoline tanks, chemical containers, etc.) are removed. The refuse is then
pushed by a front-end loader onto the horizontal feed conveyor for the pri-
mary shredder. The shredded refuse is removed from under the primary shred-
der by a vibratory conveyor which transports the refuse under a magnetic
belt conveyor before depositing the non-ferrous refuse on the inclined con-
veyor which feeds the secondary shredder. A second vibratory conveyor
removes material from under the secondary shredder and discharges the mate-
rial onto an inclined conveyor for transport to a surge bin prior to air
classification. An inclined conveyor, which acts as a leveling control,
removes material from the surge bin, passes under a scalping roll, and
deposits the refuse on a third vibratory conveyor which feeds the shredded
refuse into the air classifier through a rotary air lock.
11
-------
The light fraction is de-entrained from the airstream by a cyclone.
From the cyclone discharge, the light fraction is conveyed to a rotary air
lock via a twin screw conveyor and conveyed pneumatically under positive
pressure to the power plant. The heavy fraction from the air classifier
passes through another ferrous recovery stage before the non-ferrous mate-
rial recovery system. The ferrous component separated after primary shred-
ding is magnetically classified a second time to remove trapped non-ferrous
material before joining the ferrous material recovered from the air classi-
fied heavy fraction. The final ferrous product is loaded directly into a
trailer for removal from the plant. Since the non-ferrous heavy material
recovery system is currently inoperative, the non-ferrous heavies are
transported to the landfill instead.
Cockevsville
The Baltimore County Resource Recovery Facility (Figure 8) in Cockeys-
ville, Maryland shreds approximately 320 tons of refuse per day through two
Tracer Marksman Model A60 horizontal hammermills (Figure 9) operating in
parallel. Although the capability exists for both material and energy
recovery, currently only magnetic separation of ferrous material is carried
out and the remainder of the shredded refuse is taken to the landfill.
Table 4 lists the specifications of the shredders at Cockeysville, while
Figure 10 presents a flow schematic of the Cockeysville operation.
Incoming refuse trucks are weighed at a scale house before proceeding to
dump their contents into one of four hydraulic push pits (two pits for each
shredder). Refuse is pushed by a hydraulic ram onto an inclined conveyor
which feeds the shredder. As the refuse falls onto the feed conveyor, it is
scanned for hazardous material which may be removed by a small set of grap-
ples controlled by the operator in the control pod. Whenever the capacity
of the push pits to receive refuse is exceeded, the trucks may dump into a
large raw refuse storage pit from which an overhead crane may remove the
material during slack operating periods. Shredded material is removed from
under the shredder by a short rubber belt conveyor and discharged onto a
long, inclined rubber belt conveyor for transport to the ferrous removal
station. After ferrous removal, the remaining refuse is loaded into trans-
fer trailers and taken to the landfill.
Great Falls
The Great Falls Solid Waste Reduction Facility (Figure 11) in Great
Falls, Montana, operates two Heil vertical shaft hammermills in parallel.
The larger of the two is a Heil model 42F (capable of shredding approxi-
mately 20 TPH), and the smaller is a Heil model 42D (15 TPH). The combined
daily tonnage of the two shredders is typically close to 150 tons. Although
the plant is equipped to recover ferrous material, the ferrous recovery
stage was not in operation during the course of the tests; hence, all the
shredded waste was compacted into transfer trailers and transported to the
landfill. Table 5 summarizes the major features of the shredders which are
utilized in Great Falls. Since the vertical hammermill has no grates, both
12
-------
Figure 8. Baltimore County Resource Recovery Facility;
Cockeysville, Maryland.
13
-------
Figure 9. Shredder #1 at Cockeysville.
(shown partially disassembled for hammer change)
14
-------
TABLE 4. COCKEYSVILLE SHREDDER SPECIFICATIONS
Shredder Summary
Item
Manufacturer
Model
Grate opening (cm)
Number of hammers
Hammer mass (kg)
Hammer type
Tip to tip diameter (m)
RPM
Drive
Tip velocity (m/sec)
Freewheeling power (lew)
II
Tracor-Marksman
AGO
20.3 x 35.6
24 (4 rows of 6)
73
Manganese steel
14% manganese
1.52
880
direct
70.0
64.3 forward,
67.8 reverse
Tracor-Marksman
A60
20.3 x 35.6
24 (4 rows of 6)
73
Manganese steel
14% manganese
1.52
880
direct
70.0
Item
Manufacturer
Model
RPM
Rating (kw)
Voltage (volts)
Current (amps)
Motor Summary
£L
Toshiba
Type TIM, wound rotor
880
746
4160, 3 phase
130
£2
Toshiba
Type TIM, wound rotor
880
746
4160, 3 phase
130
15
-------
lower
Screened lights
Truck
gangway
Storage pit
Truck
gangway
Shreddei
02
^ To landfill
>- Fe
Shreddei
II
Fe
Air
Solids
Potential Solids
Figure 10. Flow Schematic of Cockeysville Plant
Figure 11. Great Falls Solid Waste Reduction Facility;
Great Falls, Montana.
16
-------
TABLE 5. GREAT FALLS SHREDDER SPECIFICATIONS
Item
Manufacturer
Model
Smallest hamner clearance
Discharge opening (cm)
Number of hammers
Hammer mass (kg)
Hammer type
Tip to tip diameter (m)
RPM
Drive
Tip velocity (m/sec)
Freewheeling power (kw)
Item
Manufacturer
Model
RPM
Rating (kw)
Voltage (volts)
Current (amps)
Shredder Summary
20 ton 15
Heil
42F
(cm) 2.5
91 x 30
38
6.5
1060 steel
1.22
1155
V-belt
73.8
21.7
Motor Summary
20 ton
Howell Electric
Motor
type BA10 3
1775
186
480, 3 phase
258
ton (not tested)
Heil
42D
2.5
91 x 30
34
6.5
1060 steel
1.22
--
V-belt
—
--
15 ton
Howell Electric
Motor
—
—
149
480, 3 phase
~
the clearance between the hammers and the lining as well as the size of the
discharge opening are listed instead of the usual grate opening dimensions.
All testing of the shredders in Great Falls was done on the Model 42F shred-
der (Figures 12 and 13). Figure 14 provides a schematic of the Great Falls
plant.
As refuse trucks enter the facility, the gross weight is recorded. At
the conclusion of each day, the pre-recorded tare weights of the trucks are
subtracted from the gross weights, thus providing a record of the daily
17
-------
Figure 12. Exterior view of Great Falls vertical hammermi11
Figure 13. Interior view of Great Falls vertical hammermill
18
-------
Fe
Transfer
trailer
packer
Figure 14. Flow schematic of Great Falls facility.
tonnage. Refuse is dumped by the trucks onto a tipping floor and then
pushed by a rubber-tired front end loader into either of the two horizontal
steel piano-hinged conveyors. The horizontal conveyors transport the refuse
to the inclined shredder input conveyors which are also of steel piano-
hinged construction. As the refuse travels up the inclined conveyors, it is
visually inspected by the plant personnel so any hazardous or difficult to
shred items may be removed. After shredding, the" output from both shredders
drops onto a horizontal rubber belt discharge conveyor which transports the
shredded product to an inclined rubber belt conveyor for removal to the fer-
rous recovery stage. After removal of the ferrous metal, the remaining
material is loaded into transfer trailers for transport to the landfill.
Tinton Falls
The Monmouth County Reclamation Center (Figure 15) in Tinton Falls, New
Jersey, operates two Carborundum vertical shaft ring shredders (formerly
Eidel Model 1000). Table 6 provides a summary of the major features of the
shredders utilized in Tinton Falls. Since vertical shredders do not use
grate bars, the clearance between the individual grinders and the lining
along with the size of the discharge opening are listed. A central tipping
floor serves both shredders (which are located in separate rooms on opposite
sides of the tipping room). The separation of the shredders permits main-
tenance and repair of one shredder to be accomplished in a relatively clean
19
-------
Figure 15. Monmouth County Reclamation Center;
Tinton Falls, New Jersey
20
-------
TABLE 6. TINTON FALLS SHREDDER SPECIFICATIONS
Shredder Summary
Item
Manufacturer
Model
Smallest grinder clearance (cm)
Discharge opening (on)
Number of grinders
Grinder mass (kg)
Grinder type
Tip to tip diameter (m)
RPM
Drive
Tip velocity (m/sec)
Freewheeling power (lew)
Plant 2 (Plant I Identical)
Carborundum
1000
5 (at time of test)
66 x 35.6
60
28.1
n'ickel -manganese steel
1.5
369
gear
29.0
38.5
Uem
Manufacturer
Model
RPM
Rating (kw)
Voltage (volts)
Current (amps)
Motor Summary
Starter motor
Allis-Chalmers
124
1770
75
460 , 3 phase
116
Drive motor (2 required)
Westinghouse
ABOP
1746
373
460, 3 phase
563
and quiet environment while shredding is done in the opposite end of the
plant. Normally, a shredder is operated for approximately five weeks before
it is taken off line. During the five weeks one shredder is operating, the
other shredder is idle so maintenance can be performed. Nine hundred fif-
teen tons is typical of the total refuse received in one week. Figure 16
shows the interior of Plant 2 (where the tests were done) from the entrance
of the feed conveyor to almost the exit of the rubber discharge conveyor. A
view of the interior of the shredder in Plant 2 is presented in Figure 17.
Figure 18 shows a schematic of the entire Reclamation Center.
Trucks are weighed both as they enter and as they leave the facility in
order to maintain a record of the daily tonnage. The trucks dump onto a
tipping floor from which a small rubber-tired loader can push the refuse
onto a horizontal steel conveyor serving one of the shredders. The horizon-
tal conveyor loads an inclined section of the steel feed conveyor. As shown
21
-------
Figure 16. Interior of Monmouth County
Reclamation Center, Plant 2.
Figure 17.
Interior of vertical
ring shredder.
22
-------
To landfill
To landfill
Transfer
trailer
packers
Transfer
trailer
packers
Magnet
Fe
d
Control
room #1
Control
room #2
Tipping floor
Figure 18. Flow schematic of Tinton Falls facility.
in Figure 16, the inclined section of the feed conveyor is followed by a
horizontal section prior to depositing refuse in the shredder. After shred-
ding, the output of the shredder is transported by a rubber belt conveyor
through a ferrous recovery stage to the hopper used to load the transfer
trailers, which carry the shredded product to the adjoining landfill.
Energy, size, and wear measurements were collected for shredder #2 only.
Odessa
The City of Odessa Solid Waste Management Facility (Figure 19) in Odes-
sa, Texas, has a single Newell model 68 horizontal hammermi11 (Figure 20) in
operation. Approximately 230 tons of refuse are shredded on a typical day.
The plant is equipped to recover ferrous material and pack the shredded
23
-------
Figure 19. City of Odessa Solid Waste Management Facility;
Odessa, Texas
24
-------
a. upper portion including feed conveyor
b. lower portion
Figure 20. Odessa horizontal hammermill
25
-------
refuse into transfer trailers for transport to the landfill. The major fea-
tures of the Newell shredder are summarized in Table 7. A schematic of the
Odessa facility is provided in Figure 21.
Records of the net tons of refuse entering the facility are maintained
in the scalehouse. After weighing, trucks dump their contents onto the
tipping room floor. A rubber-tired front end loader pushes the refuse from
the tipping floor into either one of two opposing hydraulic push pits. As
the moving wall of the push pit approaches the rubber belt feed conveyor,
refuse is forced onto the belt. A small grapple is situated near the junc-
ture of the push pits and feed conveyor in order to pick out hazardous items
and help control the flow of refuse from the push pits. After shredding,
the refuse is discharged onto a second rubber belt conveyor which transports
the refuse to the ferrous recovery station. After ferrous removal, the ref-
use is dropped into a packer unit which packs the shredded product into
transfer trailers for transport to the landfill.
HAMMER MAINTENANCE PROGRAMS
The Appleton, Cockeysville, and Great Falls plants utilize a program of
hammer retipping as a means of extending hammer life (Figures 22-23). At
Appleton, welding wire is normally used instead of electrode rods. Stoody
110 is used for building up the hammer surface prior to an application of
one or two passes of Stoody 134 hardfacing. Stoody 134 is the wire equiva-
lent of the Stoody 2134 electrode which was used as part of the wear experi-
ments at Appleton.
The Cockeysville facility used McKay 118 for buildup of the hammers,
followed by an application of one or two passes of McKay 55. Both of these
alloys are applied in electrode form. McKay 55 was one of the alloys used
in the wear experiments at the Cockeysville plant.
In order to reduce manpower and time requirements for hammer changes, a
pin puller (Figure 24) is used at Cockeysville for removing the hammer
pins. This hydraulic unit eliminates the need for various "brute force"
methods, icluding the use of sledge hammers and come-alongs.
The Great Falls facility normally uses Amsco Super 20 electrode to retip
the 1060 carbon steel hammers. Stoody 2134 was also used in the wear
experiments in Great Falls in order to compare with the Appleton results.
Hammer retipping is not practiced at either the Ames or Odessa plant.
At the time of the tests, the operating procedure at Ames and Odessa was to
wear the hammers in the shredders until refuse was no longer shredded effec-
tively (indicated by lower throughput and larger particle size). The worn
hammers were then removed, scrapped, and replaced by new hammers (Figures
25-28). The plant located in Tinton Falls scraps the used ring grinders
(Figure 29) when the product particles become excessively large. Contrary to
the case of horizontal hammermills, the throughput increases when the shred-
ding elements wear out. Unlike the grinders, the shredder breaker bars were
regularly resurfaced in order to extend their life as were the sweep plates
in the bottom of the shredder.
26
-------
TABLE 7. ODESSA SHREDDER SPECIFICATIONS
Shredder Summary
Item
Manufacturer
Model
Grate opening (cm)
Number of hammers
Hammer mass (kg)
Hammer type
Tip to tip diameter (m)
RPM
Drive
Tip velocity (m/sec)
Freewheeling power (kw)
Motor Summary
Item
Manufacturer
Model
RPM
Rating (kw)
Voltage (volts)
Current (amps)
Newell
68
35.6 x 24.1
14
67
manganese steel
1.5
880
direct
69.1
101.7
Toshiba
TIM-VCK-V
880
373
2400, 3 phase
295
27
-------
Tipping floor
Fe
t
Magnet
•*.
Trailer
packer
To landfill
Shredder
Push pit
Tipping floor
Push pit
Tipping floor
Control room
Maintenance
area
Tipping floor
Scale
house
Figure 21. Flow schematic of Odessa facility.
28
-------
Figure 22. Hammers undergoing hardfacing at Appleton,
Weld deposits are shown on hammer tips.
Figure 23. Hardfacing application at the Cockeysville facility.
-------
Figure 24. Utilization of pin puller for hammer changes
Figure 25. View of hammers in Ames primary shredder.
(note uneven pattern of wear among hammers and
pencil resting on hammer for reference of scale)
30
-------
Figure 26. Inside view of Ames secondary shredder.
(note severe wear of hammers shown in right center of photograph)
Figure 27. New hammer for Ames secondary shredder.
31
-------
Figure 28. Worn hammers from Ames secondary shredder.
(tape is extended to length of new hammer)
Figure 29. Used and new ring grinders for
Tinton Falls vertical shredder.
32
-------
SECTION 2
FINDINGS AND CONCLUSIONS
In the study reported herein shredders at six large-scale refuse proces-
sing plants are evaluated. In-plant operational data were collected from
nine shredders: namely seven horizontal hammermills, a vertical hammermill,
and a vertical ring shredder.
The main objectives of this study were to: 1) determine shredder per-
formance, and 2) develop theoretical relationships that describe shredder
operation and performance.
Emphasis was placed on the development of analytical relationships such
that the shredders could be compared on the same basis. Consequently, many
of the findings are necessarily mathematical in nature. The mathematical
relationships make possible a comparison of shredders on the hypothetical
basis that the products of all the shredders have the same average particle
size. In this manner shredders can be compared even though under actual
circumstances each shredder may produce a somewhat different range of par-
ticle sizes.
Where possible, shredder performance was explained by deriving governing
relationships, utilizing key parameters of size reduction such as energy
consumption, throughput, and particle size. These relationships would be of
interest primarily to those concerned with shredder design, operation, and
optimization.
Some results of the study are of interest not only to those peripherally
involved with shredding as a unit operation but also to those involved in
the overall requirements of a refuse processing plant, e.g., Directors of
Public Works and Plant Supervisors.
General findings that are of interest to those involved in plant manage-
ment (i.e. the user community) are:
1. Costs associated with primary size reduction of refuse to nominal
product size in the range of 6 to 12 cm are estimated to be as follows:
a. Energy = $0.08 to $0.20 per ton of refuse
b. Hammer Maintenance = $0.40 per ton of refuse.
2. On the basis of the data collected in the investigation, and assuming
an equivalent average particle size production, only a few minor differences
33
-------
could be found in the energy consumption by and the hammer wear in the
following four shredders:
a. Allis Chalmers horizontal hammermill;
b. Tracor-Marksman horizontal hammermill;
c. American Pulverizer horizontal hammermill; and
d. Heil vertical hammermill.
3. In comparison to the preceding four shredders, the Newell horizontal
hammermill and the Carborundun vertical ring shredder had relatively low
energy requirements. On the basis of the production of an average charac-
teristic particle size of 2.1 cm, the Newell shredder at Odessa uses 1.1
kwh/T. A similar shredder, the Tracor-Marksman located in Cockeysville,
requires 8.9 kwh/T. It seems as though the relatively low energy require-
ment by the Newell shredder is due primarily to the number of hammers used
for the shredding rather than to an inherently superior shredder design.
The Newell shredder uses 14 hammers as opposed to the 24 to 48 hammers used
by the other hammermills.
4. For a given shredding operation and from the standpoint of cost, the
two following hammer maintenance programs are judged equivalent:
a. build-up and/or hardfacing of hammers (build-up method); and
b. the use of hammers until they no longer effectively
shred refuse, after which they are replaced (wear-and-scrap
method).
Since the cost of the two maintenance programs are equal to each other,
the build-up method of hammer maintenance is to be preferred. The rationale
for this choice is that regularly built-up hammers produce a uniform shred
size throughout the course of shredding large quantities of refuse, and yet
do not unduly affect energy consumption and throughput capacity.
In addition to the preceding general findings, a number of specific
findings deal with energy requirements of shredding, hammer wear, and with
functional >elationships among shredding parameters. These specific results
are of interest primarily to those concerned with shredder design, opera-
tion, and optimization, and they are the subject matter of the remainder of
this section.
ENERGY AND PARTICLE SIZE
5. Mathematical expressions that describe the functional relationship of
power draw, throughput, and moisture content of shredded waste can be
established for each shredder. These relationships can be expressed in the
general form:
34
-------
PN - air£ (1 - MC)S (1)
where: PN = net power
m .throughput on a wet weight basis
MC = air dry moisture content,
and a, r, and s are experimentally determined constants. The range for
these constants for the shredders tested are 0.14 <. a <. 47.04,
0.27 < r 11.92, and -12.64 £ s 1 4.79.
6. Mathematical expressions have been developed that relate average
specific energy consumption to average size of the shredded product. These
relationships have the form:
EQ = bXu (2)
where: E0 = specific energy,
X = product size, expressed as either characteristic or nominal
product size;
and b and u are analytically determined constants. Values of u were found
to be in the range, -0.92 1 u 1 -0.81.
