United States
Environmental Protection
Agency
Industrial Environmental
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-84-090 Sept. 1984
Project Summary
Homer City Multistream Coal
Cleaning Demonstration:
A Progress Report
D. W. Carey, S. T. Higgins, A. A. Slowik, R. D. Stoessner, C. W. Sypult,
J. H. Tice, M. E. Till, and E. A. Zawadski
This report gives an overview of
ongoing testing and evaluation of the
Homer City Coal Cleaning Plant, built to
enable the Homer City Power Complex
to meet sulfur dioxide (SO2) emission
levels mandated by the Pennsylvania
and Federal governments.
The plant was constructed as a result
of an extensive comparative evaluation
of flue gas desulfurization (FGD) and
physical coal cleaning. The Homer City
System, The Multistream Coal
Cleaning System (MCCS), was chosen
as an economical alternative to FGD.
The plant contains circuits for clean-
ing coarse, medium, and fine coals and
for recovering fine and very fine coals.
The dominant type of cleaning
equipment used in the plant is the dense
medium cyclone.
The original "93 plant" configuration
was never able to clean coal to the
conditions specified in the plant design.
An extensive test and evaluation
program was begun to identify and
correct the causes of plant operating
problems. After extensive pilot plant
equipment tests and engineering
studies were completed, recommen-
dations were made for plant modifica-
tions neccessary to correctthe design
and operating deficiencies of the plant.
Extensive modifications were made to
one of two parallel processing trains in
the plant (the "B" circuits), and a test
program was initiated to evaluate these
corrective measures.
The recently modified "B" circuits
have not yet met design conditions.
Presently, the fine and medium coal
circuits are undergoing an evaluation to
determine why the cyclones are not
operating at predicted levels and to test
further proposed corrective actions in
actual operation.
This Project Summary was developed
by EPA's Industrial Environmental
Research Laboratory, Research
Triangle Park. NC, to announce key
findings of the research project that is
fully documented in a separate report of
the same title (see Project Report order-
ing information at back).
Introduction
The Homer City Power Complex
consists of two 600 MW boiler/turbine/
generator units and one 650 MW unit, fed
primarily from two dedicated mines and
from other sources within 30 mi (48 3 km)
of the plant.
Units 1 and 2 (600 MW) at the genera-
ting station must meet an SO2 emission
limit of 3 7 lb/106 Btu (1590 ng/J)
mandated by the Pennsylvania Depart-
ment of Environmental Resources
(PA.DER). Unit 3 must comply with an
EPA New Source Performance Standard
(NSPS) limiting SO2 emission levels to
1.2 lb/106 Btu (51 6 ng/J).
To meet both emission levels using the
captive coal reserves on-site, two
compliance strategies were initially
considered- the first considered a coal
cleaning system in Units 1 and 2 and a
FGD system for Unit 3; the second, the
MCCS, involved constructing and
operating a coal preparation plant to
produce two grades of compliance coal
capable of meeting both emission regula-
tions. Evaluating both strategies led to
the conclusion that the MCCS would offer
-------
substantial capital, operating, mainte-
nance, effluent disposal, and boiler oper-
ating costs savings over FGD.
This report chronicles the history of the
MCCS coal preparation plant, describes
its design and operating problems, sum-
marizes corrective measures, and
provides some preliminary results of tests
on the modified portion of the plant.
Coal Preparation Technology
Operations and Equipment
Coal preparation or benefication is
used to remove mineral matter from coal.
Since the advent of sulfur emission
regulations, many development activities
have focused on coal desulfurization.
Several different unit operations are
used in coal preparation, including size
reduction, size classification, coal
cleaning, dewatering and drying, and pol-
lution control/waste disposal. Modern
commercial coal preparation plants crush
coal and separate the particles into
several size ranges which may be defined
as coarse, medium, or fine. Each size
fraction is usually processed in a separate
circuit using equipment suitable to the
size range and cleaning objectives.
