PERFORMANCE CHARACTERISTICS OF A VERTICAL

HAMMERMILL SHREDDER

P. AARNE VESILIND Department of Civil Engineering

Duke University

Durham, North Carolina

ALAN E. RIMER MICA

Durham, North Carolina

WILLIAM A. WORRELL Brown and Caldwell, Engineers

Atlanta, Georgia

ABSTRACT

This paper reviews the results of an acceptance test performed on a vertical shaft hammerrnill

shredder installed at the Pompano Beach Solid Waste Reduction Facility, Pompano Beach, Florida. It was found that the shredder met the required

acceptance condition, producing a refuse of at least 70 percent less than 3 in. (76 mm). The power requirements were low and the hammer wear was found to be linear with the amount of refuse processed.

DESCRIPTION OF PLANT AND PROCESS

The Pompano Beach Solid Waste processing facility consists of a shredder and landfill system, as well as an experimental refuse to methane facility (REFCOM). The overall processing facility layout is shown in Fig. 1.

The refuse trucks enter the solid waste reduc­tion center and dump refuse on the tipping floor. The deposited refuse is stockpiled (the height may reach 20 ft (6 m) by a rubber-tired front-end loader until the refuse can be pushed onto the variable speed horizontal pit conveyor, which takes it to a variable speed inclined conveyor, which in turn feeds a 1000 hp (746 kW) motor vertical shaft hammer mill shredder (see Fig. 2). The shredded refuse is deposited onto a covered

5 ft (1.5 m) wide, troughed rubber belt conveyor and transpor.ted to the belt type magnetic separa­tor which removes the ferrous from the waste

199

stream. The nonferrous refuse (or all of the shred­ded refuse if the ferrous magnet is not operating) continues down the conveyor system and is either diverted to the methane production facility or the landfillioadout area. At the landfillioadout area, the conveyor dumps the shredded refuse through a bifurcated chute into the transfer trailers which are then driven the short distance to the landfill and dumped.

After an initial shakedown period, an accept­ance test was conducted. The objectives of this test were:

1. To determine whether the shredder satisfied the specification that 70 percent (by dry weight) of the shredded refuse has a nominal size of less than 3 in. (76 mm).

2. To determine whether the solid waste shred­ding system would operate at a throughput of 60 tons/hr (54 t/h) for twelve consecutive days with an average minimum daily throughput of 850 tons (770 t).

In addition to these requirements, the following parameters were measured: 1. particle size distri­bution, wet and dry; 2. power consumption; 3. moisture content of refuse; and 4. hammer wear.

SAMPLING PROCEDURE AND SAMPLE ANALYSIS

The determination of the proper location and procedure for obtaining a sample of shredded re­fuse is dependent on several criteria. First, the most

FROM REFCOM

LANDFILL LOADOUT

R MAGN

-SHREDDE

- INCLINED

�- HOR IZONT L PIT

TO LANDFILL

. FIG.1 SITE PLAN FOR THE POMPANO BEACH SOLID WASTE REDUCTION FACILITY

200

DRIVE :--___ � MOTOR

o o

o o

OISCHARGE PORT

o

o

FIG. 2A

o

o

FIG. 28

MSW ENTRY PORT

HAMMER

FIG.2 THE HEIL 928 SHREDDER

important is that any sample obtained has' to ac­curately represent the output of the shredder dur­ing steady-state operation. In addition, the sampl­ing site and procedure must not interfere with the normal operation of the facility, and finally, the sampling site should be safe and conveniently located and the procedure as uncomplicated as possible.

201

SAMPLE COLL ECTION

Based on the above constraints, it was decided that the best sampling site was the transfer point

at the end of the conveyor exiting the ferrous separation building. Access to the conveyor head pulley area was gained through doors on either side of the conveyor (see Fig. 3). A canvas "stretcher" was placed through the doors and when a sample was required the stretcher was quickly opened into the freely falling waste

1

\

FIG.3 SAMPLING SITE AND THE COLLECTION

CANVAS "STRETCHER"

stream, a sample collected, the two stretcher poles quickly closed, and the canvas stretcher containing the collected sample pulled out. Not only did this provide a representative sample under steady-state operating conditions, but it also captur­ed the whole waste stream during the sampling, thus assuring that the sample was truly representa­tive of the distribution of the shredded product across the conveyor belt. The contents of the stretcher, generally 5 ft3 (0.14 m3) of material were then placed into a 50 gal (190 1) can and carried to the storage and sieving bUilding.

