«>EPA
United States
Environmental Protection
Agency
Industrial Environmental Research EPA-600/7-80-103
Laboratory May 1980
Cincinnati OH 45268
Research and Development
Wood Waste as a
Power Plant
Fuel in the Ozarks
Interagency
Energy/Environment
R&D Program
Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-80-103
May 1980
WOOD WASTE AS A POWER PLANT FUEL
IN THE OZARKS
by
V. J. Flanigan
Mechanical Engineering Department
University of Missouri - Rolla
Rolla, Missouri 65401
Grant No. R804270-010
Project Officer
Harry Freeman
Incineration Research Branch
Industrial Pollution Control Division
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmen-
tal Research Laboratory-Cincinnati, U. S. Environmental Protec-
tion Agency, and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the U. S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
11
-------
FOREWORD
When energy and material resources are extracted, processed,
converted, and used, the related pollutional impacts on our
environment and even on our health often require that new and
increasingly more efficient pollution control methods be used.
The Industrial Environmental Research Laboratory-Cincinnati
(lERL-Ci) assists in developing and demonstrating new and
improved methodologies that will meet these needs both efficiently
and economically.
In an attempt to solve the problem of wood waste disposal,
the study described here was initiated to study the possibility
of using the waste as a supplement to present coal fire facili-
ties. The overall goal was to promote the economic and energy
conservation and pollution free characteristics of this material.
This report will be of interest to those considering supplementary
coal with wood chips. Requests for further information should
be addressed to the Incineration Research Branch, lERL-Cincinnati.
David G. Stephan
Director
Industrial Environmental Research Laboatory
Cincinnati
iii
-------
ABSTRACT
The report discusses the testing program conducted on a
chain-grate stoker boiler with a blended coal and wood waste
fuel. The boiler was designed to produce 18,000 Ib/hr (8,200
kg/hr) of saturated steam at 150 psig (103.4 N/cm2). The
objective of the tests was to determine the difference, if any,
in the performance and the emissions of the boiler co-firing
wood and coal as compared to firing coal alone. Four different
coals with different sulfur contents were fired with the wood
waste. The wood waste content was varied up to 2/3 by volume.
Particulate loadings, opacity and concentrations of sulfur
dioxide, nitrogen oxide and total hydrocarbons in the flue gas
were measured. Also, cyanide concentrations and particulate
polycyclic organic emissions were determined for some of the runs,
The baseline for all data comparisons was obtained by running
the boiler with a particular pure coal.
The boiler performance was reported by plotting the boiler
efficiency as a function of the wood percentage in the fuel and
by plotting a performance index defined to be the pound of steam
produced per pound of fuel fired versus weight percent wood. The
efficiency curve showed a slight increase with increasing wood
content while the performance index naturally decreased due to
the lower heating value of the wood. The performance index was
used to predict the operational savings from supplementing the
coal with the wood waste.
Results indicate that particulate and nitrogen oxide
emissions are not substantially altered by using wood waste as
supplemental fuel with coal for the test conditions adopted in
the program. Sulfur dioxide emissions decreased with increased
proportions of wood waste in the coal-wood mixture, whereas total
hydrocarbon concentration increased when wood waste content was
increased. Cyanide concentrations were not substantially
affected by substitution of wood waste for coal; particulate
polycyclic organic emissions were below the detectable limits for
all fuels used in the tests.
IV
-------
CONTENTS
Foreword iii
Abstract iv
Figures vi
Tables viii
Abbreviations and Symbols ix
Acknowledgment xi
1. Introduction 1
2. Conclusions 3
3. Recommendations 5
4. Materials and Methods 6
Fuel Characteristics 6
Fuel Availability 8
Fuel Handling 13
Fuel Firing 15
5. Experimental Procedures 22
Fuel Test (MRI) 22
Sampling and Analysis Methods 2<
6. Results and Discussion 28
7. Experimental Procedures 46
Fuel Testing (UMR) 46
Analysis 48
8. Results and Discussion 52
References 67
Appendices
A. Facilities Statement 69
B. Data Collected and Calculated 70
C. Report of Analysis 73
-------
FIGURES
Number Page
1 U.S. Forest Service Regions for Missouri 9
2 Fuel Handling Diagram 14
3 Boiler Sketch 16
4 Boiler Modifications 19
5 Fuel Supply Sketch 20
6 Power Plant Breeching System 23
7 Section of the Sampling Location 25
8 Particulate Loading as a Function of the
Excess Air 43
9 Particulate Correction as a Function of Weight
Percent Coal 44
10 Sulfur Dioxide Emissions as a Function of Fuel
Blend and Coal Sulfur Content 53
11 Normalized Sulfur Dioxide Output as a Function
of Wood Percent 54
12 Nitrogen Oxides Emissions as a Function of
Fuel Blend 55
13 Opacity of Flue Gas as a Function of Excess
Air and Coal Sulfur Content 56
14 Opacity as a Function of the Weight Percent
of the Coal 57
15 Opacity Correction Curve as a Function of the
Weight Percent Coal 59
16 Dry Gas and Unburned Combustible Losses as a
Function of Fuel Blend 60
VI
-------
Number Page
17 Moisture Losses Due to Fuel Hydrogen, Air
Moisture, and Fuel Moisture as a Function
of Fuel Blend 61
18 Net Efficiency Excluding Radiation and Unac-
counted Losses as a Function of Fuel Blend 62
19 Performance Factor (Pound Steam per Pound Fuel)
as a Function of Fuel Blend 63
20 Economic Results from the Blended Fuel Use 66
VI1
-------
TABLES
Number Page
1 Percentage of Sulfur Contents on the Dry Basis .... 6
2 Fuel Ultimate Analysis (% by weight) and
Heating Values (Dry Basis) 7
3 Residue Production Pounds per Year 8
4 Missouri Forest Products Industries Near
Rolla, MO 11
5 Test Results (Uncorrected) 29
6 Test Results (Corrected) 35
7 Comparison of Measured and Calculated Flue
Gas Flow Rates 41
8 Comparison of 1974 Compliance Test with Coal-Only
Results of Current Programs 45
B-l Test Data 70
B-2 Test Data Plotted 71
Vlll
-------
ABBREVIATIONS AND SYMBOLS
SYMBOLS
A — Weight percentage of ash in the net fuel fired
C — Weight percentage of carbon in the net fuel fired
C — Weight percentage of carbon actually burned in the net
fuel fired
FR — Fuel firing rate
G — Volume ratio of wood to coal in the blended fuel
H_ — Weight percentage of hydrogen in net fuel fired
H — Weight percentage of hydrogen in coal
G
H — Weight percentage of hydrogen in wood
Wf
HHV — Dry higher heating value of coal
C
HHV_ — Dry higher heating value of net fuel
HHV -- Dry higher heating value of wood
WF
L — Loss due to moisture in air
r\
L — Loss due to dry flue gases
L-. — Loss due to hydrogen in fuel
ri
L — Loss due to moisture in fuel
m
L — Loss due to unburned combustible in refuse
M — Weight of coal added along sides of stoker
MF — Weight of blended fuel fired
MC — Moisture content of coal
c
MC — Moisture content of wood
w
MW — Molecular weight of flue gas
N_ — Weight percentage of nitrogen in net fuel fired
0_ — Weight percentage of oxygen in net fuel fired
0 — Weight percentage of oxygen in coal
G
0 — Weight percentage of oxygen in wood
Vr
P — Pressure
R — Ideal gas constant
ix
-------
S — Weight percentage of sulfur in net fuel fired
S _ — Weight percentage of sulfur actually burned in net
fuel fired
T — Temperature
T — Air temperature
T — Flue gas temperature
V— Actual volume of dry flue gas per 100 pounds of fuel
VAIR — Actual volume of air entering boiler per pound of
fuel
V.™ — Theoretical dry flue gas volume per 100 pounds of
Ci fuel
— Average volume of excess air per 100 pounds of fuel
VEA| — Volume of excess air per 100 pounds of fuel from
C02 carbon dioxide analysis
vEA|o — Volume of excess air per 100 pounds of fuel from
2 oxygen analysis
V — Volume of nitrogen in theoretical air per 100 pounds
iN of fuel
V n — Volume of oxygen in theoretical air per 100 pounds
of fuel
WDG — Weight of dry gas per pound of fuel
WH — Weight of moisture per pound of fuel
W — Weight of air moisture per pound of fuel
WF — Weight fraction of coal in blend
x — Weight ratio of wood/weight of fuel
y — Weight ratio of coal/weight of the fuel
Z — Weight percentage of a constituent (excluding oxygen,
hydrogen) of net fuel fired
Z — Dry weight percentage of a constituent (excluding
oxygen, hydrogen) of coal
Z — Dry weight percentage of a constituent (excluding
oxygen, hydrogen) of wood
%C02 — Volumetric percentage of carbon dioxide from Orsat
%02 — Volumetric percentage of oxygen from Orsat
3 — Weight fraction of carbon in the refuse
Y — Weight fraction of sulfur in the refuse
p — Density of the constituent
-------
ACKNOWLEDGMENTS
The author wishes to express his appreciation to the
Environmental Protection Agency for the financial support of this
project and to Mr. Harry Freeman for his help and guidance during
the tenure of the project.
The author also would like to thank the Chancellor for his
encouragement and confidence in the work undertaken. A special
thanks also goes to the power plant operating and maintenance
personnel who were always helping and to Elmer Doty, the graduate
student who assisted in obtaining the data.
The author also thanks Art Spratlin and Jim Kelly of the
EPA regional office in Kansas City for the loan of the Method 6
sampling equipment, which greatly helped in the testing.
XI
-------
INTRODUCTION
The report defines the experience at the University of
Missouri-Rolla with a boiler fuel, which consisted of a blend of
wood residue and coal. This study was undertaken to save an un-
renewable resource, coal, and at the same time conserve energy
by using a waste product.
The conversion of waste products to energy has become an
attractive and necessary means of disposing of waste material.