Size reduction liberates mineral
impurities and produces the range of
sizes needed in subsequent cleaning
activities. Size reduction is mechanical:
equipment breaks the coal by impaction,
compression, splitting, shearing, or attri-
tion. Primary size reduction usually
involves rotary breakers or crushers;
second and third stages, crushers or
mills
Size classification is commonly used to
remove fines prior to size reduction a nd to
separate size ranges prior to further
processing or sale. The major
classification technique is wet or dry
screening. The most common screens
used in coal preparation are-
— Perforated plate and square-opening
wire screens that shake or vibrate.
— Curved stationary screens (sieve
bands) to separate fine and
intermediate size particles in water
slurries.
— Classifying cyclones to separate fine
coal particles from coarser fractions.
Coal is most commonly cleaned in
equipment that relies on differences in
the size, shape, and specific gravity of
particles for separating the organic and
mineral coal particles.
Jigs remove mineral matter and mining
refuse from coarse and intermediate size
coal, using hydraulic pressure to stratify a
bed of coal.
Wet concentrating (Deister) tables use
water to separate or wash intermediate
and fine size coal on a vibrating ribbed
surface.
Hydrocyclones process a slurry of
medium or fine size coal in large
quantities at relatively low cost.
Dense medium separators include
dense medium vessels and dense
medium cyclones. Both types of
separators are fed a slurry of sized coal,
water, and magnetite. The amount of
magnetite can be adjusted to control the
separating specific gravity. The vessels
are static baths in which the less dense
clean coal particles float and the heavier
refuse particles sink. The cyclones
separate coal and refuse by centrifugal
force: the heavier refuse particles are
forced to the outside of the cyclone and
are removed through the underflow
orifice at the cyclone apex; the less dense
coal particles are transported to the
center of the cyclone where they are
removed via the vortex finder as the clean
coal overflow. In addition to the vessel or
cyclones, the dense medium circuit
contains equipment for controlling the
slurry density and recovering magnetite
for reuse.
Performance Criteria
Several criteria are used to evaluate
the performance of coal cleaning
equipment and are classified according
to the degree to which they depend on the
characteristics of the coal being cleaned.
Some criteria are dependent, either
directly or indirectly, and some are, under
certain circumstances, essentially
independent. Dependent criteria include:
weight yield, Btu recovery, emission
parameter (of clean coal), emission
reduction, recovery reduction, misplaced
material, yield error, and ash error.
Independent criteria include: probable
error, error area, and imperfection.
Design of the 93 Plant
In 1975, owners of the Homer City
Power Complex decided to proceed with
the design and construction of a complex
coal cleaning system using dense
medium cyclones as the primary
component of deep coal cleaning for Unit
3 NSPS compliance.
The basic coal cleaning technology
initially implemented at Homer City was
well established; however, several new
process design features and innovations
were incorporated into the system
including:
— Raw coal crushing was controlled to
minimize production of very fine
coal.
— Dense medium cyclones, commonly
used to perform gravitational clean-
ing of sized coal, were used to
separate coal and ash at smaller
particle sizes and lower specific
gravity than normally used in com-
mercial practice.
— Scavenging equipment was incor-
porated in the fine coal processing
circuits to recover about 95 percent
of the energy in the coal fed to the
preparation plant.
— Efficient pollution control devices
and methods of residue disposal
were used to minimize the environ-
mental impact of coal cleaning. The
process water was recirculated to
the coal cleaning operation to pro-
vide a closed water circuit.
The 93 plant contained circuits for
cleaning coarse, medium, and fine coal as
well as circuits for recovering fine and
very fine coal. The dominant type of
cleaning equipment used in the plant was
the dense medium cyclone (DMC).
Hydrocyclones and concentrating
(Deister) tables were used to reclean fine
coal from the underflow of the fine-coal,
low-gravity DMC circuit. Thickeners were
used to close the water circuit and to help
recover fine and very fine coal.
The coarse coal, 1 % x Vt in. (31.8 x 6.3
mm), was processed through 24 in. (61.0
cm) diameter heavy media cyclones at 1.8
specific gravity (s.g.). The overflow from
this separation was screened to remove
water, and the resultant material was
used as a component of the middling
product for Units 1 and 2. The underflow
was discarded as refuse.