At the storage area the sample was either quar­tered or halved, depending on the amount of sample collected. The approximately 3 ft3 (0. l m3) sample was then wet sieved through 8, 4,3, and 2 in. (203, 102,76 and 51 mm) sieves, weighed, divided into four % ft 3 (0.02 m3) pans and placed in a 217 F (103 C) oven to dry. After approximately 10 hr, the oven-dried samples were weighed on a single arm balance accurate to one

FIG.4 SI EVES USED FOR ANALYZING COARSE

SHREDDER REFUSE

202

tenth of a gram and then placed back in the oven. The weighing was repeated about every 45 min un­til the weight of the samples stabilized, which was considered to be the point at which all free water had evaporated, and was therefore the dry weight. The dry and wet weights were used to calculate the moisture content. The dry samples were again sieved through the same sieves and weighed. Finally the fines from the dry sample sieving were boxed and shipped to Duke University to be sieved through a set of standard engineering sieves. Figure 4 shows the four sieves used in analyzing coarse materials, and one of the tests in progress.

The sieving was done by shaking the sieve vertically and horizontally until by visual inspec­tion, no further particles would fall through. This method turned out to be simple and reproducible, and corraborates the results of other investiga­tors [1 ,2] .

POW ER CO NSU MPTION

The power consumption of the shredder was determined by a watt-hour meter. Due to the placement of the meter, the power was recorded only when the shredder ran in the forward direc­tion (every other day). The variability of the amperage was recorded with an amp meter which produced a continuous strip chart at a speed of 3 in. (76 mm) per hour.

HA M M ER W EAR

The hammer wear was determined by weighing all hammers and arm assemblies on a platform scale accurate to lIb (0.45 kg), at the begin-ning and end of the test period. One hammer from each level was also weighed after 3 days of testing and again after 10 days of testing.

THRO UGHPUT

Three different types of throughput were cal­culated and used in this study:

1. Shredder throughput = tons of refuse shred­ded divided by the actual shredding time.

2. System throughput = tons of refuse shred­ded divided by the sum of the actual shredding time and any down time due to a system failure.

3. Total plant throughput = tons of refuse shredded divided by the sum of the actual shred­ding time, the system down time, and any other interruptions including break and lunch time.

c: 0 ,

I-� Q) c: . -

LL I c: Q) u � Q)

Cl..

100�----------------------------� 80 t-60t-

40t-

20 t- Sample 12

10 � __________ � __________ � ____________ � __ �

100r---------�----------��� 80t-60t- •

40t-

20 J-- Sample 13

10 I

60

40

20 Sample 14

10 100

I

80 60 40

20 t- Sample 15

10� _________ �� ____________ � __________ �� __ �

60 J--

40 J--

20t-

•

Sample 16

10� __________ � ____ _______ ' ____________ � __ __

/0 0./ /.0 Particle Size x (inches)

/0

FIG. 5 PARTICLE SIZE DISTRIBUTION IS TYPICAL OF SHREDDED REFUSE. (mm = in. x 25. 4)

203

-

-c: Ci) v .... Ci) Q.

.. . 'Ci) ..., . -U') "0 Ci) -0 -

U') c: 0

.c. ......

.... Ci) c: . -

LL. V> Ci) -v

. --.... 0

0-

100

90

80

70

60

50

40

30

20

1 0

4 Inches

3 Inches

2 Inches

I Inc h

,

0.0937 Inches (fines)

2000 4000 6000 8000 10 .000 12 .000 14.000 16.000 Refuse Processed (tons )

FIG. 6 PARTICLE SIZE O F SHREDDED PRODUCT DID NOT VARY DURING THE TEST PERIOD. (mm = in. x 25.4) (t = tons x 0.907)

The different times were determined by using the daily shredder log maintained by the shredder operator, while the tons of refuse shredded were obtained by adding the weight of the refuse de­posited on the tipping floor as recorded by the scale operator.

PARTICL E SIZ E

Results of particle size analyses are commonly presented as semi-logarithmic plots such as the five representative samples shown in Fig. 5. These particle size curves are typical of the type of data commonly obtained for shredded municipal re­fuse [3, 4, 5,6] .

A major concern during the shredder evaluation procedure was the potential deterioration of per­formance, as measured by particle size distribution.