7. Through the testing and evaluation program valuable data were
obtained regarding the efficiency of single versus multiple-stage size
reduction. In particular, evaluation of the data collected for the two
shredders in series at Ames showed a significantly greater gross specific
energy consumption in comparison to that in the single-stage size reduction
practiced at the Cockeysville facility. However, this result must be viewed
with reservation, since the shredding lines at Ames and Cockeysville have
not been shown to represent an optimum configuration for either multiple-
stage or single-stage shredding.
8. The fact that the secondary shredder at Ames used approximately twice
as much energy as the primary shredder supports the assumption that the
shredder line at Ames has not been optimized. Through appropriate manipu-
lation of grate openings in both shredders, it might be possible to develop
a more energy efficient shredding operation at the Ames facility.
9. Apparently a definite inverse relationship exists between the size of
the shredded material and the rate of material flow through each shredder.
This relationship, though only qualitatively established in the present
research, supports previous data from research conducted at the University
of California, Berkeley. From the standpoint of efficient and reliable pro-
cessing and separation of materials by downstream unit processes, the pre-
diction, manipulation, and control of shredded product size are of obvious
importance. Results of this research indicate that for a particular set of
shredder characteristics (e.g. grate opening, rpm, etc.), an increase in
throughput will result in a decrease in product size.
35
-------
HAMMER WEAR
10. The optimum hardness range for hardfacing alloys appears to be in
the range of 48 to 56 Rockwell C. Severe hammer wear is encountered at the
lower levels of hardness for base hammer material and hardfacing alloys.
11. Proper testing and evaluation of the performance of hardfacing
alloys requires a standardized test procedure that evaluates the hardfacings
under identical operating conditions.
12. Prudent selection of hardfacing alloys can result in significant
reductions in hammer wear (typically 25 percent) arid can lead to an increase
in the amount of time required between hardfacing applications and hammer
changes, or both.
13. The upper limit of alloy hardness for size reduction of MSW is
governed by the chipping tendency of alloys under impact loading.
14. The degree of chipping of hard alloys appears to be a function of
the particular composition of each alloy and of the type of wastes being
shredded. Consequently, the proper selection of hardfacing requires that
several hard alloys be tested and evaluated in order to ascertain the best
alloy for a particular shredding application.
15. Proper hammer maintenance calls for the measurement of wear and a
calculation of rates of wear such that in the interval between hardfacing
applications, hammers are limited to the maximum shredded tonnage at which
the hardfacing weld deposits will be removed. With such a program, wear
will be confined primarily to the hardfacing, and consequently minimum rates
of overall wear will occur because of the preventing of wear of the softer
base metal of the hammer with its concomitant higher rate of wear.
16. To take into consideration the different degrees of size reduction
between the test sites, a means of normalization and comparison of hammer
wear data was formulated that accounts for variations in size of the feed
material and size of the shredded product for different shredders.
36
-------
SECTION 3
RECOMMENDATIONS
Several recommendations can be presented with regard to areas of
research that would extend the knowledge of the science of size reduction of
MSW, while at the same time result in a potential reduction of operational
and maintenance costs for shredding operations. These recommendations are
presented, along with the rationales for them. A number of the recommen-
dations are aimed specifically at the control of the shredding operation.
1. From the standpoint of efficient energy utilization, the single most
important outcome of this study involves the finding that the number of
hammers appears to have a significant influence upon the specific energy
(kwh/T) required for size reduction of refuse. The data indicate that the
possibility exists for minimizing energy consumption through selection of
the appropriate number of hammers. Consequently, further testing is needed
to establish the influence of the number of hammers upon energy consumption
and throughput. Such testing would involve energy and throughput measure-
ment for several different hammer complements, for example, 48, 24, and 12
hammers. Testing at one site should provide enough data to allow a determi-
nation of the effect of the number of hammers upon energy consumption and
throughput.
2. An evaluation of the data collected at the city of Ames, Iowa,
indicated that the primary shredder uses significantly less energy than does
the secondary shredder. Moreover, a comparison of the data from Cockeys-
ville with that from Ames shows that single shredding can be more energy
efficient than two-stage shredding. However, it is not known whether or not
the processing line at Ames is optimized with regard to energy consumption.
The discrepancy in energy requirements between primary and secondary
shredders leaves room for doubt that the shredding operation is optimized.
Consequently, a test program to determine the best operating conditions for
the Ames shredding process is recommended. Suggestions for this program
include the varying of the following parameters: 1) grate openings in each
shredder, 2) number and length of hammers, and 3) throughput under different
test conditions. In addition, comparative data from another site that has a
multiple stage size reduction, e.g. Lane County, Oregon, would be useful.
At least two sets of data are necessary in order to determine whether or not
multiple stage size reduction is more energy efficient than single stage
size reduction. Such test programs are needed before a definitive answer
can be provided for settling the argument concerning single versus multiple
stage size reduction.
37
-------
3. Inasmuch as in the present study only the hammer wear resulting
from primary shredding was characterized, an extension of hammer wear evalu-
ation to include secondary shredding would serve to establish possible ave-
nues for minimization of hammer wear. In particular, the selection of pro-
per grate openings for multiple stage shredding operations needs to be
addressed. One of the important consequences of grate opening manipulation
in primary and secondary shredders could be a significant reduction in ham-
mer wear. The hammer wear test programs for secondary shredders could be
combined with the test programs for evaluating energy requirements for mul-
tiple stage size reduction (per item 2). Possible sites include the Ames
and Lane County facilities. Upon completion of the hammer wear and energy
consumption test programs for both single and multiple stage size reduction,
process optimization for shredding, based upon considerations of hammer wear
and energy consumption, could be accomplished, while simultaneously, predic-
tive criteria could be established.
4. In light of the fact that shredder holdup (material within the
shredder at any instant in time) can be related analytically to throughput
and energy consumption, additional collection of shredder holdup data is
warranted. Research on shredder holdup would make possible a determination
of the influence of machine characteristics upon energy utilization and
throughput. Holdup data is seen as the link between the power equation,
PN = am£ (1 - MC)S
and the basic physical characteristics of shredders, as for example, grate
size, internal machine geometry, and number and geometry of hammers.
38
-------
SECTION 4
MATERIALS AND METHODS
POWER MONITORING INSTRUMENTATION
The power monitoring equipment includes a Scientific Columbus Model
DL34-2K5A2-AY-6070 watt/watt-hour transducer, a Houston Instruments Model
3000 chart recorder, and a digital dividing circuit (Figure 30). Since the
nominal inputs to the transducer are 75 to 135 volts and 0 to 6.5 amperes,
transformers are required to reduce the line voltage and shredder current to
these levels (Figure 31).
The transducer itself has three separate elements and is adaptable to
single phase, three phase-three wire, and three phase-four wire systems.
Two output signals are provided by the transducer. The first is an analog
current signal which is directly proportional to power and may range from
zero at no load to one milliamp at full load with all three elements in
use. A full load one mi 11iamp signal corresponds to 500 watts per element
or 1500 watts total on the secondary (transducer) side of the current and
potential transformers. The analog current signal is passed through a pre-
cision 1000 ohm resistor which converts the current signal to a voltage sig-
nal related to the current by Ohm's Law, E = IR. The voltage drop across
the resistor is recorded on the chart recorder to provide both a time base
and a permanent, continuous record of the power requirements. The power
required by the shredder can be found by multiplying the power on the secon-
dary side of the monitoring equipment by the product of the potential and
current transformer ratios.
The second transducer output is a digital signal directly proportional
to the integral of power over time or, in other words, energy. The trans-
ducer has a constant internal count rate of 2160 counts per hour with full
load. Dividing the full load rated analog output by 2160 indicates the num-
ber of watt-hours each output pulse represents. Again, multiplying by the
product of the potential and current transformer ratios determines the
energy used on the primary side of the transformers. The digital signal is
sent to the dividing circuit, which counts the pulses and triggers an event
marker on the chart recorder after a predetermined number of pulses have
been accumulated. The divider may be set manually to count between 1 and
9999 input pulses before triggering the event marker.
The general layout of the power monitoring equipment is presented in
Figure 32. Since the voltage and current requirements of the shredders var-
ied from site to site, the transformer ratios used at each site are listed
in Table 8.
39
-------
Figure 30. Power monitoring equipment; Appleton.
(shown left to right: dividing unit, chart recorder, transducer)
Figure 31. Installation of voltage and current transformers; Appleton.
(motor control box)
40
-------
Shredder
Motor
PT,
Analog
Signal
v
Transducer
Chart
Recorder
Digital
Signal
Dividing
Circuit
Figure 32. Schematic layout of power monitoring equipment.
-------
TABLE 8. TRANSFORMER RATIOS USED AT EACH SITE
Site Voltage Voltage Product
ratio, Rv ratio Rc RVRC
Appleton
Ames
Cockeysville
Great Falls
Tinton Falls
Odessa
2.4:1
20:1
20:1
2.4:1
2.4:1
20:1
100:1
40:1
20:1
60:1
120:1
60:1
240:1
800:1
400:1
144:1
288:1
1200:1
OTHER EQUIPMENT
A Chronos model 3-ST digital chronometer was used for making any neces-
sary measurements of time. The 3-ST has two modes of operation, Taylor/
Sequential and Split/Cumulative. Under Taylor/Sequential operation, the
elapsed time between each successive activation of the master control button
is displayed. In the Split/Cumulative mode, the first time the master con-
trol button is pushed the clock and display begin to accumulate time start-
ing at 0.0 seconds. The second time the master control button is pushed the
display stops accumulating time (thus indicating the elapsed time from when
the master control button was first pressed) while the internal clock con-
tinues to accumulate time. Each succeeding activation of the master control
button will then cause the elapsed time between the moment the chronometer
initially started and the moment of the most recent operation of the master
control button to be displayed.
Conveyor belt speeds were determined with a Power Instruments, Inc.
TAK-ETTE model 1707 digital rpm gauge. A disc, exactly one foot in circum-
ference, allowed reading the belt speed in feet/minute directly from the
display. An alternative method for obtaining the belt speed was to measure
the time necessary for the belt to traverse a known distance, and then
divide the distance by the elapsed time.
In order to obtain size distributions for the shredded refuse samples,
the samples were first screened on a set of manually held screens having
square wire mesh openings of 20.32, 10.16, 5.08, and 2.54 centimeters.
Screening of the undersize from the 2.54 centimeter hand held screen was
carried out on a SWECO model LS 18533333 Vibro-Energy Rotary Screen. The
SWECO was equipped with a series of square wire mesh screens having openings
of 2.54, 1.59, 0.95, 0.51, 0.27, and 0.13 centimeters.
42
-------
SIZE DISTRIBUTION ANALYSIS
The samples were weighed and air dried to a constant weight in a drying
room (ambient conditions, 22° C and 65 percent relative humidity) after
which the samples were reweighed to permit determination of the moisture
content. Size analysis of the samples was performed using both manual and
mechanical screens. The dried refuse was placed on the largest of the man-
ually held screens and shaken until no further refuse was observed to pass
through the screen. The oversize from the screen was collected and weighed,
and the undersize was placed on the next smaller screen. The process was
then repeated until all four manual screens had been used.
The undersize from the 2.54 centimeter manually held screen was next
processed through the SWECO screens. Since the total mass of material in
each sample was much larger than could be accommodated by any individual
SWECO screen at one time, two screens were placed on the vibrating base
simultaneously, and a series of small batches were fed to the screens until
the entire sample was processed. After each small batch was screened for 15
minutes, the oversize on the screens was collected, the next batch of refuse
placed on the screens and the process repeated until the entire sample had
been processed. The total mass of screen oversize for each of the two
screens was weighed and tabulated before processing through the next two
pair of screens in the same manner. Finally the undersize from the 0.13
centimeter screen was collected, weighed, and tabulated along with the rest
of the data.
43
-------
SECTION 5
EXPERIMENTAL PROCEDURES
GENERAL PROCEDURES FOR POWER-FLOW RATE DATA COLLECTION
Installation of the potential and current transformers was done at each
site by qualified electricians and connected to the power monitoring equip-
ment under the direction of Cal Recovery personnel. After installation, the
system was tested to insure all the connections were correct and nothing had
been damaged during shipment. On actual days of testing, all the monitoring
equipment was turned on and allowed to warm up for at least half an hour
before data was collected. Several times during each day of testing, free-
wheeling power measurements were taken, including one which was obtained
shortly after shredder startup with the rest obtained as shredding opera-
tions permitted throughout the day. Freewheeling power is defined as the
power required to maintain constant rotational velocity of the shredder
rotor under a no-load condition (i.e. during idling).
Flow rate samples and power level data were collected under a number of
different operating conditions for the purpose of characterizing the shred-
ders. To guarantee that the flow rate sample coincided exactly with the
interval during which the power was monitored, it was necessary to accu-
rately determine both the distance from the center line of the shredder dis-
charge to the center line of the segment of the discharge conveyor from
which the flow rate sample was gathered and the speed of the discharge con-
veyor itself. With the shredder sampling distance and conveyor speed known,
it was generally possible to calculate the elapsed time needed for the sam-
ple to travel from the shredder to the point of sample collection from
Zh1/2
(3)
where: t = elapsed time (sec);
di = shredder center line to sample center line distance (m);
V = speed of the conveyor from which the flow rate sample was
collected (m/sec);
h = distance from the bottom of the grates at the center line
of the shredder to the top of the discharge conveyor; and
g = acceleration due to gravity (m/sec2).
The term 2 h^/2/g in equation 3 is used to take into account the time
required for shredded material to drop from the grates of the shredder to
the conveyor belt.
44
-------
When the actual power-flow rate test was being done, the time at which
the discharge conveyor stopped was marked on the chart and a value for the
elapsed time was calculated. The point on the chart recording of the power
level corresponding to the time during which the flow rate sample was just
emerging from the shredder was found by measuring the distance, d, from the
mark identifying when the discharge conveyor was stopped. A value for d was
found from
d - Vct (4)
where: d = distance from discharge conveyor stop mark on the chart
recording to the center of the chart interval during which the
flow rate sample was emerging from the shredder (m);
Vc = chart recorder speed (m/sec); and
t = elapsed time (sec).
The complete interval on the chart recording during which the flow rate
sample was being discharged was identified by measuring the distance ;+ A
from the point marking the center of the sample interval. A value for A is
obtained from
& • \ V vc
where: L = length of the conveyor segment from which the flow rate sample
was collected (m);
V = speed of the conveyor from which the flow rate sample was
collected (m/sec); and
Vc = chart recorder speed (m/sec).
The mass flow rate through the shredder was calculated from
mw = 3.6 ^ (6)
where: fi^, - mass flow rate (TWPH);
M = mass of the total sample removed from the discharge
conveyor (kg); and
L = length of the conveyor segment from which the flow rate sample
was collected (m); and
V = speed of the conveyor from which the flow rate sample was
collected (m/sec).
Preparation of the refuse samples for shipment included bagging and
weighing the samples. The bagged samples were tagged with the date, weight,
and sample number, placed in a second plastic bag to insure against puncture
damage to the inner bag, and packed into shipping containers for transport
to CRS for size and moisture analysis.
In order to ensure that a representative sample for the size distribu-
tion analysis was obtained, the sample size was varied according to the
grate spacing of the shredder undergoing testing. In general, the larger
the grate spacing, the larger the sample that needed to be gathered. The
45
-------
reason for the selection of this method for determination of sample size
follows from the fact that, in general, relatively large particles issue
from grates with large openings. Consequently in order to assure a suffi-
cient number of particles for determination of the particle size spectrum, a
large particle size dictates a large sample size.
An estimation of the sample size was made from Table 9:
TABLE 9. SAMPLE SIZE
Grate
cm
20
15
10
5
Spacing
(inches)
(7.9)
(5.9)
(3.9)
(2.0)
kq
10
7.5
5
2.5
Sample Size
(pounds)
(22.1)
(16.5)
(11.0)
( 5.5)
The actual sample used for the size distribution and moisture analysis
was obtained by thoroughly mixing the flow rate sample and dividing it into
equal portions of the correct size. One of the portions so obtained was
then prepared and shipped for subsequent size distribution and moisture
analysis.
APPLETON PROCEDURE
Since both shredders at Appleton use the same discharge conveyor, only
one shredder at a time was fed refuse during the tests. In addition, the
water spray used to control dust inside the shredders was turned off several
minutes before all but the first test.
The motor control room in Appleton is located directly underneath the
main control room. The person monitoring the power requirement of the
shredder under test stood at the entrance to the motor control room in sight
of the monitoring equipment. An operator was in the main control room,
while another member of the testing team was stationed at the end of the
horizontal discharge conveyor. After recording an interval of relatively
constant power the person monitoring the power signaled both visually and
audibly to the operator in the control room and marked the time of the sig-
nal on the chart recording. Upon receiving the signal, the control room
operator stopped both the feed and discharge conveyors and allowed the
shredder to empty. While the shredder was emptying, the final five meters
of refuse on the discharge conveyor was collected and weighed in order to
obtain flow rate data. Correlation of the flow rate and power data was
accomplished through utilization of the methods outlined under General Pro-
cedures.
46
-------
After collecting the flow rate sample, the discharge conveyor was
advanced so the shredder holdup (the mass of material actually within the
shredder at the instant the feed conveyor was stopped) could also be collec-
ted and weighed.
AMES PROCEDURE
The motor control room at the Ames facility is also located directly
under the main control room, and,a telephone intercom system connects the
two control rooms. Because the shredders operate in series, somewhat dif-
ferent techniques from those used at Appleton were utilized for collecting
the necessary data for each shredder. The presence of vibratory discharge
conveyors with no readily measurable transport speed made it impossible to
use equation 3 to determine the elapsed time needed for a particle of refuse
to travel from the center line of the shredder to the center line of the
conveyor segment from which the flow rate sample was to be collected.
Instead the elapsed times were determined by measuring the travel time of
refuse samples with a digital chronometer as the refuse was transported
along the conveyors from the shredders to the respective sampling stations.
The sampling station for the primary shredder was approximately at the mid-
point of the inclined conveyor which feeds the secondary shredder, while the
secondary shredder sampling station was in the surge bin at the end of the
second inclined discharge conveyor.
To obtain a sample for the primary shredder, the magnetic belt over the
discharge conveyor was turned off and one of the test personnel began to
monitor the power level. After recording a period of relatively constant
power, the operator in the main control room was signaled over the intercom
to stop the feed and discharge conveyors serving the primary shredder. The
time at which the conveyors were stopped was noted on the chart recording.
Three meters of inclined conveyor were swept clean of refuse, and the refuse
sample thus collected was weighed and the entire sample prepared for ship-
ment. Since both the belt speed and the time of travel between the shredder
and sample were known, the flow rate determination and sample correlation
with power requirements were done as described in the section on General
Procedures with the exception that the elapsed time required for refuse to
travel from the shredder to the sampling station was measured directly
rather than calculated.
Sampling the secondary shredder in the same manner as the primary shred-
der would have hindered plant operation (due to material accummulation
between the shredder grates and the vibratory discharge conveyor), so an
alternative sampling method was developed. A large box was constructed and
suspended inside the air classifier surge bin such that, when pulled into a
retracted position by ropes affixed to the corners of the box, the box was
held clear of the refuse stream; yet when the ropes were released, the box
would swing forward and catch the entire flow of refuse for as long as
necessary to fill the box. Since it was no longer necessary to halt any of
the equipment in order to obtain samples using such a technique, plant oper-
ation was not affected by secondary sample collection.
47
-------
The'exact procedure used to obtain flow rate and power data for the sec-
ondary shredder was as follows:
1. A period of relatively constant power was recorded on the chart
recorder.