The medium coal circuit produced two
products. The 14-in. x 9 mesh (6.3 x 2.0
mm), was processed in 14 in. (35.6 cm)
s.g. The overflowof this separation (deep-
cleaned product) was split; 57 percent
was used in the deep-cleaned product, 43
percent in the middling product. The
underflow of this separation was
reprocessed at 1.8 s.g. to recover usable
coal for Units 1 and 2.
The fine coal, 9x 100 mesh (2.0x0.15
mm), was processed in 14 in. (35.6 cm)
DMC's after being reclassified to remove
any very fine material to avoid an adverse
-------
effect on product quality. The underflow
was recleaned (using hydrocyclones and
Deister tables) and used in the middling
product. Material finer than 100 mesh
(0.15 mm) was not cleaned, but by-
passed to the middling product.
Evaluation and Modification
of the 93 Plant
Because of several design and
operating problems, the plant never
produced the projected quality and
quantity of product coal. The three major
problems identified with these circuits
were:
— Poor water feed to the plant and poor
water distribution in the circuits
made it impossible to run the plant
for any sustained time and maintain
conditions needed for proper equip-
ment operation.
— The low gravity DMC circuits failed
to achieve design performance even
when design operating conditions
were attained.
— Magnetite losses in the DMC fine
coal cleaning circuits were
excessive.
To rectify these problems, a
cooperative pilot plant test program was
conducted by the Department of Energy
(DOE), EPA, and the Homer City owners.
Thirty-six tests were conducted on an 8
in. (20.3 cm) DMC with the cyclone
operated at a nominal feed rate, at 1.3s.g.
with 9 x 100 mesh (2.0 x 0.15 mm) coal.
Statistical evaluation was performed by
Bituminous Coal Research, Inc., and the
following conclusions characterize
cyclone operation at low gravity with fine
coal:
— Dense medium cyclone performance
improves as particle size of the feed
coal increases.
— Cyclone performance, as measured
by the dependent criteria and per-
cent sulfur reduction, is statistically
related to flow rate level.
— Cyclone performance, as measured
by the dependent criteria, is related to
the size of the inlet orifice.
— Dense media cyclone performance
improves as the percent of coal in the
slurry decreases.
— The cyclone operating parameters
investigated are, in general, more
highly correlated with independent
measures of cyclone performance
than with dependent or sulfur-based
criteria.
— For the test matrix under investiga-
tion, cyclone operating conditions
corresponding to a 1.5 in. (3.81 cm)
orifice size, 120 gpm (454.3 l/s)
flow rate, and 7:1 media to coal ratio
produced best overall cyclone per-
formance.
In addition to the pilot plant tests,
several cleaning plant consultants and
engineering firms were requested to
survey the existing plant and propose
corrective measures.
As a result of these efforts, the "B" side
of the 93 plant was extensively modified
and started up in 1982.
Although the coarse coal circuit
cleaning equipment was not changed, it
now processes a minimum coal size of Ve-
in. (3.15 mm) instead of 1/4-in. (6.35 mm).
The coarse coal is separated at 1.8 s.g. in
24 in. (61.0 cm) cyclones into a middling
product and refuse.
The medium coal circuit was changed
to process coal in the size range of Vs-in. x
16 mesh (3.15 x 1.0mm) instead of Vi-in.
x 9 mesh (6.35 x 2.0 mm) and the low
gravity DMC underflow is recleaned in
hydrocyclones and concentrating tables
instead of only in dense medium cyclones.
The major changes in the fine coal
circuit involve redefining the coal size
from 9 x 100 mesh (2.0x0.15 mm)to 16x
100 mesh (1.0 x 0.15 mm), the use of
fifteen 8 in. (20.3 cm) cyclones to replace
the former bank of eight 14 in. (35.6 cm)
cyclones, the use of a low head static feed
tube to stabilize inlet pressure, and the
use of special screening devices to elimi-
nate the 100 mesh (0.15 mm) "slimes"
before processing.