This might occur when hammers wear excessively and thus allow progressively larger particles to leave as product.

Fig. 6 is a plot of the percent fmer than several sizes as measured over the entire trial period. The refuse processed (tons) is the value at the end of the working day.

There seems to be no deterioration of particle

204

size reduction during the trial period. In fact, the shredder performed well enough to\have been able to meet a guarantee of 70 percent less than 2 in. (50.8 mm) in size if that was required.

One major drawback in the analysis of particle size distributions is that no single parameter can be used to describe a distribution. A solution to this problem is to plot the data in such a way as to obtain a straight line, which then allows for a complete description using only the slope and one point. The Rosin-Ramrnler equation [7] is such an empirical model, and is stated as

Y = 1 - exp [-x/xo] n

where

Y = cumulative fraction of particles by

weight less than size x

n = a constant

Xo = a constant, known as "characteristic

size" defmed as the size at which 63.2 percent of the particles are smaller

This equation can be linerized by noting that

1 x n In ( )= [ ]

l-Y Xc and taking the log of both sides. A plot of

><

Q) 10 N - -

c.n c::: 0 �

� ..... <L> c:::

. -

LL en Q) I - -------------

u I ---..... I 0

CL I c::: I 0

--- n =0.62 I u 0 I .....

LL 0.1 I II

I >-<L> I ..... <L> I c:::

� I I I I

>- I I I xo = I . 30 I ,

0.01 0.1 I 10

Particle Size, x (inches)

FIG.7 THE PARTICLE SIZE ANALYSIS FITS THE ROSIN-RAMMLER RELATIONSHIP.

(mm = in. x 2 5.4)

In (l/1-y) vs x on log-log paper yields the slope n and the characteristic size Xo at [I n (1/1-Y)] = I, since the log I = 0, and hence x = Xc.

Plots for all of the particle size analyses were constructed and n and 'b values determined. Figure 7 is a typical plot. It is noted that an ac­ceptable straight line is defined, and that for this sample, the characteristic size is read as 'b = 1.30 in. (33.0 mm) and the slope as n = 0.62. For the most part, all of the data plotted as straight lines, indicating good agreement with the Rosin­Rammler empirical. particle size distribution equation.

Physically, as the characteristic size Xo increases, the particles increase in size. Further, if n increases, the particles sizes will tend to be most uniform (all of one size). A low value of n would denote a large size distribution, with many very small and very large particles.

Fig. 8 is a plot of both Xo and n over the test period. AltJ:j.ough Xo varies from day to day, there

205

seemed to be little if any overall change in Xo or n over the test period.

There was no correlation between characteristic size (xo) and n when plotted against the shredder throughput. This is partially explained by the fact that throughput is calculated on a daily basis, and the values are thus gross averages of the necessarily variable feed rates. Much more controlled test conditions, which would have violated the imposed acceptance test conditions, would be necessary to establish particle size variability with throughput.

EFF ECT OF R EF US E MOIST UR E CONT E NT

The moisture concentration of the refuse did not change appreciably over the test period. Both wet and dry samples were analyzed for particle size distribution, and Fig. 9 shows the percent fmer than 3 in. (76 mm) for wet and dry samples. There is no discernable difference be­tween the wet and dry samples .. This is interesting

c: 1.8 -

I./") ><

<Ll 1.6 J::; I./") u > Characteristic Size Xo c: r:::::-

1.4 =>-o I

>< -

...... 1.2 <Ll N"="

en u

-I./")

� <Ll

-u 0 � 0

J::; u

I./") <Ll

J::; u c:

r0 c: 0

J::; I-

� Q) c:

w:: -c: Q) u � C1.)

CL

c: -

-0

<Ll 0. 0

en "1:) c: 0

1.0

0.8

0.6

0.4

0.2

00

7"'O::;;:OO-'_-=iL"- - - - - - - - - - - -

Slope of In [I/( I-YIJ vs x ,n

2000 4000 6000 Refuse

FIG. 8 THE ROSIN-RAMMLER COEFFICIENTS DID NOT SEEM TO SHOW A TREND DURING THE TEST

PER 100. (metric tons = 0.907 x tons)

110

100

90

80

70

60

50

40

30

20

10

0 0

� �

• Wet o Dry

2000 4000 6000 8000 10,000 12,000 14,000 16,000

Ref.use Processed (tons)