The success of this expedient was attested to by the programs at
Union Electric in St. Louis, Mo. (1), and the program in Ames,
Iowa (2). The facilities both burned solid waste derived fuel to
produce electricity. The wood industry in the Northern States of
the U.S. has long used waste wood as an energy source.
The wood industry in the state of Missouri is represented by
approximately 500 sawmills, most of which are too small to use
their waste as an energy source, but the waste represents a huge
amount of material and a severe problem. The problems arise from
the danger of fires in the stockpiles, difficulties with run-off
and simply the large amount of land required. Previously, the
waste was simply stockpiled or burnt in a teepee burner, but
neither of these options are presently satisfactory, leaving the
operator with the thoughts of land-filling the material. The
land-fill solution becomes especially critical when one examines
how depressed the industry has been and how important it is to
the states' economy.
With these facts one immediately begins to consider the
energy content of the waste and its value as a fuel.
The University of Missouri-Rolla is located at the edge of
the Eastern Ozark Region, the most heavily forested area of the
state. At this location, the University has been investigating
the possibility of using wood waste to fuel their central power
plant, which provides the heating and cooling load for the
campus.
The estimated coal need at the plant for FY 1977-78 was
12,000 tons at an estimated cost of approximately $450,000. A
savings here would be of great benefit to the University and the
people of the state. The coal normally used in the plant is an
Illinois blend of high and low sulfur coal providing a maximum
sulfur content of 2%; the wood was a hardwood blend mostly oak
-------
with zero sulfur content.
The primary objective of this report is to present and dis-
cuss experimental results obtained from the monitoring of the
flue gases and the performance characteristics of a boiler fired
with blends of coal and hardwood waste (primarily oak chips).
The research was performed using a boiler located in the
University of Missouri-Rolla Power Plant from January through
August 1977.
Results are presented for sulfur dioxide emissions, nitrogen
oxides and particulate effluents. The gaseous pollutants,
presented as weight per unit energy input, are treated as a
function of both the sulfur content of the coal and the weight
percentage of wood in the fired fuel. Particulate effluents are
presented as opacity and particulate versus percent excess air
and weight percent coal.
The report also discusses potential hazardous organic
materials that may be omitted with the use of wood waste.
The boiler performance was addressed in terms of the boiler
heat balances for the different blends of each of the coals. The
overall efficiency of the unit was obtained from this data for
each run providing the data necessary to suggest a yearly cost
with that particular fuel blend. The blended fuel costs were
compared to the all coal operating cost and the dollar savings
presented.
-------
CONCLUSIONS
In this investigation, the relative performance of coal and
coal/wood blends as fuel was investigated experimentally. The
conclusions arising from this work are as follows:
1. Wood waste burned in conjunction with coal exhibits heat
losses that vary as predicted, and result in relative
efficiencies somewhat higher than pure coal. Simply
stated, the utilization of the heat released per unit
weight of fuel is slightly improved.
2. The performance factor tended to decrease slightly with
the addition of wood to the blended fuel. Whether or
not this decrease is significant depends on the size of
the plant and the extenuating economic factors. For
this particular unit, the use of a composite coal/wood
fuel appears as an attractive proposition. In fact,
savings of up to $35,000/year were suggested with large
decreases in the coal used.
3. The boiler response is affected by the blended fuel.
The boiler cannot respond to a change in load in the
same sense as when firing all coal. The blended fuel
requires more attention and skill from the fireman.
In this report, the emission characteristics of various
blends of coal and hardwood waste were also investigated experi-
mentally in conjunction with the boiler performance. The
conclusions arising from this work may be stated as follows:
1. Sulfur dioxide emissions expressed unit weight per unit
heat input were shown to be measurable with a relatively
high degree of accuracy. Emissions decreased as
expected when the wood content of the blend was
increased. In fact, the sulfur dioxide emissions may
be satisfactorily approximated from the sulfur content
and the heating value of the coal.
2. Emissions of nitrogen oxides were not visibly dependent
on the wood to coal ratio of the fired fuel. However,
the lack of flame temperature data does not permit
complete resolution of these effects.
-------
3. The examinations of the particulate emissions and
opacity data for the different, fuels, with the correc-
tions for excess air, point out that the particulate
wasn't substantially altered by the use of the wood
waste. This conclusion was based on the zero slope of
the corrected opacity data and the slight decrease in
the particulate data with increased coal content. The
opposite slope of the opacity data (uncorrected) as
compared to the particulate data also leads to this
conclusion.
4. Total hydrocarbon concentrations increase with increased
substitution of wood waste for coal. For coal-only
tests, hydrocarbon emissions averaged 4.25 ppm, and for
one volume wood waste plus one volume coal, hydrocarbon
levels were 6.2 ppm. When two volumes wood waste were
used with one volume coal, the average hydrocarbon
emissions were 8 ppm.
5. Cyanide concentration was not substantially affected by
substitution of up to approximately 50% of coal with
wood waste.
6. Particulate Polycyclic organic emissions were below
detectable limits for all fuels used.
-------
RECOMMENDATIONS
It is recommended that from the results of this study the
University consider firing one of the two small boilers on a
blended fuel. It is recommended that the firing be on the con-
tinuous basis and for a period of one year.
In order to reduce the difficulties experienced with
ignition problems, high stoker speeds, high gate settings, and
slow response the University should consider a higher energy
blend. A blend of 20% by weight wood, 39% by volume, would be
a possibility. The blend suggested would still provide a 12%
savings in fuel cost and would provide the higher Btu content per
pound of fuel.
The demonstration of an extended operation on blended fuel
at some reasonable boiler load would provide valuable operating
data, including the demonstration of any slagging and ash prob-
lems, if they exist. The demonstration would also provide
emphasis to encourage other users to study the possibilities of
adapting wood waste to their combustion system.
-------
FUEL CHARACTERISTICS
The fuel used consisted of either a pure coal or a blend of
wood and coal, a requirement of boilers equipped with chain-
grate stokers. The operating instructions of these stokers
state a requirement of a minimum of a 6 to 7% ash content in the
fuel used. The ash content of the wood chips was less than 1%,
and required that the chips were always blended with the higher
ash coal. This ash requirement provides protection for the
moving chain from the radiant energy of the fire.
The stoker's manufacturer also recommended these stokers
for a wide range of screened bituminous coals, suggesting a
maximum 1 inch top size and a minimum of 28 mesh for fines. The
coal normally used at the plant is a blended Illinois coal
normally providing a 2% sulfur content on the as received basis.
A sample screen test and proximate analysis of the blended coal
is given in Appendix C. This analysis meets the coal specifica-
tions suggested by the manufacturer.
The coal used in the test program was as follows; the normal
blended Illinois coal, Ziegler coal, Orient coal, and River King
coal. The blended coal normally used was a blend of 60% Orient
and 40% Ziegler. The sulfur content of these fuels is given in
Table 1. The ultimate analysis for these coals and the wood,
as measured by an industrial laboratory, is given in Table 2.
TABLE 1. PERCENT SULFUR CONTENTS ON THE DRY BASIS
Orient Ziegler Blended (Orient & Ziegler) River King
1.91 3.12 2.44 3.29
The wood used for the most part was hardwood chips. The
chips on an average were 1 inch square by 1/8 inch thick
(6.54 cm? x 0.3174 cm). They were produced by chipping debarked
slabs of mixed hardwoods. The sulfur content of course was very
low, typically zero, though the laboratory analysis reported
some trace of sulfur in the wood fuel. Other fuels used on a
limited basis were sawdust and bark. The bark was the result
of the debarking of the hardwood logs and contained a large
percentage of wood fiber. The bark's size distribution was
-------
TABLE 2. FUEL ULTIMATE ANALYSIS (% BY WEIGHT) AND
HEATING VALUES (DRY BASIS)
Blended
Coal No. 1
Ziegler
Coal No. 2
Orient
Coal No. 3
River King
Coal No. 4
Wood
Carbon
Hydrogen
Oxygen
Nitrogen
Sulfur
Ash
Higher Heat-
ing Value
Btu/lb
MJ/KG
72.515
4.957
8.560
1.585
2.444
9.939
11,830
27.5
69.114
5.227
6.997
1.176
3.120
14.366
12,560
29.2
71.955
5.104
9.706
1.477
1.911
9.847
12,680
29.5
67.67
4.4
11.04
1.02
3.29
11.98
12,633
29.4
50.18
5.06
43.00
0.30
0.15
1.31
8,770
20.4
typically a mixture varying from long fibers of approximately 3
inches (7.62 cm) to a considerable concentration of fines. The
sawdust was all fines and the dust was collected off the header
saw in the mill.
-------
FUEL AVAILABILITY
Looking at the Missouri Sawmills, it is of interest to note
that approximately 50% of the incoming log ends up as sawdust,
edgings or possibly bark or chips, depending on the mill. The
waste percentage was taken from the work accomplished by the
Missouri Conservation Department for Missouri Sawmills (3). Of
course the actual percent of waste was dependent on the size of
the logs, material to be produced, and the actual mill operation.
To give an indication of the availability of wood waste a survey
of existing stock piles and annual production of wood waste in
the Eastern Ozark Region of the State was taken (4).
The Eastern Ozark Region, shown in Figure 1, is the most
heavily forested area of Missouri. The figure was taken from
the NCFES bulletin, U. S. Forest Service, USDA, 1972 (5). The
percentage of forested area is given below the name of each
region in Figure 1.
This Eastern Ozark Region is naturally the area representing
the most active forest industry in the state. The survey not
only looked at the residue from the mills, but also from
manufacturing operations like pallet plants, flooring mills,
etc. The annual production for this region was 1,764,563,000
pounds of residue, broken up as in Table 3.
TABLE 3. ANNUAL RESIDUE PRODUCTION
EASTERN OZARK REGION (pounds)
Sawdust 571,800,000
Trim 82,210,000
Shavings 70,200,000
Slabs/Edgings 882,080,000
Bark 154,400,000
Shreds 3,860,000
Total 1,764,550,000
This represents an annual production of 10,587,300 x 10
Btu, based on a Btu content of 6,000 Btu/lb, and amount to
880,000 tons. At the same time, in just this region, there are
2,317,880,000 pounds (1,150,000 tons) of residue existing in
8
-------
PRAIRIE REGION
9% For res ted
r--—,
MISSOURI
FOREST SURVEY REGIONS
vo
N—NRIVER BORDER
/ REGION
|~-T~' 26
NORTHWESTERN OZARKS
REGION 38 %
UNIVERSITY.