2. The sample box was swung into position and the digital chronometer
started in the Split/Cumulative mode (see Other Equipment for a
discussion of the Split/Cumulative mode).
3. When the sample box was full, it was removed from the refuse
stream and the digital chronometer was stopped.
4. The sample box was removed from the surge bin.
5. The member of the testing team with the chronometer returned to the
power monitoring equipment and wrote the time displayed by the
chronometer (in this case, the time needed to fill the box) on the
chart recording without stopping the chart.
6. Next, the master control button on the chronometer was pressed
again while simultaneously removing the pen from the chart paper.
7. The chart recorder was stopped and the new number displayed by the
chronometer was written on the chart. The final number displayed
represented the total time elapsed since the sample box was swung
into position.
The period during which the sample was emerging from the shredder could
be located on the chart recording by measuring the distance d +_ A from the
chart recording when d is found from
d = Vr (t? + t3 - h) (7)
c t J 2
where: Vc = the chart recorder speed (m/sec);
t] = the amount of time needed to fill the sample box (sec);
t2 = the elapsed time between swinging the box into position and
removing the pen from the chart paper (sec); and
t3 = the amount of time needed for particles of refuse to travel
from the shredder to the surge bin (sec).
and A is found from
The sample of shredded refuse that was collected in the box was weighed,
and a two to five kilogram subsample was prepared for shipment to CRS for
further analysis.
48
-------
The flow rate for the secondary shredder was calculated from
= 3.6^ (9)
rl
where: mw = flow rate (TWPH);
M = total sample mass (kg); and
t-j = time necessary to fill the box (sec).
COCKEYSVILLE PROCEDURE
The Cockeysville facility has separate motor and main control rooms for
each shredder with neither intercom nor line-of-sight communication existing
between the motor and main control rooms of either shredder. Therefore,
short-range radios were used for communication between the test and operat-
ing personnel. Since the shredder motors at the Cockeysville facility are
reversible and the shredders are run in both forward and reverse directions
on an alternating day basis, data was:collected for both directions of motor
rotation.
To obtain a sample of shredded refuse, one of the test personnel sig-
naled the shredder operator to stop the feed and discharge conveyor after an
interval of relatively constant power was recorded. At the same time a ref-
erence mark noting the point at which the conveyors stopped was made on the
chart recording. After stopping the conveyors, a three-meter segment of
conveyor located at approximately the mid-point of the inclined rubber belt
conveyor was swept clean of refuse and weighed. From the collected mater-
ial, a representative sample of 8-13 kilograms was gathered and prepared for
shipment to Richmond, CA, for further analysis.
Since the Cockeysville shredders discharge onto short discharge convey-
ors which, in turn, empty onto the long inclined conveyors from which the
flow rate samples were collected, it was necessary to modify equation 3 in
order to correctly determine the time needed for refuse to travel from the
shredder to the point of sample collection. Equation 3, as modified for the
Cockeysville calculations, becomes
l
-
l
2
-
2
2h
1/2
2h,
1/2
(10)
where: d\ =
V2 =
horizontal distance from the shredder center line to the
point below which the refuse stream impacted upon the inclined
conveyor (m);
velocity of the discharge conveyor (m/sec);
distance from the point at which the refuse stream impacted on
the inclined conveyor to the center line of the segment from
which the flow rate sample was taken (m);
the velocity of the inclined conveyor (m/sec).
49
-------
hi = vertical distance from the bottom of the grates at the center
line of the shredder to the top of the horizontal discharge
conveyor;
\\2 - vertical distance between the top of the discharge conveyor
and the point at which the refuse stream impacted upon the
inclined conveyor (m); and
g = acceleration due to gravity (m/sec );
With the use of equation 10 instead of equation 3, flow rate determina-
tion and correlation with power requirements were done as described under
General Procedures.
GREAT FALLS PROCEDURE
Both shredders at Great Falls discharge onto a common conveyor, so feed
to the shredder which was not being tested was stopped and sufficient time
allowed for the untested shredder to empty before starting the test.
Since the power monitoring equipment was installed in a location from
which observation of the rest of the plant was impossible, the method for
obtaining data as described under General Procedures was modified slightly
to better suit the circumstances. The distance from the center line of the
shredder discharge to the center line of the segment of conveyor from which
the sample was obtained was carefully measured as were the velocity of the
horizontal and inclined discharge conveyors. From the distance and velocity
data the time, t, needed for refuse to travel from the shredder to the sam-
ple point could be calculated from
d, d, 2h,1/2
* =
where: dj = horizontal distance from the shredder discharge center line
to the point below which the refuse stream impacted upon the
inclined conveyor (m);
Vi = velocity of the horizontal discharge conveyor (m/sec);
d2 = distance from the point at which the refuse stream impacted
upon the inclined conveyor to the center line of the segment
from which the flow rate sample was taken (m);
V2 = velocity of the inclined discharge conveyor (m/sec);
\\l = vertical distance from the midpoint of the shredder discharge
opening to the top of the horizontal discharge conveyor;
h2 = vertical distance between the top of the horizontal discharge
conveyor and the point at which the refuse stream impacted
upon the inclined conveyor; and
g = acceleration due to gravity (m/secz).
At the beginning of each experiment a member of the test team started the
monitoring equipment and, after returning to the control room to stand with
one of the plant operators, started the chronometer. Approximately t sec-
onds later the operator was told to stop both the feed and discharge belts.
Since the feed and discharge belts are interlocked, a single control button
50
-------
was able to stop them both simultaneously. With the chronometer still accu-
mulating time, the test team member returned to the location of the monitor-
ing equipment and stopped the chronometer simultaneously with removing the
pen from the chart paper. The total elapsed time, tt, displayed by the
chronometer was then recorded.
The midpoint of the section of chart paper corresponding to the midpoint
of the flow rate sample was located by measuring the distance, d, from the
end of the inked line on the chart. The value of d was found from
d = (tt - t)Vc (11)
where: t^ = total elapsed time (sec);
t = time for refuse to travel from shredder midpoint to sampling
midpoint (sec);
and
Vc = velocity of the chart (m/sec).
Determination of the complete interval on the chart recording during which
the flow rate sample was being discharged, calculation of flow rate,
correlation of power and flow rate, and sample shipment were performed as
described in General Procedures.
TINTON FALLS PROCEDURE
The tipping floor is the only part of the Tinton Falls facility which is
common to both shredders. However, the layouts of the two shredding lines
are virtually identical, so the same testing procedure could be used on
either line. A three meter section near the end of the discharge conveyor
belt was selected as the most convenient point at which to obtain flow rate
samples. The ferrous recovery magnets were turned off before running any
tests so the entire flow rate could be measured.
Prior to testing, the distance from the shredder discharge to the center
line of the sampling segment and the discharge conveyor belt speed were
carefully measured. The speed and distance data were used in conjunction
with equation 3 to obtain a value for t, the time needed for refuse to tra-
vel from the shredder to the midpoint of the sampling segment. Due to the
design of the shredder, the value of h in equation 3 is zero. For each
test, the monitoring equipment was started and allowed to record the power
level for at least one minute before the conveyors were stopped.
The power monitoring equipment was located in the transfer trailer com-
pactor room near the controls for the discharge conveyor and the location
selected for obtaining flow rate samples. Since the shredder and the feed
conveyor are controlled from the main control room which cannot be seen from
the compactor room, coordination between the operator in the control room
and test personnel in the compactor room was affected through an intercom
system. If the discharge conveyor had been stopped at the same time as the
feed conveyor, jams in the shredder discharge could have resulted. By wait-
ing approximately t seconds (the length of time necessary for refuse to tra-
vel from the shredder discharge to the center of the sampling area) ample
51
-------
time was provided for the shredder to empty and thus avoid discharge jams.
The operator in the control room was told when to stop the feed conveyor.
Simultaneously the chronometer was started. Roughly t seconds after the
feed conveyor was stopped, one of the test personnel simultaneously stopped
both the discharge conveyor and the chronometer before returning to the mon-
itoring equipment. The time, tj., displayed by the chronometer was recor-
ded on the chart itself and the Start/Stop switch on the chronometer activa-
ted a second time simultaneously with removal of the pen from the chart.
Since the chronometer was operating in the Split/Cumulative mode (see Other
Equipment for details) the time, tt, displayed by the chronometer after
the second activation of the Start/Stop switch was the total time accumula-
ted since the chronometer was initially started. The value of t^ was also
recorded, and all the refuse was removed from the three meter section of the
conveyor which had been selected for flow rate sampling. The refuse was
weighed, and a representative subsample was removed and prepared for ship-
ment as discussed under General Procedures.
Measurement of the distance, d, from the end of the inked line on the
chart located the midpoint of the section of chartpaper which corresponded
to the midpoint of the flow rate sample.The distance, d, was calculated from
d = (tt - tx + t)Vc (12)
where: t^ = total elapsed time between stopping the feed conveyor and
removing the pen from the chart (sec);
ti = elapsed time between stopping the feed and discharge
conveyors;
t = time required for sample to travel from discharge to centerline
of sample (sec); and
Vc = chart speed (m/sec)
Again, determination of the complete interval on the chart during which the
flow rate sample was being discharged, calculation of flow rate, and corre-
lation of power and flow rate were performed as described in General Proce-
dures.
ODESSA PROCEDURE
A 2.4 meter section of belt just above the point at which the discharge
conveyor reached ground level was selected as the best flow rate sampling
area. The belt speed and distance from the centerline of the sampling area
to the centerline of the shredder were carefully measured and used in con-
junction with equation 3 to obtain a value for the time, t, needed for
refuse to travel from the shredder to the midpoint of the sampling area.
The power monitoring equipment was placed in the motor control room
(located under the feed conveyor and just behind the shredder discharge).
Due to the potential danger of falling objects and explosions, personnel
were not permitted in the shredder pit while refuse was being fed to the
shredder, thus limiting the opportunities for sampling to those times when a
full transfer trailer was being replaced by an empty one. Although the
shredder at the Odessa plant is reversible, all the sampling was done when
52
-------
the shredder was running in the forward direction. The chart recorder was
started as the full transfer trailer was being removed. The hydraulic push
pit into which shredded refuse was discharged had enough capacity to hold
roughly the amount of refuse which could be shredded in five minutes of nor-
mal operation. The feed conveyor was restarted and refuse fed to the shred-
der until the shredder motor current monitor in the control room indicated a
relatively constant load. The feed conveyor was then halted and the chro-
nometer simultaneously started. Allowing some time between stopping the
discharge conveyor reduced the possibility of jamming the shredder dis-
charge. Approximately t seconds after stopping the feed conveyor, the dis-
charge conveyor was also stopped and the elapsed time, t, between stopping
the two conveyors was recorded. A reference mark on the chart was obtained
by removing the pen from the paper. The total elapsed time, t^, from
stopping the feed conveyor to removing the pen from the chart paper was also
recorded. The point on the chart corresponding to the flow rate sample was
located by measuring a distance, d, from the point at which the pen was
removed. The distance, d, was calculated through use of equation 12. Flow
rate determination, location of the complete sample interval on the chart,
and power-flow rate correlation were performed as described in General
Procedures.
HAMMER WEAR EXPERIMENTAL PROCEDURE
General Procedure
Hammer wear investigations were conducted on base hammer materials as
well as on a number of hardfacing alloys. The experimental procedure con-
sisted of cleaning and weighing hammers prior and subsequent to shredding a
measured amount of solid waste. The quantity of hammer material lost and
tonnage of refuse shredded were used to ascertain the degree of wear (W0)
for each type of hammer base material or hardfacing alloy tested. As is
common in the industry, the degree of wear is expressed as the weight of
material lost due to wear per unit weight of refuse shredded.
Hammer wear tests were conducted by CRS for each site except Ames, Tin-
ton Falls, and Odessa. Since hammer retipping is not practiced at Ames,
Tinton Falls, or Odessa, wear data for these sites were provided by the
plant operators. Tonnage for all the wear experiments were monitored by the
truck scales located at each facility. A general overview of the wear
experiments conducted at each site is provided in the following paragraphs.
Appleton
Two sets of hammer wear data were collected at the Appleton facility.
One data set involved hammer wear in the east mill where 12 hammers hard-
faced with two passes of Stoody 134 electrode were cleaned, weighed, and
placed in the mill in two rows of six hammers each. After shredding 398
tons of waste, the east mill hammers were removed, cleaned, and reweighed.
The other data set was obtained from the west mill. Each of the four
rows of 12 hammers in the west mill was hardfaced with two passes of hard-
facing alloy. The four alloys tested were McKay 48, McKay 40 TiC, Stoody
53
-------
134, and Amsco Super 20. Hammers were cleaned, weighed, and installed after
the hardfacing was applied (Figure 33). In order to facilitate proper iden-
tification of test alloys, each row of hammers containing a particular
application of hardfacing alloy was marked in the shredder and noted on the
data sheets. After shredding 141 tons, the hammers were removed, cleaned
and reweighed.
Hammers used in both mills were built-up prior to hardfacing application
with Stoody 110 electrode.
Ames
The wear data for the Ames site was supplied by the city of Ames. The
data covered hammer wear for the primary shredder over the four-month period
from October 11, 1977, to February 11, 1978. Since the hammers in the pri-
mary shredder were neither hardfaced nor built-up, these data give the wear
of the base hammer material (manganese steel). Each of four rows of 12 new
hammers was weighed and installed in October. After four months of opera-
tion in which 14,254 tons of waste were shredded, each row of hammers was
pulled and reweighed. The hammer weight and tonnage information was subse-
quently transmitted by letter to CRS by the city of Ames.
Cockeysville
Two sets of hammer wear data were collected at the Baltimore County
facility. The first set consisted of wear determination of manganese steel
hammers installed in shredder #2. These hammers were installed new and
without any deposition of hardfacing or build-up material. Hammers were
cleaned and weighed prior to installation. After shredding 305 tons of
waste, these hammers were removed, cleaned, and reweighed.
The second set of wear experiments were conducted using the same hammers
which were worn-in during the first data set and subsequently built-up with
McKay 118 electrode before application of the hardfacing alloys. Each of
the four rows of hammers (six hammers per row) was coated with one pass of a
hardfacing alloy. The four alloys tested were: 1) Lincoln Mangjet, 2) Lin-
coln Abrasoweld, 3) McKay 55, and 4) Lincoln Faceweld 12. After application
of the hardfacing alloys, the hammers were cleaned, weighed, and installed
in shredder #1 (Figure 34). So the test alloys could be identified, each
row of hammers containing a particular application of hardfacing material
was marked in the shredder and noted on the data sheets. After shredding
255 tons of waste, the hammers were removed, cleaned, and reweighed.
Great Falls
Hammer wear data was obtained from the Model 42F shredder at the Great
Falls facility. The 42F vertical shaft hammermill has three distinct zones
within the shredder (Figure 35). The top zone, in which refuse undergoes
preliminary breakage, is known as the breakage zone and extends from the top
of the shredder to the bottom of the section with the cone-shaped walls.
There were 20 hammers in the breakage zone. The middle or grinding zone
54
-------
Figure 33. Cleaned hammer being weighed prior to installation in
west mill.
Figure 34. Set of newly hardfaced hammers ready for installation
at the Cockeysville facility (note pen resting on hammer
for reference of scale).
55
-------
•<- Breaker zone
Cutting zone
Sweep zone
Figure 35. Pictoral of vertical hamrnermill
56
-------
contains nine hammers and, as the name implies, is where most of the shred-
ding takes place. The bottom area is the exit zone in which shredded refuse
is forced from the shredder by a series of nine hammers arranged in three
groups of three hammers each. The hammers in the three zones are known as
breakers, cutters, and sweeps respectively.
Twenty-six new hammers were used in the wear tests. Nine hammers were
hardfaced with Stoody 2134, nine with Amsco Super 20, and eight were left
bare. All twenty-six hammers were cleaned and weighed prior to installa-
tion. The first group of sweeps was replaced with three Amsco surfaced ham-
mers, the second with three Stoody surfaced hammers, and the third with
three bare hammers. Each vertical row of cutters was similarly replaced so
that the wear of the different surfaces could be compared for the same
height above the bottom of the shredder. Only the lower eight of the break-
ers were replaced with new hammers since the upper breakers experience very
little wear. After shredding 126.9 tons, the hammers were reversed so as to
present the opposite face to the refuse and, after shredding an additional
51.4 tons, were removed, cleaned, and reweighed.
Tinton Falls
The wear data for the Tinton Falls facility was provided by the plant
manager. The data represented grinder wear for the period from October,
1978, through January, 1979, during which 16,353 tons of refuse were
shredded. The ring grinders were never resurfaced so the data represents
the wear of the base grinder material, in this case nickel-manganese steel.
The Tinton Falls shredders utilize 60 grinders in 5 flights of 12 grinders
each. After shredding 16,353 tons of refuse, the'grinders were removed,
cleaned, and weighed. The tonnage shredded and the initial and final
weights were then sent to CRS for analysis. The internal configuration of
the ring shredder showing the locations of the ring grinders is presented in
Figure 36.
Odessa
Wear data for the Odessa site was provided by the Sanitation Department
of the City of Odessa. The data were collected during the winter of 1978
during which 5,201 tons of refuse were shredded. Again, since hammer retip-
ping is not practiced at this site the wear data represents the wear of the
base material. The 14 new hammers (in rows of four, three, four and three
hammers each) were weighed and the weights recorded before installation.
When the hammers were removed, they were weighed a second time and the
weights again recorded. The initial and final weights and the total tonnage
shredded were turned over to CRS for analysis.
57
-------
Figure 36. Internal configuration of the
Tinton Falls ring shredder.
58
-------
SECTION 6
RESULTS AND DISCUSSION
MOISTURE CONTENT OF SHREDDED REFUSE SAMPLES
Average air dry moisture contents of the shredded refuse samples exhibi-
ted a range of approximately 16 to 36 percent (Table 10). The Appleton
refuse had the highest moisture content, averaging in the neighborhood of 30
to 36 percent. Although a water spray is sometimes used at Appleton to
control dust dispersion, the spray was turned off five minutes before
beginning the test runs. Consequently the moisture contents for Appleton
are average innate values. None of the other sites used a water spray
system during the test programs.
The average moisture contents of shredded refuse at the other five sites
tended to be rather low, (in the range of 16 to 22 percent) when compared to
"typical" refuse. A range of 25 to 30 percent air dry moisture content is
typical of shredded refuse. The primary reason for the low values is the
time period of the tests, namely late fall and winter. A lack of green lawn
and garden debris exhibited during this time period accounts for the rela-
tively low moisture contents. Appleton on the other hand was tested during
September when significant quantitites of green lawn and garden debris were
present in the refuse brought to the plant.
Based upon visual observation, the refuse processed at the Ames and
Cockeysville plants was characterized as "commercial" waste, that is waste
tending to have a large particle size and significant quantities of paper
and plastic. Refuse at the other four sites was characterized as "residen-
tial". Such waste contains lawn and garden debris, food waste, and tends to
contain particles that are smaller than those typical of commercial waste.
MEASURED THROUGHPUTS
The primary purpose of this study was to obtain performance and operat-
ing data on large-scale shredders, that is, shredders capable of handling
large throughputs of solid waste. Shredders evaluated in this study covered
the full spectrum of throughput capacity ranging from an average of approxi-
mately 20 TPH at the Ames plant to 82 TPH at the Odessa facility (wet weight
basis). The minimum throughput measured during the tests was 3.5 TPH at the
Appleton facility while the maximum throughput was 127.4 TPH measured at the
aFor shredded refuse, bone dry moisture content (percent) equals air dry
moisture content (percent) plus 8 to 9 percent.
