-------
The glass and aluminum rich fraction of the refuse separated
from the input waste by the liquid cyclone is partially
dewatered and delivered to a surge bin for temporary storage
and feed rate control. This material is then metered at
a fixed rate into the glass plant. Since many of the
separation processes are size dependent, the first operation
in the glass and aluminum recovery processing is a washing
and sizing step to remove particles smaller than 1/4 in.
From there the process stream passes a rotating drum magnet
which removes the magnetic materials from the process
stream. The magnetically cleaned process stream is then
delivered to a heavy media separator where materials with
a specific gravity less than 1.8 (mostly organic materials)
float and are removed for landfill disposal. Materials with
a specific gravity greater than 1.8 (glass, stones, metals)
sink in the heavy media separator and are carried to a jig
where the aluminum is separated from the glass and stones.
The aluminum rich fraction prepared by the jig can be stored
for market, or upgraded to improve its market value.
The glass rich process stream is dewatered on a vibrating
screen and dried in a kiln dryer. The dried material is
then conveyed to a high-tension electrostatic separator for
the removal of any remaining conducting materials, such as
-------
metals or any hygroscopic materials which might reabsorb
sufficient moisture to show a high surface conductivity.
The materials passing the electrostatic separator are
conductors which consist mostly of glass. They enter the
final separation process that utilize differences in optical
properties to segregate glass and stones. The material first
passes through a transparency sorter. Here, the opaque
materials are rejected and removed for landfill disposal.
The transparent materials are then color sorted into a flint
and a mixed color glass product. Both of these final glass
products are commercially salable glass cullets. If desired,
the mixed color glass product can be further color sorted
into an amber and a green product, both of which have a
higher market value than the mixed color glass cullet.
1.4 TECHNICAL EVALUATION
The Glass Recovery Plant is a major subsystem of the Franklin
Solid Waste and Fiber Recovery Facility. The feed to the
glass plant originates in the pulping and separation system
of the solid waste processing facility. After the solid
waste is pulped, the material is pumped through a cyclone
where the slurry is separated into heavy and light fractions.
The light fraction from the cyclone is processed in the fiber
-------
recovery system, and the heavy fraction from the cyclone
is fed to the glass plant.
During the evaluation period, 228 pounds of cyclone rejects
were generated for every ton of refuse deposited on the
tipping floor. These cyclone rejects would normally
constitute the feed to the glass plant. Because of size
limitations in some glass plant components, the glass plant
did not process all of the cyclone rejects that were available
Instead, the glass plant operated on a slip stream. The
cyclone rejects have the following composition:
Percent*
Flint Glass 38
Colored Glass 18
Magnetics 9
Aluminum 5
Other Metals . 0.3
Organics 8
Plastic and Rubber 8
Ceramics and Stones 12
Miscellaneous 3
*Percentages may not add to 100 due to rounding of individually
determined values.
-------
The following diagram shows the mass balance for the glass
plant assuming that all the cyclone rejects were processed.
The percentages refer to the ultimate disposition of input
to the glass plant. The numbers in parentheses are pounds
of material recovered per ton of refuse deposited on the
tipping floor given in pounds per tipping floor ton
(abbreviated to #/TFT).
INPUT
100%
(228#/TFT)
GLASS AND
ALUMINUM
RECOVERY
SYSTEM
MAGNETIC MATERIALS
13% (30#/TFT)
ALUMINUM RICH MATERIALS
4% (8.5#/TFT)
FLINT GLASS
15% (35#/TFT)
COLORED GLASS
10% (23.5#/TFT)
TO LANDFILL
58% (131#/TFT)
The separation efficiency of any unit operation, involves
both the efficiency of removal (the ratio of removed material
to available material) and the cleanliness of the product
in terms of the fraction of the recovered product that is
contaminanted. This data was obtained for the product streams
emanating from the glass plant.
-------
The magnetic separator recovered 86 percent of the magnetic
materials in the glass plant feed, but produced a product
which was 59 percent magnetics and 41 percent contaminant.
The jig separator recovered 50 percent of the aluminum
from the glass plant feed and produced a product which was
62 percent aluminum and 38 percent contaminant.
The optical sorters produce a flint glass product and a
mixed color glass product. The flint glass product contains
39 percent of the flint glass in the glass plant input
feed and is 96.0 percent flint glass, 2.9 percent green glass,
0.7 percent amber glass, 0.3 percent ceramics and stones and
0.1 percent other contaminants. The optical sorters recover
58 percent of the colored glass available in the glass plant
feed as a mixed color glass cullet. This product is 99.2
percent glass, 0.7 percent ceramics and stones and 0.1 percent
other contaminants.
1.5 ENVIRONMENTAL EVALUATION
The glass plant evaluated in this study was designed specifi-
cally as a back-end to the pulping and separation system.
As such, it would be meaningless to discuss the environmental
impact of the Glass Recovery Plant alone. Therefore, the
evaluation investigated the environmental impact of the total
-------
solid waste plant which was presented in the report of
Wittmann et al.2 However, where possible and meaningful,
independent environmental assessments were made exclusively
on glass plant operations.
1.5.1 Environmental Impact On Air
Those portions of the glass plant which could impact on air
quality include the dryer off-gases and the dust created at
several locations in the plant. In-plant dust control is
accomplished with a hooding system with the dusty air being
cleaned in a Venturi scrubber prior to atmospheric discharge.
The dryer exhaust is vented through the same Venturi scrubbing
system. Tests performed indicate that the system exhaust
gas has a composition similar to normal air with a particulate
emission rate equivalent to 3 percent of the allowable
particulate emission in the State of Ohio for such a system
on a process weight basis.
1.5.2 Environmental Impact On Water
There is no water emitted from the glass plant to the environ-
ment. All water used in the glass plant is returned to a
common sump and pumped to the Whitewater sump in the pulping
and separation system where it is used as dilution water.
The dilution water contaminant level is lower than that
2 Ibid.
10
-------
existing in the Whitewater system and, when so used, does
not degrade the Whitewater quality. Thus, there is no
significant environmental effect on water quality due to
the operation of the glass plant.
Glass 'plant water usage was 1930 gallons per hour of
operation. This usage was dependent on equipment size (much
of which is oversized) and does not relate directly to
throughput tonnage. For example, the jig could process
5 tons of material per hour with no significant increase
in water usage.
1.5.3 Environmental Impact On Land
A result of operating the glass plant is a 24 percent decrease
in the amount of material to be landfilled. Furthermore,
the solid materials emanating from the glass plant have not
been degraded by that processing; thus, they are equivalent
in composition to the totality of the liquid cyclone effluent
stream which otherwise would be landfilled. On a dry weight
basis, approximately 6.5 percent of the input to the tipping
floor is landfilled via the glass plant.
1.5.4 Noise Pollution
Noise measurements taken throughout the glass plant indicate
consistent noise levels in excess of 90 dBA which is the OSHA
limit for 8 hour exposure. A major cause of noise in the
11
-------
glass plant is the operation of vibratory conveyors and
bucket elevators filled with primarily nonresilient solids,
i.e., metal fragments and glass particles. Any new plant
of this type would require a different type of conveyor
system or enclosures around the bucket elevators and vibratory
conveyors to achieve compliance with present OSHA noise
regulations.
1.5.5 Odor Analysis
No appreciable odor is present in the glass plant since the
material being processed is washed in the first stage of
the processing. This washing removes the majority of
putrescible matter contained in the process stream. Thus,
as long as the system continues to operate and large quantities
of cyclone rejects are not retained within the plant, odor
is not expected to be a significant problem.
1.6 ECONOMIC EVALUATION
Economic data presented are based on the operation of the
present glass and aluminum recovery system, scaled to 500 TPD
and 1,000 TPD plants and with system changes indicated for
improved operation.
12
-------
Economic data is not given for the present system since it
includes equipment with different throughput capacities.
Some equipment is considerably oversized and others under-
sized for the plant's processing requirements. Thus, any
economic analysis of the existing glass and aluminum recovery
system shows disproportionately high capital costs for the
oversized equipment, disproportionately low capital costs
for the undersized equipment, and generally disproportionately
high operating costs for the total system because of this
disparity in equipment sizes. The net operating costs and
income for a glass plant serving 500 and 1,000 TPD solid
waste processing plants are displayed in Table 1.1. The
costs are normalized per ton of solid waste entering the
plant. Note that the loss associated with the glass plant
operation indicates that the plant cannot pay for itself
or make a profit for the investor.
The glass plant alone is not a money maker at 500 TPD; but
at 1,000 TPD, it is projected as a break even operation. It
should be noted that the revenue from the sale of aluminum
is twice that received from the sale of glass.
A glass plant requires a front-end system to provide the
glass plant process stream. Hence, the total economics of
both the front-end system and the glass plant should be
13
-------
TABLE 1.1 GLASS PLANT COST BALANCE
500 TPD 1,000 TPD
Capital Costs* $1,874,000 $3,290,000
Facility Expense** $1.70/T $1.49/T
Operating Expense $2.19/T $1.48/T
Income
Magnetic Sales (125/T)"*" $0.02/T $0.02/T
Aluminum Sales ($300/T) $1.95/T $1.95/T
Glass Sales
Flint ($20/T) $0.60/T $0.60/T
Amber ($20/T) $0.22/T $0.22/T
Green ($20/T) $0.13/T $0.13/T
Total Income $2.93/T $2.53/T
Net Savings (Loss) ($0.96/T) ($0.04/T)
+Assumed Sale Price for each Recovered Product - $/T
*Includes Financing Costs
**Based on Capital Recovery Factor of .11683 (15 years
at 8 percent).
14
-------
considered. Table 1.2 summarizes the economics of a Franklin
type facility with a glass recovery subsystem. Note that
the analysis excludes fiber recovery but it does include the
sale of all the fiber from the system as fuel. Revenues
from sludge disposal and magnetic and nonmagnetic metal
sales are included. The glass plant data is separated
from the pulping and separation system economics for easy
assessment of the glass plant's relative contributions.
The net savings for a 500 TPD plant is $1.28 and for 1,000 TPD
the net savings if $3.19/T. No tipping fee is included in
this analysis. One can quickly realize that the wet process
is more financially attractive without a glass and aluminum
recovery system. However, at 1,000 TPD, the analysis shows
that a glass plant becomes less risky.
15
-------
TABLE 1.2 ECONOMIC SUMMARY OF PROJECTED FRANKLIN
PLANT WITH NO FIBER RECOVERY, WITH GLASS
PLANT, AND SELLING REJECTS AS A FUEL
Franklin Projected Franklin Projected
Costs to 500 TPD to 1,000 TPD
Income*
Pulping and
Separation
Magnetic Metals $ 2.40/T $ 2.40/T
Sludge Disposal . 1.75/T 1.75/T
Fuel 10.71/T 10.71/T
Glass Plant .02/T .02/T
Magnetics 0.25/T 0.25/T
Aluminum 1.95/T 1.95/T
Glass 0.95/T 0.95/T
$17.78/T $17.78/T
Operating Expenses
Pulping and
Separation $ 6.04/T $ 5.50/T
Glass Plant 2.19/T 1.48/T
$ 8.23/T $ 6.98/T
Facility Expense
Pulping and
Separation $ 6.57/T $ 6.12/T
Glass Plant 1.70/T 1.49/T
$ 8.27/T $ 7.61/T
Net Savings $ 1.28/T $ 3.19/T
*Based on following revenue factors:
Magnetic metals selling at $25.00/T.
Sludge disposal—disposed of at a rate of 0.07/T
Input—$25.00/T.
Fuel—$1.25/1,000,000 Btu.
Aluminum— $ 3 0 0.0 0/T.
Glass—flint, amber and green—each at $20.00/T.
16
-------
The annual before-tax return on investments for 500 and
1,000 TPD total facilities, including the operation of the
glass plant, is presented in Table 1.3*.
TABLE 1.3 BEFORE TAX AROI FOR 500 AND 1000
TPD TOTAL FACILITY
$0.0/T Tipping Fee
Total Before-Tax
Net Savings Investment AROI
500 TPD
1,000 TPD
$165,000
$823,000
$ 7,966,000
$14,676,000
2.1%
5.6%
The before-tax AROI does not meet industrial investment
standards.
However, addition of a tipping fee to provide more revenue
changes the picture. Figure 1.2 illustrates the effect
adding a tipping fee to the revenue side of the balance
sheet and its effect on the before-tax AROI for a 1,000 TPD
facility.
Thus, with a tipping fee of approximately $8.00/T added to
the net income of $3.19/T, a before-tax annual return on
investment of 20 percent can be realized for a 1,000 TPD
plant. For a capital intensive and high risk business,
industry would require a before-tax AROI of at least 40
percent.