SOUTHWESTERN
OZARKS REGION
42%
Figure 1. U.S. Forest Service Regions for Missouri
(Courtesy of North Central Forest Experiment
Station, Forest Service, USDA)
-------
dumps and stockpiles. The Residue specialist for the Conserva-
tion Commission approximated that another 1,500,000 tons are
produced annually in the other regions of the state. The total
quantity of waste comes to some 2,400,000 tons. The tonnage does
not consider residue in the woods in so far as 50% of the tree
is left in the woods as slash. A portable chipper and skidder
could quickly convert the slash to chips and at the same time
remove poor quality timber for stand improvements (6). The
portable chipper system could provide even larger quantities of
available Biomass. Even though the developers and users of
these total tree harvesting systems need to exercise care in
their use because of the problems and the lack of understanding
in regards to the nutrient removal (7) from the soils. This
care is especially critical in clear cutting operations. The
Biomass was definitely available in residue form for this
experiment, but it was also apparent that much more was available
in the way of slash and cull timber if a need could be
established.
Further documentation of the available wood waste in this
area can be derived from wood industry data. The first consider-
ation might be the sawmill sizes and production. In 1975 the
Missouri Department of Conservation published the following
sawmill distribution (8);
Production per year Number of Sawmills
Under 100 Million Board Feet 152
100 MBF - 249 MBF 99
250 MBF - 499 MBF 77
500 MBF - 749 MBF 41
750 MBF - 999 MBF 24
1,000 MBF - 1,999 MBF 68
2,000 MBF - 2,999 MBF 33
3,000 MBF - 3,999 MBF 10
4,000 MBF - 4,999 MBF 5
5,000 MBF - 7,499 MBF 3
7,500 MBF - and greater 3
517
University of Missouri-Rolla1s location as a potential user
was also established through the following mileage table (9)
(Table 4). The table shows the concentration and activity of
the wood industry in the area surrounding the University.
In this work, only the available fuel at the sawmills was
used. The wood chips were obtained from the Reed Lumber Co.,
approximately 60 miles from Rolla. They were delivered in dump
bed trailer trucks at a cost of $12.50/ton. Typically the wood
10
-------
TABLE 4. MISSOURI FOREST PRODUCTS INDUSTRIES
NEAR ROLLA, MISSOURI
WITHIN X MILES
Key:
TOTAL AND
TYPE OF INDUSTRIES
50
40
30
25
20
10
A
63
40
25
22
9
5
B
15
9
2
1
1
1
C D
9 1
6
6
5
3
— —
E
1
1
1
1
—
A Sawmills (100,000 board feet per year minimum)
B Pallet Plants
C Stave Mills
D Heading Mills
E Flooring Mills
chips are hauled in standard cargo trailer but the University
doesn't have unloading facilities for the trailers. With this
in mind, the bid requested they be hauled and delivered in a
self-unloading trailer. This requirement added to the price of
the product and also restricted the bids to those people having
self-unloading equipment.
Since the time of the bid for the wood chips, the industry
has started using trailers with live bottoms. They are much
cheaper than the dump beds, and their use could lower the
delivery price for the wood chips. The wood chip price, as
suggested by the Green Mountain Study (7), was $10 to $15/ton
projected so the cost was in the proper range.
At the same time, bid prices for bark delivered in the same
fashion were obtained by the University. They were $3.90 ton/
delivered. The hauling distance was approximately 45 miles.
The coal used in the study, as well as the coal used in the
everyday running of the plant is delivered by self-unloading
trucks. The coal is pre-blended in St. Louis and delivered to
Rolla ready for use. The coal cost is $37.53/ton. The cost of
the fuels per million Btu was;
coal
wood chips
$1.56/10b Btu
$1.04/106 Btu
11
-------
bark $0.35/106 Btu
If the wood could be used with approximately the same efficiency
as the coal, the potential for savings is pointed out in the cost
per unit energy quantity given. The efficiency and firing
characteristics are defined in the preceding chapters. The
availability of the wood product discussed in this chapter seems
to be without problem.
12
-------
FUEL HANDLING
Once the fuel, either coal or wood, is delivered, it is
stockpiled separately in the coal yard to be used as needed.
The fuel used for study was either coal or a blend by volume of
wood chips and coal. The volume measure was simply the bucket
of the front-end loader used to transport fuel to the power
plant from the stockpile. The initial preparation for a test
was the decision of the volume ratio to be used and the quantity
necessary for the test. When the ratio was given to the front-
end-loader operator he obtained the necessary volumes of wood
and coal. These volumes were dumped in an open area of the coal
yard and then the volumes were wind rowed. The wind rows were
passed through several times to mix the fuel. This operation
seemed to work well and there weren't any problems with in-
consistent blending, at least not in terms of the mixing
operation.
The blended fuel was then moved to the coal pit where it
was carried by a bucket conveyor to a segrated bunker in the top
of the plant. The normal bunker, which has approximately a
60-ton capacity, was divided to allow for the storage of
approximately 15 tons of the blended fuel.
At this point the fuel was available to the weigh lorry
of the plant to be weighed and supplied to the boiler as fuel.
This pattern of fuel handling is defined by the block diagram,
Figure 2.
The weights of the fuel were defined from the volume ratios
and from weighing representative samples of the fuels. The fuel
supplied was always weighed and records of these weights kept.
13
-------
Wood
SIorage
Loader
Blended
by
Volume
Front End
L oad er
Coat
Storage
Loader
Coal
Pit
Bucket
Conveyor
Segrated
Coal
Bunker
Dump
Weigh
Lorry
Gravity
Chute
Boiler
Hopper
Figure 2. Fuel Handling Diagram
14
-------
FUEL FIRING
The power plant's function is to be a heating plant for the
University, supplying steam to heat and cool the buildings of
the campus. The plant at one time was responsible for providing
both the electrical and heating needs of the campus and still has
that capability. The plant's role was important to the project
because the test runs were only made when there was adequate
heating load. The test boiler was always run in conjunction
with another boiler in the plant. The two boiler combination
insured that a loss of boiler load on the test boiler would not
interrupt the campus heating system. The two boilers also
required an increased plant load to adequately load both boilers.
This load condition restricted the testing to fixed periods of
time and weather. A boiler load of greater than 5,000 Ib/hr
(2277 Kg/hr) was the goal in all the testing.
All the tests performed were undertaken on boiler No. 2.
The No. 2 boiler was a small 18,000 Ib/hr (8,200 Kg/hr), 150
psig (103.4 N/cm2), saturated steam boiler equipped with a
chain-grate stoker. The boiler has no heat traps, superheaters,
or pollution equipment. A sketch of the boiler is shown in
Figure 3. The fuel was supplied by the fireman from the coal
lorry. Before he dumped the fuel into the hopper of the boiler,
he weighed it and recorded the weight. The fuel, which was in
the hopper, flowed onto the flat bed or chain as it moved into
the boiler. The speed of the chain was determined by the steam
demand or, if the boiler was set to manual control, the speed
was determined by the stoker speed setting. The energy supplied
to the boiler was completely a function of the speed of the chain
and the depth of the fuel bed, assuming that amounts of combus-
tion air were satisfactory. The fuel bed depth was controlled
by the gate shown in Figure 3. The operation manual recommends
a thick bed of fuel and a slow chain, but with the blended fuel
it could only be partially realized. The situation was apparent
when an equivalent fuel bed of blended fuel was calculated from
a ratio using the normal coal bed thickness of 4 (10.16 cm) to
4.5 (11.43 cm) inches. The calculation was made assuming the
same chain velocity for both fuels
t • L • HHV p
c cpc
b L • (HHV x + HHV y)p
c wu m
15
-------
GATE
STEAM
CHAIN
AOITATORl
BLEND
vV
HOPPER
CHAIN DIRECTION
WATER
ASH
AIR
Figure 3. Boiler Sketch
16
-------
where:
t, - thickness of the blend bed (in)
t - thickness of the normal coal bed (in)
L - width of the chain (in)
HHV - Higher Heating Value of coal (Btu/lb)
c
HHV - Higher Heating Value of wood (Btu/lb)
W
x - pounds of coal/pound of mixture
y - pounds of wood/pound of mixture, and
P0/Pm - density ratio coal to mixture.
\*> III
For a blend of 1 wood by 1 coal, x was 72% and y was 28%, so t,
would be;
4" x 60" x 11,500 Btu/lb
fcb = 60" x (11,500 Btu/lb x .72 + 7000 Btu/lb x .28)
=6.5 inches (16.51 cm) (1)
The boiler was limited to a 7-inch (17.78 cm) opening of the
gate so it was quite apparent that the gate was close to the
limit of bed thickness. For a fuel blended 2 wood by 1 coal,
the bed thickness would be;
4" x 60" x 11,500 (1§§°)
b 60" x (11,500 Btu/lb x 0.54 + 7,000 x .46)
= 8.23 inches
This thickness wasn't realizable so the only other way to
increase the energy input was to run the chain faster. In all
cases where the blend was used even with large gates 6 to 7
inches, the chain was always run faster than with all coal with
a 4 inch gate. The ignition of the fuel was provided by the
ignition arch (Figure 3) of the boiler and the mechanism for
ignition was the radiant energy from the arch. In the operation
it was desirable to provide for ignition as soon as the fuel
flowed out from under the gate, but it was also very important
to keep it external from the gate itself.
When the blend was first run with these high gates, the
results and experience pointed to ignition problems from both
the gate settings and the low ignition temperature of the wood.
The problem most frequently encountered was burning under the
gate of the boiler.
As one examines the ignition difficulties it is apparent
that the problem was most pronounced in the entrance corners.