59
-------
TABLE 10. SUMMARY OF AVERAGE VALUES OF IMPORTANT PARAMETERS
MEASURED DURING THE SHREDDER PERFORMANCE EVALUATION
•
Throughput
Shredder
Apple ton East
Appleton East
Appleton West
Ames Primary
Ames Primary
Ames Secondary
Ames Secondary
Cockeysville
Shredder #1 (F)
Shredder #1 (R)
Combined 11
Great Falls 20TPH
Tinton Falls
Odessa (F)
HC
(*).
35.6
35.0
30.1
15.9
16.3
21.8
22.2
16.3
17.9
17.1
21.1
19.2
17.8
Sw
(TPH)
27.0
24.8
18.1
19.6
18.6
20.9
22.4
39.3
59.8
49.5
14.8
60.8
82.0
"d
(TPH)
17.1
18.1
11.7
16.5
15.6
16.3
17.4
32.9
48.9
41.1
11.5
48.6
66.7
Characteristic
Size
If
0
(cm)
3.8
3.7
5.6
4.6
5.0
1.4
1.3
2.2
2.0
2.1
2.4
3.8
3.2
°o
(cm)
1.0
0.9
0.9
1.5
0.9
0.2
0.2
0.5
0.3
0.4
0.8
0.8
1.2
Nominal
Characteristic Specific Energy
Size
*90 'go
(cm) (cm)
9.9 1.0
9.8 1.0
11.2 1.2
11.7 3.5
12.1 3.5
3.4 0.6
3.3 0.5
6.4 1.3
6.1 1.0
6.2 1.2
s.e1 1.7
9.6 1.8
9.2 4.0
r ov F:
0 EQ 0
fKwh [Kwh fKw
F F k,
Or Notes*
°w
h Kwh
~ 17
4.6 2.5 3.0 1.9 a
4.7 2.6 3.4 1.7 b **
5.3 2.0 3.
9 2.0 e,h
6.1 3.2 5.1 2.6 e
6.6 3.0 5.5 2.4 c **
13.1 6.3 9.
14.2 5.5 10.
12.9 2.6 10.
11.0 1.6 9.
11.9 2.4 10.
8.2 2.5 6.
2.8 0.7 2.
0.8 0.7 1.
9 4.1 e
8 3.3 d **
8 2.4 e,f **
1 1.8 e,g **
0 2.4 e
4 2.0 e **
3 0.6 e **
1 0.7 e,f **
Notes: (test numbers refer to tests listed for each site and shredder in Appendix A)
a Excludes test #3, which was aborted due to extreme water content
b Excludes tests #3 and #9; #9 was conducted at abnormally low flow rate and results
are not indicative of normal operation
c Excludes test #10, which was largely dirt and demo debris; judged non-representative
of normal waste
d Excludes test 05, which was conducted at an abnormally low flow rate
e all test runs Included
f F = Forward Rotation (see text for explanation)
g R = Reverse Rotation (see text for explanation)
h 3 samples only to establish representative product size for wear experiments in
West Mill
** Data used in report as representative of average operating conditions at each plant
Odessa plant, again on a wet weight basis. It must be pointed out that both
of the above extreme values were special test runs and therefore are not
indicative of normal plant operation.
Average values for throughputs that were measured at each plant, along
with minimum and maximum values, are summarized in Table- 11. Data are shown
on both wet and dry weight bases. Scrutiny of the data in the table shows a
wide spectrum of throughputs about the average value for each site. These
wide variations are a consequence of obtaining performance and operational
data over as wide a throughput base as plant operation would permit.
60
-------
TABLE 11. THROUGHPUTS MEASURED DURING SHREDDER PERFORMANCE EVALUATIONS
mw md
Shredder
Appleton East
Ames Primary
Ames Secondary
Cockeysville tl (F)a
Cockeysville #2 (R)b
Great Falls 20 TPH
Tinton Falls
Odessa (F)a
Min
3.5
10.6
7.0
22.6
39.4
10.0
20.7
13.3
(TPH)
Max
47.5
35.1
27.2
52.3
95.7
19.4
99.5
127.4
Avg
27.0
19.6
20.9
39.3
59.8
14.8
60.8
82.0
Min
2.1
8.3
5.7
18.8
34.6
8.1
17.2
11.9
(TPH)
Max
34.0
28.6
22.2
43.7
81.9
15.3
76.5
108.3
Avg
17.1
16.5
16.3
32.9
48.9
11.5
48.6
66.7
F = forward rotation
R = reverse rotation
SIZE DISTRIBUTION OF SHREDDED SOLID WASTE
In keeping with information available in the literature and the experi-
ence of CRS in developing relationships among the parameters governing the
process of size reduction, the results of the size distribution analysis are
reported chiefly in terms of characteristic size, i.e., the screen size cor-
responding to 63.2 percent cumulative passing. At the same time, the nomi-
nal product size, or that screen size corresponding to 90 percent cumulative
passing, is accepted for use by the solid waste industry in general. Both
size designations, where appropriate, are included for the reader's consid-
eration. We must emphasize, however, that studies in size reduction of
solid waste have shown that there is a greater degree of correlation between
characteristic size and other variables of size reduction (most signifi-
cantly specific energy) than exists between nominal size and other variables
of size reduction.
As a matter of scientific interest and also as a means for developing a
method of interchangeability between characteristic (X0) and nominal
(X ) product sizes, a relation has been developed between these size
parameters based upon average values of the data obtained from the test
sites and University of California data. This relationship is presented in
Figure 37 and can be expressed as,
X9fJ = 2.23 X0 + 0.69 (13)
with a correlation coefficient of 0.93.
61
-------
H
12
o 8
r- 6
'©A
X9Q = 2.23 XQ + 0.69
Great Falls
Tinton Falls
Odessa
Appleton
Ames
Cockeysville
A,8,0 Un. of Ca. (Berk.)
10
Characteristic Size, XQ (cm)
Figure 37. Relationship between nominal and characteristic
product sizes for shredded MSW.
Average values of the characteristic and nominal sizes for shredded
refuse samples collected at the test sites along with other pertinent shred-
ding and operational parameters are shown in Table 10. The reader should
witness the notes to the data since some of the average values include data
which was not typical of normal plant operation. As part of the research,
some tests were conducted at very low throughputs. These special tests and
their results are discussed later. The data corresponding to normal plant
operation are designated by (**) in the table. The data used to develop
Table 10 are included in Appendix A.
Under typical operating conditions, the average values of characteristic
product sizes varied from a maximum of 5.6 cm for the Appleton west mill to
a minimum of 1.3 cm for the Ames secondary shredder while the average values
of nominal product size ranged from a maximum of 12.1 cm from the Ames
primary shredder to a minimum of 3.3 cm from the Ames secondary mill.
Reference should also be made to the average characteristic (nominal)
sizes encountered at the Cockeysville facility, namely on the order of 2.1
(6.2) cm. These values are not far removed from the average product sizes
of material discharged from the Ames secondary shredder. Despite the fact
that the Ames facility uses primary and secondary shredding, the character-
istic si^e of the product is only marginally smaller (approximately 38 per-
cent) than the size of the product resulting from the single shredding at
62
-------
the Cockeysville plant. The reasons that the product sizes from each of
these plants are comparable involves a combination of factors which will be
explained and discussed after presentation of the rest of the experimental
data.
Because of the number of variables that affect the degree of size reduc-
tion, there is, in general, a wide dispersion of data about the average
value of product size as evidenced in some instances by the large values of
the standard deviation. Consequently when talking about product sizes from
a particular shredder, one must be aware that there can be considerable
instantaneous variations in size of the product. The variables that contri-
bute to this dispersion of product sizes include moisture content, composi-
tion and size of refuse, and flow rate of material through the shredder.
The effect of variation of throughput upon product size will be covered in
more detail later in this report.
as a means of visualization of the results of the size distribution
analysis, the range of size distributions for several different sites is
plotted in Figure 38 along with typical size ranges for residential and com
mercial solid waste3. The wide band of the size spectrum for each shred-
der (as alluded to earlier) is apparent from the figure. The average size
distribution for shredded material encountered at each site under normal
operating conditions is shown in Figure 39, again with representative size
distributions for residential and commercial solid waste also shown. In
order to avoid congestion, the curve for Tinton Falls has been omitted from
Figure 39. The average curve for Tinton Falls lies between those for
Odessa and Appleton.
Screen Size (cm)
:igure 38. Ranges of size distribution of shredded MSW
measured at Appleton, Ames, and Cockeysville
aThe range of size distributions for raw residential and commercial solid
waste were determined from University of California data and that resulting
from studies conducted for private clients.
63
-------
100
80
60
•= 40
20
Cockeysville
Great Falls-
Odessa
Appleton
Ames
Secondary
0.5 1 5 10 50
Screen Size (cm)
Figure 39. Average size distribution of raw & shredded solid waste
*
MEASURED POWER
Along with size, power is one of most important variables affected by
throughput and moisture content. During the course of the study, power
measurements were found to range from a minimum of 4.5 kw (Odessa) to a max-
imum of 957.8 kw (Cockeysville). The minimum, maximum, and average values
encountered at each site are tabulated in Table 11. As indicated in Table
12, at least one data point was obtained at Cockeysville for which the net
power momentarily exceeded the rating of the shredder drive motor. All the
sites had motor protection circuits which would temporarily cut off the feed
of refuse to the shredder if the motor current rating was exceeded for more
than a few seconds.
The average power draw of the shredders that were tested ranged from
12.6 to 73.2 percent of full load motor rating.
THROUGHPUT EFFECTS ON SIZE
As some shredder operators are aware, product size tends to increase
when a shredder is emptying or when the throughput to the mill is reduced
significantly. Some tests were conducted at Appleton, Ames, and Tinton
Falls to quantitatively verify the above observation. These tests included
a run at each of the aforementioned sites in which the throughput to the
shredder was significantly reduced below the average throughput value found
under normal operating conditions. In each test, energy requirements and
throughput were measured and the size distributions analyzed for later com-
parison with the data collected under normal operating conditions.
64
-------
TABLE 12. POWER MEASURED DURING SHREDDER PERFORMANCE EVALUATION
Net Power, Pn Full Load
(kw) Motor Rating
Min. Max. Average (kw)
Appleton East
Ames Primary
Ames Secondary
Cockeysville #1 (F)a
Cockeysville #1 (R)b
Great Falls 20TPH
Tinton Falls
Odessa (F)a
7
.8
4.8
15
213
316
.5
.5
.8
32.8
61
4
.5
.5
167
213
376
560
957
128
281
162
.3
.5
.2
.7
.8
.3
.5
.9
76.
93.
217.
417.
0
7
3
Z
546.2
92.
134.
95.
4
9
0
298
746
746
746
746
187
746
373
Average as a
% of Rating
25
12
29
56
73
49
18
25
.5
.6
.1
.0
.2
.4
.1
.5
?F=Forward
R=Reverse
cAverage of all measurements taken at site
The experiments involving size reduction at low throughputs confirmed
the observation of larger product size when compared to those average values
encountered under regular operating conditions (Table 13). For example, at
Appleton the average values for throughput (wet weight basis) and character-
istic product size were 24.8 TPH and 3.7 cm, respectively, whereas data for
a test at 3.5 TPH (wet weight basis) showed the product size to be 5.5 cm.
Similarly, a test of the Ames secondary shredder at a low throughput of 7.0
TPH (wet weight basis) produced a characteristic product size of 1.7 cm
while the product size under normal operating conditions, 22.4 TPH (wet
weight basis), was 1.3 cm. The data collected at Tinton Falls exhibited the
same trends as those shown for Appleton and Ames (Table 13).
The shift in product size due to the effects of variation of throughput
is manifested in those tests conducted at low flow rates of refuse through
the shredders. Other data (1), which were obtained under strict experimen-
tal procedures, are available that indicate product size is a function of
throughput for a given set of operating conditions. Due to the concern for
maintaining normal plant operation and the concomitant loss of rigid
experimental control, it was not possible to develop reliable relationships
between product size and throughput other than to substantiate that such a
relationship does exist at the low end of throughput values.
As mentioned above, over the full range of throughput it was not possi-
ble to develop empirical relations for product size that were statistically
significant. However, the data from four cases, namely the Ames primary and
secondary shredders, the Odessa shredder, and the Tinton Falls shredder, are
interesting from the standpoint of a potential avenue of further research.
Although there is a good deal of scatter, plots of product size versus flow
rate indicate a trend of decreased particle size for high flow rates,
Figures 40, 41, 42, and 43. To substantiate these trends, a more comprehen-
sive study of each shredder is needed and could be a topic for further
research.
65
-------
TABLE 13. EFFECT OF VARIATION OF THROUGHPUT
UPON CHARACTERISTIC PRODUCT SIZE AND SPECIFIC ENERGY
Shredder
Shredder
Operation
m
(TPH)
rti
d
(TPH)
X
(cm)
E
Kwh/Td
Appleton East Average 24.8 18.1 3.7 4.7
Appleton East Low Flow Rate 3.5 2.1 5.5 3.7
Ames Secondary
Ames Secondary
Tinton Falls
Tinton Falls
Average
Low Flow Rate
Average
Low Flow Rate
22.4
7.0
60.8
20.0
17.4
5.7
49.1
17.2
1.3
1.7
3.8
5.0
14.2
2.7
2.8
3.6
With regard to the data presented in Figures 40 through 43, it should be
noted that the reduction of product size appears more noticeable for the
Xo data as opposed to the Xgg data. Data scatter also appears to be
greater for the latter size data.
The resolution of the effects of throughput on size and energy consump-
tion is important from the standpoint of control of the process of size
reduction. For example, in a resource recovery facility incorporating
shredding and recovering RDF, the particle size of the fuel will change
somewhat for variations in the throughput to the shredders. If the particle
size of the fuel is specified by the user, then a model or relationship
between product size and throughput would allow the plant operator to
establish the allowable variations in throughput values to maintain the
recovered fuel within specification.
The argument can be expanded to include the input of shredded refuse
into air classifiers, screens, and other processing equipment. These pro-
cessing devices will obviously exhibit more efficient performance if they
are provided with a uniform particle size. Establishment of the particle
size consequently dictates the size range of the product discharged from the
shredder, which in turn indicates the degree of variation in throughput that
can be tolerated.
Although process control in resource recovery plants is now practiced to
only a very limited degree, the successful and efficient operation of such
facilities will eventually be dependent upon effective control of each unit
process. Consequently, the development of a relationship between product
size and refuse throughput is deemed as a necessary first step in the con-
trol of the size reduction process.
66
-------
25
20
15
A
A
CTl
--J
10
A A
X90
10 20
Flpw Rate, mw (T PH)
w w
30
Figure 40. Relationship of product size and
and throughput for Ames Primary
Shredder.
TO _ 20
Flow Rate, m, (T PH)
30
Figure 41. Relationship of product size and
throughput for Ames Secondary
Shredder.
-------
CM
00
15
10
O
A
A90
O
O
25
50
75
Flow Rate, mw (TWPH)
100
125
Figure 42. Relationship of product size
and throughput for Tinton
Falls shredder.
25
20
§ 15
01
"o
4-f
<2 10
50 100
Flow Rate, 111 (T PH)
w w
150
Figure 43,
Relationship of product size
and throughput for Odessa
shredder.
-------
ESTIMATION OF SPECIFIC ENERGY REQUIREMENTS
Data collected in the test program along with that from other sources
(2) enables the development of relationships for the prediction of specific
energy on a dry weight basis, E0, (kwh/Tj), as functions of both charac-
teristic and nominal product sizes (X0 and Xgg, respectively) for hori-
zontal hanmermills. Average values of E0, and X0, and Xgg and the
range of one standard deviation for each variable are plotted in Figure 44.
The trend of an increasing energy requirement as a consequence of producing
smaller and smaller part.icle sizes through size reduction is quite apparent.
An attempt to develop a functional relationship between E0 and the
size parameters using standard curve fitting techniques yielded the
following equations:
Ert = 23.25 Xrt
o o
-0.92
(14)
and
EQ = 49.94 Xgo
-0.86
(15)
60 -
SO
40
£ 30
20
10
-0.92
•f Un. of Cal.
© Appleton East
B Ames Primary
• Anies (Pri.+Suc.)
A Cockeysville (Reverse)
a Cockeysville (Forward)
i 1 One Standard Deviation
0.1
1.0
IVcirtuct Size (cm)
10
Figure 44.
Specific energy consumption (A.D.B.)
as a function of product size.
69
-------
where EQ is expressed in kwh/Tj and X0 and Xgo are expressed in cm.
The correlation coefficients of equations 14 and 15 are 0.93 and 0.91
respectively.
Equations 14 and 15 indicate that specific energy on an air dry weight
basis (A.D.B.) is approximately proportional to the -0.9 power of the
characteristic size or nominal size.
These equations were developed with the aid of data from facilities
handling anywhere from four to ninety tons of refuse per hour and
consequently represent the full gamut of operating conditions found in
actual practice. The curves should be valuable in estimating net power
requirements for shredders in general. The term "in general" should not be
overlooked since the actual power requirement for a particular shredder is a
function of many variables including internal machine configuration, hammer
tip speed, throughput, refuse size, composition, and moisture content. The
dispersion bars about each average value in Figure 44 allow the range of
specific energy that could be used as a consequence of producing a particle
of a particular size to be estimated.
The preceding discussion was formulated on a dry weight basis, and
consequently some estimation of moisture content is necessary before energy
requirements can be calculated on a wet weight basis (W.B.). Although
correlation between specific energy on a wet weight basis (kwh/Tw) and
average product size is not as good as that between specific energy on an
air dry weight basis and average product size, the relationship between the
former variables is presented in Figure 45. The data represent samples
ranging in moisture content from 6.9 to 47.5 percent (A.D.B.) with an
average moisture content of 22.9 percent (A.D.B.). The appropriate
governing equations on a wet weight basis are
EQ = 17.91 X0'°-90 (16)
EQ = 35.55 X90-°-81 (17)
w
Equations 16 and 17 expressing specific energy on a wet weight basis as
a function of- product size are similar to those calculated for specific
energy on a dry weight basis. In addition to the similarity of the
exponents, (all range from 0.81 to 0.92) the values of the constants for the
equations containing Xgo are approximately twice the values of the
constants of the equations containing X0 as would be expected when
equation 13 is considered. For equations 14 and 15, the ratio of the
leading constants is 2.15, while the ratio of the leading constants for
equations 16 and 17 is 1.98.
With regard to the correlation of data, specific energy (A.D.B.) and
average product size have correlation coefficients in the range of 0.91 to
0.93 while those for specific energy (W.B.) and average product size are in
70
-------
30
20
10
17.91 X0
-0.90
= 35.55 X
-0.81
90
Un. of Cal.
G Appleton East
m A;.ies Primary
• A:ies (Pri . fSec. )
A fockeysville (Reverse)
& Cockeysville (Forward)
0.1
1.0
Product Size (cm)
10
Figure 45. Specific energy consumption (W.B.) as a function of product size.
the range of 0.87 to 0.89. Consequently, correlation of product size with
energy on an air dry weight basis is considered slightly more accurate.
The energy and size data for the vertical shredders located at Great
Falls and Tinton Falls and the horizontal hammermill located in Odessa were
not used to develop the relationships shown in Figures 44 and 45. The Great
Falls and Tinton Falls test results were omitted because these data were for
vertical, as opposed to horizontal, shredders. The Odessa test results were
omitted because the data showed a lower specific energy consumption than
would be predicted from the data collected from the other four horizontal
hammermills and the University of California hammermill.