*This analysis assumes that all cash flows remain relatively
constant in time. This is effectively a "Unicost" assessment.
17
-------
10-
14 15 16 17 18 (9 20 21 22 23
•
I
BEFORE-TAX AROI (PERCENT)
FIGURE 1.2 EFFECT OF TIPPING FEE ON AROI FOR A
1000 TPD TOTAL FRANKLIN FACILITY
13
-------
Industry and municipalities use different techniques to
assess the merit of an investment. What may not be attractive
to industry could very well be attractive to a municipality
since the community might apply a break even analysis to
solve a pressing waste disposal problem where private
industrial participation (funding) would make the venture
unattractive.
19
-------
2.0 INTRODUCTION
2.1 PROGRAM EVALUATIONS
The objective of this study was to perform a technical,
economic, and environmental evaluation of the Franklin Glass
Recovery Plant, which is a subsystem of the Franklin Solid
Waste and Fiber Recovery Facility located in Franklin, Ohio.
The evaluations were performed over a six-month period
(September 1975 to March 1976). Because of the changing
nature of the facility, the data presented in this technical
evaluation was collected during February and March of 1976.
2.1.1 Technical Evaluation
The technical evaluation included:
1. A description of the Franklin Glass Recovery Plant
2. Power and water consumption
3. Stream characteristics
4. Material balances
5. Equipment description, theory of operations, and
efficiencies.
The results of the technical evaluation are presented in
Section 4.
20
-------
2.1.2 Environmental Evaluation
The environmental evaluation included an analysis of the
potential for polluting the air, land, and water. It also
evaluated the noise levels in the plant and observations of
odor and industrial hygiene requirements. Results of the
environmental evaluation are presented in Section 5.
2.1.3 Economic Evaluation
The economic evaluation of the glass recovery subsystem
included a determination of the economic viability of the
technology being used at Franklin. Cost and performance
data collected at the plant provides a measure of the economics
that would be experienced if the glass plant were to be
incorporated into a different type of solid waste/energy
recovery facility.
The data is presented, where possible, in non-dollar value
terms so that it can be readily used by all interested parties.
Results are projected for facilities scaled up to 500 and
1,000 TPD. Results of the economic evaluation are presented
in Section 6.
21
-------
2.2 HISTORY AND BACKGROUND
2.2.1 Franklin Plant
In 1967 the City of Franklin assessed its solid waste problems
and realized that its landfill would be complete in three to
four years. At the same time, employees of the Black Clawson
Company, Middletown, Ohio, conceived the idea of using paper
mill machinery to pulp solid waste, eject the nonpulpable
items, separate the finely chopped noncombustibles from the
paper fibers, recover them and burn the organic residual in
a fluidized bed reactor. The Black Clawson Company investigated
this concept further and, to prove its feasibility, con-
structed a pilot plant at their Middletown facility. This
pilot plant showed that municipal solid waste could be pulped,
that separation of the inorganic fraction from the organic
material could be accomplished, and that the organic material
could be burned with the remaining material placed into a
landfill.
Based on the results obtained from the pilot plant and the
City of Franklin's need to find an alternate solution to
their solid waste problems, a solid waste processing
demonstration grant was requested from the Public Health
Service under the Solid Waste Act of 1965.
22
-------
The City of Franklin made this grant request to design and
construct a full-scale plant which would demonstrate this
new and innovative concept of solid waste disposal recycling
with the recovery of magnetics. The grant was awarded to
the city on September 24, 1970. Land was acquired for the
project from the Miami Conservancy District near the new
waste water treatment plant. The solid waste plant was
completed in May 1971, and the fiber recovery plant was
completed in June 1971.
While the Franklin Solid Waste and Fiber Recovery Facility was
being constructed, the Glass Container Manufacturers Institute
(GCMI) announced that, under their sponsorship, the Sortex
Corporation of North America had completed a series of test
and trial operations using the glass rich fraction separated
from the pulped refuse by the liquid cyclone at Black Clawson's
Middletown pilot plant. Using a series of screening and
classifying steps to separate the extraneous material, Sortex
was able to recover a stream of color sorted glass cullet
and an aluminum rich stream. Based upon the laboratory
bench/pilot studies on the glass rich fraction, GCMI proposed
to the City of Franklin that the City apply for a supplemental
grant to add a glass and aluminum recovery line. GCMI indicated
that they would reimburse the City of Franklin for the matching
funds required for the demonstration grant. This grant
23
-------
request was submitted and approved. Construction of the
glass plant began in the summer of 1972 and was completed
by the end of 1972.
2.2.2 Contract History
As part of Contract No. 68-01-2211 with the U.S. Environmental
Protection Agency, SYSTECH was to perform a nine-month
evaluation of the Franklin Solid Waste and Fiber Recovery
System including the glass recovery system. However, during
much of the technical evaluation period (January through
September 1974) of the total plant the glass recovery system
was nonoperational. In mid 1974 it became apparent that the
glass plant would not be commercially operational during the
remainder of the nine-month evaluation. The major difficulty
stemmed from the use of air classification as a primary
technique for separating organics from the remainder of the
glass concentrate stream.
After several modifications to the system, start-up of the
modified glass recovery subsystem and this evaluation began
in November 1974. However, it became apparent during the
beginning of the evaluation period that the system would
again not meet end product objectives (yield, quality).
Equipment malfunction, low quality of salable output, and
24
-------
relatively high maintenance costs necessitated further
changes in the process. Hence, the evaluation was postponed
pending these changes.
The Black Clawson Company offered to "turnkey" the modifications
based on technology which they had originated and previously
piloted at the Institute of Minerals Research, Houghton,
Michigan. The GCMI and the EPA agreed to provide the funding,
and the new technology was integrated into the system during
May and September 1975. The original concept and the modified
concept of the glass plant are discussed in the following
section.
Other problems (e.g., plugging of lines and screens, etc.)
further delayed the evaluation. Finally in February 1976,
the glass plant was modified sufficiently to produce an
acceptable color sorted glass product on a continuous basis.
The data presented in this report represent the data collected
during February and March of 1976, e.g., once the plant was
made operational.
2.3 PLANT DESCRIPTION
Before a description of the glass plant is presented, it is
important that the reader understand the operation of the
front-end system that prepares the feed to the glass plant.
25
-------
material. The cleaned process stream contains mainly
fibrous organic materials and can be directed to a fiber
recovery system which produces a paper pulp marketable for
the manufacturing of roofing shingles and other low
grade fiber applications. This material can also be disposed
of by first dewatering and then using it in a combustion/
heat recovery system.
The cyclone reject stream (the heavy gritty materials removed
by the liquid cyclone) is the feedstock to the glass recovery
plant. This fraction of the refuse processing stream is
rich in glass and small pieces of metal, especially aluminum
when the nonmagnetic junk is recycled. It represents
approximately 11 percent of the original waste stream.
2.3.2 Glass Plant - 1974
Figure 2.1 is a schematic of the glass recovery subsystem
"as built" in 1974. The outputs from this configuration
were not acceptable and the items encircled in dotted lines
were removed and replaced with newer technology. Several
of the reasons for renovating this subsystem are as follows:
1. Air classifiers did not effectively remove heavy
organics from the glass rich fraction.
27
-------
2. Excessive glass breakage was encountered in
bucket drops, dryers, etc., resulting in a decrease
in potentially recoverable materials.
3. The high tension electrostatic separators did not
operate effectively because the process stream
reabsorbed moisture from the atmosphere after
leaving the dryer. The moisture caused the
separators to operate ineffectively.
These three problems caused excessive process losses and
resulted in unacceptable contaminant levels in recovered
products.
Other minor problems also contributed to production of
unmarketable end products. Hence, the glass recovery sub-
system was modified to the configuration shown in Figure 2.1.
2.3.3 Modified Glass Plant - 1976
Figure 1.1 displays the evaluated glass plant configuration.
A detailed description of this revised configuration follows.
28
-------
CO
o
UJ
CD
UJ
o
o
UJ
gc
<
UJ
CO
o
Z
<
O
cc
O
00
UJ
co
CO
cc
y coco
CO CO CO
•cozcDco5!55^ J
< CD GJ UJ (3 u 3 CD CD
• ;£—I>Z
<£^
5 < in
05 0 ^
CO
cc
O
CC
Q.
UJ
CO
z z
UJ CO UJ
LU UJ UJ
CC ^ OC
OHO
co x co
CM CO •*
cc
UJ
t .-.-
r"~
cc
o
CO
>• oc
0 LU
UJ CC
cc O
< co
1 §
"• Q
°O LU
Z LU
mC3
LU y
o s
< ^
si
>
LU
_J
•D
a.
cc. o
OP
£ m
< z
> (Tj
_^<
co > uj 2
cc
LU
QC
LU
LU
CD
Z
r>
CM co
LU s: cc <
LU LU UJ >
OC CC > Lu
o o oc _j
CO CO Q LU
IO CO N^ 00
2
H
Q
w
£-•
u
D
CO
2
o
u
CO
LO
en
<
,4
o
w
cu
o
EH
o
CO
2
o
M
Pi
2
o
CN
w
Di
D
o
29
-------
3.0 OPERATIONAL CAPABILITY
The glass recovery system at Franklin, Ohio demonstrates
the technical feasibility of recovering marketable glass
and aluminum products from the heavy rejects of a "wet
process" resource recovery system. The glass plant utilizes
a variety of separation techniques to recover a clean,
color-sorted glass product and an aluminum product. The
separation steps employed include methods utilizing the
magnetic properties, material size and density, electrical
conductivity and optical properties of the various components
of the glass plant feed stock.
3.1 EQUIPMENT DESCRIPTION
This section describes only .the major pieces of equipment
installed in the glass plant. No attempt will be made to
describe all the conveyors, bucket elevators, and dewatering
conveyors.
3.1.1 Surge Bin
The surge bin is a 10 cubic yard cylindrical hopper with a
conical bottom. Cyclone rejects from the pulping and
separation system are dropped in the top of the hopper.
Material to be processed is removed at the bottom by a
rotary feeder.
30
-------
3.1.1 Initial l.izing Ar.-i Cleaning
Many of the separation operations in the glass plant depend
upon having a consistent size for the particles being
processed. Thus the first step in the glass and aluminum
recovery operation involves screening. A 48-inch Sweco
Incorporated "Vibroenergy" vibrating screen separator rated
at 3,000 Ib/hr and fitted with screens that will pass
articles smaller than ;. in. is used. A rotating wash
removes fines which adhere to larger particles. Particles
which do not pass the screen are scraped into a feed control
bin which regulates the rate of feed to the following
separation processes. The +'; in. fraction becomes the
feed stock to the rest of the plant. The -\ in. fraction
is dewatered and landfilled.
3.1.^ Magnetic Separation
Oversized material from the screen is passed to a drum
magnot which draws off magnetic materials and deposits them
in a container for recovery and sale. The rest of the glass
and aluminum rich fraction passes on to the next unit
operation.
31
-------
3.1.4 Ht-avy Media Separator
The magnetic free +% in. material enters a heavy media
separator to purge it of residual organic materials. The
dense media separator is a Wemco HMS laboratory unit with
a rated capacity of up to 500 Ib/hr. The HMS is a rotating
drum partially filled with a pool of water whose apparent
specific gravity has been adjusted to 1.8 by the addition
of magnetite. The feed enters one end of the drum just
below the surface of the pool. Materials with a specific
gravity less than 1.8 will then float to the top and
eventually overflow a circular opening at the discharge
end of the drum. Particles with a specific gravity greater
than 1.8 sink to the bottom of the drum and are picked up
by lifters, elevated out of the pool and dropped into a
sinks hopper. Both the floats and sinks are spray washed
to remove residual magnetite which is recovered for reuse.
The floats are mostly heavy organic materials. These are
collected for subsequent -landfill disposal.
The sinks include glass, nonmagnetic metals, rocks, dirt
and ceramic materials.
32
-------
3.1.5 Jigging
The dense media sink material is processed in a Wemco-Remer
jig. In this device, liquid is pulsated vertically through
a horizontal bed of material being processed. This causes
the heavier material to work its way to the bottom of the
jig bed as the lighter material rises to the top. The Wemco-
Remer jig uses a double stroke jig mechanism for improved
separation efficiency.
Aluminum rich material comes off the top of the jig and can
be further refined for sale. The bottoms consisting primarily
of glass are passed on to dewatering and drying. The
middlings may be reprocessed or rejected to the landfill.
3.1.6 Electrostatic Separator
The sink material from the jig is gravity dewatered then
dried in a rotary drum drier. It then passes to a Carpco
high tension electrostatic separator. The electrostatic
separator is a device consisting of rotating, electrically
grounded drums on which the feed material is dropped. Near
the feed entry point, a high voltage ion source charges
the particles in the process stream. Conducting materials,
once beyond the influence of the ion source, rapidly dissipate
their charge to the grounded drum. These conducting
materials then drop from the drum since there is no electrical
33
-------
force holding them. (A charged body creates an "image"
force which causes it to adhere to a grounded surface).