17
-------
The burning under the gate would start in the corners and then
spread to the total length of the gate. There were probably
several reasons for this. One of the reasons was the boiler
construction itself; there was a point of stagnation in the fuel
flow for gates above 5 inches. This point was on the entrance
corners of the boiler as the fuel flowed under the gate. The
ignition problem was also experienced with coal but not as
frequently. With this in mind, a modification was undertaken
to solve the problem in the corners. The boiler hopper was
segrated by placing subdivisions in each end of the hopper shown
in Figure 4. This modification provided three compartments:
dimensions looking at the front of the boiler are 3 inches, 54
inches, and 3 inches. The pure coal was placed in the outside
compartments and the blend in the middle compartment, the fuel
as supplied to the boiler would look like Figure 5. With this
fuel-firing pattern the corners had pure coal with ignition and
flow properties of coal. As experience was gained, coal would
only be fired in the corners if the fuel was difficult to burn,
quite often the corners would be fired with the blend or left
empty cooling the corner areas.
Another problem relating to the ignition problem was the
fuel bridging in the boiler hopper. The converging shape of the
hopper was fine for coal but was difficult for the flow of wood.
As the bridging occurred, the fuel fell away from the gate pro-
viding a low energy input. The lack of fuel on the chain also
allowed for ignition under the gate since it wasn't sealed with
fuel. In order to prevent the bridging condition, a cam shaft
was supported by bearings across the top of the hopper (Figure
4). This cam shaft was driven by a variable speed, 3/4 horse-
power electric motor through a 10/1 speed reducer. There were
six cams on the shaft and from each cam a heavy chain was hung.
As the shaft rotated, it provided the chains with a 1 1/4 inch
travel. The-chains were hung in the fuel hopper providing
agitation to the blended fuel. The agitation prevented the
bridging in the hopper. Again as experience was gained the
chains would frequently not be used, especially if the fuel was
wet and difficulty was experienced providing for the combustion
air.
The firing was normally started with a 6 to 6.5-inch gate
setting and stoker speed was a function of the load desired.
The boiler was always fired with the boiler controls in the
manual position. With manual control, the loading was a base
load condition requiring the other boiler to accept most of the
load swings.
The input combustion air was controlled by the fan and
locked into a constant setting. The boiler was equipped with a
force-draft fan and loovers on the intake to the fan controlled
the amount of air. At the same time, the damper on the boiler
stack was closed more than when operating on coal. In
18
-------
SIX CAMS-DRIVING
CHAINS
fl
BAFFLES
BLENDED
FUEL
COAL
COAL
CHAIN
Figure 4. Boiler Modifications
19
-------
GATE
54"
COAU
BLEND
COAL
CHAIN
Figure 5. Fuel Supply Sketch
20
-------
combination, the force-draft fan and damper were controlled to
maintain the boiler at the point where it would almost be at
zero gauge pressure. This pressure was sensed by keeping the
boiler on the verge of smoking. It typically related to a
pressure of -0.01 to -0.05 inches of H20 vaccuum. With the
damper control, the fire could be moved away from and back to
the gate, but at the same time, opening the damper tended to
cool the boiler allowing for a loss of load. The firing re-
quired a delicate balance between stoker speed, damper setting,
and gate height, and most of the time a less porous fuel bed
would have been desirable. The excess air was almost always
high even when firing with pure coal. The overfire air was
normally kept constant and didn't seem to strongly influence
the firing.
Typically the fuel was dry and ignition very rapid. The
fireman had to be very careful not to obtain ignition under the
gate or a very short fire with little ability to increase or
hold the boiler load. As the fireman gained experience with the
fuel the firing problems tended to become easier to handle.
The boiler consultants made the measurements to cut a cam to
run the boiler system automatically but the cam was not
fabricated.
21
-------
FUEL TEST (MIDWEST RESEARCH INSTITUTE)
A test program was undertaken to describe effects of the use
of blended fuel on the effluents and the efficiency with varying
blends and different coals (10).
The first test was undertaken during the last week of
January and the first week of February of 1977. Field testing
was performed by Midwest Research Institute, field sampling team
to determine particulate, sulfur dioxide, and nitrogen oxide
emissions from the boiler firing the blended fuel and to identify
potential hazardous organic materials that may be emitted. In
all cases, the goal was to determine if differences occurred in
boiler emissions from different blends and as compared to all
coal.
For this series of tests, a high sulfur (>3%) coal (River
King) was fired alone to establish a baseline, and blends of wood
and coal in ratios of 1:1 (30% wood, 70% coal) and 2:1 (46% wood,
54% coal) by volume were compared to the coal baseline. The
ultimate analysis of the coal and wood used for this test was
given in Table 2.
The sampling of the unit was accomplished through the
breeching as shown in Figure 6. Note the No. 2 boiler is
connected to the common breeching.
22
-------
ro
to
Breeching \ Breeching
Tranemissometer
POWER HOUSE
I.D. FAN
Figure 6. Power Plant Breeching System
-------
SAMPLING AND ANALYSIS METHODS
The sampling and analysis methods used in the test program
are individually discussed below along with problems encountered
on site.
Sampling
The first task was to perform an EPA Method 1 procedure to
ensure the extraction of a representative sample. The sampling
locations were numbered 1 to 30 as shown in Figure 7. Locations
31 through 36 were not tested because, on opening the test port,
fly ash was found 2 inches deep, 18 inches inside the duct. The
nearest upstream disturbance was five equivalent duct diameters
away, and the nearest downstream disturbance was two diameters
away. By using six sampling locations at the remaining five
ports, the conditions for sampling by Method 1 were fulfilled.
Method 2 (EPA) was then performed to determine stack gas
velocity and flow rate at the sampling sites. Method 3 (EPA),
using an integrated sampling train, was employed to obtain a
representative stack gas sample which was analyzed in an Orsat
analyzer.
Method 4 (EPA) was initially used to determine moisture
content in the stack gas. The measured moisture content of gas
on January 26, 27, and 28 was both low and constant; therefore,
Method 4 was not performed on subsequent test days (January 31,
February 1, 2, 3, and 4). However, the back half of EPA Method
5 was used to measure the moisture content of the stack gas on
the latter days. The moisture content remained low enough and
consistent enough to insure that all of these determinations
were well within isokinetic limits.
To determine particulate emissions, EPA Method 5 was used.
On January 26 and 27, isokinetic results were not obtained due
to the difficulty of reading the inclined manometer at the very
low stack velocities found. Subsequent test runs utilized a
more precise manometer to read pressure drop. The level of the
precision manometer was checked every time the probe was moved
to a new sampling port. Although this ameliorated the difficul-
ties somewhat, precise velocity pressure readings were still
difficult to obtain.
24
-------
3.5'
O 0
© Q
D ®
D ®
O O O
o o o
o o o
^® Q
V"*^ T
• '•' i •••'•"'•••••!
0 ©
(jo) (jj)
o o
0 0
0 O
© (35)
v.^' ^ ^_x
''&&•:!:?'.& &i::::"rj ~'.
®_. _
o
o
o
<$
>«— X j
'i^'-.r..**" j-i*. '.]
6.75"
-^-
13.25^'
1
Ports
80* x 42 Inside Dimensions of the
Breeching
Section A-A
Flyosh
Figure 7. Section of the Sampling Location
25
-------
As soon as the Method 5 probe cleared the second sampling
port, then EPA Method 6 samples for sulfur dioxide were taken
with the probe extending halfway into the stack from the second
sampling port (third actual port from the bottom of the duct).
At the conclusion of the sulfur dioxide sampling, four samples
were taken according to EPA Method 7 for determination of oxides
of nitrogen.
After the oxides of nitrogen samples were taken, a HCN
sample was taken. A sampling train similar to that used in
Method 6 was employed, but the midget impingers contained 0.05 N
KOH instead of the absorbents used in Method 6. The entire
train consisted of midget impingers. The probe was inserted in
the second sampling port with its tip at the center line of the
duct.
Analysis
Particulate samples from the EPA Method 5 strains were
assayed by the analytical precedures specified in Method 5.
However, glass fragments were observed in residues of the probe
rinses from some samples. In order to correct for this con-
tamination, observed glass fragments were physically separated
from the residues, rinsed with acetone, and allowed to dry before
weighing. The weights of the glass fragments were subtracted
from the corresponding probe-rinse residue weights.
Following gravimetric assays of the Method 5 filter catches
for particulate weights, each filter was extracted with three
50-ml portions of cyclohexane. The cyclohexane was added to the
filter in a 250-ml beaker, the beaker placed in an ultrasonic
bath for 20 min, and then the extract decanted into a vial. All
extracts from runs with the sample fuel mixture were composited.
Hence, three composited extracts were obtained, which represent
the coal-only runs, the 1 wood/1 coal runs, and the 2 wood/1 coal
runs. These composited extracts were concentrated to 5.0 ml with
a gentle stream of dry nitrogen. The extracts were assayed for
four POMs by gas chromatography with electron capture detection.
The chromatographic column, 3% Dexsil 300 GC on 100/120 mesh,
was operated isothermally at 290°C with a nitrogen carrier flow
of 30 ml/min. Injector and detector temperatures were maintained
at 245 and 248°C, respectively. A mixed POM standard containing
7,12-dimethylbenz[a]anthracene, benz[a]pyrene, 3-methylchol-
anthrene, and dibenz[a,h]anthracene was utilized.
Samples were prepared for cyanide assay by distillation ac-
cording to standard procedures and were assayed with a cyanide
ionspecific electrode.
Samples were assayed for NOX by procedures described in
EPA Method 7.
26
-------
Samples were assayed for S02 by procedures described in EPA
Method 6.
Aliquots of coal, wood chips, and 1:1 and 2:1 mixtures of
wood chips and coal were submitted to Industrial Testing Labora-
tories, Kansas City, Missouri, for proximate and ultimate
analyses.
27
-------
RESULTS AND DISCUSSION
The results of the test program are presented in Tables 5
and 6. Table 5 displays the measured or uncorrected values while
Table 6 displays results corrected for anisokinetic test condi-
tions and low-flue gas velocities. These corrections are
discussed next.