Statistical analysis of the data supports the conclusion that the
vertical shredders and Odessa shredder need to be examined separately. An
attempt to develop functional relationships predicated upon the results of
all six test sites, i.e. both vertical and horizontal hammermills, results
in correlation coefficients that are poorer than those developed solely for
horizontal hammermill (Odessa data omitted). As previously discussed,
correlation coefficients are approximately 0.9 for the curves developed for
the horizontal hammermills whose test results are shown in Figures 44 and
45. On the other hand, if the test results for the vertical shredders and
Odessa shredder are included in the functional relationships for specific
energy versus product size, the correlation coefficients drop to
approximately 0.7 as shown in Table 14.
The suspected reason for the different behavior of the Newell hammermill
as compared to the other hammermills is addressed in the section, "Compari-
son of Cockeysville and Odessa Results".
71
-------
TABLE 14. SPECIFIC ENERGY REQUIREMENTS VERSUS PRODUCT SIZE UTILIZING
TEST RESULTS FROM VERTICAL AND HORIZONTAL HAMMERMILLS
Air Dry Weight Basis
Equation Correlation Coefficient
-1.20
EQ = 21.77 XQ 0.71
-1.14
EQ = 60.17 Xgo 0.72
Wet Weight Basis
EQ = 16.30 X ~1J5 0.69
w
E = 42.92 X." ' 0.69
0 0
w
MOTOR SIZING
With regard to the estimation of the proper size motor for shredding
refuse at a specified rate, it must be remembered that gross power
requirements are composed of the net power required for size reduction plus
the freewheeling power (otherwise known as the idle power or power required
when not shredding refuse). Typically, the freewheeling power represents
about 10 percent of the full load rating of the motor thus leaving the
remaining 90 percent available for size reduction.
An estimation of the net power required to produce a specified particle
size can be found by multiplying the specific energy (EQ ) by the
anticipated flow rate (rt^). Dividing the product by 0.9wwill then produce
a rough estimate of the motor power which would be required.
This derivation assumes a service factor of 1.0. The actual motor size
chosen should account for normal variations in throughput. The tests showed
that service factors were anywhere in the range of 12 to 73 percent based
upon average power usage (Table 12).
POWER-FLOW RATE-MOISTURE RELATIONSHIPS
Although the trend for increased power requirements as throughput
increases is generally recognized, there is a paucity of data available in
the literature to support this contention. In this section data are presen-
ted to quantify the trend indicated above. Plots of net power versus flow
rate (on a wet weight basis) show the dependence of net power requirements
on throughput (Figures 46 through 52).
72
-------
A second commonly made observation regarding the shredding of municipal
solid waste is that higher moisture contents often lead to reduced power
requirements due to a combination of the effects of lubrication and a degra-
dation in the strength of the fibrous components of refuse.
To show how flow rate and moisture affect power requirements for the
systems tested, a multiple regression analysis was carried out for each set
of data. The multiple regression analyses were based on the assumption that
the curves which which would best describe the experimental results, shown
graphically in Figures 46 through 52, could be represented by an equation of
the form
Pn = A mwB(l-MC)C
where Pn = net power (kw);
mw = flow rate of refuse through the shredder (TWPH); and
1C = fractional moisture content of the refuse.
(18)
200r
150
TOO
50
Curve shown is for
average moisture content
of 35.0",.
1
10
20
30
Flow Rate, m. (T PH)
W
40
Figure 46. Net power draw of Appleton East Mill
73
-------
Curve shown is for
average moisture content
of 16:3--..
_ 150 _
10 20.
Flow Rate, mw (TWPH)
30
40
Figure 47.
Net power draw of Ames primary
shredder.
400_
350-
30C-
2SC-
20C-
15C
IOC
5C
Curve shown is for
average moisture content
of 22.2s;.
10 20
Flow Rate, m 0,/H)
30
Figure 48.
Net power draw of Ames secondary
shredder.
-------
01
1000
900
o
a.
.U
Q
Curve shown is for
average moisture content
of 16.3° (Forward) and
17.93 (Reverse)
0 10 20 30 40 50 60 70 80 90 100
Flow Rate, mw (TWPH)
Figure 49. Net power draw of Cockeysville #1
shredder.
Curve shown is for
average moisture content
10
15
20
25
Flow Rate, m (T PH)
w w
Figure 50. Net power draw of Great Falls
20TPH shredder.
-------
250
cn
Curve shown 1s for
average moisture content
25
50 75
Flow Rate, m (T PH)
TOO
125
Figure 51. Net power draw of Tinton Falls
shredder.
250
200
150
100
50
Curve shown is for
average moisture content
50
100
Flow Rate, mw (TW?H)
150
Figure 52. Net power draw of Odessa shredder.
-------
The exponents B and C in equation 18 determine how power requirements
change with flow rate and moisture content respectively, while the
coefficient A affects the magnitude of the power requirements. Positive
values for the exponents B and C indicate that power increases with flow
rate but decreases as moisture content increases. The equations resulting
from the regression analyses are presented in Table 15.
An examination of the information contained in Table 15 reveals that the
assumed form of the curve fits the experimental data relatively well except
in the case of Great Falls. The goodness-of-fit is indicated by the
correlation coefficient, R. An exact match between the experimental data
and the assumed form of the curve will yield a value for R of +1 while no
correlation will yield a value of zero. The Cockeysville reverse data was
fit best by the curve (R = 0.96) while the Great Falls data had the worst
fit (R = 0.28).
The poor fit of the Great Falls data is speculated to be a consequence
of the design and internal configuration of the shredders. To understand
the rationale behind such speculation, one must understand the assumptions
that are implicit in the development of equation 18, namely:
1. The net power is a function of the mass and moisture content of the
material actually held within the mill;
2. The flow rate and mass of material within the shredder (the holdup)
are related by a power curve or, in other words, the holdup equals some
TABLE 15. RESULTS OF MULTIPLE REGRESSION ANALYSIS
NET POWER AS A FUNCTION OF FLOW RATE AND MOISTURE CONTENT
Correlation Number of
Shredder
Appleton east
Ames primary
Ames secondary
Cockeysville #1 , forward
Cockeysville #1, reverse
Cockeysville #1, combined
Great Falls 20 TPI1
Tinton Falls
Odessa, forward
coefficient
0.85
0.74
0.93
0.92
0.96
0.91
0.28
0.85
0.91
data points
11
10
10
6
6
12
12
12
12
Equation
0.94 1.54
P =6.33 lii (1-HC)
n w
P^.62,!,1-20 (l-MC)-7-04
n w
1.92 -2.84
P =.28 in (l-MC)
n w
0.82 4.79
P =47.04 m (l-MC)
n w
1.18 2.01
P =6.40 m (l-MC)
n w
0.90 2.25
P =21.85 iti (l-MC)
n w
0.27 -0.63
P =39.97 m (l-MC)
n w
0.94 1.66
P =4.05 m (l-MC)
n w
0 84 -12-64
P =0.14 in (l-MC)
n w
77
-------
constant factor multiplied by the flow rate which has been raised to some
power; and
3. The load is distributed uniformly along the axis of the shredder.
It is evident that in the case of a vertical hammermill such as the one
used in Great Falls, the load is not uniformly distributed along the shred-
der axis. The hammers in the upper portion of the mill (the breakers) and
the hammers at the bottom (the sweeps) contribute relatively little to the
size reduction of material within the mill. This observation is corrobora-
ted by the low rates of hammer wear exhibited in these areas when compared
to the wear of the hammers in the throat section (cutters). Most of the
size reduction is done in the latter section where the clearance between the
hammers and the mill liners is least. Although the power requirements of
the narrow clearance area may indeed be fit by equation 18, it is not possi-
ble to obtain moisture and flow rate data for material within this section
of the mill without knowing substantially more about how material is trans-
ported through the mill.
Values of the exponent B range from 0.27 (Great Falls) to 1.92 (Ames
secondary). The very low value for Great Falls stems from the problem
mentioned above. The relatively high value obtained for the Ames secondary
shredder is believed to be derived from the fact that, at the time of the
tests, the hammers had been worn to less than half the original length and
were not shredding refuse very effectively. In most cases, however, the
value of B is close to unity thus indicating that power requirements are
almost directly proportional to the flow rate.
The exponent C ranges from -12.64 (Odessa) to 4.79 (Cockeysville,
forward). Positive values for C indicate that the net power required
decreases as moisture increases, while negative values indicate that power
increases with moisture content. Contrary to expectations, some negative
values of C can be noted. Since such a result runs counter to most previous
experience, it is difficult at the present time to offer a reasonable
explanation of the phenomenon without performing a more extensive series of
tests.
The curves drawn in Figures 46 through 52 represent the power equations
listed in Table 15 with the value of the moisture content (MC) in the equa-
tions replaced by the average value calculated for the respective sites.
DISCUSSION OF APPLETON HOLDUP DATA
The Appleton site was the only one at which holdup data could be gath-
ered. Holdup, as mentioned earlier, is the mass of material actually held
within the shredder at any given moment. It has been demonstrated that the
power requirement of a shredder will remain relatively constant regardless
of flow rate and grate spacing as long as the shredder holdup also remains
constant (1). In other words, if the grate spacing of a shredder is changed
and the flow rate is adjusted so as to provide the same holdup as a differ-
ent flow rate did before the grate spacing modification, then the power
requirements of the shredder will be unchanged. The implication is that,
78
-------
although power is a function of holdup, the shredder holdup is a function of
flow rate and the design of the shredder.
A regression analysis of the power-holdup and holdup-flow rate data
(Appendix A) shows that the power-holdup data may be described by an
equation of the form
P = 1.60284H0'95716 (19)
n v '
and the hold-up flow rate data may be described by
m = 0.50749H1'04402 (20)
w
where: Pn = net power (kw);
H = shredder holdup (kg);
mw = flow rate (TWPH).
The regression coefficients are 0.82 and 0.90'for equations 19 and 20
respectively.
Flow rate is shown as a function of holdup in equation 20 because the
flow rate data gathered were output flow rates while, strictly speaking,
holdup is a function of both input and output flow rates. Moisture is
neglected as a parameter in equation 19 since the moisture contents were
determined from the flow rate samples and hence may not have been
representative of the holdup as well.
DISCUSSION OF THE RESULTS FOR THE AMES PRIMARY AND SECONDARY SHREDDERS
Although both the Ames primary and secondary shredders are American Pul-
verizer model 6090 shredders and are driven by identical motors, the inter-
nal configuration of the hammers and grates differ greatly as evidenced by
the different equations developed for each shredder. Principal among the
internal shredder variations are size and spacing of grate bars and hammer
design. It has been demonstrated at the University of California Solid
Waste Processing Laboratory in Richmond, Ca. (1,3) that:
1. The power requirement of a shredder is dependent on the mass of
material held within the shredder (the shredder holdup);
2. For a constant grate spacing the shredder holdup increases with flow
rate;
3. For a constant flow rate the shredder holdup increases as the grate
openings are decreased in size; and
4. Power requirements decrease with hammer rotational velocity.
It is thus expected that the power requirements of the secondary shred-
der with its substantially smaller grate openings would be significantly
higher than that of the primary shredder. Examination of the power-flow
rate plots for the two shredders while inserting trial flow rate and mois-
ture data into the equations developed for the Ames shredders show that the
79
-------
secondary shredder does indeed require more power than the primary shred-
der. It has already been noted that reducing the rotational velocity of the
shredder (causing lower hammer tip velocities) results in decreased power
requirements (3). Since shortening the hammer length for a fixed rotational
velocity also reduces the hammer tip velocity, decreased power requirements
may be expected for shorter hammers also. At Ames the hammers are allowed
to wear out before replacing them (rather than regularly retipping them),
and it is reasonable to expect the ratio of primary to secondary power
requirements to vary as the difference in primary and secondary hammer
lengths changes due to wear. At the time of the tests the hammers on the
secondary shredder were already substantially worn as shown in Figure 13.
Consequently the only data collected is for a "worn down" condition of the
secondary hammers. From the standpoint of optimization of two-stage size
reduction, it would be interesting to obtain power, flow rate, moisture con-
tent and size distribution data for the condition of new hammers in the
secondary mill and then compare the energy requirements of primary and
secondary shredders.
At this point it is not possible to predict what the power ratio would
have been had the primary and secondary hammers been the same length. Due
to the lack of data for other hammer lengths on either shredder, it is
impossible to quantitatively predict the power requirements as the hammers
wear (other than predicting the power should decrease) even though the plant
personnel claim that the power requirements are in fact less for the well
worn hammers than for new hammers.
Although low power requirements are desirable from a processing stand-
point, a potential pitfall of a design based on short hammers is limitation
of throughput capacity of the shredder, i.e. as the hammers get shorter the
maximum flow rate through the shredder can be expected to decrease.
DISCUSSION OF BI-DIRECTIONAL ROTATION OF COCKEYSVILLE #1 SHREDDER
The Cockeysville shredders are reversible Tracor-Marksman model A60 ham-
mermills. Since the interior of the shredder is axially symmetric, the
power requirements ideally should be identical for either forward or reverse
operation although the plant operating personnel claim that operating in the
forward direction requires more power. From the plot of the Cockeysville
power-flow rate data in Figure 49, it is not readily apparent which direc-
tion actually requires more power to process the same flow rate. The equa-
tions resulting from the regression analysis, however, show that forward
operation does indeed require more power to process the same flow rate.
The probable cause of this difference was identified by the plant mana-
ger, Ken Cramer, as grate blockage during forward operation caused by shred-
ded refuse adhering to the inclined back of the discharge hopper. A cross-
sectional view of the shredder is presented in Figure 53. While operating
in forward rotation, shredded refuse adheres to the inclined back of the
discharge hopper, particularly in the corners. Eventually enough material
becomes packed into the region labeled A to block the openings between the
grates which effectively reduces the area through which shredded refuse may
be discharged. As a result more material resides in the shredder for the
80
-------
Figure 53. Cross-sectional view of discharge under #1
shredder at Cockeysville
same flow rate (the holdup is increased) and the power required to process
refuse at the same rate is increased as well.
DISCUSSION OF GREAT FALLS VERTICAL HAMMERMILL RESULTS
The vertical hammermill in Great Falls exhibited substantially different
behavior than the rest of the shredders which were tested. As mentioned
earlier, the difference is believed to stem from the fact that the shredding
of refuse is highly non-linear along the length of the mill, while the model
assumes uniform or at least linear shredding along the mill.
To gain a better understanding of the operation of the vertical hammer-
mill, the specific energy on a wet basis, E0 , was plotted against wet
flow rate, rt^, as in Figure 54. Three of the points may be identified as
outlyers, and if these points are neglected and a regression analysis done
on the remaining data, a curve drawn through the remaining points is best
described by the equation
= 89.26 m,
-0.99
w
w
(22)
The correlation coefficient for equation 22 is -0.88 which indicates that 78
percent of the variation in specific energy is explained by the variation in
flow rate. Equation 22 implies that the specific energy is inversely
proportional to the flow rate. Recalling that the definition of specific
energy is net power divided by the flow rate, multiplying the left side of
equation 22 by the flow rate would then predict the net power. Multiplying
the right hand side of equation 22 by the flow rate yields
Pn = 89.26
0.01
(23)
81
-------
In other words, the net power is seen to be nearly constant and only
slightly affected by the flow rate. Such a result is born out by the plot
of net power versus flow rate in Figure 50, where points are congregated
about 90 kw.
The relatively constant net power over a wide range of flow rates at
Great Falls is directly opposed to the results obtained at the rest of the
sites indicating that a vertical hammermill operates on substantially
different principles than horizontal hammermills and vertical ring grinders,
and that it may be impossible to apply the usual model, as described by
equation 18, to any portion of the vertical hammermill grinding zone.
DISCUSSION OF TINTON FALLS VERTICAL RING SHREDDER RESULTS
The vertical ring shredder tested in Tinton Falls exhibited behavior
very similar to that of horizontal hammermills indicating that the size
reduction process may be relatively linear along the axis of the mill. One
notable aspect of the ring grinder's operation was the relative freedom from
large spikes in the power requirements. A typical recording of the power
requirements of a horizontal hammermill shows large power fluctuations as
£
3
?
01
12.5
10.0
7.5
5.0
Z.5
a Outlyers
10 15
Flow rate, ifiw (TWPH)
20
25
Figure 54.
Relationship between flow rate and specific
energy for Great Falls vertical hammermill.
82
-------
material is fed to the mill while the recording of the power for the ring
shredder shows a more uniform power level. Figure 55 shows a comparison
between representative power recordings for the ring shredder (top graph)
and a horizontal hammermill (bottom graph). The vertical scale is the same
in both graphs as is the horizontal scale. The more uniform power level
indicated by the recording for the ring shredder indicates that shock
loading of the mechanical components of the ring shredder will be less than
corresponding shock loads in hammermills.
As indicated by the relatively low value of the leading coefficient in
the power equation for Tinton Falls (shown in Table 15), the power
requirements for the Tinton Falls shredder are relatively low when compared
to the other shredders (omitting the Odessa shredder). Part of the reason
for the low power requirements is explained by the fact that the grinders
inside the shredder were well worn at the time of the tests. The plant
manager, John Gray, indicated that power requirements do temporarily
increase when new grinders are installed and that as the grinders wear, the
power requirements gradually decrease.
COMPARISON OF COCKEYSVILLE AND ODESSA RESULTS
The specific energy required to produce an equivalent particle size was
significantly lower for the Odessa shredder than for any other shredder tes-
ted. In view of the similarity in physical characteristics between the
shredder in Odessa and those at the Cockeysville site, such a result is
highly significant. The differences between the Odessa and Cockeysville
shredders are:
1. Mill length: Odessa is 2.59 meters while Cockeysville is 2.74 meters;
2. Motor size: Odessa has a 373 kw motor while Cockeysville has 746 kw
motors; however, the type of motor (wound rotor) is the same for both
shredders;
3. Grate spacing: Odessa has 24.1 x 36.5 cm openings while Cockeysville
has 20.3 x 35.6 cm openings;
4. Total grate opening: Odessa has 3.97 m2 of grate openings while
Cockeysville has 3.54 m2 of grate openings; and
5. Number of hammers: Odessa has 14 while Cockeysville has 24.
Each of the above factors may contribute to the power requirements of a
shredder, although the prime contributing factor is suspected to be the dif-
ference in number of hammers for the case of Cockeysville and Odessa.
In order to show how the power requirements of the Odessa and Cockeys-
ville shredders differ, the characteristic product size was chosen as a'
basis for comparison. Both the Odessa and the Cockeysville data were exam-
ined for relations between characteristic size and flow rate. Although no
statistically significant relation for the Cockeysville data was found, the
83
-------
Vertical Ring Grinder
R"
Horizontal Hammermil
Figure 55. Representative hammermill and ring shredder power
recordings.
84
-------
Odessa data clearly showed an inverse relationship between characteristic
size and flow rate.
The relation between the wet flow rate, it^, and characteristic size,
Xg, obtained at Odessa is best described (correlation coefficient equal to
-0.86) by the equation
Xn = -0.034 ft + 6.03 (24)
u w
Eauation 24 and the data it describes are plotted in Figure 56.
To compare the Odessa results with the Cockeysville results, an auxil-
iary axis for the Cockeysville flow rates was drawn below the Odessa flow
rate axis in Figure 56, and the Cockeysville data was plotted such that the
average Cockeysville characteristic size fell on the line describing the
Odessa data. The average characteristic size and flow rate found at
Cockeysville were 2.1 cm and 49.5 TWPH, respectively. Examination of the
Odessa rate axis shows that a characteristic size of 2.1 cm would be expec-
ted from the Odessa shredder if the flow rate were 114.3 TWPH.