Nonconducting materials, however, do not readily lose their
electrical charge. Thus they adhere to the drum longer than
the conductors.
Some nonconductors drop off with the conductors in the
electrostatic separator, but they tend to be principally
stones and ceramic materials. Theories as to why this
occurs are presently in dispute, but for our purposes the
result is fortuitous since it tends to aid in cleaning up
the glass product. The phenomena does result in contamination
of the mixed non-ferrous concentrate.
3.1.7 Opacity Sorter
The process stream, having been cleansed of conducting
materials and some stones and ceramics, passes to an optical
sorter for the removal of opaque materials. This is a
Sortex Model 962M Optical Sorter with a rated capacity of
400 Ib/hr per channel. In the optical sorter, each piece
of material passes through a beam of light. If the light
beam is cut off (the particle appears opaque) a small air
jet blows that particle aside and it is rejected. Transparent
materials are not subjected to the air blast and thus report
with accepted material. It should be noted that because of
34
-------
the irregular surface of the glass materials in the process
stream, transparent materials will be rejected because they
happen to pass through the light beam at such an angle that
their reflectivity makes them appear opaque. Rejected opaque
materials are removed for landfill disposal. Transparent
materials are passed on for further sorting into colored
and flint glass products.
3.1.8 Color Sorting
The final step in the glass plant processing is to sort the
glass product into a clear or flint glass cullet and a mixed
color glass cullet. This sorting is also performed by a
Sortex Model 962M Optical Sorter with colored filters
installed on the optics so that colored glass appears to
be "opaque". The principle of operation is exactly as
before, with the resultant products being clear glass cullet
and mixed color glass cullet. The colored glass mixture
could be further sorted into a green and an amber product,
but that is not presently done at the Franklin Glass
Recovery Plant.
35
-------
4.0 TECHNICAL EVALUATION OF THE GLASS RECOVERY PLANT
4.1 EVALUATION METHODS USED
This chapter presents the data gathering and analysis method-
ologies used in the evaluation of the glass recovery system.
Results of the data analysis and the overall glass plant
technical evaluation are also presented. These include:
recovery efficiencies, material balances, power and water
usages and some maintenance history.
4.1.1 Development Of A Study Matrix
To organize the principal areas of evaluation and to aid in
defining the data to be measured, a study matrix was developed.
The study matrix is a graphic representation of all possible
measurements. This allows the experimenter to rapidly assess
the necessity of any particular measurement, the relationships
between measurables and the completeness of his final experiment.
This study matrix is presented in Figure 4.1. Along the
horizontal axis, the unit operations of the glass recovery
system are listed. The vertical axis is divided into the
three evaluation categories. Each of the evaluation categories
is subdivided into its component parameters. A mark is placed
at the intersection of each unit process/parameter relation-
ship that was evaluated during this effort.
36
-------
EVALUATIONS
TECHNICAL
ECONOMIC
>
o J
K <
*H H
> z
2 U)
U) Z
N. UNIT
N. PROCESSES
PARAMETERS N.
DESCRIPTION OF EQUIPMENT
THEORY OF OPERATION
MAINTENANCE HISTORY
POWER REQUIREMENTS
STREAM CHARACTERIZATION
MATERIAL BALANCE
EFFICIENCY
RECOVERED PROD. CHARACT.
OPERATING LABOR COSTS
OPERATING POWER COSTS
MAINTEN. LABOR COSTS
MAINTEN. PARTS COSTS
GENERAL CONSTR. COSTS
EXTENDED CONSTR. COSTS
REVENUES
AIR
WATER
LAND
IN-PLANT NOISE
ODOR
INPUT SYSTEM 1
0
0
0
0
0
WASH SCREEN
0
0
0
0
0
0
MAGNETIC SEP'N I
0
0
0
0
0
0
0
z.
0,
H
V)
<
w
o
u
X
>•
<
u
X
0
0
0
0
0
0
•
u
M
^
0
0
0
0
0
0
0
ce
u
X
a.
o
M
*
0
0
0
0
0
0
SCREEN I
0
0
0
0
0
0
ELECTROSTATIC SEP'N I
0
0
0
0
0
0
TRANSPARENCY SEP'N I
0
0
0
0
0
0
0
Z
a
M
tt
$
o
o
0
0
0
0
0
0
0
SYSTEM
0
0
0
0
0
0
0
0
Q
0
0
0
0
0
0
0
0
0
0.
0
FIGURE 4.1 GLASS RECOVERY SYSTEM STUDY MATRIX
37
-------
4.1.2 Sampling And Analysis Methods
Of the parameters listed in Figure 4.1 under technical
evaluation, brief descriptions of the equipment and theory
of operation have already been presented in Section 3.
Since so much of the technical and economic evaluation of
the plant hinges on the separation efficiencies of the unit
processes, the weights of the output streams and the
characterizations of the output and process streams are
of great importance.
Process streams within the plant were sampled daily and
these daily samples were composited for weekly analysis.
Figure 4.2 shows the points at which the process stream
samples were collected and the points at which the output
streams were weighed.
The weekly composite samples were used to characterize
individual process and reject streams within the plant. The
weekly composites were also used to determine the recovered
product characteristics and individual unit efficiencies.
Analyses of the collected samples were performed by Systech
personnel by hand sorting into the characterization categories
appropriate for the stream under analysis. Magnetic metals
were separated from other metals using a small lab magnet.
38
-------
H
MAGNETIC
SEPARATOR
HEAVY
MEDIA
SEPARATOR
sauTj
soTueBao^O'
n
h|m|
VIBRATING
SCREEN
SEPARATOR
i
sauT jp
DE WATER
AND
DRY
$
MAGNETIC
SEPARATOR
$
ELECTRO-
STATIC
SEPARATOR
$
TRANS-
PARENCY
SORTER
!j SAMPLE TYPES
SOT}SU6EWV^/ KS
sauTj td
O- Composition
Q~ Weight
FIGURE 4.2 GLASS PLANT SAMPLING POINTS
C 2
OS H
3 CQ
e/3
UT9
COLOR
SORTED
39
-------
There are no in line scales for measuring the input to the
glass plant, nor are there any scales on any of the conveyors
within the process. Therefore, the analysis of material
flows had to be accomplished by network analysis using the
weights of the output streams from the plant and recognizing
that their sum had to be the total input. Each output
stream was collected in barrels and weighed periodically
and these weights were recorded. In addition, spillages
and other losses were collected and weighed so that they
might be accounted for. When all outputs had been measured,
they were appropriately added together to determine total
input weights. The operating time for the plant was used
to compute average flow values per unit time based on the
assumption that plant operation was quasi steady state.
The maintenance history for the plant was obtained from
operating logs kept by plant operating personnel. These
logs were verified by SYSTECH personnel who monitored the
operation at the glass plant during the study.
Power requirements were determined for the system as a whole
by using electric power meters. Power requirements per unit
time were determined by using a timer attached to the power
supply of the washing and sizing screen. Since this unit
was required for all subsequent operations, it was felt that
-------
its time of operation effectively measured the time of
operation of the whole plant. There were operations in the
glass plant which did not run as long as the washing and
sizing screen (e.g., the color sorters), but no unit ran
longer.
Section 4.2 of this report presents the results of the
technical evaluation phase of the study. The data represents
a compilation of six weeks data extending from 3 February 1976
through 12 March 1976. Findings presented in the following
sections of this report represent average operating and
characteristics data.
4.2 RESULTS OF THE TECHNICAL EVALUATION
The items covered in this section of the technical evaluation
of the glass plant include: (1) material balance data;
(2) characterization of the input, process and reject streams
and oatput products of the plant; (3) power and water usage
data; and (4) efficiencies of the separation equipment used
in the glass plant. Actual data collected with raw material
balance data, periodic test data, and data collection forms
are too numerous to include in this report. Hence, only
summaries are presented.
41
-------
4.2.1 Mass Balance Data For The Glass Plant
Figure 4.3 is a material balance diagram for the glass
recovery system. The input is for a typical 5-day operating
week at the Franklin Solid Waste and Fiber Recovery Facility.
About 43 percent of the processed input to the glass plant
is recovered; so only 57 percent of the cyclone rejects are
landfilled.
Mass balance data for the glass plant can be related to the
refuse received at the tipping floor by assuming that the
glass plant is appropriately sized to accept all the cyclone
rejects. For every ton delivered to the tipping floor,
228 pounds of cyclone rejects would be received at the glass
plant in a community with Franklin waste characteristics.
Figure 4.3 shows the different output streams from the glass
plant. Percentages shown in that figure refer to the fraction
of input to the glass plant recovered, while the weight
numbers refer to pounds of material removed per ton of
refuse at the tipping floor. Thus, for example, 30 pounds
of magnetic materials are recovered in the glass plant for
every ton of Franklin refuse at the tipping floor. The jig
top product, which is the aluminum rich stream, is 8.6
pounds per Franklin tipping floor ton. The flint glass
product is 35 pounds per Franklin tipping floor ton, while
the mixed color glass product is 23.5 pounds per Franklin
42
-------
0)
s: 01
4J '0
c
O 3 5)
VOX
V
3 01 U
1)
O 01
4— f- rH
m rH rH
r^ O n p^ CD rH
*r ff* o i"* r- m
o-riinomoin ootnm
• ^« in ^r 1—4 '
(N 00
IH (U
•4-1 0 M
C HH
CJ (U -4J
U IH 3
01 Ul C
a u -H
O
M
EI 01
M 01
ui m
O -
s c
O -^
o
—
CM
01 01 £1
01 4) p
in c 01 ix
rH O 3
U JJ 01 0 "»
in £ o u 01
T3 3 --H w O U
0 » C JJ 41 -4 -H
U
3M> 4k C7O
O 00 cH
O CN CN
I
UJ
5 t-O
A\- uj — cO
i x
en <
i
i
^"T
O
m
04
rH
-r O
H"
il uj
5 °
to <
1
ct
o
ct
UJ
(O
01 01 01 O D^>4H
QJ O 01 4J -H O
01 c c 01 m 4)
3 OO01i-4 Ui5c
01 ooijjvmo do
01£ EOUl r4£U01O3UlrH 14HT3-4J
O3 3 -HO UiS'HO O"O DC-
••HC C 4J -I (UC-U-H^^aO) MftJMrj
CH- -rtQJCOl- h-HOICOl • 4J 14 • (DO
03 E O S C 03 Ul O I6C03U1.CJ.COU M -U a O
O^ 30} 3 CP T* in 01 C 3 ^ (7* fl M t(l IH *H rH Ul CJ C — H
|0<£| < fi 0 3 SB | S < I S 3 iS SEll 0 fc 5 1 1 ra^«H «
3 M Id i O t'l
b. A > U 5 =
O rH Q O
Ul J3 U OS Z <
£ £ W O a,
s j: os cj O
1 (N I-- rH
~l
_
f
in
^t
00
i 1
^-
f*i
&*
rH
1 1 1
afe --,
\O ^*
. v^
00 rH
OP 0*
1
^^
-^
i"O
r-H
cJP
^O
' 03
< 01
J 01
O EH C
U O
— Q D V
=•» S3 Q 01
in Q* O
30 - O ^4 ^
o m J o. 01
rH 04 O Ul • O
— CJ 03 n 01
Q ^
^0
^ <
UJ LU
rH - — OCC — O~ — -Jrr
CC S^ *^O ^O
UJ ^ O ^^^^ va O r4 C
C (U <1) • O
•H jQ (3J Ul 01
rH S Ul OJ -H
t. 3 u u £
2
O
H
W
EH
<
G
D
CM
EH
O
O
EH
H
S
.2
o
w
u
o
"3"
W
O
M
&H
O c
rH <0
43
-------
tipping floor ton. Because some processing equipment in
the glass plant is undersized (especially the heavy media
separator) only a fraction of the available cyclone rejects
were processed by the glass plant. Total cyclone rejects
were measured at 11.4 percent of the incoming refuse during
the test period. However, because of this undersizing, the
material handled by the glass plant was only 1.9 percent
of the refuse received at the tipping floor or a slip stream
of approximately 20 percent of the available cyclone rejects
was actually processed.
4.2.2 Glass Plant Stream Characterization
The characteristics of the streams in the glass plant were
developed from composite samples collected at various points
in the process stream and all the output points (see
Figure 4.2). The input to the glass plant consists of the
water-saturated cyclone rejects which are rich in glass,
magnetic materials, and aluminum. The majority of this
material (70 percent) is -3/4 in. and +^ in. in addition
to the fraction of the material entering the glass plant
which is separated at various points within the process,
Figure 4.3 also shows the characteristics of those output
products. All the compositions are reported as appropriate
weight percents.