Discussion of Test Results
Natural draft-flue gas circuits are inherently unable to
overcome large resistances to gas flow. As a consequence, gas
velocities must be kept very low in order to minimize the
pressure drop due to skin friction. The flue gas velocity in
the breeching at Rolla was only about 500 ft/min which
corresponds to a velocity pressure of = 0.01 in of water. These
low values as well as the turbulent, fluctuating nature of low
in real systems, make the manometer readings unreliable.
In order to overcome these difficulties, the flue gas flow-
rate was calculated for each test from the Orsat analysis of the
specific fuel mixture used for the test. It is necessary to use
the Orsat analysis of the sampling location because the flue gas
composition is not constant throughout the system due to in-
leakage of ambient air. Table 7 compares measured and calculated
flue gas flow-rates at the sampling location.
The calculated flue gas flow-rates were used as the basis
for correcting the measured pollutants. . Because it is not
necessary to sample gaseous pollutants isokinetically, the
measured gaseous concentrations are unaffected by errors in flue
gas flow-measurement; however, in calculating the total weight
rate of pollutant from the concentration, the corrected flue
gas flow-rate does make a difference.
The measured particulate concentration was corrected for
anisokinetic samplings by using a technique suggested by Badzioch
(11). His method is summarized by Equation 2.
|J + (1 - a)] 1
c = Cs iLL + (1 - a) where: (2)
C = true fly ash concentration
Cs = sampled fly ash concentration
28
-------
TABLE 5. TEST RESULTS (UNCORRECTED)
Coal only
Particulate
lb/106 Btu
gr/dscf
Ib/hr
Ib/lb steam
so2
lb/106 Btu
ppm
Ib/hr
(expected)
Ib/hr
(measured)
measured/
expected
Ib/lb steam
NO
x
lb/106 Btu
ppm
Ib/hr
Ib/lb steam
HCN*(yg/ft3)
Methane (ppm)
Acetylene,
ethylene,
and ethane
(ppm)
Formaldehyde
(ppm)
Propane (ppm)
Methanol and
acetaldehyde
(ppm)
1/28/77
0.1776
0.0312
2.21
3.09 x
, A-4
10
2.26
340
81.7
28.11
0.34
3.9 x
io-3
0.19
40.85
2.43
3.4 x
io-4
_
1.3
0.15
0.6
1.3
0.5
Run No. 1
2/2/77
0.371
0.0463
3.04
5.28 x
—4
10
2.7
289
53.8
22.21
0.41
3.8 x
io-3
0.28
41.6
2.29
4.0 x
io-4
M
1.7
0.2
0.6
1.5
trace
Run No. 2
2/2/77
0.159
0.0207
1.31
2.27 x
in'*
10
2.03
225
53.8
16.65
0.31
2.9 x
io-3
0.20
31.5
1.67
2.9 x
io-4
M
1.4
0.1
0.9
0.8
* "
Avg
0.236
0.033
2.19
3.55 x
— 4
10
2.33
285
63.1
22.32
0.35
3.5 x
ID'3
0.22
38
2.13
3.4 x
io-4
_
1.5
0.15
0.7
1.2
<0.5
* Below detection limits
29
-------
TABLE 5 (continued)
Coal only
Ethanol,
butanes, and
butenes(ppm)
Total HC (ppm)
7,12-dimethyl-
benz[a]
anthracene*
(ng/scf)
benz[a]pyrene*
(ng/scf)
3-methylcho-
lanthrene*
(ng/scf)
dibenz[a,h]*
anthracene
(ng/scf)
Boiler
1/28/77
0.2
4.05
Run No. 1
2/2/77
trace
=4
Run No. 2
2/2/77
Avg
<0.2
=4.25
<5
<7
<37
Ib coal/hr
Ib wood/hr
Ib steam/hr
106 Btu/hr in
% excess air
Steam enthalpy
(Btu/hr)/
Fuel heat in
(Btu/hr)
flue gas flow
(dscfm)
flue gas flow
(acfm)
1,083
0
7,154
12.45
40
0.69
8,266
13,722
713
0
5,757
8.2
133
0.84
7,660
11,989
713
0
5,757
8.2
139
0.84
7,402
11,484
836
0
6,223
9.6
104
0.79
7,776
12,398
* Below detection limits
30
-------
TABLE 5 (continued)
Particulate
lb/106 Btu
gr/dscf
Ib/hr
Ib/lb steam
so2
lb/106 Btu
ppm
Ib/hr
(expected)
Ib/hr
(measured)
measured/
expected
Ib/lb steam
NO
X
lb/106 Btu
ppm
Ib/hr
Ib/lb steam
HCN*(yg/ft3)
Methane (ppm)
Acetylene,
ethylene,
and ethane
(ppm)
Formaldehyde
(ppm)
Propane (ppm)
Methanol and
acetaldehyde
(ppm)
1:1
Run No. 1
2/1/77
0.460
0.0776
5.56
6.70 x
io-4
1.57
263
65.9
19
0.29
2.3 x
io-3
0.39
77.9
4.69
5.6 x
io"1
«.
2.5
trace
0.4
2.2
1.0
wood to coal
Run No. 2
2/1/77
0.303
0.05736
3.67
4.42 x
io-4
1.49
262
65.9
18.10
0.27
2.2 x
ID'3
0.28
62.8
3.37
4.1 x
io-4
_
1.8
0.5
0.7
1.8
0.1
volume ratio
2/4/77
0.363
0.0606
4.14
6.42 x
ID'4
2.35
334
61.6
26.88
0.44
4.2 x
io-3
0.19
37.8
2.17
3.4 x
io-4
—
1.5
0.3
0.9
2.8
trace
Avg
0.375
0.065
4.46
5.85 x
io-4
1.80
286
64.5
21.32
0.33
2.9 x
ID'3
0.29
59.5
3.41
4.4 x
10-"
—
1.9
<0.4
0.7
2.3
<0.6
* Below detection limits
31
-------
TABLE 5 (continued)
Ethanol,
butanes, and
butenes(ppm)
Total HC (ppm)
7,12-dimethyl-
benz[a]
anthracene*
(ng/scf)
1:1 wood to coal volume ratio
Run No. 1
2/1/77
0.3
=6.4
Run No. 2
2/1/77
0.2
5.1
2/4/77
= 5.5
Avg
<0.3
= 6.2
benz[ a] pyrene*
(ng/scf)
3-methylcho-
lanthrene*
(ng/scf)
dibenz[a,h]*
anthracene
(ng/scf)
Boiler
Ib coal/hr
Ib wood/hr
Ib steam/hr
10 6 Btu/hr in
% excess air
Steam enthalpy
(Btu/hr)/
Fuel heat in
(Btu/hr)
flue gas flow
(dscfm)
flue gas flow
(acfm)
-
863
326
8,303
12.1
156
0.82
8,365
14,218
-
863
326
8,303
12.1
180
0.82
7,463
12,426
-
806
326
6,438
11.4
125
0.68
7,977
12,921
<5
<7
<38
844
326
7,681
11.87
154
0.77
7,935
13,188
* Below detection limits
32
-------
TABLE 5 (continued)
Particulate
lb/106 Btu
gr/dscf
Ib/hr
Ib/lb steam
S02
lb/106 Btu
ppm
Ib/hr
(expected)
Ib/hr
(measured)
measured/
expected
Ib/lb steam
NO
x
lb/106 Btu
ppm
Ib/hr
Ib/lb steam
HCN* (yg/f t3)
Methane (ppm)
Acetylene,
ethylene,
and ethane
(ppm)
Formaldehyde
(ppm)
Propane (ppm)
Methanol and
acetaldehyde
(ppm)
1/31/77
0.269
0.0443
3.50
4.11 x
io-4
2.19
310
61.6
28.56
0.46
3.3 x
1Q-3
0.24
47.9
3.17
3.7 x
— 4
10
_
6.0
5.4
1.6
1.8
0.3
2 : 1 wood to
Run No. 1
2/3/77
0.360
0.0478
3.49
5.31 x
io-4
3.29
373
46.6
31.93
0.68
4.9 x
io"3
0.35
56
3.44
5.2 x
•1-4
10
_
1.3
0.2
0.8
1.3
0.8
coal volume
Run No. 2
2/3/77
0.331
0.04865
3.21
4.81 x
ID'4
2.94
334
46.6
28.57
0.61
4.3 x
1Q-3
0.27
47.4
2.62
4.0 x
— 4
10
^
1.6
0.5
0.7
1.0
trace
ratio
Avg
0.32
0.0469
3.40
4.74 x
io-4
2.81
339
51.6
29.69
0.58
4.2 x
ID'3
0.29
50.4
3.08
4.3 x
lrt-4
10
_
3.0
2.0
1.0
1.4
<0.6
*Below detection limits
33
-------
TABLE 5 (continued)
Ethanol,
butanes, and
butenes(ppm)
Total HC (ppm)
7,12-dimethyl-
benz[a]
anthracene*
(ng/scf)
benz[a]pyrene*
(ng/scf)
3-methylcho-
lanthrene*
(ng/scf)
dibenz[a,h]*
anthracene
(ng/scf)
Boiler
Ib coal/hr
Ib wood/hr
Ib steam/hr
106 Btu/hr in
% excess air
Steam enthalpy
(Btu/hr)/
Fuel heat in
(Btu/hr)
flue gas flow
(dscfm)
flue gas flow
(acfm)
1/31/77
= 15.1
2:1 wood to coal volume ratio
Run No. 1 Run No. 2
2/3/77 2/3/77 Avg
797
573
8,518
13
81
0.78
9,202
16,068
= 4.4
603
418
6,567
9.7
341
0.81
8,538
13,648
trace
= 3.8
603
418
6,567
9.7
181
0.81
7,690
12,231
trace
=8
<4
<7
<35
668
470
7,217
10.8
201
0.80
8,477
13,982
* Below detection limits
34
-------
TABLE 6. TEST RESULTS (CORRECTED)
Coal only
Particulate
lb/106 Btu
gr/dscf
Ib/hr
Ib/lb steam
S02
lb/106 Btu
ppm
Ib/hr
(expected)
Ib/hr
(corrected)
corrected/
expected
Ib/lb steam
NO
x
lb/106 Btu
ppm
Ib/hr
Ib/lb steam
HCN*(ug/ft3)
Methane (ppm)
Acetylene,
ethylene,
and ethane
(ppm)
Formaldehyde
(ppm)
Propane (ppm)
Methanol and
Acetaldehyde
(ppm)
1/28/77
0.204
0.0265
2.54
3.55 x
IO"4
3.05
340
81.7
38.03
0.47
5.3 x
10"3
0.26
40.85
3.29
4.60 x
m-4
10
_
1.3
0.15
0.6
1.3
0.5
Run No. 1
2/2/77
0.390
0.0438
3.20
5.56 x
io-4
3.11
289
53.8
25.48
0.47
4.4 x
io"3
0.32
41.6
2.63
4.57 x
~4
10
—
1.7
0.2
0.6
1.5
trace
Run No. 2
2/2/77
0.170
0.0194
1.39
2.41 x
ID'4
2.29
225
53.8
18.80
0.35
3.3 x
io"3
0.23
31.5
1.88
3.27 x
—4
10
mm
1.4
-
0.1
0.9
0.8
•*
Avg
0.255
0.0299
2.38
3.84 x
io-4
2.82
285
63.1
27.44
0.43
2.97 x
io"3
0.27
38
2.60
4.15 x
iA-4
10
—
1.5
0.15
0.7
1.2
<0.5
* Below detection limits
35
-------
TABLE 6 (continued)
Coal only
Ethanol,
butanes, and
butenes(ppm)
Total HC (ppm)
7,12-
dimethylbenz
[a]*, anthra-
cene (ng/scf)
benz[a]pyrene*
(ng/scf)
3-methylcho-
lanthrene*
(ng/scf)
dibenz[a,h]*
anthracene
(ng/scf)
Boiler
1/28/77
Run No.