Assuming a constant moisture content of 20 percent, the net power
required to produce a characteristic size of 2.1 cm would be 124.9 kw and
441.7 kw for Odessa and Cockeysville, respectively. The specific energy
required to produce a characteristic particle size of 2.1 cm is thus 1.1
kwh/Tw for Odessa and 8.9 kwh/T for Cockeysville. If it is assumed that
the Odessa refuse is not substantially different than that at Cockeysville,
then most of the variation may be attributed to one or all of the aforemen-
tioned physical differences between the shredders.
There has been some speculation (1) on the effect of various physical
characteristics of a horizontal hammermill on the operating parameters, but,
as yet, there exists no definitive answer to such speculation. From a
design standpoint, the effect of the physical characteristics of a hammer-
mill on its operation is one of the least understood areas and most in need
of further research.
Due to the fact that significant energy savings are possible, further
research is needed in order to discern the reason for the low energy con-
sumption exhibited by the Odssa shredder. Such research could include vary-
ing the number of hammers, the grate spacing, and the hammer geometry while
measuring energy consumption and flow rate.
SPECIFIC ENERGY COMPARISON OF SINGLE VERSUS MULTIPLE STAGE SIZE REDUCTION OF
MSW
Some of the data collected at the Ames and Cockeysville sites can be
used to compare single and multiple stage size reduction of solid waste.
This comparison is, of course, site specific and due to the nature of the
study, only a limited amount of data is available for comparison. In order
to reach any general conclusions regarding the efficacy of single stage
versus multiple stage size reduction, a rigidly constructed test program is
85
-------
10
9
8
i. ,
« 3
o
A Odessa
A Odessa average
O Cockeysville
^ Cockeysville average
Odessa mw(TwPH) >
Cockeysville
50
100
150
srr
85.2
Figure 56. Comparison between Odessa and Cockeysville
flow rate-size relationships.
needed. The nature of such a program (for instance, requiring changes of
grate openings in each shredder tested) would probably preclude testing at
commercial plants due to interruptions in normal plant operation and
operating procedure.
Bearing the preceding comments in mind, an analysis of the Ames and
Cockeysville data allows a comparison of a specific case of single versus
multiple stage size reduction. The comparison involves examination of gross
and net power consumption (Pg and PN, respectively), mass flow rates of
refuse (mw), and product size (X0). For the purposes of comparison of
processing alternatives, the criterion of equivalent product size from both
single shredding at Cockeysville and secondary shredding at Ames was used.
The data allow the comparison of three such cases for which an average
characteristic product size (X0) of 1.6 cm was calculated, as shown in
Table 16.
In order to compare the power requirements of single versus multiple
stage size reduction, the moisture content (MC) must be specified as well as
the product size (already chosen as 1.6 cm). For purposes of comparison,
the average moisture content was taken as 20.5 percent for a hypothetical
refuse encountered at Ames and Cockeysville. As previously discussed, net
power requirements for size reduction can be expressed as
PN = ArfiwB(l-MC)C
where A, B, and C are constants determined by curve fitting techniques.
86
(25)
-------
Since we have specified MC as 20.5 percent, equation 25 reduces to
PN = AmwB(0.795)C (26)
The constants A, B, and C have been calculated for equation 26 for the Ames
primary and secondary shredders and for the Cockeysville shredder.
For the Ames primary and secondary shredders, the most appropriate equations
for net power are
Pw = 0.619 rti ^'^'(0.795)'^ (27)
N w
and
PN = 0.278 mw1-922(0.795)"2*843 (28)
respectively, where rtiw = 21.6 TWPH since that throughput produces the
required average produce size (X0) of 1.6 cm. For shredder #1 at
Cockeysville, the equation for net power (calculated by combining data for
both forward and reverse directions of rotation) is
0 899
PN = 21.85 m^'^(Q.795r' (29)
wherein mw = 52.2 TWPH since that throughput produces the desired aver-
age product size (X0) of 1.6 cm.
TABLE 16. SHREDDING DATA USED TO DEVELOP COMPARISONS
BETWEEN SINGLE AND MULTIPLE STAGE SIZE REDUCTION
Shredder m X Test
w o
(TPH) (cm) No.
Cockeysville 52.3 1.5 6-F
63.2 1.6 7-R
41.1 1.6 8-R
Average 52.2
Ames Secondary 15.8
22.2
26.9
Average 21 . 6
1.6
1.5
1.6
1.6
1.6
7
6
2
aTest numbers refer to tests in data summaries in Appendix A.
87
-------
The solution of equations 27, 28, and 29 allows calculation of the total
net power required to produce the required product size. The total net
power required at Ames and Cockeysville are 319.0 and 456.8 kw, respec-
tively. The specific energy (on a wet weight basis, PN/I\) for Ames and
Cockeysville are 14.8 and 8.8 kwh/Tw, respectively. Consequently, the
specific energy required to produce the required product size at Ames is 168
percent of the energy requirement at Cockeysville. Table 17 contains a sum-
mary of the net power and specific energy reqired to produce a specified
size of 1.6 cm at the Ames and Cockeysville sites.
TABLE 17. NET POWER AND ENERGY REQUIREMENTS FOR
SINGLE AND MULTIPLE STAGE SIZE REDUCTION
Shredder
MC mw Pn
(*) (T>) (kiJ)
Ames Primary
Ames Secondary
Cockeysville
20.5
20.5
20.5
21.6
21.6
Totals
52.2
123.1
195.9
319.0
456.8
5.7
9.1
14.8
8.8
Calculation and scrutiny of the gross energy requirements(i.e. net spe-
cific energy plus freewheeling energy) for both sites allows establishment
of the overall energy requirement to produce an equivalent product size from
each site. As an example, size reduction of 100 tons of refuse is consi-
dered and the appropriate calculations are carried out, as shown in Table
18. When the freewheeling energy contribution (column G) is added to the
TABLE 18. CALCULATION OF REQUIRED GROSS SPECIFIC ENERGY TO OBTAIN
EQUIVALENT PRODUCT SIZE FOR AMES AND COCKEYSVILLE
(A)
Mass
Flow
Site Rate
(<«„)
(B)
Tonnage
Shredded
-------
net energy for size reduction (column E), the result is the gross energy
required to yield an average product size of 1.6 cm at each site. These
gross energy requirements are 1911 and 1006 kw for size reduction of 100
tons at Ames and Cockeysville, respectively (column H). Division of these
gross energy requirements by 100 tons yields the gross specific energy,
namely 19.1 kwh/Tw for the former site and 10.1 kwh/Tw tor the "latter
site (column I). On the basis of overall energy consumption, Ames would
require almost twice as much energy as Cockeysville to produce an average
product size of 1.6 cm.
Although this result is startling, one should not jump to any hasty con-
clusions with respect to the effectiveness of single versus multiple stage
size reduction. In particular, it must be emphasized that the preceding
comparison is only for a specific case. The effects of other variables upon
single and multiple stage size reduction, for example, throughput, size of
grate openings, etc., cannot be ascertained from these data. Neither can it
be judged whether or not the two shredder stages at Ames are optimized with
respect to energy consumption. It is a fact that at Ames the secondary
shredder uses considerably more energy per unit mass of refuse than the pri-
mary shredder. This fact alone points to the possibility of optimizing the
shredder operation at Ames. Proper optimization of the shredder combination
at Ames could potentially reduce the present energy requirement as developed
in this example such that the overall energy consumption would approach that
estimated for the single shred at Cockeysville.
RESULTS OF THE HAMMER WEAR STUDIES
Since each processing facility had its own particular type of solid
waste and set of operating conditions, the results of the hammer wear tests
are reported here by individual test site. Later, generalizations concern-
ing hammer wear will be presented and discussed.
Appleton (Refer to Table 19)
Test results of hammer wear in the east mill (where the T-l, type B ham-
mers were hardfaced with Stoody 2134) show an average weight loss per hammer
per ton (denoted as WT) of refuse processed equal to 9.44 x 10-* kg.
Wear of a full complement of hammers (W0), i.e. 24 hammers, is 0.023
Test results of hammer wear in the west mill elucidate the wearability
of McKay 48, McKay 50 TiC, Stoody 2134, and Amsco Super 20. These alloys
exhibit values of wn- of 4.84 x 10-4, 3.30 x io-4) 3.30 x 10'4, 3.60
x 10-4, and 2.73 x lO'4 kg/Hammer-Tw, respectively. When these values
are corrected to a full hammer complement of 48 hammers, the W0 values are
0.023, 0.016, 0.017, and 0.013 kg/Tw for McKay 48, McKay 40 TiC, Stoody
2134, and Amsco Super 20, respectively.
Two important results can be gleaned from the data. First, as-deposited
hardness appears to correlate inversely with W0. From Table 19 and Figure
57, W0 can be seen to be least for the hardest hardfacing alloy and great-
est for the softest alloy. The second important result stems from compari-
son of W0 values for Stoody 2134 obtained in the east and west mills. The
89
-------
TABLE 19. APPLETON HAMMER WEAR EXPERIMENTS
Alloy
EAST MILL
Stoody 2134
WEST MILL
McKay 18
McKay 40 TIC
Stoody 2131
Amsco Super 20
Hardness
(Re)
48
38
45
48
5fi
Number of
Hammers
Weighed
12
12
12
12
7?
Test
Tonnage
(Tj
398.32
141.30
141.30
141.30
141.30
Total
Height Loss
(Kg)
_4.51
0.82
n.56
0.61
0.27
'Wg Mt Loss
(w,)
(Kq/Hammer-Tu)
9.44 x 10"4
ft
4.84 x 10 '
3.30 x 10'"
3. en x in'4
2.73 x 10'4
Wear of Full
Hammer Complement
("o>
(Kg/Tw)
0.123
n.023
n.me
0.017
o.ms
j;As-deposi ted hardness per manufacturer's literature
Five harmers failed due to brittle fracture of hanroer at midsection, reason for failure undetermined
two values are significantly different which possibly points out the unknown
effects of refuse composition and operational variables, such as hammer geo-
metry, base material of hammer, rpm, etc. Each of the above was different
for the mills at Appleton (refer to Table 2). Because of the aforementioned
4
\
\
Wear Due
TgJ\brasion
Wear Due
To Chipping
Under Impact
© Appleton West
A Cockeysville =
Q Great Falls 20TPH
10 20 30 40u 50 60 70
Alloy Hardness (Re)
Figure 57. Hanmer wear as a function of alloy hardness
90
-------
variations in refuse composition and operating parameters, comparisons of
wear characteristics of different alloys are best accomplished by simultan-
eous testing, that is, employing the same, or similar methods, used in
securing the Appleton west mill data.
Ames (Refer to Table 20)
After four months of shredding, the manganese hammers in the primary
shredder exhibited an average weight loss per hammer per ton (w-j)of 5.50 x
10~4 kg. Corrected to a full complement of 48 hammers, this degree of
wear (W0) corresponds to 0.026 kg/Tw.
TABLE 20. AMES HAMMER WEAR EXPERIMENTS
Hear of Full
a Number of Test Total Avg Wt Loss Hammer Complement
Hardness Hammers Tonnage Weight Loss (wj) (W0)
AJJ_o_y_ (Re) Weighed (Ty) (Kg) (Kg/Hammer-Tj (Kg/Tw!
Manganese Steel 14 48 14,253.83 376.59 5.50 x 10~4 0.026
aflardness per manufacturer's literature
Cockeysville (Refer to Table 21)
Average weight loss (w-j) of manganese steel hammers installed in
shredder #2 was found to be 4.40 x 10'3 kg/Hammer-Tw. This value cor-
responds to a value of 0.106 kg/Tw for a full complement of 24 hammers.
This value of W0 is significantly higher than the value of 0.026 kg/Tw
that was obtained for manganese steel hammers used in the primary shredder
at Ames. In addition, the average W0 value experienced at Cockeysville
over a one and a half month period in early 1978 was 0.020 kg/Tw. The
high rate of wear measured at Cockeysville (over 500 percent of normal) is
attributed to the use of new hammers in the experiments and as a consequence
of accelerated wear of the sharp corners during initial run-in.
Experimental results for wear of different hardfacing alloys applied to
hammers installed in shredder #1 show values of w-j for Mangjet, Abraso-
weld, McKay 55, and Faceweld 12 to be 2.88 x 10~3, 1.87 x 10'3, 0.89 x
10~3, and 1.57 x 10'3 kg/hammer-Tw, respectively. When corrected to a
full complement of 24 harrmers, the degrees of wear (W0) are 0.069, 0.045,
0.021, and 0.038 kg/Tw for Mangjet, Abrasoweld, McKay 55, and Faceweld 12,
respectively.
91
-------
TABLE 21. COCKEYSVILLE HAMMER WEAR EXPERIMENTS
Wear of Full
Number of Test Total Avg Wt Loss Hammer Complement
Hardness Hammers Tonnage Height Loss (w«) (W )
(Re) Weighed (Tw) (Kg) (Kg/Hammer-Tj (Kg/?.,)
SHREDDER #2
Manganese Steel 14 12 305.36 16.11 4.40 x 10"3 0.106
SHREDDER »1
Mangjet
Abrasoweld
McKay 55
Faceweld 12
21
36
•49
57
6
6
6
6
255.09
255.09
255.09
255.09
4.41
2.86
1.36
2.41
2.88 x 10"3
1.87 x 10~3
0.89 x 10"3
1.57 x 10"3
0.069
0.045
0.021
0.038
a
'As-deposited hardness per measurements at site
As in the case of the Appleton data, a general trend of descending wear
exists for an ascension of as-deposited hardness values of the test alloys
(shredder #1, Table 21 and Figure 57). The lone exception occurs for Face-
weld 12 (the hardest alloy tested at 57 Rc) which exhibits greater wear
(0.038 kg/Tw) than the next hardest alloy, McKay 55 (49 Rc), 0.021
kg/Tw). Reference is made to the fact that visual observation of the weld
deposits after the test run showed that some of the Faceweld deposits had
chipped away from the hammer base metal. The chipping phenomena could be
due to poor weld deposition, incompatability with manganese steel and/or
buildup alloy, and/or unsuitability of this alloy for refuse shredding.
Although the reason for chipping was not resolved quantitatively, the suspi-
cion is that the weld deposits of Faceweld 12 were too hard to absorb the
shock loads that are associated with the size reduction of refuse. Hence,
the welds of Faceweld 12 chipped under impact, whereas those of McKay 55
(which averaged eight points softer than Faceweld 12) were capable of with-
standing the high impacts.
Great Falls (Refer to Table 22)
Results of the hammer wear experiments conducted in the Heil vertical
shredder in Great Falls show the average weight loss per hammer (w-j) to be
0.080, 0.037, and 0.036 kg/Tw for bare 1060 hammers, hammers tipped with
Stoody 2134, and hammers tipped with Amsco Super 20, respectively. Once
again, the results show that the hardest materials wear the least, as gra-
phically depicted in Figure 58.
The results in Table 22 are for the hammers in the bottom nine rows of
the shredder. The hammers in the upper stage are changed infrequently and
consequently were not used in the wear experiments. CRS estimates that the
hammers in the lower nine rows undergo 90 to 95 percent of the total hammer
wear.
92
-------
St oody
2134 "
Amsco
Super
20
Bare
Figure 58. Hammers removed from Great Falls
vertical hammermill . Note how wear
visibly increases from left to rig It
(hammer location within mill found
i n Figure 55) .
93
-------
The pattern of hammer wear showed that the hammers occupying the stages
in the throat of the shredder (the cutting zone) and at the same level as
the top of the discharge opening exhibited the greatest wear (see Figures 58
and 59), typically four to six times the wear of the sweep hammers (those
hammers in the bottom stage which sweep material out of the shredder). Con-
sequently, it may be concluded that most of the size reduction occurs in the
cutting zone of the vertical shredder.
TABLE 22. GREAT FALLS HAMMER WEAR EXPERIMENTS
Hardness3
Alloy (Re)
1060 Manners 11 c
Stoody 2134 41
Amsco Super 20 51
Wear of Full
Number of Test Total Avg Wt Loss Hammer Complement
Manners Tonnage Weight Loss (w.) (W0)
Weighed (T.) (Kg) ( Kg/Hammer-T w) (Kg/T..)
8 178.3 4.40 3.08 x 10"3 0.080b
9 178.3 2.26 1.41 x 10"3 0.037b
9 178.3 2.19 1.37 x 10"3 0.036b
aAs deposited hardness as measured
bottom three stages of shredder only (4 stages total); CRS estimates the hamners in the bottom
three stages undergo 90 to 951 of the total hanrner wear.
Estimated, too soft to measure accurately
Odessa (Refer to Table 23)
Data supplied by the plant supervisor, Dwane Dobbs, were used to cal
culate the wear of hammers used in the Odessa mill. These bare hammers
shredded 5,201 tons of refuse. The resultant average wear per hammer (w
was 2.13 x 10~3 kg/Tw. For a full complement of 14 hammers, the hammer
wear was 0.030 kg/Tw, as shown in Table 23.
TABLE 23. ODESSA HAMMER WEAR EXPERIMENTS
Alloy
Hardness
(Re)
Number of
Hammers
Weighed
Test
Tonnage
> ^
(Kq/Ty/)
Modified Hatfields
Mn Steel Hammers 33
14
5,201
156.7
2.15 x 10
-3
0.030
As deposited hardness as measured
94
-------
}
1
f
9S
r 8A |
1
^•^^^
1 ' 8B ! |
1 7A | | | |
9A
—
las |
I
, i .
1 7B 1
1
$
L 7S i
1
{ \
Lower three rows of breakers
6B
L
58
| J
1 5A
b
i
4i
i
1
4B
1 _
L
1
4A I
r
Cutters
I
)
3S
2S
IS
38
28
IB
3A
2A
1A
\
S = Stoody 2134 SweepS
A = Amsco Super 20
B = Bare
Figure 59. Hammer installation pattern for Great Falls
wear experiments.
Tinton Falls
Data supplied by the plant manager, John Gray, were used to determine
the wear of the ring grinders used within the Carborundum vertical ring
shredder. The ring grinders, which were not hardfaced, shredded 16,353 tons
of refuse. The resultant average wear per hammer (w-j) was 2.13 x 10"3
kg/Tw. Wear for a full complement of sixty ring grinders was 0.027
kg/Tw, as shown in Table 24.
TABLE 24. TINTON FALLS WEAR EXPERIMENTS.
Alloy
Hardness
(Rr)
a)
Number of
Grinders
Weighed
Test Total
Tonnage Weight Loss
Avg Wt Loss
(wi)
Wear of Full
Hammer Complement
(Wo)
Nickel-Manganese
43
60
16,353
445.4
4.5x10
-4
0.027
As measured on new ring grinders
95
-------
COMPARISON OF HAMMER WEAR DATA AMONG SITES
Some general observations can be drawn from comparisons of the wear data
among the sites. These observations are described below and are divided
into the following headings: optimum alloy hardness range, mechanisms of
hammer wear, and normalization of hammer wear data.
Optimum Alloy Hard?ess Range
The greatest rates of wear were sustained by hammers that exhibited low
values of hardness. For the hardfacing alloys that ware tested, as the
hardness of the alloys increased, the degree of wear correspondingly
decreased. The limiting factor at the high range of alloy hardness appears
to be chipping of the welds under high impact loads. The trend of decreased
hammer wear for harder alloys is apparent if one examines the data from the
Appleton west mill, the Cockeysville #1 shredder, and the Great Falls 20 TPH
shredder, as shown in Figure 57.