FIG. 9 THERE WAS VERY LITTLE DIFFERENCE IN THE PARTICLE SIZE ANALYSIS FOR DRY AND WET

SAMPLES. (mm = in. x 25.4)

206

N

0

-.l

To

tal

Oper

atin

g T

ime

lin

ho

urs

& m

inu

tes)

Dat

e

19

78

P

lan

t S

yst

em

11

/2

7:5

1

6:3

1

11

/3

14

:55

1

2:4

1

11

/4

9:1

4 7

:56

11

/6

19

:33

1

4:1

4

11

/7

17

:29

1

5:1

7

11

/8

11

: 11

9

:01

1

1/9

1

4:2

7

11

:49

11

/10

1

4:3

8

12

:20

11

/11

1

0:09

9

:47

11

/13

1

6:5

6 1

4:1

3

11

/14

1

5:5

3

10

:06

11

/15

1

7:2

8

14

:29

To

tal

169

:44

1

37

:57

TA

BL

E 1

P

OM

PA

NO

BE

AC

H P

LA

NT

TH

RO

UG

HP

UT

AN

D D

OW

NT

IME

Th

rou

ghp

ut

Pla

nt

tota

l P

lan

t S

hre

dder

B

reak

s E

lect

rica

l

Shr

edd

er

ton

s/d

ay

ton

s/hr

. to

ns/

hr.

• %

•

%

min

. m

in.

6:1

8

687

8

8

109

2

6 6

52

1

1

12

:02

1

18

9

80

9

9

99

1 1

0

-

7:4

5

83

2

90

1

07

58

1

0

3

1

14

:10

1

47

0

75

1

04

14

9

13

1

13

9

12

: 19

1

28

6

74

10

5

10

3

10

3

-

7:5

9

108

7

97

1

36

83

1

2

7

1

11

:36

11

45

7

9

99

11

4

13

1

2

0

11

:52

1

31

5

90

1

11

1

20

1

4

3

0

8:2

8

96

3

95

1

14

1

8

3

4 1

13

:01

1

62

9

96

12

5

13

9

14

1

2

0

8:3

7 f

25

30

1

19

� 13

2

13

1

50

1

4

12

:56

76

11

7

11

3

5

2

12

7:0

3

Note

: m

etric

tons

a

0.90

7 X

tons

Do

wn

tim

e ,

Co

nve

yo

r D

iver

ter

Oth

er

• %

•

%

• %

m

in.

min

. m

in.

4 1

9

2

2

0

0

-1

2

1 6

2

7

0

0

8

1 2

0

4

0

0

2

0

59

4

4 1

171

17

2

9

3

0

-5

5

8

47

7

0

-

1 0

4

4

5

1 1

1 1

4

2

18

2

73

1

2

2

0

4 1

31

3

2

9

3

24

2

32

2

5

5

3

67

4

68

4

25

2

2

7

3

in terms of future shredder analyses, since dry samples may not be necessary, thus saving signi­ficant effort and perhaps allowing for a larger

number of particle size analyses to be performed

instead.

PLANT THROUGHPUT

During the twelve day test period, 14,133 tons (12,800 t) of waste were processed by the shred­der. The total waste processing system was in

operation for over 176 hr, and the shredder

operated for more than 127 hr. The plant through­

put was 80.3 tons/hr, (73 t/h) but the shredder processed nearly 111.5 tons/hr (10 1 t/h) of waste. This significantly exceeds the contractual require­

ment to process at least 60 tons/hr (54 t/h) and for the discharge conveyor to handle 96 tons/hr (87 t/h). Table 1 summarizes the plant through­put and downtime.

About 25 percent of the plant operating time was devoted to dealing with problems or taking scheduled breaks (the latter was about 11 percent

of the overall system operating time or 50 percent of the total downtime). Eight percent of the down­

time was caused by conveyor or diverter problems. An additional five percent of the downtime was

related to electrical problems or which the shred­der manufacturer was not responsible.

During this test period, the plant was able to process 80 percent more waste than the design

capacity. Following the acceptance test, the sys­

tem throughput was further increased by making minor adjustments to the diverter and conveyor.

POWER CONSUMPTION

An analysis of power consumption was complet­ed based on the throughput during the test period.