2/2/77
1
Run No.
2/2/77
2
Avg
0.2
4.05
trace
=4
<0.2
=4.25
<5
<7
<37
Ib coal/hr
Ib wood/hr
Ib steam/hr
106 Btu/hr in
% excess air
Steam enthalpy
(Btu/hr)/
Fule heat in
(Btu/hr)
flue gas flow
(dscfm)
flue gas flow
(acfm)
1,083
0
7,154
12.45
40
0.69
11,186
18,569
713
0
5,757
8.2
133
0.84
8,546
13,756
713
0
5,757
8.2
139
0.84
8,358
12,967
836
0
6,223
9.6
104
0.79
9,363
15,097
* Below detection limits
36
-------
TABLE 6 (continued)
Particulate
lb/106 Btu
gr/dscf
Ib/hr
Ib/lb steam
so2
lb/106 Btu
ppm
Ib/hr
(expected)
Ib/hr
(corrected)
corrected/
expected
Ib/lb steam
NO
x
lb/106 Btu
ppm
Ib/hr
Ib/lb steam
HCN*(yg/ft3)
Methane (ppm)
Acetylene,
ethylene,
and ethane
(ppm)
Formaldehyde
(ppm)
Propane (ppm)
Methanol and
Acetaldehyde
(ppm)
Run No.
2/1/77
0.519
0.0675
6.28
7.57 x
-4
10 *
2.04
263
65.9
24.67
0.37
2.97 x
io"3
0.50
77.9
6.09
7.33 x
10"4
mm
2.5
trace
0.4
2.2
1.0
1 : 1 wood to
1 Run No.
2/1/77
0.354
0.0478
4.28
5.15 x
-4
10 4
2.09
262
65.9
25.25
0.38
3.04 x
10~3
0.39
62.8
4.71
5.67 x
io"4
^
1.8
0.5
0.7
1.8
0.1
coal volume
2
2/4/77
0.411
0.0527
4.69
7.27 x
— 4
10
3.07
334
61.6
34.99
0.57
5.43 x
10~3
0.25
37.8
2.82
4.38 x
io"4
^m
1.5
0.3
0.9
2.8
trace
ratio
Avg
0.428
0.056
5.08
6.66 x
-4
10
2.4
286
64.5
28.30
0.44
3.81 x
10~3
0.38
59.5
4.54
5.79 x
ID'4
^
1.9
<0.4
0.7
2.3
<0.6
* Below detection limits
37
-------
TABLE 6 (continued)
1:1 wood to coal volume ratio
Ethanol,
butanes, and
butenes (ppm)
Total HC (ppm)
7,12-
dimethylbenz
[a]*, anthra-
cene (ng/scf)
benz[a]pyrene*
(ng/scf)
3-methylcho-
lanthrene*
(ng/scf)
dibenz[a,h]*
anthracene
(ng/scf)
Boiler
Run No. 1
2/1/77
0.3
= 6.4
Run No. 2
2/1/77
0.2
5.1
2/4/77
= 5.5
Avg
0.3
=6.2
<5
<7
<38
Ib coal/hr
Ib wood/hr
Ib steam/hr
10 6 Btu/hr in
% excess air
Steam enthalpy
{ Btu/hr)/
Fuel heat in
(Btu/hr)
flue gas flow
(dscfm)
flue gas flow
(acfm)
863
326
8,303
12.1
156
0.82
10,863
18,464
863
326
8,303
12.1
180
0.82
10,442
17,386
806
326
6,438
11.4
125
0.68
10,383
16,818
844
326
7,681
11.87
154
0.77
10,563
17,548
* Below detection limits
38
-------
TABLE 6 (continued)
Particulate
lb/106 Btu
gr/dscf
Ib/hr
Ib/lb steam
lb/106 Btu
ppm
Ib/hr
(expected)
Ib/hr
(corrected)
corrected/
expected
Ib/lb steam
NO
x
lb/106 Btu
ppm
Ib/hr
Ib/lb steam
HCN*(yg/ft3)
Methane (ppm)
Acetylene,
ethylene,
and ethane
(ppm)
Formaldehyde
(ppm)
Propane (ppm)
Methanol and
Acetaldehyde
(ppml
1/31/77
0.260
0.0457
3.38
3.97 x
io-4
2.06
310
61.6
26.76
0.43
3.14 x
10~3
0.23
47.9
2.97
3.49 x
io"4
_
6.0
5.4
1.6
1.8
0.3
2:1 wood to coal
Run No. 1
2/3/77
0.341
0.0502
3.31
5.06 x
1C'4
2.96
343
46.6
28.79
0.62
4.38 x
10~3
0.32
56
3.10
4.72 x
ID'4
—
1.3
0.2
0.8
1.3
0.8
volume
Run No.
2/3/77
0.352
0.0455
3.41
5.11 x
ID'4
3.35
334
46.6
32.52
0.70
4.95 x
io"3
0.31
47.4
2.98
4.54 x
l
-------
TABLE 6 (continued)
Ethanol,
butanes, and
butenes (ppm)
Total HC (ppm)
7,12-
dimethylbenz
a *, anthra-
cene (ng/scf)
benz a pyrene*
(ng/scf)
3-methylcho-
lanthrene*
(ng/scf)
dibenz a,h *
anthracene
(ng/scf)
Boiler
Ib coal/hr
Ib wood/hr
Ib steam/hr
106 Btu/hr in
% excess air
Steam enthalpy
(Btu/hr)/
Fuel heat in
1/31/77
2;1 wood to coal volume ratio
Run No. 1 Run No. 2
2/3/77 2/3/77 Avg
= 15.1
797
573
8,518
13
81
= 4.4
603
418
6,567
9.7
341
trace
= 3.8
603
418
6,567
9.7
trace
=8
<4
<7
<35
668
470
7,217
10.8
201
(Btu/hr)
flue gas flow
(dscfm)
flue gas flow
(acfm)
0.78
8,624
15,058
0.81
7,698
12,305
0.81
8,753
13,921
0.80
8,358
13,761
* Below detection limits
40
-------
TABLE 7. COMPARISON OF MEASURED AND CALCULATED FLUE GAS FLOW RATES
Run No.
Date
Measured*
Calculated*
1/28
8,266
11,186
Coa^
No. 1
2/2
7,660
8,546
No. 2
2/2
7,402
8,358
No. 1
2/1
8,365
10,863
1:1
No. 2
2/1
7,463
10,442
2/4
7,977
10,383
1/31
9,202
8,624
2:1
No. 1
2/3
8,538
7,698
No. 2
2/3
7,690
8,753
* Flue gas flow rate - dscfm
-------
V = true duct velocity at sampling location
Vs = sampling velocity
a = parameter (0.5 recommended by Badzioch for stoker-
fired boilers)
The correction of the measured particulate concentration is
predicted on the reasonable assumption that, on the average, the
ratio of true velocity to sampling velocity at each point will
be equal to the ratio of corrected flue gas flow-rate to measured
flue gas flow-rate. Therefore, Equation 2 becomes:
C = Cs[22| + (1 - a)]'1 where: (3)
C = true fly ash concentration
Cs = sampled fly ash concentration
Qc = corrected flue gas flow rate
Qs .= sampled flue gas flow rate
a = parameter (as before)
These values of particulate were plotted in Figure 8 versus
excess air. A least square fit was drawn through the data and
the correlation was quite good. A large part of the change in
particulate loading can be related to the large changes in excess
air. This was especially true in the 1:1 ratio wood to coal.
The particulate was also plotted versus weight ratio of coal in
the fuel in Figure 9. A least square fit was constructed through
this data, and then to correct for excess air, the excess air
versus weight ratio was plotted. These data were also fitted
with a first order, least square fit. Then by taking the excess
air change for the weight ratios of the data times the slope of
the particulate-versus-excess air curve, a correction was made
to the particulate-versus-weight ratio plot.
The increase in particulate from all coal to 59% coal was
approximately 14%, on the corrected curve a change from 0.2916
lb/106 Btu to 0.33 lb/106 Btu, respectively.
No consistent increases or decreases were apparent in the
NOX, S02, or POMs when increasing amounts of wood chips are
substituted for coal. There does appear, however, to be a
consistent increase in total hydrocarbon (THC).