The optimum range of alloy hardness appears to lie within the range
48 £ Rc <. 56 (shown in Figure 57) with the upper value limited by the
chipping tendency of the welds under impact loading. The exact hardfacing
alloy for a given shredding operation would have to be determined experi-
mentally by testing a group of different hardfacings having as-deposited
hardness values within the optimum range. For reasons that were mentioned
before, testing a number of alloys under identical operating conditions
(e.g. one alloy per hammer row) is the only valid method for comparing wear
of alloys, and consequently this method is recommended for comparison
purposes.
Mechanisms of Hammer Wear
Visual observation of worn hammers after testing showed abrasive wear to
be the dominant mode of wear for alloys with Rc less than about 50 (Figure
57). For these alloys, impact loading appeared to damage at most 10 to 20
percent of the hammer surface. Whether these impact loads actually removed
a significant degree of hammer material or only plastically deformed the
hammer is unknown (particularly since abrasive wear would tend to smooth out
any sharp edges resulting from impact chipping).
Impact chipping on the other hand was the dominant mode of wear for the
hardest alloys tested (Rc greater than about 50) as shown in Figure 57.
Consequently, the brittleness of the deposited alloy and its union with the
base hammer material are the determining factors with regard to maximum wear
of extremely hard alloys. The impacts apparent on the hammers coated with
the softer alloys consequently represent the blows that cause the harder
alloys to fail by means of brittle fracture.
Normalization of Hammer Wear Data
Comparisons of hammer wear data among sites must take into account the
fineness of shred since operating experience has shown that wear increases
as product size of shredded refuse decreases. In addition to size of the
96
-------
product, size of the feed material also may have a bearing on the degree of
hammer wear. It has already been pointed out that the composition of the
refuse at Ames and Cockeysville was different than that processed at the
other four plants. The fact that refuse that was processed at Appleton,
Great Falls, Tinton Falls, and Odessa tended to be more residential in
nature than that processed at either Ames or Cockeysville implies that the
characteristic size of the feed for those plants processing residential
waste should be on the order of 12.7 cm. On the other hand, the commercial
character of the waste encountered at Ames and Cockeysville implies that the
feed size should be larger, on the order of 20.3 cm. These are average val-
ues for raw residential and commercial waste which are based upon previous
experience and several waste composition and sizing studies conducted by CRS.
In order to account for variations of the size of the feed material and
the shredded product size among different sites, a parameter termed the
degree of size reduction, Z0, is introduced. This term is defined as
Z0 = (po - XO)/FO (30)
where F0 and X0 are characteristic feed size and product size,
respectively. .
Values of the degree of size reduction range from a value of zero
corresponding to no size reduction to a maximum limit of 1.0 corresponding
to a product size of zero, or in other words, an infinite amount of size
reduction. The latter limit is, of course, unachievable in actual practice.
Normalization of the wear data collected at Appleton's west mill, Great
Falls' vertical shredder, and Cockeysvilie's shredder #1 using average Z0
values and data from the three curves drawn in Figure 57 (summarized in
Table 25) is demonstrated in Figure 60. The solid lines represent the data
gathered during the wear experiments conducted at Appleton and Cockeys-
ville. Although only two points are present for each curve, the general
trend of increased wear at large values of Z0 can be discerned nonethe-
less. Also apparent is the parametric effect of alloy hardness (Rc),i.e.
for a particular value of Z0 wear is greatest for the softest hardfacing
alloy.
For comparative purposes, the wear data gathered at Great Falls has been
normalized using the curve drawn for Great Falls in Figure 57. The normal-
ized data, which is presented in Table 25, has been added to Figure 60 so
that the hammer wear in a vertical shredder can be compared to that in hori-
zontal shredders. For an equivalent degree of size reduction, the wear data
for the hammers in the vertical mill tends to be slightly above the corres-
ponding values for the horizontal hammermills. Since the data from only one
vertical mill is available, the reason for the slight difference in wear
depicted in Figure 60 cannot be ascertained at this point in time. The dif-
ference may be due to experimental error or an actual difference between the
wearing mechanisms in vertical and horizontal hammermills.
97
-------
TABLE 25. NORMALIZATION OF HAMMER WEAR MEASUREMENTS
Alloy
Shredder Hardness
(RC)
Appleton West Mill
(Horizontal Hammermill)
Cockeysville Shredder 11
(Horizontal Hammermill)
Great Falls
(Vertical 20 TPH
Hammermill )
28
38
48
56
28
38
34
56
28
38
48
56
Average Average
Characteristic Characteristic
Feed Size (Fo) Product Size (Xo)
(cm) (cm)
12.7
12.7
12.7
12.7
20.3
20.3
20.3
20.3
12.7
12.7
12.7
12.7
5.6
5.6
5.6
5.6
2.0
2.0
2.0
2.0
2.4
2.4
2.4
2.4
Average
Degree of Size
l Reduction
(Z0)
0.56
0.56
0.56
0.56
0.90
0.90
0.90
0.90
0.81
0.81
0.81
0.81
Hammer
Wear
(W0)
0.034
0.023
0.016
0.013
0.057
0.043
0.031
0.023
0.056
0.044
0.033
0.025
Figure 60 represents a convenient method for representation of test data
gathered at different sites and serves to allow intelligent comparison of
wear data that may have been collected from shredders where'different types
of solid waste and product sizes are experienced. The general conclusion
that can be drawn from the data in the figure is that hard alloys yield sig-
nificant reduction in hammer wear, for example, on the order of 60 percent
if an alloy with a hardness of 56 Rc is used instead of an alloy with a
hardness of 28 Rc. For an equivalent amount of material worn from the
hammers, this 60 percent reduction in wear corresponds to a running time for
hammers that are coated with the harder alloy that is 250 percent of that
for hammers coated with the softer alloy.
0.08
0.06
0.04
s_
to
HI
i 0.02
Appleton West '-1111
Cockeysville «
Great Falls Vertical
Shredder
0 0.2 0.4 0.6 0.8 1.0
Degree of Si-ze Reduction (Z0)
Figure 60. Hammer wear as a function of alloy
hardness and degree of size reduction.
98
-------
SITE SPECIFIC COMMENTS WITH REGARD TO HAMMER MAINTENANCE
Some general comments can be made about the present hammer maintenance
procedures followed at each plant. Beginning first with the Appleton plant,
consideration could be given to testing several hardfacing alloys in the
range of 48 to 56 Rc (as deposited). Although Stoody 2134 (and therefore
its wire equivalent Stoody 134) exhibited good wear characteristics, the
harder Amsco Super 20 (56 Rc) exhibited 24 percent less wear than Stoody
2134 (0.013 kg/T versus 0.017 kg/T, respectively). For an equivalent amount
of hammer material lost due to wear, Amsco Super 20 will wear approximately
31 percent longer than Stoody 2134. Any hardfacing alloys that exhibit
as-deposited hardness values in the range of 48 to 56 Rc, as well as being
compatible with manganese steel, would be satisfactory for testing purposes.
The McKay electrode that is used for hammer retipping at the Cockeys-
ville facility provides good wear resistance. However, as noted for Apple-
ton, possible improvements in wearing characteristics might be obtained if
several harder alloys (e.g. 50 to 56 Rc) were tested. The estimated upper
limit for reduction in hammer wear is approximately 25 percent.
SUMMARY OF SHREDDER PERFORMANCE
Based upon an evaluation which considers energy consumption and hammer
wear as a consequence of shredding refuse to an equivalent particle size, no
major differences could be discerned among the following shredders:
a. All is Chalmers horizontal hammermill
b. Tracer-Marksman horizontal hammermill
c. American Pulverizer horizontal hammermill
d. Heil vertical hammermill
This observation follows mainly from analysis of the data depicted in
Figures 44 and 60. First hand observation of each of these shredders in
operation and undergoing maintenance provided no major discernible differ-
ence from an operational or maintenance point of .view.
When compared to the four shredders listed previously, two shredders,
namely the Newell horizontal hammermill and the Carborundum vertical
shredder, exhibited low specific energy requirements (kwh/T) for shredding
refuse to an equivalent product size. In the case of the Newell shredder,
it is thought that the low energy requirements are derived mainly from the
fact that only 14 hammers were utilized (as opposed to the 24 to 48 hammers
used in the other horizontal hammermills that were tested) rather than from
an inherently superior shredder design.
99
-------
SECTION 7
COSTS ASSOCIATED WITH REFUSE SIZE REDUCTION
ENERGY COSTS FOR SHREDDING
Energy costs for size reduction can be estimated utilizing equation 15,
which presents energy consumption (wet basis) as a function of nominal
product size. If net specific energy is assumed to be 90 percent of the
gross energy utilized (to account for freewheeling energy), equation 15
becomes
Eo (kwh/Tw} = 39-50 X90~°'81 (cm) <32>
w
If energy costs are estimated at $0.02/kwh, the cost of energy ($E) for size
reduction can be estimated from
$E ($/T) = 0.79 X90"°'81 (cm) (33)
The cost of energy versus nominal product size is shown in Figure 61. As
shown in the figure, energy costs for primary shredding are in the range of
$0.08 to $0.20 per ton. Energy costs rise steeply for nominal product sizes
that are less than 3 cm, denoted as the "fine shred" region in the figure.
COMPARISON OF ALTERNATIVE HAMMER MAINTENANCE PROGRAMS
The hammer maintenance programs used at the six sites can be separated
into two categories: hammer buildup and wear-and-scrap. The first method
involves building up the hammers with buildup and/or hardfacing electrode
after a given refuse tonnage has been shredded. In the second method, as
the name implies, the hammers are worn until they no longer effectively
shred refuse before being scrapped and replaced by new hammers.
By coincidence, of the six sites visited, three sites use the method of
hammer buildup and three sites utilize the wear-and-scrap method. Conversa-
tions' with the plant managers revealed some-interesting information regard-
ing the utility of each type of hammer maintenance program. Since each
plant manager felt that the method used at his plant was the most cost
effective, the issue should be put in its proper perspective.
In the following analysis, cost alone will be considered. Among those
factors thus ignored are variations among sites with respect to operational
procedures, availability of proper equipment for hammer changing and weld-
ing, proper training of maintenance personnel, and the fact that badly worn
100
-------
c
o
100
80
60
O
CJ
UJ
20
Primary
Shred
Six
Test
Sites
1 2 4 6 8 10 20
Nominal Product Size (cm)
Figure 61. Cost of energy associated
with size reduction.
hammers (which are inevitable in the wear-and-scrap method) affect particle
size, energy requirements, and throughput. Neglect of such considerations
does not mean that they are inconsequential, but rather that it is difficult
to assign specific monetary values to them. Perhaps the two most effective
means of minimizing the costs associated with either type of hammer
maintenance program are proper training of maintenance personnel and
availabi-lity of proper maintenance equipment.
Bearing in mind the preceding discussion and assuming a well-run hammer
maintenance program, the two types of programs may be compared.
Three scenarios are examined based upon information provided by the
plant managers and the experience of CRS. All three scenarios are developed
assuming a horizontal hammermill with a nominal rating of 50 TPH and having
a hammer complement of 24 hammers, each weighing 73 kg. All operational
parameters (e.g. size of shredded product, input size and composition, rotor
rpm, etc.) are assumed to be identical for all scenarios. Shred size is
assumed to be approximately 90 percent passing 9 cm. In addition, each
scenario assumes that both properly trained personnel and proper equipment
are available. Proper equipment is considered to include mechanical or
hydraulic pin pullers for pin removal and installation, overhead crane for
hammer removal and installation, and a hydraulically actuated system for
opening up the shredder.
101
-------
These assumptions are necessary in order to compare programs for hammer
maintenance under similar conditions. A site by site comparison was not
used specifically because the different operational procedures used at each
site significantly influence the time, and hence cost, required for hammer
maintenance.
The cost breakdown for three different programs for hammer maintenance
are summarized in Table 26. Case I assumes buildup of hammers once per week
(every five days of operation). Case II assumes daily buildup of hammers
during a second shift. Case III covers the wear-and-scrap alternative.
TABLE 26. TYPICAL COSTS FOR HAMMER MAINTENANCE
Man- $ Per
Cost Items Hours Man-Hour
CASE I: BUILDUP ONCE PER WEEK
Labor
remove and install 12 7
welder3 40 11
Materials ,
hardfacing .
hammers (2 sets)D>c
TOTALS
CASE II: DAILY BUILDUP
Labor .
welder" 9 13
Materials .
hardfacing
hammers (1 set)c'J
TOTALS
CASE III: WEAR AND SCRAP
Labor
remove and install 12 7
Materials .
hammers (1 set)*
Salvage credit1 •m
TOTALS
Total
$
84
440
126
3,538
117
35
1,769
84
3,154
(235)
Tons
J
1,800°.
l,800d
j
l,800d
104.0006
400
400
104,000
7,200
7,200
7,200
Unit Cost
$ Per Ton
0.047
0.244
0.070
0.034
0.395
0.292
0.088
0.017
O97
0.012
0.438
(0.033)
6T4TT
Annual
Cost9
$
4,900
25,400
7,300
3,500
41,100
30,400
9,100
1,800
41,300
1,200
45,600
(3.400)
43,400
Percent
of
Cost
-11.9
61.8
17.8
8.5
100.0
73.6
22.0
4.4
T6O
2.6
97.4
TooTo
Welder for re-tipping, regular time charge, 40 hours per set of 24 hammers
bone set = 24 hammers x 73 kg x $1.80/kg = $3,154; two sets needed, one for install-
ation, one for retipping
cAmortized 2 years 0 8% per annum, capital recovery factor = 0.5608
Tonnage per hammer change
eAnnual tonnage: 50 TPH x 8 hr/day x 260 days/year
f!800 T x 0.020 kg/T x $3.50/kg = $126
9Rounded dollars
Welder for re-tipping, overtime charge, 9 hours to re-tip 24 hammers in place
?50 TPD x 8 hr/day x 0.020 kg/T x $4.40/kg = $35
Jone set = 24 hammers x 73 kg x S1.80/kg = 3,154; one set only, hammers welded in place
kone set = 24 hammers x 73 kg x 51.80/kg = $3,154; worn and scrapped
'Scrap value: $150 per metric ton = $0.15/kg
"Credit = (1,752 kg/set - (7.200T x 0.026 kg/T))($0.15/kg) = $235 per set
102
-------
Major items of cost are explained in the footnotes listed in Table 26.
Some of the significant assumptions used in developing the costs include:
1. Hammer wear of 0.020 kg/Tw and 0.026 kg/Tw for hardfaced hammers
utilized in Cases I and II and bare hammers utilized in Case III, res-
pectively. As indicated previously, wear for bare hammers is greater
than wear for properly hardfaced hammers.
2. Labor costs for maintenance personnel and welders are best estimates
based upon current maintenance practice and CRS experience.
3. Hammer life for Case III (wear-and-scrap) is 7,200 tons, which cor-
responds to wearing away approximately 10 percent of each hammer.
4. Hammer life for Cases I and II is estimated to be two years. Despite
buildup and hardfacing, hammers eventually wear out (e.g. hammer
thickness decreases, pin holes elongate, etc.). Two years of life
for hammers may be a liberal estimate.
5. Two sets of hammers are required for Case I, one set is built up
while the other is run in the shredder.
6. Only one set of hammers is required for Case II since these hammers
are rebuilt during a second shift.
7. The credit for scrapped hammers in Case III is assumed to be S150/T.
The unit cost ($/T shredded) and the annual cost for hammer maintenance
for all the cases are summarized in Table 27. Under the assumptions for
these cases, the unit costs and annual costs are identical for all practical
purposes. The unit cost is found to be in each case approximately $0.40/T,
while the annual cost is approximately $42,000 based upon 104,000 TRY. The
unit cost of S0.40/T is in the range of $0.25 to $0.75 per ton often quoted
by shredder manufacturers for hammer maintenance allowance.
TABLE 27. COST SUMMARY FOR HAMMER MAINTENANCE PROGRAMS
CASE
I
II
III
DESCRIPTION
Build-up once per week
Daily build-up
Wear-and-scrap
1978
UNIT COST
($/Ton)
0.395
0.397
0.417
ANNUAL COST
(1978 $)
41,100
41,300
43,400
103
-------
In Cases I and II the majority of the cost is a result of the actual
buildup and/or hardfacing operation (refer to "Percent of Cost" column in
Table 26). In Case III practically all of the total cost is a consequence
of purchasing replacement hammers.
The similarity in cost for the three alternative programs for hammer
maintenance may explain the fact that the opinions of the plant managers
were split evenly on the buildup versus wear-and-scrap question. Since
buildup and wear-and-scrap alternatives are practically identical from the
standpoint of cost, the buildup alternative should be the preferred method
for hammer maintenance. The rationale for this judgment stems from the fact
that particle size, throughput capacity, and energy consumption will remain
relatively constant and predictable if a regular schedule for rebuilding
hammers is followed.
SUMMARY OF COSTS
Based upon previously developed costs for actual size reduction of ref-
use and hammer maintenance, energy and hammer maintenance costs can be com-
pared for the case of a nominal product size of 9 cm, which is characteris-
tic of primary shredding. The energy cost for size reduction for such a
case would be approximately S0.19/T, while hammer maintenance costs would
run approximately S0.40/T. Consequently, hammer maintenance cost is approx-
imately twice the cost of the energy required for size reduction, based upon
production of a nominal product size of 9 cm.
104
-------
REFERENCES
1. Shiflett, G. R. A Model for the Swing Hammermill Size Reduction of
Residential Refuse. Ph.D. Thesis, University of California at
Berkeley, Berkeley, California, 1978.
2. Trezek, G. J. and Savage, G. M. Size Reduction in Solid Waste
Management. EPA-600/2-77-131, U.S. Environmental Protection Agency,
Cincinnati, Ohio, 1977.
3. Trezek, G. J. and Savage, G.M. Size Reduction in Solid Waste
Processing, Progress Report 1973-1976. U.S. Environmental Protection
Agency, Cincinnati, Ohio.
105
-------
APPENDIX
The following summaries of data present test data from each shredder in tabular form. The summaries are
arranged by site, and tabulated data within sites is organized by each shredder that was evaluated.
Shredder Performance Data Summary
Site: Appleton - East mill
Date: October 2-10, 1977
Test
1
2
3
4
5
6
7
8
9
10
11
*w
(TWPH)
18.3
33.4
29.2
39.8
15.4
47.5
39.1
30.4
3.5
18.0
24.5
*d
(TdPH)
13.7
17.5
-
23.5
12.7
34.0
22.2
18.0
2.1
12.1
15.0
MC
(X)
24.9
47.5
-
40.9
17.4
28.4
43.1
40.9
41.1
32.7
38.8
H
(kg)
134.5
-
-
39.5
43.3
57.4
52.7
49.3
6.4
51.4
69.9
PG
(Kw)
200.0
100.0
187.5
214.3
90.9
125.0
125.0
117.6
54.8
100.0
102.6
PFW
(Kw)
47.0
47.0
47.0
47.0
47.0
47.0
47.0
47.0
47.0
47,0
47.0
PN
(Kw)
153.0
53.0
140.5
167.3
43.9
78.0
78.0
70.6
7.8
53.0
55.6
E°w
(Kwh/Tw)
8.4
1.6
4.8
4.2
2.9
1.6
2.0
2.3
2.2
2.9
2.3
Eo
(Kwh/Td)
11.2
3.0
-
7.1
3,5
2.3
3.5
3.9
3,7
4.4
3.7
Xo
(cm)
2.2
3.6
-
5.5
3.1
3.4
3.5
4.5
5.5
3.9
3.2
X90
(cm)
8.7
8.9
-
12.0
10.1
10.5
9.7
9.7
10.5
9.8
8.6
-------
Shredder Performance Data Summary
Site: Appleton - East mill
Date: October 2-10. 1977
Test
1
2
3
Avg
\
(TWPH)
-
-
-
-
*d
(TdPH)
-
-
-
-
MC
(%)
-
-
.