44
-------
Magnetic materials comprise approximately 9 percent by weight
of the Franklin cyclone rejects*. Eighty-six percent of these
magnetic materials are recovered by the magnetic separator.
However, contaminant levels are high (41 percent) in this
product stream. This contaminant level is caused primarily
by wet materials sticking together; thus, the wet magnetics
carry contaminants along with them on the magnetic drum and
into the recovery bin. This magnetic material would require
further cleaning for sale. Once it is dried, it can be
easily processed through another magnetic separator to
achieve a much higher purity and enhance its value.
The cyclone rejects at Franklin are the input to the glass
plant. They have an aluminum content of 4.7 percent. Due
to the high economic value of this metal, an attempt is
made to recover this fraction of the cyclone rejects.
Fifty-one percent of the aluminum in the cyclone rejects was
recovered at the jig. The remainder of the input aluminum
is removed at the magnetic separation step, the heavy media
separation step, and past the jig by the electrostatic
separator.
*The cyclone rejects are a poor fuel with 94 percent ash and
417 Btu/lb higher heating value. It is evident from this low
Btu content and high ash content that this material is not
considered a useful fuel source and that little fuel value is
lost in the cyclone rejects. These numbers are consistent with
the characterization data of the input which indicates a total
organic content of 15.5 percent on a dry weight basis.
45
-------
The Franklin cyclone rejects have a glass content of about
54 percent, 53 percent of which is recovered as glass cullet.
This is equivalent to approximately 29 percent of the feed
coming into the glass subsystem (i.e., 36 percent of the
glass received on the tipping floor in Franklin).
4.2.3 Utilities Requirements
During the study period the electric power consumption was
recorded for the glass plant. The average power usage of
the glass plant was 63.3 kilowatt hours per ton. This is
equivalent to 215 kilowatt hours per ton of material processed
in the glass recovery system, or to 26 kilowatt hours per
ton of solid waste delivered to the tipping floor. Power
usage for individual unit processes is not available since
no individual units were metered.
There are four areas of water usage in the glass recovery
system: they are the surge bin conveyor, the washing and
sizing screen, the heavy media separator, and the jig. Flow
meters were not available to measure water used by the jig
and the heavy media separator. For the most part they used
their own recycled water and needed only makeup water. The
surge bin conveyor uses approximately three gallons per
minute and the washing and sizing screen uses approximately
30 gallons per minute of operation. This represents a
46
-------
total flow of 1,980 gallons per hour or 6,712 gallons per
ton processed. Water used in the glass plant at Franklin
is "city water", i.e., clean potable water. At another
installation, non potable process water could be used.
All the water used in the glass plant is collected in a
common drain system and is pumped to the Whitewater sump in
the pulping and separation system of the main solid waste
plant. Thus, all water in the glass recovery system is used
as makeup water in the Whitewater system. No Whitewater
(recycled process water from the pulping and separation system)
is used in the glass recovery system. The primary contamination
of the water used is an increase in total suspended solids.
This results because fine particles are rinsed off the feed
material going to the glass recovery system.
4.2.4 Efficiency Of Equipment
Separation efficiencies for unit operations within the glass
recovery system are calculated for the following unit
operations: washing and sizing screen, wet magnetic separator,
heavy media separator, the jig separator, the rotary kiln, the
electrostatic separator, the transparency sorter, and the
color sorter. The overall material balance for the glass
plant is presented in Figure 4.4.
47
-------
TOTAL
CYCLONE
REJECTS
48,868 Ibs.
CYCLONE REJECTS BYPASSING GLASS PLANT
41,868 Ibs.
INPUT
7,000 Ibs.
GLASS AND
ALUMINUM
RECOVERY
SYSTEM
OPERATING
TIME
11.9 hrs.
MAGNETIC MATERIALS
931.4 Ibs.
ALUMINUM RICH
MATERIALS
T68.2 Ibs.
FLINT GLASS
1078.6 Ibs.
COLORED GLASS
719.1 Ibs.
TO LANDFILL
4002.7 Ibs.
FIGURE 4.4
MATERIAL BALANCE DIAGRAM FOR GLASS AND
ALUMINUM RECOVERY SYSTEM*
*Figures represent operating data for an average week;
all weights are given on a dry basis.
48
-------
4.2.4.1 Washing And Sizing Operation
The input to the glass plant from the liquid cyclone is
directed to a surge bin having a rotating feed table which
meters the solid flow to the initial separation operation.
This first operation in the separation process is the
vibrating screen separator which rejects materials smaller
than -; in. from the remainder of the process stream. To
aid this removal, material on the vibrating screen is flushed
with water to remove small particles adhering to the surface
of the large particles. The rejected material from the
screen separator is fed to an inclined dewatering screw and
landfilled.
Twenty-one percent of the total input to the glass plant
is removed as undersized rejects at this point. Of the
material passing as process stream, only 3 percent (equal
to about 2 percent of the input) is less than % in. in
size. This indicates an efficiency for the washing and
sizing operation of 90 percent.
4.2.4.2 Magnetic Separation
The process stream from the washing and sizing operation is
fed to an electromagnetic drum separator which scalps off
the easily removable magnetic materials. The materials
removed by this magnetic scalping of the wet feed are moist
49
-------
(12 percent water). As a result, materials stick together
so that this recovered stream is only 59 percent magnetic
material. However, 86 percent of the available magnetic
materials are removed at this step. Even though the removal
efficiency is 86 percent, since the product is 41 percent
contaminants, the market value of this product is questionable
without further cleanup.
4.2.4.3 Heavy Media Separator
After removal of the wet magnetics, the process stream flows
to the heavy media separator. The function of the heavy
media separator is to remove organic materials from the
glass rich stream. By blending material with a solution
of water and magnetite, which is held at a specific gravity'
of 1.8, the organic materials are floated off and the heavy
fraction is passed on to the next operation. Eight percent
of the input to the glass plant is removed at this point
and contains 93 percent of the organic materials remaining
in the process stream at this point.
4.2.4.4 Jig Separator
The primary function of the jig is to separate the aluminum
fraction from the process stream. The jig is fairly efficient
since it removes 73 percent of the aluminum in the jig feed.
50
-------
The jig output product contains only 62 percent aluminum.
Further cleanup of the jig output product can be achieved
by drying the material and then subjecting it to an electro-
static separation process because the contaminants are
primarily organics which bypassed the heavy media separator.
4.2.4.5 Rotary Drum Dryer
After leaving the jig, the process stream passes a dewatering
screen for the removal of excess surface water; then it
passes to a rotary drum dryer. The moisture content of the
influent material averages 2.9 percent, and the moisture
content of the stream leaving the dryer averages 0.3 percent.
This indicates an average reduction in moisture of approximately
90 percent.
4.2.4.6 Electrostatic Separator
Following the rotary dryer, the process stream is passed
through a magnetic scalping operation to remove any residual
magnetic materials. It is then fed to the electrostatic
separator which separates materials according to their ability
to hold a static charge. Removal efficiency for metallic
material is nearly 100 percent. Also, almost all of the
remaining organics and 66 percent of the ceramics and stones
are removed by this device. About 6 percent of the available
glass is lost from the process stream at this point.
51
-------
4.2.4.7 Transparency Sorter
The process stream from the electrostatic separator is
conveyed to the transparency sorter. The removal efficiency
of opaque materials was determined to be 90.5 percent. The
transparency sorter also rejects 17 percent of the incoming
glass. This results in a glass contamination level of
77 percent in the sorter rejects. This percentage varies
with the feed rate to the sorter and the quantity of opaque
materials in the feed. During the evaluation, the feed rate
to the sorter was held at 400 Ib/hr.
4.2.4.8 Color Sorter
The last unit operation in the glass recovery system is the
color sorter. The function of this device is to separate
the flint glass from the colored glass. Efficiency in this
operation has been computed as the contaminant content in
the flint glass product, where contaminant means any constituent
which is not flint glass. The flint glass product was
measured to be 96 percent flint glass, 2.9 percent green
glass and 0.7 percent amber glass. Other contaminants include
0.2 percent ceramics and stones, and 0.1 percent unidentifiable
tramp fine material.
52
-------
The colored glass product was 99.2 percent glass, 0.7 percent
ceramics and stones, and 0.1 percent unidentifiable fine
materials.
The GCMI specifications for recovered glass are presented
in Appendix B of this document. Although the recovered
glass products do not meet GCMI specifications, they are
marketable glass cullets. Gullet samples from Franklin
were evaluated by glass users and served as the basis for
long term purchase contracts for the glass to be produced
by a similar plant in Hempstead, New York.
4.2.5 Comments On Maintenance History
Four major unit operations were improved prior to data
collection. The first change was the replacement of the
vacuum tube electronic systems in the electrostatic separator
with solid state components. Solid state components improved
the operating life and reliability of the system as well as
enhancing overall operating efficiencies.
The second improvement involved the magnetic separator. The
unit in use at the beginning of the testing program was not
of sufficient size and strength to give adequate magnetics
recovery. The heavy media separator output was not sufficiently
free of magnetic materials thus resulting in a lowering of
53
-------
the operating efficiency both of the heavy media and jig
separators. The present magnetic separator resulted in the
reported separation efficiencies and in improved operation
both in the heavy media separator and the jig.
The third item to be improved was the Sortex Optical Sorting
System. During the testing period the Sortex Model 962 Optical
Sorter was replaced with the Sortex Model 962M System which
has an improved, high-speed ejector device. This has resulted
in an improved quality of output product as well as an increase
in throughput capacity up to 500 Ib/hr.
The fourth item to be improved was the bucket elevators.
Friable materials dropping into the buckets caused flying
chips and dust which accumulated in the bucket pivots and
made the elevators bind at turns. The resulting stress
increase caused some of the buckets to crack and necessitated
system shutdowns to remove and replace the cracked buckets.
The problem was minimized by replacing exposed drive gears
with enclosed gears. A maintenance program requiring ongoing
cleaning of the buckets has effectively controlled this
problem.
54
-------
There were some minor problems involving the operation of
the heavy media separator due to the fact that it was con-
siderably undersized. The unit has a throughput rating of
only 500 Ib/hr, whereas a rate of about 1 T/hr is required
to process all the cyclone rejects. Thus, the unit required
considerable operator attention which would not be necessary
if the unit were sized with a 4-ft. diameter or larger drum.
These problems with the heavy media separator are responsible
for much of the contamination in the aluminum product coming
from the jig.
It should be noted that the glass recovery system in the
Franklin Solid Waste and Fiber Recovery Facility is a large
test operation which represents the first plant of its type
in the world. As such, many equipment modifications and
changes were incorporated as an ongoing process. Thus, a
long-term history of the maintenance requirements of the
equipment in the plant is not available. Based on the
limited knowledge gained at the Franklin plant, routine
maintenance requirements are not excessive. During the
evaluation, maintenance time never exceeded 10 manhours
per 40 operating hours.
55
-------
5.0 ENVIRONMENTAL EVALUATION OF THE FRANKLIN
GLASS PLANT
5.1 INTRODUCTION
The major environmental impact of the Franklin Glass Recovery
Plant is that it reduces the land required for disposal of
solid waste and recovers valuable resources. These benefits,
of course, must be weighed against the environmental impact
of various waste streams resulting from the operation of
the plant.
The environmental impact of the Glass Recovery Plant on air
quality can best be characterized by the evaluation of the
particulate emissions from the glass plant. Particulates
are produced by various operations (moving of the glass) of
the glass plant. Tests were conducted to quantify the air
emissions from the glass plant. The test results are discussed
in Section 5.2.
The environmental impact of any system with respect to water
quality is usually evaluated by characterizing the influent
and effluent flows from the plant. However, since the glass
plant is not a "stand alone" system, it must be evaluated
in terms of its additional effect on the environment over
and above the effect of the front-end solid waste plant.
56
-------
Finally/ an occupational safety and health evaluation was
also conducted with respect to industrial hygiene and noise.
Sections 5.4 and 5.5 will present the results of this
evaluation.
5.2 ENVIRONMENTAL IMPACT ON AIR QUALITY
The input to the glass recovery subsystem contains friable
material (a material that will fracture and crumble easily).
The separation and handling techniques used in the glass
recovery process tend to fracture the friable materials
so that dust is generated.
Hoods are used at those locations within the glass plant
where dust might be generated. The collected dust and air
are exhausted through a venturi scrubber located outside
the Glass Recovery Plant. This exhaust is the only major
source of air emissions to the outside environment. Thus,
the effect of the glass plant on air quality can be determined
by measuring the particulate emissions from the venturi
scrubber system.