It was expected that SC>2 would decrease substantially with
increasing amounts of wood fuel. The absence of this expected
trend coupled with the low ratios of measured SC>2 to expected S02
has cast doubt on the reliability of the S02 results. It was
initially thought that perhaps the S02 concentration was non-
uniform across the breeching at the sampling location due to
42
-------
£ 0.4
-------
UJ
o
X
UJ
190
170
150
130
110
90
70
50
30
oo
£
O
^
*
.0
_l
-0.4
-0,3
-0.2
- 0.1
0.0
ILb./IO6 Btu = 4.30xlO"4 Kg/MJ
O
A PARTICULATE
(% WEIGHT vs
PARTICULATE O)
(CORRECTED CURVE)
(% WEIGHT vs EXCESS AIR D)
n
I I
50
60 70 80 90
WEIGHT PERCENT COAL
I | I
100
Figure 9. Particulate Correction as a Function of Weight
Percent Coal
44
-------
stratified in-leakage of ambient air. However, a check of the
temperature profile and Reynolds number in the breeching at the
sampling location indicates that mixing did take place as
demonstrated in calculations of these profiles and Reynolds
Numbers (12).
An additional basis for comparison of particulate and SC>2
results is provided by a compliance test performed on the boiler
in 1974 while firing coal only. Table 8 gives the average of
this compliance test compared to the average of the coal-only
results from this test program. As can be seen from Table 8,
there is agreement on the particulate results, but again, it
appears that the current SC>2 results are invalid.
TABLE 8. COMPARISON OF 1974 COMPLIANCE TEST WITH COAL-ONLY
RESULTS OF CURRENT PROGRAM
Test dates
1974
1977
Heat input (MBtu/hr)
Steam production (Ib/hr)
Particulate (Ib/MBtu)
Particulate (gr/scf)
SO (ppm)
SOX (Ib/hr)
Measured SOx/Expected SOX
Sulfur in coal (%)
Ash in coal (%)
16.95
12,310
0.29
0.065
744
65.4
1.05
2.14
11.17
9.6
6,223
0.255
0.0299
285
27.44
0.43
3.17
11.53
Particle size determination was attempted with impactors
during each test; however, the results were invalidated by the
build-up of a static electricity charge on the collected sample.
This charge build-up prohibited accurate weighing of the small
amount of particulate collected (in fact, negative weight gains
were detected on some stages).
45
-------
FUEL TESTING (UNIVERSITY OF MISSOURI-ROLLA)
Further testing of the fuel was undertaken during the months
of June, July, and August of 1977 to better define the SC>2
concentrations and nitrogen oxides and particulate effluents.
In order to try to avoid the problems of dilution of effluents
with air leakage into the breeching, all the measurements were
taken at the top of the boiler. The sampling location was the
section of breeching connecting No. 2 boiler to the main
breeching serving all three boilers; see Figure 6.
Basically three different blends were again fired; 2:1, 1:1,
and all coal, for three different coals; see Table 2. An attempt
was made to duplicate the steam flows and flue gas temperatures
for each run. A steam load of approximately 6000 Ib/hr (2733
Kg/hr) was the normal goal. To accomplish this goal, manual
control was normally used as with the previous test except for
the occasional automatic mode when running with all coal.
During each run, three categories of data were taken.
First, a performance comparison of the blended fuels was made
using the boiler charts and integrators, and periodic Orsat
analyses (every 30 minutes). Second, emissions of sulfur
dioxide, nitrogen oxides, and particulate matter were found by
continuous monitoring of the flue gases. Finally, daily tests
were conducted to determine heating values of the fuels and
refuse (bomb calorimeter), sulfur contents of the fuels and
refuse (titration), and fuel moisture contents.
The basic system employed to determine the gaseous
pollutants in the flue gas involved an electrochemical gas monitor
that provided sulfur dioxide and nitrogen oxides concentrations
in parts per million on a continuous basis. Upstream, removal
of particulate matter was accomplished by a paper filter thimble,
and the water vapor removed by an ice bath condenser. Calibra-
tion of the instrument using standard sulfur dioxide and nitric
oxide calibration of the instrument was performed at least once
daily using standard sulfur dioxide and nitric oxide calibration
gases. Periodic sampling per EPA Method 6 was performed to veri-
fy the accuracy of the instrumentation.
Particulate data was obtained in the form of opacity with
a stack-mounted transmissometer. The transmissometer was a
fully automated system using a modulated light beam with 3%
accuracy. The unit provided a zero and span calibration to the
46
-------
system automatically every hour. The output was read and
recorded every 30 minutes.
47
-------
ANALYSIS
The analytical technique used to reduce the data was
basically a combustion analysis described in many tests on the
subject (12-15), including a heat balance of each run made.
Modifications to the analysis included a totally dry basis con-
sideration and the use of both carbon dioxide and oxygen Orsat
results to determine excess air. The dry basis analysis is
desirable, since both the gas monitor and the Orsat apparatus
operate essentially on the dry basis.
First/ the net composition of the fuel is determined from
the volumetric ratio of the wood to the coal and the respective
ultimate analyses and moisture contents. By weighing several
samples, it was determined that for the coal and wood sizing
and moisture contents experienced, a mixture of 50% coal/50%
wood by volume corresponds to 72% coal/28% wood by weight.
Accordingly, weight fractions for other blends may be calculated,
given the volumetric ratio and considering any coal added along
the sides of the blended fuel, by the relationship
WF =
72
M
+ M
M
(4)
where WFC is the weight fraction of coal in the mixture burned,
Mp is the weight of the blended fuel, Mc is the weight of any
additional coal that may be added along 'the sides, and G is the
volumetric ratio of wood to coal.
To calculate the effective composition of the fuel fired,
the following expression is used for all components except
oxygen and hydrogen:
Z = Z WF (1-MC ) +
C^ *•* v
Z (1-WF ) (1-MC )
Vr t^ Wr
(5)
where Z, Zc, Zw are the weight percentages of a component of the
net fuel, coal, and wood respectively, MCC is the moisture
content of the coal, and MCw is the moisture content of the wood.
For the oxygen and hydrogen contents,
48
-------
02 = [QC(1-MCC) + S^L MCc]WFc + [Ow(l-MCw) -f
(l-WFc) (5a)
H2 = EHC(1-MCC) + ^ MCc]WFc + CHW(1-MCW) + ^ MCw]
(1-WP ) (5b)
The symbols are the same as in the previous case with 02
representing the weight percentage of oxygen and H2 representing
the weight percentage of hydrogen.
The refuse contained often times considerable unburned
carbon and sulfur. To account for this correction, the carbon
and sulfur weight percentages were modified by
CAB - c - l=y A
SAB = A - I=fr A <6b)
In these expressions, CAB and SAB denote the weight
percentages (pound constituent per 100 pounds fuel) actually
burned, while C, S, and A are the weight percentages of carbon,
sulfur, and ash respectively in the composite fuel. As
calculated previously, 6 symbolizes the weight fraction of carbon
in the refuse, and y symbolizes the weight fraction of sulfur
in the refuse.
Next, the theoretical amount of oxygen required per 100
pounds of fuel is found by a molar balance and the ideal gas law.
¥ < - +
where VTO is the volume of theoretical oxygen required per 100
pounds of fuel, R is the ideal gas constant (1,545 Ibf-ft/lb-
mole-°F) , T is the absolute temperature in degrees Rankine, and
P is the pressure, in pounds per square foot. The corresponding
volume of nitrogen accompanying the oxygen (per 100 pounds of
fuel) is given by VTN/ where
VTN = IT VTO (8)
The theoretical volume of dry combustion products plus the
volume of nitrogen from the fuel plus the volume of nitrogen
from the combustion air is then equal to the theoretical dry
49
-------
flue gas volume per 100 pounds of fuel, or
.. _ RT ,CAB , SAB N2. . ..
V - { + ~32 + > + V
CT 2 32 2 TN
Here, Vc«r represents the theoretical dry flue gas volume
per 100 pounds of fuel, and N2 represents the pounds of nitrogen
per 100 pounds of fuel.
The actual volume of dry flue gas per 100 pounds of fuel
can be written as V , where
V = v + V (10)
ACT VCT EA UU'
Here, Vg, is the volume of excess air introduced into the
system per lufj pounds of fuel. VCT is calculated assuming
stoichiometric fuel mixtures, but VEA maY ^e calculated from
the Orsat analysis of the flue gas. The amount of excess air
in this boiler is considerable, often over 120%, and consequently
the amount of carbon monoxide present is extremely small, and
may be neglected. Determination of VEA was done considering
both the carbon dioxide and oxygen percentages, and results
averaged. The following formulas were used.
RT ,CAB SAB>
P 19 "39'
a po = - _ _ c\ i =1 ^
^U2 ^v T^J U-l-a)
^ VCT EA|
U2
CO,
21
% °2 = V - + V | (12a)
^ VCT + VEA|
2
V (% 0 )
VEA = 2>% 0 <12b>
VEA
50
-------
Here, % CC>2 and % 62 are the volumetric percentages of
carbon dioxide and oxygen respectively from the Orsat analysis
of the flue gas. VEA|CO and VEA|O represent the volume of
excess air present per 100 pounds of fuel from carbon dioxide
and oxygen analyses respectively. Note that the weight
percentage of sulfur actually burned appears in Equations (lla)
and (lib). The reason for this is that the Orsat apparatus
absorbs both carbon dioxide and sulfur dioxide during the carbon
dioxide measurement.
It was found in this research that the combined results of
a carbon balance and an oxygen balance tended to average but
errors made in reading the Orsat, and gave satisfactory results.
Now that the volume of dry flue gas per 100 pounds of fuel
has been calculated, the fuel rate in pounds per hour may be
used to calculate the volumetric flow rate of the dry flue gas.
When the pressure and temperature are taken as instrument
conditions, the volumetric flow rate may be combined with the
concentration of the pollutants and the ideal gas law to yield
the mass flow rate of the pollutant.
The heating rate may be computed by considering the dry,
higher heating values of the coal and wood, the fueling rate,
and the moisture contents and weight fractions.
