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

-------