ASME Power Test Code No. 27 type particulate emissions tests
were conducted to measure the emissions from the dust collection
system in the Glass Recovery Plant. The EPA Method 5 sampling
57
-------
procedure was not employed because the ambient stack
temperature and low moisture content did not require a
heated sampling probe. The results of these tests are
included in Figure 5.1. It was determined from these tests
that the particulate emission rate is 0.089 pounds per hour.
The stack temperature (since there is no combustion device
other than the dryer) was approximately room temperature or
74°F. The flow rate at standard conditions was approximately
8500 dry standard cubic feet per minute. An Orsat analysis
conducted on the exhaust stack indicated that the gas
composition was similar to air.
At the process weight rate of 20 pounds per minute, the
allowable emission rate in the State of Ohio is 2.90 pounds
per hour*. Thus, the emissions from the glass plant (0.089
Ib/hr) are only 3 percent of the allowable emission rate.
There are no Federal regulations limiting emission rates
in terms of the process weight rate of an industry or system.
Thus, the Glass Recovery Plant is in compliance with applicable
regulations.
*0hio Regulation AP-3-11, Table 1
58
-------
O
*
rH
|
"*" 1
r
en
CN
O
rH
•xl"
r«.
\
CN
CN
O
rH
ri
x^
CN
^_,
rH
en
**
=**
O
"ty*
o
en
r-
r-
CM
CN
r*~
f^
cn
CN
O
rH
0
rH
O
rH
j
j
I
CO
in
•
rH
00
in
rH
00
in
rH
CO
in
rH
00
00
irt
in
rH
CN
in
vo
vo
m
CM
CN
00
O
^*
r-
rH
00
00
0
rH
CM
o
^*
r~
o
^J*
f*
1
;
o\
rH
_,^
00
rH
r^
00
r-«.
CM
00
|
a
CJ
d
m
CN
in
00
••
4->
•H
TJ
l-l
a
u
•H
•H
O
cu
Q
to
*— *
•
w
£1
\
*
CO
j^
1— 1
>H_I>
1 J
4-*
CO
cu
EH
CT
C
•rH
'H
•3
Q
f
(U
T!
05
0)
cr
aJ
^
TJ
(0
0
E '
TJ E
CU '
•P £
CO
r
CD C
Cn ,
T5 C
CU C
£> )
< c
1
j
1
i
1
s~*>
S
' CU
u*
*•— »>
i CQ
c
o
! -rH
4J
G
O
CJ
y
U
T3
CO
CN 4J
4-* T3
-. — >
^t 4-1
3 TJ -H
CU O
S >-l O
3 < rH
J CU
H .* >
U •
o TJ cr
< 4J >
. — .
2
Cu
CJ
•*— *
CO
c
•H
•H
TJ
C
0
CJ
Stack
14J
CU
TJ
» O
r-H
Cl4
^~,
CM
o
•— •
CU
>H
3
4J
TJ
r-i
CU
a
cu
EH
U
TJ
4J
to
^^
s
CL<
CJ
CO
~-^
CO
C
O
4-1
d
0
CJ
T)
,_i
TJ
Stand
4J
TJ
CU
4->
TJ
05
0
rH
Qu
<
EH
Q
Z
0
MISSI
a
w
EH
CJ
05
<;
Oi
Jf^>
D
g
"•-—-i
a
4J
CJ
cu
rH
rH
O
CJ
CU
4J
TJ
rH
3
CJ
•H
4J
nj
rH
TJ
4J
O
EH
t,
CJ
to
»-«
•
TJ
C
O
• CJ
1
T3
» *\
UJ
TJ
T3
CU
rH
a
g
to
cn
5
>
Q
CU
3
rH
O
-—•
CJ
co
. ^-»
TJj pL|
Cj CJ
o] to
rjj * — '
• •
T| T)
•M C
CO| O
CJ
*T5
Tl S
cu to
r-l
CU4J
S tl
TJ
CO CU
. g
VH 3
O rH
CUO
($ ^r*
TJ
iH CU
CU rH
4-» a
TJ g
2 TJ
CO
cu
g rH
rH 4J
o o
> EH
*••— •
1
0
rH
X
P-,
CJ
to
V.
l_l
D
B
C
4J
(TJ
k_l
4J
oncen
u
cu
4->
(TJ
rH
3
CJ
1-1
TJ
a.
^-.
r*.
1
O
rH
X
PM
CJ
to
\
CC
_Q
IrH
C
o
•H
4->
TJ
M
4J
oncen
CJ
cu
TJ
rH
3
O
•rH
4J
TJ
5
^~*
.
i_i
_c
.\
•
CO
j3
rH
CU
4-1
TJ
05
C
O
missi
Cd
TJ
r-H
3
U
4J
T3
Oi
9
*—*
ID
EH
03
VO
O
rH
\
CO
_c
r-
CU
4J
TJ
05
C
0
missi
Cd
CU
4-1
TJ
rH
3
O
•rH
4->
(C
Oi
^«.
OP
0)
rH
a
g
TJ
CO
netic
•H
X
O
co
M
C
cu
CJ
IH
a,
X
\
• S
o
EH
a>
4->
TJ
05
4-)
CT
•H
CU
CQ
CQ
CD
CJ
O
rH
PH
i
j
1
rH
hyH
X)
rH
' CU
4J
ed
05
C <
0 E-
•H rt
CO C
cn
•H CO
'e 3
Cd C
cu *z
rH rt
X) r^
rd i-:
o c.
rH V.
rH t-
i^C ZE
r
tf
C
*
f]\
us
TJ
Jj
f-1
r*-l
rr
w
.,_!
(1)
VA/
*
ssaoc
^
rH
d.
T)
CU
4-)
g
•H
4-)
cn
•H
CQ
rH
in
IVJ
s
4J
T5
CQ
O
dP
in
o
i
0
CJ
in
Q
CN
1
CM
O
dP
O
0
1
O
CJ
dP
O
£
1
f\f
z
EH
CO
IS
Q
H
rH r-3
K O
\ CO
O £2
O M
CN 1-3
i K^
t-T
II (=£
C fc.
•H
s
=**= rH
O •
CN LO
CU Cd
4J pj
TJ D
05 O
M
4J Cn
CT>
• H
CU
CO
CQ
CU
0
P-I
-------
It should be noted that the moisture content of the exhaust
gases from the wet scrubbing system is only 1.8 percent by
j
volume. This low moisture content is due to the high volume
of air being pulled through the system.
5.3 ENVIRONMENTAL IMPACT ON WATER QUALITY
Water used in the glass plant consists of city water and
process water (firal clarified water from the adjacent waste-
water treatment plant). The city water is used in the wash
screen for the purpose of washing and separating the various
size fractions of glass and organic material. This washing
step is necessary because the glass plant feed material is
saturated with lower quality water (i.e., Whitewater, which
is water that is used and recirculated in the pulping and
dewatering subsystems) containing particulate matter that
needs to be removed.
Process water is used in the glass plant as makeup water
for both the jigging operation and for the heavy media
separator. All the water used in the glass recovery system
is returned to a common sump and pumped to the Whitewater
sump in the pulping and separation system where it is used
for dilution water in the Whitewater system. There is no
direct discharge from the glass plant to a receiving stream
60
-------
or wastewater treatment plant. Thus, the environmental
effect on water quality solely due to the operation of the
glass plant is not significant from a water quality standpoint.
Characteristics of the process water used in the glass plant
are given in Figure 5.2.
5.4 ENVIRONMENTAL IMPACT ON LAND
One of the side benefits of the Franklin Solid Waste and
Fiber Recovery Plant with the Glass Recovery Plant addition
is a reduction in the amount of material that would normally
require landfilling. The glass plant yields the following
landfillable material: wet fines, heavy media separator
fines, heavy media separator floats, dry fines, conductors,
opaques, and transparency and color sorter fines.
During the evaluation period, the average input tonnage to
the glass plant was 590 pounds per hour. Of this amount,
approximately 252 pounds per hour is recovered and is not
landfilled. This amounts to a 43 percent recovery rate of
the incoming feed to the glass plant. If the glass plant
were sized to accept all of the cyclone rejects available
(11.4 percent of the tipping floor tonnage at Franklin),
then 43 percent of this material would be recovered. This
is equivalent to a recovery of an additional 98 pounds from
61
-------
PROCESS WATER CHARACTERISTICS
PARAMETER
Water Temp.
Dissolved
Oxygen
Turbidity
Lab
Conductivity
Lab
oH Lab
\lkalinity
Hardness
Chlorides
Sulfates
Total Solids
Dissolved
Solids
Suspended
Solids
Ammonia
Nitrogen
Organic
Nitrogen
UNITS
°C
rag/1
JTU
Micro-MHO
S.U.
mg/1
rag/1
rag/1
rag/1
mg/1
rag/1
mg/1
rag/1
mg/1
MAXIMUM
28.1
11.4
60.0
2173.0
8.5
544.0
710.0
224.0
220.0
1525.0
1449.0
121.0
44.0
3.1
MINIMUM
6.3
5.1
6.0
1440.0
7.1
142.0
196.0
162.0
70.0
1037.0
987.0
1.0
0.0
0.1
AVERAGE
20.5
8.2
21.4
1603.9
7.8
407.0
451.9
193.0
156.2
1188.4
1138.4
51.0
7.7
2.3
Nitrate
Nitrogen
Total
Phosphorus
C-BOD2
C-BOD5
C-BOD7
C-BOD10
C-BOD15
C-BOD20
N-BOD5
TOC
COD
mg/1
22.0
0.0
5.3
mg/1
rag/1
rag/1
mg/1
rag/1
mg/1
rag/1
rag/1
rag/1
mg/1
3.7
110.0
176.0
182.0
214.0
282.0
292.0
9.8
140.0
431.0
0.1
0.6
2.5
5.0
7.3
9.0
10.2
8.5
25.0
70.0
1.2
8.4
18.4
25.5
35.2
53.7
70.0
9.2
50.9
174.3
FIGURE 5.2 PROCESS WATER CHARACTERISTICS
62
-------
every ton of waste at the tipping floor. This is a net
reduction of 24 percent of the material to be taken to the
landfill from the Franklin plant.
The glass plant rejects taken to the landfill are low in
putrescible organics and may not require covering, depending
upon the location. Nuisance conditions resulting from
landfilling this material are minimal. Some flies were
detected in warm weather where the material was landfilled
and a slight but unobjectionable odor is detectable. No
negative environmental impact exists because of landfilling
this material.
5.5 NOISE
Systech performed a noise survey of the plant during normal
operations to determine the noise level in the glass plant.
The Occupational Safety and Health Act requires that no
employee be exposed to noise levels greater than 90 dBA for
a period 8 or more hours a day. As the noise level increases,
the allowable exposure is reduced.
The glass plant has noise levels in excess of 90 dBA. The
average noise level was approximately 94 dBA. Thus, personnel
cannot work in the Glass Recovery Plant full time without
exposure to levels exceeding Federal standards.
63
-------
A second generation plant could easily make provision for
reducing the noise levels. This could be accomplished
through the use of sound-absorbing materials and partial
enclosures about some of the noisy items (air ejectors and
optical sorters).
A major source of the noise in the glass recovery system is
the operation of the vibratory conveyors and bucket elevators
It would also be necessary in a new plant to enclose the
bucket elevators and the vibratory conveyors to achieve
compliance with Federal noise level standards. Another
approach to noise control would be to replace the bucket
elevators with another type of conveyor.
5.6 COMMENTS ON INDUSTRIAL HYGIENE
5.6.1 In-Plant Dust
The dust collection system at the Franklin Glass Recovery
Plant appears to function efficiently and, as a result, no
appreciable in-plant dust problem exists. Dust is removed
from the kiln dryer, electrostatic separator, transparency
sorter and color sorter, so that dust accumulations do not
occur. If a less efficient dust collection system was used,
64
-------
dust problems could be significant. Excess glass dust in
the air could contribute to silicosis and other respiratory
problems.
5.6.2 Bacteriological Comments
In a previous study, samples were taken of several streams
within the pulping and separation system and the Whitewater
system and analyzed for bacterial contamination. Of these,
only Whitewater passes to the glass plant. The Whitewater
was found to contain a total coliform count of 2.0 x 108/
100 ml, fecal coliform count of 13 x 108/100 ml, and a total
plate count of 0.28 x 109/100 ml. The input to the glass
plant coming from the liquid cyclone is saturated with
Whitewater, and thus is highly contaminated from a bacteri-
ological viewpoint.
Water added at the washing/sizing screens is city water.
The water used by the heavy media separator and the jig is
process water. Except for the jig (most of its water is
recycled to the jig), this water is collected in the sump
and used as makeup water for the pulping and separation
system. Thus, there is no water effluent, per se, from the
glass plant to the outside environment.