For tl).e entire plant operation time, an average of 5.33 kWh were consumed per ton of processed re­fuse (5.87 kWh/t). This figure must be reduced, however, since the shredder was not fed contin­uously, and consumed power during the time it was idling. Subtracting the total idling power con­sumption as determined from the log sheets, from the total recorded power use, total useful power consumption can be calculated (Table 2). The fmal net shredder power consumption is calculated as 5.25 kWh/ton (5.79 kWh/t).

Fig. 10 illustrates the impact of varying power costs on the power costs per ton of refuse shred­ded and can be used as a design tool to help in­

dicate expected costs for other shredder installa­tions.

0.30

v;c: 00 <-)to- 0.20 _ ..... 0 -...... 0 f-

0.10

4 5 6 7 Power Used in Shredding ( KWH Iton )

FIG. 10 A RANGE OF SHREDDER POWER COSTS

MAY OBTAINED FOR VARIOUS POWER RATES.

(metric tons = 0.907 x tons)

TABLE 2 SUMMARY OF POWER USE

Refuse Total Power Time Shredder Idling Net Shredder

Date Processed Consumed was Idling Power Power Efficiency

(tons) (kWh) (min) (kWh) (kWh) (kWh/ton)

2/11 687 3640 93 47 3593 5.23

4/11 832 4200 89 44 4156 5.00

7/11 1286 6580 310 155 6425 5.00

9/11 1145 7000 171 86 6914 6.03

11/11 963 4872 101 51 4821 5.01

Average: 5.25

Note: metric tons = 0.907 x tons kWh/metric ton = 1.1 x kWh/ton

208

The work performed by a shredder can be analyzed using several empirical relationships, the most useful of which is the Bond work index [8] , defined as

1 E = 10 E-1 - v'LF

where

E = work performed, kWh/ton, in reducing the size of a product from a feed with a size 80 percent fmer than LF (in microns) to a product 80 percent fmer than Lp (in microns)

Ej = work index, representative of the effi­ciency of a process, with a low number indicating high efficiency, and a high index indicating poor efficiency.

Using the data for the first day, where Lp = 2.18 in. (55.4 mm), and a reasonable value of LF = 10 in. (254 mm), the Ej index can be calculated as equal to 230 kWh/ton (253 kWh/t).

This value can be compared to published work indexes by Stratton and Alter [9] , who calculated Ej values for a number of shredders, also using LF = 10 in. (254 mm). On the average, the Ej for refuse shredding is reported by Stratton and Alter as about 400 kWh/ton (440 kWh/t), which is some­what higher than that calculated above, indicating efficient operation of the vertical shaft shredder.

The value of LF (80 percent finer size of feed) was not measured in this study. The value of LF = lOin. (254 mm) is not unrealistic, however, since most raw refuse has LF in the range of 7 to 10 in. (178 to 254 mm) [1,2,10, 11] and the above cal­culations thus yield a conservative value of Ei.

HAMMER WEAR

Table 3 shows the average weight loss at the hammer stations over the duration of the tests. The "vertical station" refers to the location of the hammers vertically, with station three being the uppermost hammers, and station nine being the lowermost set of hammers. At station seven, the shredder casing changes from conical to cylindrical. The hammers wear increasingly as the station ap­proaches the transition in the shredder.

Figure 11 illustrates the average weight loss of hammers at selected stations. These data demon­strate that the wear pattern for hammers is linear, at least up to about 14,000 tons (12,500 t) of refuse processed. In other words, the hammers

209

TABLE 3 AVERAGE HAMMER WEAR

Station Hammers Breaking Period, Acceptance Test, No. Installed to Nov. 1, Ib 12 days, Ib

3 2 24.5 24.5 4 4 '19.25 24.5 5 4+ 4 4.00 51.5 6 4 68.5 62.5 7 4*+ 70.88 80.9 8 4 16.25 45.0 9 4 21.25 72.5

Total 26 1283 1332

Tons processed 15,573 14,133

Tons/lb hammer 12.14 10.61

Note: kg = 0.454 x pounds

metric tons = 0.907 x tons

"Note two additional hammers installed after 3 days of

operation at th i s station

+Hammers at station five switched out to station seven

and new and used hammers placed i n station five after

nine days of operation.