-
H
(kg)
-
-
-
-
PG
(Kw)
-
-
-
-
PFW
(Kw)
50.6
42.8
47.6
47.0
PN
(Kw)
-
-
-
-
s
(Kwh/Tw)
-
-
-
-
Eo
(Kwh/Td)
-
-
-
-
xo
(era)
-
-
-
-
X90
(cm)
-
-
-
-
-------
Shredder Performance Data Summary
Site: Appleton - West mill
Date: October 2-10, 1977
Test
1
2
3
*w
(T/H)
13.9
30.1
10.5
*d
(TdPH)
9.9
16.5
8.7
MC
(%)
28.5
45.1
16.7
H
(kg)
49.9
81.2
16.1
PG
(Kw)
80.0
105.3
111.1
PFW
(Kw)
41.0
41.0
41.0
PN
(Kw)
39.0
64.3
70.1
\
(Kwh/Tw)
2.8
2.1
6.7
Eo
(Kwh/Td)
3.9
3.9
8.1
Xo
(cm)
6.4
4.4
6.1
X90
(cm)
12.8
10.2
10.5
o
00
-------
Shredder Performance Data Summary
Site: Appleton - West mill
Date: October 2-10, 1977
Test
1
2
3
Avg
*w
(T/H)
-
-
-
-
*d
(TdPH)
-
-
-
-
MC
(*)
-
-
-
-
H
(kg)
-
-
-
-
PG
(Kw)
-
-
-
-
\
PFW
(Kw)
41.7
38.5
42.8
41.0
PN
(Kw)
-
-
-
-
(Kwh?Tw)
-
-
-
-
Eo
(Kwh/Td)
-
-
-
-
"o
(on)
-
-
-
-
X90
(cm)
-
-
-
-
o
vo
-------
Shredder Performance Data Summary
Site: Ames - Primary
Date: November 25 - December 6, 1977
Test
1
2
3
4
5
6
7
8
9
10*
*w
(TWPH)
18.7
15.9
25.0
17.7
35.1
10.6
21.2
13.5
10.6
27.9
*d
(TdPH)
15.2
13.8
21.4
13.4
28.6
8.3
17.4
12.0
9.9
24.6
MC
(*)
18.5
13.2
14.5
24.1
18.4
22.0
17.9
11.4
6.9
12.0
H
(kg)
-
-
-
-
-
-
-
-
-
-
PG
(Kw)
266.7
156.9
166.7
96.6
261.4
111.1
125.8
130.7
98.0
95.2
PFW
(Kw)
53.2
53.2
53.2
53.2
53.2
53.2
53.2
53.2
53.2
53.2
PN
(Kw)
213.5
103.7
113.5
43.4
208.2
57.9
72.6
77.5
44.8
42.0
(Kwh7lw)
11.4
6,5
4.5
2.5
5.9
5.5
3.4
5.7
4.2
1/5
Eo
(Kwh/Td)
14.0
7.5
5.3
3.2
7.3
7.0
4.2
6.5
4.5
1.7
Xo
(cm)
3.81
4.70
5.08
6.35
4.70
5.08
3.68
6.60
5.33
0.86
X90
(cm)
9.65
14.22
8.89
11.94
10.92
10.67
21.08
10.67
10.67
8.64
*Test #10 was mainly dirt, etc. and was judged as unreliable data.
-------
Shredder Performance Data Summary
Site: Ames - Primary
Date: November 25 - December 6, 1977
Test
1
2
3
4
5
6
Avg
*«
(TWPH)
-
-
-
-
-
-
-
*d
(TdPH)
-
-
_
-
-
-
-
MC
(*)
-
-
-
-
-
-
-
H
(kg)
-
-
-
-
-
-
-
P6
(Kw)
-
-
-
-
-
-
™ \
PFW
(Kw)
56.8
51.1
52.4
57.4
50.7
50.6
53.2
PN
(Kw)
-
-
-
-
-
-
-
E°w
(Kwh/Tw)
-
_
-
_
-
-
-
Eo
(Kwh/Trf)
-
-
-
-
-
-
-
,
xo
(cm)
_
_
_
_
_
-
-
X90
(cm)
-
_
_
_
_
_
-
-------
Shredder Performance Data Summary
Site: Ames - Secondary
Date: November 25 - December 6, 1977
Test
1
2
3
4
5
6
7
8
9
10
*w
(TWPH)
24.8
26,9
22.4
20.3
7.0
22.2
15.8
17.9
24.2
27.2
^d
(TdPH)
18.7
21.1
14.1
16.5
5.7
17.6
12.8
13.6
20.3
22.2
MC
(%}
24.4
21.4
37.0
18.7
18.5
20.8
19.2
24.1
16.1
18.2
H
(kg)
-
-
-
-
-
-
-
-
-
-
PG
(Kw)
350.9
392.2
416.7
246.9
56.0
256.4
256.4
185.2
202.0
215.1
PFW
(Kw)
40.5
40.5
40.5
40.5
40.5
40.5
40.5
40.5
40.5
40.5
PN
(Kw)
310.4
351.7
376.2
206.4
15.5
215.9
215.9
144.7
161.5
174.6
s
(Kwh/Tj
12.5
13.1
16.8
10.2
2.2
9.7
13.7
8.1
6.7
6.4
Eo
(Kwh/Td)
16.6
16.7
26.7
12.5
2.7
12.3
16.9
10.6
8.0
7.9
Xo
(cm)
1.27
1.55
0.97
1.35
1.65
1.57
1.52
1.12
1.30
1.24
X90
(cm)
3.05
4.52
3.20
3.05
4.52
3.43
;3,43
2.97
3.10
3.12
ro
-------
Shredder Performance Data Summary
Site: Ames - Secondary
Date: November 25 - December 6, 1977
Test
1
2
3
4
5
6
Avg
*«
(T/H)
-
-
-
-
-
-
-
^d
(T/H)
-
-
-
-
-
-
-
MC
(X)
-
-
-
-
-
-
-
H
(kg)
-
-
-
-
-
-
-
PG
(Kw)
_
-
-
-
-
-
-
PFW
(Kw)
34.2
46.7
43.2
42.5
32.8
43.5
40.5
PN
(KW)
_
-
-
-
-
-
-
Eo,
(Kwh/Tw)
_
-
-
-
-
-
-
Eo
(Kwh/Td)
«.
_
-
-
-
-
-
Xo
(cm)
_
-
-
-
-
-
-
X90
(cm)
_
_
-
-
-
-
-
-------
Shredder Performance'Data Summary
Site: Teledyne #1 (F = forward, R = reverse)
Date: February 18-24, 1978
Test
1-F
2-F
3-F
4-F
5-F
'6-F
7-R
8-R
9-R
10-R
11-R
12-R
13-F
14-F
15-F
16-F
\
(T/H)
22.6
41.1
45.4
48.0
26.4
52.3
63.2
41.1
56.5
L 62.8
39.4
95.7
51.5
53.3
56.6
82.1
*d
(T/H)
18.8
35.4
35.3
41.8
22.4
43.7
53.0
35.2
48.2
40.7
34.6
81.9
-
-
-
-
MC
(X)
16.8
13.9
22.2
13.0
15.2
16.4
16.1
14.4
14.7
35.2
12.3
14.4
-
-
_
-
H
(kg)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
PG
(Kw)
277.8
625.0
416.7
625.0
434.8
512.8
666.7
476.2
- 714.3
416.7
384.6
1025.6
701.8
952.4
952.4
833.3
PFW
(Kw)
64.3
64.3
64.3
64.3
64.3
64.3
67.8
67.8
67.8
67.8
67.8
67.8
64.3
64.3
64.3
64.3
PN
(Kw)
213.5
560.7
352.4
560.7
370.5
448.5
598.9
408.4
646.5
348.9
316.8
957.8
637.5
888.1
888.1
769.0
s
(Kwh7Tw)
9.4
13.6
7.8
11.7
14.0
8.6
9.5
9.9
11.4
5.6
8.0
10.0
12.4
16.7
15.7
9.4
Eo
(Kwh/Td)
11.4
15.8
10.0
13.4
16.5
10.3
11.3
11.6
13.4
8.6
9.2
11.7
-
-
-
-
Xo
(cm)
2.3
1.9
2.3
3.1
2.0
1.5
1.6
1.6
2.2
2.3
2.3
2.1
-
-
-
-
X90
(cm)
7.8
6.3
6.1
8.3
4.9
4.9
4.3
5.3
6.4
7.4
7.0
6.2
-
-
-
-
-------
Shredder Performance Data Summary
Site: Teledyne #1 Forward
Date: February 18-24, 1978
Test
1
2
•3
4
5
6
Avg
*.
(TWPH)
-
-
-
-
-
-
-
*d
(TdPH)
-
-
-
-
-
-
-
MC
(X)
-
-
-
_
-
-
-
H
(kg)
-
-
-
-
-
-
-
PG
(Kw)
-
-
-
-
-
-
-
PFW
(Kw)
65.1
64.6
64.6
65.9
63.1
62.2
64.3
PN
(Kw)
-
-
-
-
-
-
-
s
(Kwh/Tw)
-
-
-
-
-
-
-
Eo
(Kwh/Td)
-
-
-
-
-
-
-
"o
(cm)
-
-
-
-
-
-
-
X90
(cm)
-
-
-
-
-
-
-
-------
Shredder Performance Data Summary
Site: Teledyne #1 Reverse
Date: February 18-24, 1978
Test
1
2
3
4
Avg
\
(TWPH)
-
-
-
-
-
*d
(TdPH)
-
-
-
-
-
MC
(*)
-
-
-
-
-
H
(kg)
-
-
-
-
-
PG
(Kw)
-
-
-
-
-
PFW
(Kw)
69.4
69.4
68.3
64.1
67.8
PN
(Kw)
_
-
-
-
-
Eow
(Kwh?Tw)
-
-
-
-
-
Ec
(Kwh/Td)
-
-
-
-
-
"o
(cm)
-
-
-
-
-
X90
(cm)
-
-
_
_
-
-------
Shredder Performance Data Summary
Site: Great Falls 20 TPH Vertical Hammermill
Date: October 7-13, 1978
Test
1
2
3
4
5
6
7
8
9
10
11
12
*w
(TWPH)
12.3
17.1
19.4
10.0
16.5
11.9
13.9
16.0
13.3
19.0
14.3
13.3
*d
(TdPH)
10.8
14.7
13.4
8.1
10.8
10.1
11.2
11.1
10.7
15.3
10.7
11.4
MC
(*)
12.4
14.4
29.2
18.9
34.5
15.8
19.7
30.7
19.2
19.5
24.8
14.2
H
(kg)
_
_
•V
-
-
-
_
-
_
-
_
-
PG
(Kw)
120
120
114.3
120
120
100
100
100
120
150
150
54.5
PFW
(Kw)
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
PN
(Kw)
98.3
98.3
92.6
98.3
98.3
78.3
78.3
78.3
98.3
128.3
128.3
32.8
%
(Kwh?Tw)
7.6
5.7
4.8
9.9
6.0
6.6
5,6
4.9
7.4
6.7
9.0
2.5
Eo
(Kwh/Td)
9.1
6.7
6.9
12.1
9.1
7.8
7.0
7.1
9.2
8.4
12.0
2.9
Xo
(cm)
0,9
1.4
1.5
3.0
2.3
3.4
2.3
3.6
2.5
2.7
2.2
2.4
X90
(cm)
3.2
6.3
9.3
5.1
4.1
6.7
4.1
7.5
5.2
5.4
4.9
5.6
-------
Shredder Performance Data Summary
Date:
Great Falls 20 TPH Vertical Hammermill
October 7-14, 1978
Test
1
2
3
4
Ave,
*w
(TWPH)
_
-
-
—
-
*d
(TdPH)
—
-
-
_
_
MC
(*)
^
-
-
^
-
H
(kg)
_
-
_
,
_
PG
(K«)
_
_
—
PFW
(Kw)
23.1
21.3
21,3
?1 9
21.7
PN
(Kw)
^
f
-
-
-
Es-
Uwh/Tw)
,
-
r-
-
-
Eo
«wh/Td)
^
-
_
-
-
Xo
(cm)
_
-
_
-
-
X90
(cm)
-
_
-
-
00
-------
Shredder Performance Data Summary
Site: Tinton Falls Vertical Ring Shredder #2
Date: October 21-27 , 1978
Test
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
*w
(TWPH)
83.0
34.9
91.9
34.3
76.6
62.7
61.5
36.1
82.3
78.3
99.5
70.4
67.7
37.1
20.7
36.6
^d
(TdPH)
—
_
-
MC
(*)
^
—.
-
20.5
17.8
25.4
19.8
20.9
21.0
23.1
17.2
22.0
17.4
16.9
7.9
H
(kg)
—
_
-
-
~
*•
^
_
_
^
-
_
-
.
P6
(Kw)
240.0
120.0
240.. 0
120.0
218.2
171.4
133.3
109.1
171.4
320.0
200.0
240.0
171.4
126.3
100.0
120.0
PFW
(Kw)
38.5
38.5
38.5
38.5
38.5
38.5
38.5
38.5
38.5
38.5
38.5
38.5
38,5
38.5
38.5
38.5
PN
(Kw)
?(V| 5
«1 q
?ni.R
81.5
17Q.7
132.9
94.8
70,6
132.4
281.5
161.5
201.5
132.9
87.8
61.5
81.5
Eow
(Kwh?Tw)
9 4
9 1
9 9
2.4
2.4
2.1
1.5
2.0
1.6
3.6
1.6
2.9
2,0
2.4
3.0
2.2
Eo
(Kwh/Td)
Xo
(cm)
-
-
-
-
3.7
4.6
3.7
3,5
4.4
2.4
2.9
3.1
3.5
4.1
5.0
4.8
X90
(cm)
-
-
-
10.1
13.7
10.7
8,9
10.4
6.9
7.0
8.6
9.4
9.3
10.3
10.4
-------
Shredder Performance Data Summary
Site: Tinton Falls Vertical Ring Shredder #2
Date: October 21-27, 1978
Test
1
2
3
Ave.
*w
-
-
—
_
*d
(TdPH)
-
-
w
_
MC
(*)
-
-
_
_
H
(kg)
-
-
.
_
P6
(K»)
-
-
«
PFW
(Kw)
40.0
36.8
.?R 7
38.5
PN
(Kw)
_
_
^
\
(Kwh?Tw)
-
_
—
Eo
(K«h/Td)
—
_
_
xo
(cm)
_
_
_
X90
(cm)
_
—
ro
o
-------
Shredder Performance Data Sunroary
Site: Odessa
Date: November 25-December 2, 1978
Test
i
2
3
4
5
6
7
8
9
10
11
12
*w
(TWPH)
i?7 a.
83.2
67.6
76.1
81.9
124.5
13.3
96.9
98.2
68.3
89.1
57.9
*d
(TdPH)
QQ.fi
65.1
53.1
59.2
67.6
108.3
11.9
74.8
79.3
57.9
70.7
53,6
MC
(%)
??.n
21.7
21.4
22.2
17.4
13.0
10.7
22.8
19.2
15.3
20.7
7.5
H
(kg)
-
-
_
-
-
-
-
-
_
-
-
PG
(Kw)
991 .7
263.7
182.7
250.2
264.6
147.6
106.2
178.2
170.1
162.9
297.0
108.9
PFW
(Kw)
im.7
101.7
101.7
101.7
101.7
101.7
^ 101.7
101.7
101.7
101.7
101.7
101.7
PN
(Kw)
i?fi.n
162.0
81.0
148.5
162.9
45.9
4.5
76.5
68.4
61.2
195.3
7.2
\
(Kwh?Tw)
1.0
1.9
1.2
2.0
2.0
0.4
0.3
0.8
0.7
0.9
2.2
0.1
Eo
(Kwh/Td)
1.3
2.5
1.5
2.5
2.4
0.4
0.4
1.0
0.9
1.1
2.8
0.1
Xo
(cm)
1.2
2.8
3.9
2.5
2.9
2.5
5.5
2.2
3.6
4.5
2.4
4.5
X90
(cm)
5.0
6.9
9.9
6.2
8.0
20.0
12.4
6.9
9.2
10.2
6.7
9.2
ro
-------
Shredder Performance Data Summary
Site: Odessa
Date: November 25- December 2, 1978
Test
1
2
3
4
5
6
7
Ave,
*w
(TWPH)
_
_
_
-
-
-
-
_
*d
(TdPH)
_
«.,
_
-
-
-
-
_
MC
(*)
_
„
_
-
-
-
_
.
H
(kg)
_
T
,.
.
-
-
_
_
PG
(Kw)
_
_
_
-
—
PFH
(Kw)
Q7 4
Qq 3
99.3
98.7
104.9
103.3
108.7
101.7
PN
(Kw)
-
^
^_
_
-
-
^
s.
(Kwh/T^)
-
_
T-
—
_
—
_
Eo
(Kwh/Td)
_
^
^
«
_
_
*-
Xo
(cm)
_
_
—
_
-
-
—
X90
(cm)
^
_
mm
_
„
_
_
no
ro
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1. REPORT NO.
EPA-600/2-80-007C
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
PROCESSING EQUIPMENT FOR RESOURCE RECOVERY SYSTEMS
Volume III. Field Test Evaluation of Shredders
5. REPORT DATE
July 1980 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
George M. Savage and Geoffrey R. Shiflett
8. PERFORMING ORGANIZATION REPORT NO.
4426-D (Project Number)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Cal Recovery Systems, Inc.*
160 Broadway, Suited 200
Richmond, California 94804
10. PROGRAM ELEMENT NO. 1N£624E624DWF
COS Wastes as Fuels Task 5.1
11. CONTRACT/GRANT NO.
Contract No. 68-03-2589
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory—Gin. ,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final Report. - Vol. Ill of III
14. SPONSORING AGENCY CODE
EPA/600/14
15.SUPPLEMENTARY NOTESProject off-jcer; Donald A. Oberacker 513/684-7881
*Conducted under subcontract with Midwest Research Institute, Kansas City, Missouri
See also Volume I (EPA-600/2-80-007a) and Volume II (EPA-600/2-80-007b)
16. ABSTRACT
This report presents the results of a program to test and evaluate large-scale
shredders used for the size reduction of solid waste. In all, tests were conducted
on seven horizontal hammermills, one vertical hammermill, and one vertical ring
shredder at six commercial sites (Appleton, Wisconsin; Ames, Iowa; Cockeysville,
Maryland; Great Falls, Montana; Tinton Falls, New Jersey; and Odessa, Texas). Both
two stage size reduction (Ames) and single stage size reduction were studied as
part of this work. Evaluation and interpretation of the data have resulted in the
development of analytical relationships among the comminution parameters and the
establishment of levels of performance with respect to energy consumption and hammer
wear associated with size reduction of solid waste.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATi Field/Group
Shredders
Solid waste
Size reduction
Hammermill
Energy
Wear
Research
Refuse disposal
Solid waste
Waste as Energy
13B
68
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
unclassified
21. NO. OF PAGES
137
20. SECURITY CLASS (Thispage)
unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
123
U.S. GOVERNMENT PRINTING OFFICE: 1980--657-165/0038
------- |