65
-------
The process streams within the plant and recovered products
are biologically contaminated and workers who handle this
material should do so only when required. They should wash
or shower appropriately after handling the material.
5.6.3 In-Plant Odor
No appreciable odor is present in the glass recovery system.
The material processed by the glass plant is washed in the
washing and sizing screen at the front-end of the plant.
This removes the majority of the putrescible matter which
is on the surface of the glass particles. In addition, the
material down stream of the drying operation is of a
relatively dry nature and does not produce an appreciable
odor. The wet material input to the glass plant, if allowed
to stand for a length of time, develops a sharp, rancid odor.
Mold growths will also develop. This is not a problem with
the drier material in the following sections of the plant
or a plant operating in a continuous manner. In general,
odor is not a significant problem in the Glass Recovery Plant,
66
-------
6.0 ECONOMIC ASSESSMENT
6.1 INTRODUCTION
The Franklin Solid Waste and Fiber Recovery Plant is a
facility designed to demonstrate wet processing of solid
waste for disposal and recovery of some materials. It must
be remembered that the plant was designed and constructed
for demonstration purposes. As such, the glass plant module
of the Franklin Solid Waste and Fiber Recovery Plant has
undergone extensive modifications, additions, and changes
during the past 2 years, to improve reliability and product
quality. Hence, the economics of the glass plant at Franklin
are not truly representative of glass recovery economics.
Many of the changes, additions, and modifications that have
been made at the glass plant, tend to skew the overall
economics of the system. Since work to accomplish these
changes has been classified as maintenance activities for
accounting purposes, the modification related labor and the
actual maintenance labor are not distinguishable. Hence,
an assessment of the real economics of the glass plant in
Franklin required definition of a system that would yield
the desired recovered products and then obtaining pertinent
economic data for such a system. This data was scaled from
the Franklin plant to 500 and 1000 TPD plants.
67
-------
Because of the volatile nature of the Franklin glass plant
configuration, most of the subsequently reported economic
projections were developed using engineering cost accounting
procedures. Considerable effort was expended to assure
that the results are representative of an operating plant and
not simply this developmental experience. To accomplish
the economic evaluation, it was necessary to develop the
costs associated with both the construction and operation
of the glass plant module. The glass plant includes the
unit operations discussed in Section 2.0.
The data required to perform the economic evaluation was
derived from information provided by vendors, the U.S. EPA,
Black-Clawson, GCMI, and the City of Franklin.
It must be remembered that the glass plant is indeed a module
that must be attached to a separation system which yields
a heavies fraction. When evaluating the economics of the
glass plant, it is important to simultaneously evaluate the
economics of the pulping and separation system with the glass
plant module. This is appropriate because the glass plant
requires a specific type of feed that can be obtained from
the pulping and separation system. This report presents the
economic analysis of the glass plant module by itself and
also couples this data with an integrated disposal facility
producing refuse derived fuel (RDF).
68
-------
6.2 APPROACH TO DEVELOPING THE ECONOMIC DATA
The basis of the economic analysis is the operating experience
gained at the Franklin Glass Plant. Where possible, data
incorporated in the base is representative of the actual
operating experience at Franklin. The cost factors used to
quantify the Franklin operating experience included the
following:
a) revenues - non-magnetic metals, magnetic metals,
and glass sales.
b) operating costs - operating labor, maintenance
labor, operating supplies, maintenance materials,
utilities (fuel, power, water, etc.), and other
expenses.
c) facility expense - amortization and interest.
The objective of the economic analysis is to use the data
gathered from the Franklin experience to help project the
economics of larger commercial plants. No attempt will be
made to show the economics of the Franklin facility because
it is a demonstration plant. Hence, it has not been operated
to maximize output and minimize cost. For example, there
has been no continuous sale of glass or aluminum from the
operation of the Franklin Glass Plant. Plant changes have
69
-------
been made continuously throughout the last two years to
upgrade the quality of the recovered products. Hence, to
make an economic evaluation of this facility would be in-
appropriate and misleading since it has inherently no chance
of being economic. Instead, the technical and economic data
available from the operation of this facility will be used
to project commercial facility configuration and cost for
500 and 1000 TPD plants.
6.3 DEVELOPMENT OF COST DATA AND PROJECTED ECONOMICS
In order to have a complete understanding of all factors
comprising the income and expense for the operation of the
Franklin plant and the projections to 500 and 1000 TPD,
the following sections describe the income and expense
categories used. The cyclone reject inputs to the glass
plant will be 53.5 TPD and 107 TPD, respectively for the
500 and 1000 TPD plant size.
6.3.1 Income
Income that can be realized form the operation of the
Franklin Glass Plant are from the sale of three commodities:
magnetic sales, non-magnetic sales (aluminum), and glass
(see Table 6.1). A potential credit also exists because the
amount of material taken to the landfill for disposal is
reduced.
70
-------
INCOME SOURCE
TABLE 6.1 INCOME SOURCES
PERCENT OF MATERIAL
IN RAW REFUSE
(TIPPING FLOOR)
PERCENT RECOVERED
IN GLASS RECOVERY
SYSTEM
Glass
Flint
Amber
Green
6
2
1
.0
.2
.3
50**
50
50
Magnetic Metals
1.0 (9.8) *
10
Aluminum
0.9
72
*Remaining fraction of the magnetics in the solid waste is
recovered in the pulping and separation system.
**Based on recovering approximately 50% of input glass. Actual
operating experience showed this to be lower. However, it
is estimated that new operating procedures and equipment
changes will yield a 50% recovery of input glass or higher.
TABLE 6.2 REVENUE FACTORS
INCOME
SOURCE
Glass
Flint
Amber
Green
TONS MATERIAL
RECOVERED/
PER TON INPUT
(Tipping Floor)
.03 T/T
.011 T/T
.0065 T/T
ASSUMED MARKET VALUE
IN DOLLARS PER TON
(Early 1976)
$ 20.00/T
$ 20.00
$ 20.00
INCOME PER
TON INPUT
(Tipping
Floor)
$ .60
$ .22
$ .13
Magnetic
.001 T/T
$ 25.00
$ .025
Aluminum
.0065 T/T
$ 300.00
$1.95
71
-------
The assumed market values indicated in Table 6.2 for the
aluminum, magnetics, and glass are based upon quotations
received by Black-Clawson from sources who have agreed to
purchase these materials from their Hempstead (Long Island)
facility which is currently under construction. This facility
will include design changes that are expected to improve the
purity of the recovered glass and aluminum compared to the
Franklin products. These market values were substantiated
by contacts with other sources made by SYSTECH during this
study as well as during a separate study being performed
for the Navy's Civil Engineering Laboratory3. All commodity
purchase prices are F.O.B. the plant and consequently assume
a user within 500 miles of the plant.
The revenue factors were based on an average percentage input
of 9.8 percent for magnetics, 0.9 percent for aluminum, and
9.5 percent for glass in the received refuse (tipping floor).
These fractions are typical of Franklin Refuse and the
procedure outlined in Tables 6.1 and 6.2 should be followed
to revise the revenue projections for the solid waste
characteristics in other areas. By using tonnage based
revenue factors, they remain constant for all plant sizes.
Hence, they can be used directly in the 500 and 1000 TPD
plant economics.
3Rigo, H.G. and Hausfeld, B.A., Development of Alternative
Approaches to a Small Scale Solid Waste Transfer/Resource
Recovery Station for Navy Installations, Contract No.
N68305-76-C-0025, Civil Engineering Laboratory, Port Hueneme,
California, 1977 (in press).
72
-------
6.3.2 Description Of 500 And 1000T/D Glass Plant
Before projecting costs of 500 and 1000 TPD (Tipping Floor)
glass plants, it is necessary to determine the processing
equipment required by these glass plants. The required
processing equipment was determined in the following manner:
1) unit operations data gathered at Franklin formed
a basis
2) needed improvements were identified and translated
into hardware changes using literature and vendor
information
3) consultation with Black-Clawson representatives
was coupled with our observations.
It must be recognized that many components projected to be
installed in large glass plants are bigger than any heretofore
operated. Hence, cost and performance projections are based
heavily on vendor and Franklin experience.
It is known that several pieces of equipment sized in
Figure 6.1 and 6.2 is used in minerals and minimg industries,
and is frequently capable of processing tonnages much in
excess of that indicated. Equipment that is based upon
Franklin performance data includes the jig, color sorter,
and opacity sorter.
73
-------
53.5T
STORAGE
I.2T
REJECTS r SCREEN
6.9
") MAGNET
4.3
42.3
35.4
HEAVY MEDIA
SEPARATOR
31.1
J
IG
1
Jl
X
G
>
2ii
B|
29
DRYER
DRYER
2.1
3.7
HTES
"* 3 ROLLS
REJECTS ^
1.6 *
1 k
i ~—
OWCITY
SORTER
1
COLOR
SORTER
1 1
AMBER GRE
5.5T 3.
5"
r
•EN
2T
-25.3
i
OPACITY
SORTER
1 23.7
COL
SON
/•
.
.OR
TER
\
r
i
i
FLINT
i
COLOR
SORTER
5T
ALUM REJECTS
NOTE: THRUPUT RATES
FOR MAJOR EQUIP-
MENT ITEMS ARE
PRESENTED IN
TABLE 6.5
FIGURE 6.1
PROJECTED FACILITY DESCRIPTION FOR A FRANKLIN
GLASS PLANT SIZED TO PROCESS FEED FROH
500 TPD PLANT
74
-------
STORAGE
I07T
SCREEN
22.4 „„ 0
| 84-6
l3.8rTMA^EFQ
8.6
J70.8
HEAVY MEDIA
SEPARATOR
ALUM REJECTS
AMBER GREEN
II.OT 6.4T
NOTE : THRUPUT RATES
FOR MAJOR EQUIP-
MENT ITEMS ARE
PRESENTED IN
TABLE 6.6
FIGURE 6.2
PROJECTED FACILITY DESCRIPTION FOP. A FRANKLIN
GLASS PLANT SIZED ^O PROCESS FEED FROM A
1000 TPD PLANT
75
-------
The large plants are sized to process the glass plant feed
in two shifts or approximately 12 to 14 hours.
No attempt was made to use the thru-put rates of the Franklin
Glass Plant to size the equipment required for the 500 and
1000 TPD facilities. The Franklin plant had undersized
equipment early in the processing line that adversely effected
the thru-put of the entire system. Hence, no meaningful
thru-put rates could be established for many of the components
installed at Franklin. Hence, manufacturers data and the
literature became a major technique for estimating the equip-
ment for the entire process line.
6.3.3 Operating Expense Items
The following categories were developed and used for deriving
the operating expenses for a Franklin Glass Plant.
6.3.3.1 Operating Labor
This category includes the personnel required for the operation
of a Glass Recovery Plant and projects the operating labor
force needed at a glass plant coupled to a facility receiving
500 and 1000 TPD (Tipping Floor) of solid waste. Table 6.3
summarizes the development of this data. The total operating
man-hours (MH) required per day by each of the facilities
are:
76
-------
Franklin 50 TPD - 26 MH
Franklin 500 TPD - 68 MH
Franklin 1000 TPD - 68 MH
These man-hour projections and appropriate salary rates can
be used to project operating labor costs for each size
facility. It should be noted that the proposed facilities
for 500 or 1000 TPD operations will require two shifts to
process the material. As can be seen in Table 6.3, the
operating labor force for the 500 and 1000 TPD plant is
identical. The reason for this is that operational tasks
are identical for both size facilities and there is not a
significant difference in the amount of equipment that must
be operated for the processing of the material. The data
for the 50 TPD plant was the actual operating experience at
Franklin.
TABLE 6.3 OPERATING LABOR
2 Shift Total 2 Shift Total
Franklin 50 TPD 500 TPD 1000 TPD
Supervision
Chief Operator
Assistant Operator
General Labor
Office
\
1
1
1
0
%
1
2
4
1
h
l
2
4
1
77
-------
6.3.3.2 Maintenance Labor
Very little data has been recorded for the maintenance
requirements of the Glass Plant as it exists at Franklin.
The reason for this is that the plant has been significantly
modified many times over the last two years and it is
impossible to separate actual maintenance costs from the
cost of changes in equipment that have been performed by
maintenance personnel. Table 6.4 presents an estimate of
the maintenance man-hours associated with the operation of
the Franklin plant in the evaluated configuration and also
projects the maintenance labor requirement to 500 and 1000 TPD
plants (Tipping Floor). Table 6.4 is a projection of the
maintenance hours required to maintain these facilities
based upon equipment similar to that being used at Franklin
and assuming that the maintenance problems observed for this
equipment will continue but that obvious improvements will
have been made.