-;: 80 .. E E 70 o -<=

.... :e 60 -

'" :g 50 ..J

:;; 40 "" .-.. � 30 � .. � 20 o

:t: 10

5000 10,000 15,000 Refuse Processed ( Ions)

Siolion • 3 o 4 .. 5 l> 6 • 7 o 8 • 9

20,000

FIG. 11 HAMMER WEIGHT LOSS IS LINEAR WITH REFUSE PROCESSED, UP TO AT LEAST 14 ,133

TONS PROCESSED. (metric tons = 0.907 x tons, kg = 0.45 x Ib)

seem to wear just as fast when they are new as when they are older.

Based on previous experience at other shredder installations, the hammers and arm assembly for this type of shredder are removed when they weigh between 140 and 150 lb (64 and 68 kg). The average weight of a new hammer and arm assembly is about 235 Ib (107 kg). If it is assumed that this results in an average loss of about 90 lb (41 kg) per unit, a linear extrapolation of the wear data in Fig. 11 would indicate that the hammer units at station seven would last long enough to process about 16,000 tons (14,600 t) of waste while those at station 3 would be able to process nearly 52,000 tons (57,000 t) of waste.

If it is assumed that, on the average, the ham-

mer units can process about 10.61 tons of refuse per pound of hammer weight loss (21.2 t/kg) (see Table 3) the total tons of refuse processed can be calculated as about 24,000 - 27,000 tons (21,800 -24,500 t) per one complete set of hammer units. This value agrees with the results of the Charleston, South Carolina test, where it was estimated that the average life of replaceable hammer head is about 25,000 tons (22,700 t) of refuse [12] .

SUMMARY

The primary acceptance criteria of the accep­tance test were that the shredder produce a prod­uct with 70 percent fmer (by dry weight) than 3 in. (76.2 mm) and that the shredder be able to operate at a throughput of 60 tons (54 t) per hour for twelve consecutive days at an average minimum daily throughput of 850 tons (770 t). In addition to particle size and throughput, the power con­sumption, the effect of moisture and hammer wear were also studied.

It was concluded that the shredder met the ac­ceptance criteria. The following results over the duration of the test support this conclusion:

1. A product 86 percent less than 3 in. (76 mm) in size.

2. Operation at an average shredder through­put rate of 111.5 tons (101 t) per hour.

3. Operation for 12 consecutive days. 4. Average daily plant throughput of 1178

tons (1068 t) per day.

REFERENCES

[1] Ruf, J. A.. "Particle Size Spectrum and Com­

pressibility of Raw and S hredded Municipal Solid Waste,"

Ph.D. dissertation, University of Florida, Gainesville,

1974.

[2] Trezek, G. J. "Significance of Size Reduction in

Solid Waste Management," EPA-600/2-77-131, Cin·

cinnati, O hio, 1977.

[3] Gawalpanchi, R. R., Berthouex, P. M., and

Ham, R. F., "Particle Size Distribution of Milled

Refuse," Waste Age, p. 34, 1973.

[4] Trezek, G. J., O beng, D. M. and Savage, G.,

"Size Reduction in Solid Waste Processing," EPA Grant·

#R801218, progress report, 1973.

[5] Diaz, L. F., "Three Key Factors in Size Reduc­

tion,H Res. Rec. & Cons., Vol. 1, p. 111, 1975.

[6] Rogers, H. W. and Hitte, S. J., "Solid Waste

Shredding and Shredder Selection," EPA OSWMP SW-

140,1974.

[7] Rosin, P. and E. Rammler, "Laws Concerning

the Fineness of Powdered Coal," J. /nst. of Fuel, Vol. 7,

pp. 29·36, 1933.

[8] Bond, F. C., "The Third Theory of Comminu­

tion," Trans. A/ME, 217, p. 139, 1960.

[9] Stratton, F. E. and Alter, H., "Application of

Bond Theory to Solid Waste Shredding," J. Envir. Eng.

Div., ASCE, EEl, 1978.

[10] Reinhard, J. J. and Ham, R. K., "Solid Waste

Milling and Disposal on Land Without Cover," EPA

(NTIS PB-234-930), 1974.

[11] Diaz, L. F., Savage, G. M., Trezek, G. J. ,

"Elements of Refuse Size Reduction," Com. Sci./Land

Util., Jan./Feb. 1979.

[12] Heil 92A Shredder Performance at Charleston,

South Carolina, unpublished internal report, The Heil Co.,

Milwaukee.

Key Words

Economics Materials Handling Power Process Refuse Sampling Methods Shredding

210