1982 Plant Performance
In 1982 modifications to solve major
operating problems in the MCCS were
completed on the "B" circuits within the
plant. The program to test the modified
"B" circuits has centered around the
evaluation of the revised low gravity
circuits for cleaning medium and
fine/very fine size coal. Testing has
included that needed for start-up,
conceptual design verification, and per-
formance evaluation of the low gravity
modifications. Performance evaluations
of the 1982 plant configuration have
centered on the medium and fine coal
cleaning circuits to improve the quality
and yield of the Unit 3 product.
Medium Coal Circuit
Tests were conducted during the start-
up period to optimize performance of the
14-in. (35.6 cm) diameter heavy medium
cyclones in the medium coal circuit which
operates at low specific gravity. The
cyclones fell just short of the targeted
quantity and quality requirements
necessary to meet design conditions for
producing compliance coal. In an effort to
pinpoint the problem areas within this cir-
cuit, an individual cyclone was modified
for further testing. This single cyclone
within the bank was sampled under
various operating conditions to determine
the effects of operating parameters on
performance. To date, the most dramatic
effects in cyclone performance were
observed when the plant medium was
purged and recharged with fresh E grade
magnetite. (Magnetite grade is
determined by the weight percent pass-
ing a 325 mesh or 0.045 mm screen: B
grade is 90 percent passing, E grade is 95
percent passing, and F grade is 98
percent passing the classification point.)
Under these conditions, the performance
at 1.34 s.g. separating gravity met the
design sharpness of separation criteria
(i.e., a probable error of 0.03).
Fine Coal Circuit
Optimization testing, including cyclone
bank modifications and variation of
operating conditions, could not achieve
design conditions. As a result, a detailed
performance analysis began on a single
cyclone within a five cyclone bank. As
with the dense medium cyclones above,
the most dramatic performance
improvement occurred when the plant
medium was recharged with fresh
magnetite. However,the cyclone did not
achieve design performance using fresh
magnetite alone. Future testing will be
performed to define remedial measures
which can be prescribed to achieve
design separating efficiency.
Future Testing
Present and future work is aimed at
determining why the 8 in. (20.3 cm)
dense medium cyclone at the Homer City
Coal Cleaning Plant behaves so
differently from the pilot plant cyclone
tested at the DOE facility. Future plant
testing will investigate the effects of
cyclone geometry, inlet pressure, cyclone
orientation, finer-sized magnetite,
medium-to-coal ratio, and coal size
distribution on the performance of the 8
in. (20.3 cm) cyclone.
-------
Conclusions
Experience over the past 6 years at the
Homer City Coal Cleaning Plant has
indicated that fine coal cleaning at low
specific gravity is possible, with a good
potential for recovering fuel that is
remarkably free of ash and pyritic sulfur.
To reach full potential, it is necessary to
control the feedstock to the cleaning
devices, to specially engineer the clean-
ing devices for their specific coal cleaning
function, and to closely control the
devices in the plant circuits.
In addition, successful operation of any
coal cleaning process depends on the
characteristics of the coal reserves to be
cleaned. A system that meets coal quality
Specifications when operating on coal
from one reserve may not perform as well
when operating with coal from an over- or
underlying seam or in a distant part of the
same seam.
Future application of coal cleaning to
meet extremely low ash and sulfur
criteria should be proposed with
sufficient lead time to extensively sample
the reserve bases proposed for the
operation and to test the cleaning
efficiency of the low specific
gravity device on coal that is
representative of the reserve. Plant
design should provide accurate
classification, sufficient separating
equipment capacity, and a system to effi-
ciently by-pass fine coal that cannot be
processed. Since it appears that
separating efficiency is enhanced using
fine magnetite in the system, the
magnetite recovery equipment must be
optimized for the applications and
slurries being processed.
D. W. Carey, S. T. Higgins, A. A. Slowik. R. D. Stoessner. C. W. Sypult, J. H. Tice, M.
E. Till, and E. A. Zawadzki are with the Pennsylvania Electric Company,
Johnstown. PA 15907.
James D. Kilgroe is the EPA Project Officer (see below).
The complete report, entitled "Homer CityMultistream Coal Cleaning Demonstra-
tion: A Progress Report," (Order No. PB 84-214 181; Cost: $8.50, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park. NC 27711
•(r U S GOVERNMENT PRINTING OFFICE, 1984—759-015/7827
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
------- |