TABLE 6.4 MAINTENANCE LABOR
Chief Mechanic
Helper
FRANKLIN
50 TPD
\
\
FRANKLIN
500 TPD
h
l
FRANKLIN
1000 TPD
1
1
78
-------
6.3.3.3 Operating Supplies
This category reflects all the cost for the operating supplies
and includes all consumables required to operate the Glass
Plant. Since no record for the Franklin Glass Plant has
been maintained at Franklin, this factor had to be estimated
($.03/T).
6.3.3.4 Maintenance Supplies
Again, no records exist at Franklin that adequately describe
the glass plant maintenance supply cost. However, during the
evaluation period, it was observed that little maintenance
was actually required in the glass plant.
Maintenance supplies are often estimated at 3 to 6 percent
of the installed equipment costs. It is believed that a
5 percent factor would conservatively indicate the appropriate
maintenance supply costs.
6.3.3.5 Utilities
Power - In Section 4.2.3 it was shown that approximately
26 Kwh are required to operate the glass plant per ton
input to the tipping floor. This cost factor can be
applied directly to the power costs required for larger
plant sizes.
79
-------
Fuel - Fuel oil is used to heat the building housing
the glass plant facility. No technique was available
for making a direct measurement of the building heat
fuel requirements. Since the plant area approximates
that of the fiber recovery module, the cost factor
derived for the fiber recovery module (.09 gallon/ton)
will be used. **
Additional fuel oil is required to operate the dryers.
The type and size of the dryer can vary depending upon
the material and thru-put desired.
The type used for this analysis was the rotary drum
(same as used at Franklin). The fuel estimates were
based upon vendor information supported by literature
describing the operational requirements for rotary
drums.
Water - Although some water is consumed in the washing
and screening operations within the glass plant, it
becomes part of the Whitewater system which is used
as a feed to the pulper. Hence, no charge will be
levied against the glass plant for water consumption.
If the user of the data wants to estimate the charge
"Wittmann, T.J., et al, ibid.
30
-------
for water usage, if the plant were attached to another
process, he should refer to Section 4.2.3 for water
consumption data and use his local water supply and
treatment costs to estimate this operating cost factor.
6.3.3.6 Land Disposal
With the glass plant module, some material that would normally
be landfilled is recovered. That is, approximately 43 percent
of the cyclone reject stream is recovered as magnetics,
aluminum, and glass.
Hence, the rejects from the glass plant that require land-
filling will not be defined as an operating costs to the
glass plant, but for purposes of this evaluation will be
assumed as part of the operating cost of the pulping and
separation system. However, if a potential user desires
to levy a charge against the landfilled material from the
glass plant, it can be based upon the fact that 6.5 percent
of the refuse received at the tipping floor is landfilled
from the glass plant. A credit for reduced landfill require-
ments could also be developed if residue disposal were a
significant total cost element.
81
-------
6.3.3.7 Miscellaneous Expenses
This category includes all other expenses that are required
for operating the plant. They include insurance, accounting,
taxes, security, etc. Since records at Franklin do not
separate the miscellaneous expenses for the glass plant from
the total plant account, it is estimated that the expense
will be approximately $20,000/year, e.g., a proportional
amount of the total facility expense.
6.3.4 Facility Expense Charge
In order to develop a full economic evaluation, costs for
each of the two larger facilities must be determined so
that appropriate amortization and interest charges, depreciation
or other related facility expenses can be derived.
No attempt will be made to present the total capital costs
of the Franklin Glass Plant because, for the reasons detailed
previously, no economic evaluation of the Franklin operation
was made. If the reader desires to know more about the costs
of the Franklin Glass Plant, he should refer to the report,
"Glass and Aluminum Recovery Demonstration Subsystem for
the City of Franklin, Ohio", prepared by John P. Cummings,
dated November 22, 1976, Owens - Illinois, Glass Container
Division, Toledo, Ohio.
82
-------
The cost for the 500 and 1000 TPD (Tipping Floor) glass
plant module was based on the process as modeled in Franklin
and displayed in Figures 6.1 and 6.2. Equipment costs were
developed from manufacturers literature and construction
costs were estimated using Mean's5 and Richardson6 construction
cost estimating handbooks. Tables 6.5 and 6.6 tabulate the
estimated costs for 500 and 1000 TPD glass plant modules.
The total facility costs for the 500 and 1000 TPD (Tipping
Floor) Franklin Glass Plant are $1,442,000 and $2,531,000,
respectively. However, there are other factors that contribute
to the total cost of the facility. For purposes of the
analysis, it will be assumed that the facility will be owned
by a municipality and that 15 year bonds at 8 percent interest
would be issued to finance the facility. Furthermore, it
is known that the total cost of the facility must include
monies to cover a "Debt Service Fund", a "Debt Revenue Fund",
and the "Bond Finance Charges". These items cause the total
bond issue to be approximately 30 percent higher than the
actual facility cost.7 Hence, the total capital and bond
costs are as follows for the 500 and 1000 TPD (Tipping Floor)
Glass Plant Facilities.
5Building Construction Cost Data, 35th Annual Edition
Robert: Snow Means Company, Inc., Duxbury, Mass.
^Process Plant Construction Estimating Standards, 1976,
Richardson Engineering Services, Inc., Solena Beach, California,
7Wittmann, T.J., et al, ibid.
83
-------
TABLE 6.5 CAPITAL COSTS FOR 500 T/D
FRANKLIN GLASS PLANT
(PROCESS DESCRIBED IN FIGURE 6.1)
Major Equipment Items and Cost for Processing Input To Glass
Plant From 500 T/D (Tipping Floor) Facility
1 - Screen - Vibro Energy Separator
- 8' D - Est. Capacity 5T/HR -------- 40,000
1 - Magnetic Separator ________ 12,000
1 - Heavy Media Separator 4'D x 4'L
- Est. Capacity 4T/HR ______ 100,000
2. - Jig 2 Section Jig - 26" x 26"
.5T/HR/sq.ft. ________ 40,000
1 - Dryer Rotary Kiln 4'0 x 30'
- Est. Capacity 3T/HR ________ 100,000
1 - Dryer Rotary Kiln 2'0 x 8'
- Est. Capacity 0.4T/HR - _____ 50,000
1 - HTES - 3 Rolls 1.75D Roll
- CAP. 1500#/HR/FT Of Rotor Length - - 120,000
1 - HTES - 1 Roll I'D Roll
- Cap 300#/HR/FT of Rotor Length- ----- 40,000
2 - Opacity Sorters - 6 Channel-400#/HR/Channel- - - - - 60,000
3 - Color Sorters - 6 Channel-400#/HR/Channel- _____ 90,000
- Conveyors, Silo's for Storage of Recovered
Products, Input Storage, etc. (Estimated)- ----- 200,000
- Building-5000 sq.ft. x 30 Ft. Hight, including building
electrical and plumbing (16.00/ft2) -------- 80,000
- Electrical and Plumbing for Process Equipment- - - - 60,000
- Installation of Above Equipment
Estimated Cost ________ 200,000
$1,192,000
Engineering Costs 10% of facility
Cost _____ 119,000
1,311,000
Start-Up Costs 10% of Total _______ 131,00
$1,442,000
Note - No Land Cost Included
84
-------
TABLE 6.6 CAPITAL COSTS FOR 1000 T/D FRANKLIN
GLASS PLANT (PROC2SS DESCRIBED IN FIGURE 6.2)
Major Equipment Items and Cost for Processing Input to Glass Plant
From 1000 T/D (Tipping Floor) Facility
2 - Screen - Vibro Energy Separator
- 8'D Est. Capacity 5T/HR -------- 80,000
1 - Magnetic Separator _---____ 12,000
1 - Heavy Media Separator 6'D x 5" Long
- Est. Capacity 6T/HR _-____-_ 150,000
2 - Jig - 2 Section 36" x 36" 5T/HR/ft2 - 80,000
2 - Dryers - Rotary Kiln 4'D x 30'
- Est. Capacity 3T/HR ___--___ 200,000
1 - Dryer - Rotary Kiln 2'D x 81
- Est. Capacity 0.4T/HR --____-- 50,000
1 - HTES - 3 Roll - 2.5'D Roll - 1500#/HR/ft of
Rotor Length ________ 160,000
1 - HTES - 1 Roll - I'D Roll - 300#/HR/ft of
Rotor Length ________ 40,000
3 - Opacity Sorters - 6 Channel-400#/HR/Channel- - - - - 90,000
5 - Color Sorters - 6 Channel-400#/HR/Channel- ----- 150,000
- Conveyors, Silos for Storage of Recovered Products,
Input Storage, etc. (Estimated) ________ 400,000
- Building - 10,000 sq.ft. x 30 FT High including
building electrical and plumbing (16.00/ftz) - - - - 160,000
- Electrical and Plumbing for Process Equipment- - - - 120,000
- Installion of Above Equipment (Estimated)- ----- 400,OOP
$2,092,000
Engineering Costs 10% of Facility --______ 209,OOP
$2,301,000
Start-up Costs 10% of Total ________ 230,000
TOTAL CAPITAL COSTS ------- -$2,531,000
NOTE: No land cost included.
85
-------
CAPITAL AND BOND COSTS
500 TPD 1000 TPD
Project Cost $1,442,000 $2,531,000
Total Bond Issue $1,874,000 $3,290,000
Annual A & I (11.683 percent)* $ 218,900 $ 384,300
*Based upon 15 years and 8 percent (a Capital Recovery
Factor of .11683) .
6.3.5 Estimated Economics For A Franklin Glass Plant To
Service 500 and 1000 TPD (Tipping Floor) Facilities
6.3.5.1 introduction
Before presenting the estimated operating costs for the glass
plant modules servicing a 500 and 1000 TPD (Tipping Floor)
facility, some discussion about the facilities and its
operation is necessary.
Each of the facilities are to be operated such that all
received products and/or landfilled material will be conveyed
to storage bins for easy removal by a contractor. This is
done to reduce the labor required for material handling.
Each of the facilities will operate two shifts with maintenance
being performed on the third shift. The glass plant modules
for the 500 and 1000 TPD facilities will require approximately
5000 and 10,000 sq. ft. of building space, respectively.
86
-------
Much information has been gained from the operation of the
Franklin Glass Plant regarding the processing of material
to achieve a better quality product. This information will
be used extensively in the design of the larger glass plants.
The recovered aluminum at Franklin has too high a contaminant
level to be of much commercial interest. It was determined
that by drying the aluminum rich fraction from the jigging
operation and running the material through an electrostatic
separator, a higher quality aluminum product could be
achieved. Hence, new facilities should incorporate an
additional dryer and electrostatic separator for the aluminum
recovery operation.
Recent changes, made after our evaluation, have included
new optics in the color sorters to allow color sorting of
smaller glass particles. The glass sorters during the
evaluation required particles to be \ in. The new optics
will permit sorting of particles down to 1/8 in. This
modification will increase the glass yield; however, no
yield estimate can be made until performance tests are
conducted. The revenues from glass in this analysis were
based upon a 50 percent yield (a 4 percent improvement),
although Black-Claeson anticipates even higher yields.
87
-------
The cost estimates presented for the 500 and 1000 TPD plants
do not reflect the cost of land. This value varies con-
siderably and, hence, was not considered in this analysis.
All labor rates and other schedules for determining the total
category costs are presented in the notes for each projected
facility cost.
6.3.5.2 Estimated Economics For 500 TPD Franklin
Glass Plant
Table 6.7 summarizes the estimated costs for the operation
of a 500 TPD Franklin Glass Plant.
Income from the sale of recovered material amounts to
$2.93/T (Tipping Floor). Operating expenses are 2.19/T
and the facility expense is $1.70/T. Hence, the net loss
from the operation is $.96/T). An analysis for the recovery
of magnetics and aluminum only also show a non-profitable
operation.
6.3.5.3 Estimated Economics For 1000 TPD Franklin
Glass Plant
Table 6.8 summarizes an estimated cost summary for the operation
of a upgraded 1000 TPD Franklin Glass Plant.
88
-------
Q
CU
EH
O
O
Q
W
U
§
O O 0 O O
fl> 0) 0) Hi 4)
33 333
IT) (0
(0 (0 «
0) 0)
O
IH
C
0
u
c
0
•a
0)
0)
as
E
O
(4H
01
Ul
10
rH
O
f]
4J
• rH
3
4J
O
(0
4-1
c
0
u
• c
~ 0
4-1
CJ 'O
10 CU
rH 01
Cu (0
CQ
^
o
<+H
c
o
•H
4-1
(0
a
u
o
u
r4
(I)
C
•rH
ID
C
O
o
o
CK
"- —
•0
ro
01
4J
0)
cu
0)
33
E
0
MH .
.— .
01 4-1
01 C
ffl 10
rH rH
O CU
J>,
V-i
CD
O
U
y
(0
fa
•
^-^,
C*P
00
4-)
rd
CO
CQ OQ
O
Z
w
EH
tn
o
CJ
w
cn
EH EH EH
EH EH O
n p- '
rH rH
CO-
Z
O
o o
o o
Cfl CN
VO V£
CN ON
O
(N
O O O
o o o
o m r-
U1 VO O\
ooooooooo
ooooooooo
I
U
CQ
EH
ui m
0) 01
rH rH
(0 ro Ul
cn w cu
rH
U E fl
H 3 CO 4J in C
4J C COO)
0) -H 01 -H X) 01
M C E 01 -H E M
£ CJ* 3 iD Urf C
IH O'HOUO) rHlOrH
ai v-i ja rH a 3 cuai-H
&0 rOa3O 3Sr<
o £> t-5 a.tn a. b
<0 3
4J rj cu cn oi i i
c ou
10 rH C CT C 01
4-1 rg 01 (0 C ID 0)
01 rH O C -H C -
-------
O
in
o
o
o
O
K
&4
EH
EH\
O O
O •
• o
in o
CM n
«>••
<4H <4-l
0 0
0) 01
3 3
r-l r-l
m 10
> >
0) 0)
rH rH
10 fO
to to
c c
0 O
•c n
o> oi
Ul (A
10 (0
03 03
EH EH EH
Ul
o> i« tn tn in tn
c^ ^y d) 0} o) o
c o^ o^ tr* CT* CP
•H C C C C C
!_, .H -^ -H -H -H
fit, fi^ [|i^ Jj^ Q^
ON
C &* c^ D^ o^ Cr*
•H C C C C C
*0 -H 'H 'H !H *H
rH 3 3 3 3 3
(J rH rH i-H i-l i-H
C U O U CJ U
M c e c c c
in o o o o o
• in m o o o
o
rH ON oo r- vo ON
c c c c c c
O O O 0 O 0
^0 ^3 *O *O f-0 *O
ON
ui c
0 -H
O 4J
«S
4J IH
C 01
a
3 M
crx:
•o 3
oi is
rH
rH O
(0 <0
4J i-H
(0
C 0) JC
H O> 3
*P IA V.
m 3 o
m
C fi •
0 0 *H TJ
M 10
4-> (0 0)
C T) 4J
O r-l (0
CJ 0 0.
C E
C >i 01
O 0> X
•0 E
01 T3 0
Ul C IH
Ifl (0 tH
OQ
V)
W
fT3 M CD
rH 0 O
O
01 CO
^
O s>
U
i-H U) l»H
04 (0 O
OQ
to
H
CO
O
CJ
EH EH EH
EH
\
O
CO
C/l
3
2
M
2:
o
a
00
•
vo
U
J
CQ
O O O O O
o o o o o
r^ in o o m
p"i (N O ro ON
in ON cr> i** f^*
rH tS V CM
^* rH
CO-
Ul Ul
01 01
rH rH
(0 iO Ul
to to o)
rH
O E * 4J W C
•rl 3 tO C 01 0)
4J C -H fl 0)
01 -H tn rH E >H
u c E tn &-• < cj
Z ON 3 <0
O TJ ^H ^^
U Z < 0
2
M
O OOOOOOOOO OOO
O OOOOOOOOO OOO
r-- oinoomomoo om^n
CO VOOrHr^it^rHTTinCM 'gOtN'N
oo oomot^oovoor- OOOM^S
CM rHn^trHCIOOrH POrO
V£
1
Vl Ul rH O>
O 0) -H C
4-i tn -H o -IH in
>0 IH 01 rH IH 4-i C
InkH O-rlCUO) rHIOrH
kj OOllH flr-4Q,3 OlOI'iH
0 4J£XiO lOOiaO 3Z«U1
COflCfOOfl t-4CUtOCU EIH 01
W <0 0 IH fl 3 Ul
tO>J-HO)4JrJ OICA01I 1 C
z in a. c u u oi
WCT>"HOlOrH CO">CU1 £1
04C> 4JiOOIiOCn30) X
X-iHlH>»HUlUCJC-rlC-r4 U
U4-IOI01--HOI-HOI4JOI4-I
(0 CU--H U1C«H4-»I04J-H IH
cj)u3x:tnoil*HCiHCrH oi
20)tOCJi
>4 O rH
EH 01 r-* A
M -p-l «3 4-1
r3 O 4-1 C
M In O O
CJ O< EH Z
1^
fcj
OOO
OOO
r- in o
OO i-H O
00 ON O
i— 1 rH CM
^O fl fl
CO-
01
Ul
c
0)
a
X
X W
IX
< in >
Z 01 4-1
Z 0) Ul --H
3 e C rH
to O 0) -H
u cu o
EH C X (0
O3 M a &u
o
o
._
o
.
CO
-------
Income from recovered materials are the same as in the
500 TPD case and amount to $2.93 (Tipping Floor). Operating
expenses are lower, however, due to no increase in labor
and amount to $1.48/T. The facility expense also falls on
a per ton basis to $1.49/T. Hence, the net loss from
operation of the plant is ($.04/T).
The recovery of aluminum accounts for 67 percent of the
total revenue. By installing only the equipment necessary
to recovery the magnetics and aluminum, it can be shown
that an aluminum recovery plant servicing a 1000 TPD facility
can yield a pre tax profit of $.44/T. Hence, it appears
that magnetic and aluminum recovery would be more attractive
than adding glass recovery.
6.4 COST SUMMARY OF FRANKLIN SOLID WASTE FACILITY
Table 6.9 presents an economic summary of the operation of
a Franklin Solid Waste Plant projected to 500 and 1000 TPD.
The facility costs have been adjusted to reflect early 1976
prices. This summary sheet is presented to show the complete
operation of a Franklin type facility. The summary is
presented for a plant that does not recover fiber but rather
it sells the rejects as a fuel. The projection includes the
recovery of magnetics, glass, and aluminum. This assessment
91
-------
TABLE 6.9 ECONOMIC SUMMARY OF PROJECTED FRANKLIN
PLANT WITH NO FIBER RECOVERY, WITH GLASS
PLANT, AND SELLING REJECTS AS A FUEL
FRANKLIN PROJECTED FRANKLIN PROJECTED
COSTS TO 500 TPD TO 1000 TPD
INCOME*
Pulping And Separation
Magnetic Metals
Sludge Disposal
Fuel
Glass Plant
Magnetics
Aluminum
Glass
Total Income
OPERATING EXPENSES
Pulping And Separation
Glass Plant
FACILITY EXPENSE**
Pulping And Separation
Glass Plant
$ 2.40/T
1.75/T
10.71/T
.02/T
1.95/T
.95/T
$17.78/T
$ 6.04/T
2.19/T
$ 8.23/T
$ 6.57/T
1.70/T
$ 8.27/T
$ 2.40/T
1.75/T
10.71/T
.02/T
1.95/T
.95/T
$17
$ 5.50/T
1.48/T
$ 6.98/T
$ 6.12/T
1.49/T
$ 7.61/T
Total Expenses $16.50/T
NET SAVINGS $ 1.28/T $ 3.19/T
*Based On Following Revenue Factors
Magnetic Metals - $25.00/T
Sludge Disposal At A Rate Of .07/T Input - $25.00/T
Fuel - $1.25/MBtu
Aluminum - $300.00/T
Glass - Flint, Amber And Green - Each At $20.00/T
**Based On Following Facility Costs Projected To 1976 Prices.
500 TPD 1000 TPD
Weighing & Receiving, Pulping,
Separation And Dewatering $6,524,000 $12,145,000
Glass Plant 1,442,000 2,531,000
TOTAL PLANT COST $7,966,000 $14,676,000
92
-------
is presented because the glass plant cannot exist without
some front-end system preparing the feed to the glass plant.
The economic data for the front-end system was taken from
the data reported and collected in Wittmann, et al.8
The net savings for this type of operation is $1.28/T for
a 500 TPD plant and $3.19/T for a 1000 TPD plant. The use
of the fiber as a fuel is necessary for a facility of this
type to be financially sound. Sixty percent (60 percent)
of the income is derived from the sale of the fuel product.
The data presented in Table 6.9 includes no tipping fee,
hence, the net operating costs can be compared directly to
existing disposal costs with the following exception. It
should be noted that for these examples, no charge is made
for the land on which the facility is to be established.
Individual communities may readily approximate the cost of
land locally and add the appropriate amortized capital cost
of the land to the facility expense charge to obtain a
specific projection for their community.
6.5 ECONOMIC OBSERVATIONS
Industry and municipalities use different techniques to
assess the merits of an investment. What may not be
attractive to industry could be attractive to a municipality,
Wittmann, et al, ibid.
93
-------
The following discussion presents an appraisal concept used
by industry to determine the merits of a potential investment.
This approach was used because it is our opinion that most
municipalities are looking for turn-key operation of an energy/
resource recovery system with industries providing the capital
to protect municipal banding limits.
One of the techniques used by industry to determine the merit
of an investment is to use a term called Annual Return On
Investment (AROI). This term, depending upon the industry,
can be defined in many different ways. For purposes of
our discussion here, AROI will be defined as follows:
Gross Profit (Year)
Total Investment
Analyzing a complete Franklin type facility projected to
500 and 1000 TPD with a glass plant and the sale of the
fiber as a fuel (RDF), the following table was derived.
TIPPING
FEE
500
1000
TPD
TPD
0
0
.0
.0
GROSS
SAVINGS
$165,
$823,
000
000
TOTAL BEFORE TAX
INVESTMENT AROI
$ 7
$14
,966
,676
,000
,000
2
5
.1%
.6%
94
-------
As indicated, an AROI for the 1000 TPD plant size is only
5.6% before tax, a very unacceptable AROI for industry. It
should be remembered that the data is based on net savings
which does not include a tipping fee. As an illustration,
if a $10.00 tipping fee were included in the above analysis,
a before tax AROI of 23% is realized. The AROI now approaches
a level where industry would become interested if the risks
are small.
95
-------
APPENDIX A
PICTORIAL FLOW DIAGRAMS
This section presents a pictorial flow diagram with which
the reader can visualize the various equipments used in the
Franklin Glass Plant. Section 3, Operational Capability
describes the process flow and can be used to supplement
this pictorial presentation.
96
-------
'!) Storage Vessel For
Feed To Glass Plant
(2) Rotary Screen,
Metering Hopper,
And Bucket Elevator
-------
[3) Side View Of Heavy Media Separator
(4) End View Of Heavy Media Separator
98
-------
(b) Vibrating Conveyor From
Heavy Media Separator
To Jig
(6) Aluminum Recovery Jig
99
-------
Close-up Of Jig
Separator
Dewatering Conveyor (Electrostatic
Separators Shown In Background)
100
-------
Druiii Dryej.
[ID) Outfrai: ..r Elect-roritdt ic :
] 0 i
[ -^ ra t ors
-------
(11) Close-up Of Electrostatic Separator
;i2) Close-up Of Optical Sorters
102
-------
(13) Front View Of Optical Sorter
-------
APPENDIX B
GLASS PROCESSING INDUSTRY SPECIFICATIONS
FOR GLASS GULLET*
Parameter
Original GPI
Specificat ion
Liquid Content
Organic Materials
Magnetic Metals
Non-Magnetic Inorganics
Non-Magnetic Metals
Flint Glass
Colored Glass
10%
0.1%
0.05%
0.1%
0.015%
Min. 95%
Max. 5%
Revised GPI
Specification
(11-9-76)
Do Drainage
0.2%
0.05%
0.1%
No Particles +\ in.
1 (one) -\ in +20 Mesh/
40 Ib.
Min. 90%
Max. 5.0% Amber
Max. 1.0% Green
Max. 0.5% Other
Refractory
+20 Mesh
-20
-40
+40 Mesh
+60 Mesh
No Specification
2 Particles/lb.
20 Particles/lb.
1 Particle/40 Ibs.
(no Particle +nt; in.
2 Particles/lb.
20 Particles/lb.,
NOTES
1. Flint glass containing over 0.1% Fe2O3 and/or 0.002%
Cr203/ by chemical analysis, shall be considered mixed
color glass.
2. Flint glass can contain up to 1% emerald green or 10%
georgia green, or a combination within these limits (1%
georgia green = 0.1% emerald green).
* REFERENCE:
Glass and Aluminum Recovery Demonstration Plant
Subsystem for the City of Franklin, Ohio, Report
by Dr. John P. Cummings, Owens-Illinois, Glass
Container Division, Toledo, Ohio. Extensive cullet
analysis was performed by Dr. Cummings and it is
recommended that if further information is desired
about the composition of the glass cullet produced
at Franklin, a copy of the above referenced report
be obtained.
104
-------