HR = FR [HHV (1-MC )WF + HHV (1-MC )(1-WF )] (14)
c c c w w c
where HR denotes the heating rate in unit energy input per unit
time, HHVC and HHVW represent the dry higher heating values of
the coal and wood respectively. Fueling rate (FR) is unit
weight of fuel per unit time.
51
-------
RESULTS AND DISCUSSION
The results for gaseous and particulate emissions are
graphically presented in Figures 10, 12, and 13.
Sulfur dioxide emissions, as plotted in Figure 10, show a
consistent decrease in the pounds per million Btu emissions
index as the percentage of wood, increases in the fired fuel.
The solid lines indicate the results of the experimental data
while the dashed lines represent the expected data based on the
sulfur actually burned. Sulfur accountability was rather
encouraging, deviating about 17% at the most from the ideal or
expected value. The curve in Figure 11 was an attempt to
normalize the data. Each point was divided by an all coal value
for that set of coal runs and the expected curve was the
normalized expected curve for the 3.12% sulfur coal. This was
identified as coal No. 2 in Table 2.
The nitrogen oxides data, presented in Figure 2, were
scattered as expected, since formation of nitrogen oxides is
basically a function of flame temperatures, residence time, and
percent excess air, the first two of which are unaccounted for
variables in this testing program. Due to this uncertainty,
no relationship between nitrogen oxides emissions and the wood
content of the fired fuel could be determined. The emissions
were also considered versus percent excess air, and moisture
content, with no success. A complete study of the nitrogen
oxides emissions would require additional experimental con-
siderations not included in this work.
The particulate emissions, considered as opacity, are
plotted versus excess air in Figure 13. The presentation of the
data in this manner presents an almost linear relationship
between opacity and excess air. The linear relationship to
excess air was the same as that demonstrated by the particulate
data taken by Midwest Research Institute. Again the opacity
versus weight percent was plotted in Figure 14 for all three
fuels. This data was fitted with least square linear approxima-
tion as plotted. In order to investigate the effect of the
excess air, the opacity versus weight was plotted in Figure 15
for 2.44% sulfur coal. Again, the change in the excess air
from the linear curve was found to be 63% in going from 59%
coal to all coal on the curve. Multiplying this change in
excess air by the slope of the opacity versus excess air curve
CO.211) produces a Aopacity due to excess air of 1.33.
52
-------
U1
m 5
*0
O 2.44% A 3.12% X 1.91% A 3.29%
I Lb/Btu = 4.30 xlO"4Kg/J
! 5t
o> P
A
o
Q_
*
Q 3
X
O
0 2
-------
1.2
1.0
•o
0)
LU
"o
o
o
cf
CO
0>
(O
o
0)
CM
0.8
0.6
0.4
0.2
A O
n
o
n
O-2.44 % SULFUR COAL
D-l.91%
A -3.12%
D
10
20
30
40
50
WEIGHT PERCENT WOOD
Figure 11. Normalized Sulfur Dioxide Output as a Function of
Wood Percent
54
-------
Ul
Ul
0>
Q-0.8
0.6
«*
CO
UJ
Q0.4
X
o
o
o
So
1 1 I
O 2.44 % A 3.21% X 1.91%
I Lb/Btu = 4.30 x IO"4 Kg/J
A
O
A X
X
1
I
I
A 3.29%
10 20 30
WEIGHT PERCENT WOOD
O X
40
50
Figure 12. Nitrogen Oxides Emissions as a Function of
Fuel Blend
-------
10-
Ul
a\
0 ft
2 8
o
UJ
£6
UJ
Q_
50
i i i i
O 2.44% A 3.21%
X 1.91 %
3.29%
O
1.91%
O ,
70
VOLUME
90
110
130
150
PERCENT EXCESS AIR
Figure 13. Opacity of Flue Gas as a Function of Excess Air
and Coal Sulfur Content
-------
9.0
8.0
7.0-
6.0
>•
a 5.0
4.0
3.0
A
50
D
D
A
I.9I%SULFURA)
3.12% SULFURO)
O
;2.44%SULFURQ)
D
D
60 70
90 K>0
WEIGHT PERCENT COAL
Figure 14.
Opacity as a Function of the Weight Percent of
the Coal
57
-------
Subtracting this from the curve in Figure 15 produced a constant
opacity for the weight variation in coal from 59% to 100%.
Suggesting no change in the particulate loading over the weight
percent range, it was assumed the other two fuels would have
similar behavior. This behavior was also suggested by the MRI
data. To link the Midwest Research results to the results of
the summer testing program, tests of the River King coal was
made. The opacity readings, the SC>2, NOX levels were taken for
this coal. The run produced an average opacity of 7.1 at an
excess air value of 88%. With this value of excess air, the
particulate value was found from Figure 8. The value was
0.282 lb/106 Btu. The SO2 value for the run was 4.7 lb/106
Btu accounting for 98% of the expected amount and the NOX was
2.2 lb/106 Btu.
To define the boiler performance operation on the different
coals and blends boiler heat balance data was measured and
presented. The experimental results for the various losses
considered were plotted as percent loss versus the weight
percentage of wood in the fuel fired. The results achieved
were essentially as expected, in view of the natures of the coal
and wood. It must be remembered, however, that presentation of
the results as percentages might prove deceptive since the net
heating value of the fuels change with the wood and moisture
content of the composite fuels.
Figure 16 depicts the decrease in both dry gas loss and loss
to unburned combustible as wood content increases. These are
logical results, since the weight of both combustion products
and refuse per pound of fuel decrease with increasing wood
content.
Presented in Figure 17 are the various moisture losses. As
expected, the most severe dependence on wood content appeared
in the fuel moisture losses, and the least dependence in the
air moisture losses.
The net efficiencies (excluding what are normally denoted
as radiation and unaccountable losses) appear in Figure 18. It
is interesting to notice the slight increase in efficiency as
the wood content in the fuel increases. However, in terms of
steam generation, the decrease in the net heating value resulting
from additional wood in the fired fuel offsets any increase in
efficiency.
This fact is pointed out in Figure 19, where the performance
factor, defined as pounds of steam per pound of fuel is plotted
versus the weight percentage of wood in the fuel. A slight
decrease with increasing wood content is evident, as expected.
However, considering the fact that the cost of coal is two to
three times the cost of wood waste, the results are rather
encouraging.
58
-------
QC
5
CO
CO
UJ
O
/70
f50
130
I/O
90
70
50
30
10
- 6
- 5
o
I
- 4
- 3
2.44% SULFUR COAL
n
(%COAL vs OPACITY D)
AOPAC/TY
(opacity corrected) I
a
(% COAL vs EXCESS
A/R A)
A
50
60
70
80
90
fOO
WEIGHT PERCENT COAL
Figure 15. Opacity Correction Curve Function of the Weight
Percent Coal
59
-------
25
-------
FUEL HYDR06E
AIR MOISTURE
LU
O
o:
UJ
a.
FUEL MOISTUR
0 COAL I A COAL 2 X COAL3
1
10
20
30
40
50
WEIGHT PERCENT WOOD
Figure 17.
Moisture Losses Due to Fuel Hydrogen, Air Moisture,
and Fuel Moisture as a Function of Fuel Blend
-------
0 COAL I A COAL 2 X COAL 3
10 20 30
WEIGHT PERCENT WOOD
Figure 18. Net Efficiency Excluding Radiation and Unaccounted
Losses as a Function of Fuel Blend
-------
o\
u>
o:
o
12
10
8
tu 6
o
o:
O
8
— O COAL I A COAL 2 X COAL 3
i
1
10 20 30
WEIGHT PERCENT WOOD
X
A
40
50
Figure 19. Performance Factor (Pound Steam per Pound Fuel)
as a Function of Fuel Blend
-------
As to the experience of actually burning the blended fuels,
it was concluded that while the existing automatic control
system worked relatively well for pure coal, some modifications
would be necessary for the firing of a blended fuel containing
more than 50% wood by volume. The firing of blended fuels
containing, for instance, 67% wood by volume requires significant
operator attention, especially when the moisture content is high
and when load changes are severe. This is due mostly to increased
restriction of the primary air flow caused by the faster stoker
speed and the damp fuel. This problem would not be as severe
with a spreader stoker.
In fact, during the summer test there was a period of one
week where the area received nine inches of rain. This was in
the middle of the testing and the wood stockpile was considerably
depleted, making it much more susceptable to the moisture.
The fuel for the next week was mixed and loaded during the
rainy spell, including mixing on the wet coal yard. The fuel was
so wet, it dripped water and the wood was completely saturated.
When this wet fuel was admitted to the boiler, the boiler began
cooling rapidly. Holes appeared in the air pattern of the fuel
bed and the only way the fire was maintained was with dry
kindling thrown on top of the fuel bed. The result was a total
loss of boiler load and the boiler was quickly switched back to
coal. Several days were allowed for drying and then the fuel
ran again. A similar situation occurred but by closing the
damper and increasing the air until the boiler smoked (positive
pressure) the fuel burned. Firing the wet fuel was a problem
for several weeks following this occurrence. It wouldn't have
been as severe if larger stockpiles of wood were maintained.
The outer layer of wood tends to protect the center of the pile
from absorption. The lack of ignition was quite a change from
the earlier rapid ignition problems. During the entire testing
program, close attention was given to any possibly slagging
problems. There were none. Ash and fuel samples were taken
daily and the decrease in ash was noticed as a function of
increased wood content. The combustible in the ash was also
noticed to be reduced with the addition of the wood.
There are a number of potential sources of error present in
the experimental portion of this work, and the methods of calcula-
tion are especially sensitive to some. The largest single
source of error probably resulted from the actual blending of
the fuel, maintaining a uniform blend during fuel handling, and
accounting for that blend in computation. The wood was often
very moist and tended to cling to the sides of the fuel bunker,
segregating the blend. Attempts to correct this were frequently
made, but variations did occur.
An unresolvable source of error appears in the absorption of
sulfur dioxide in the filter, stainless steel filter holder, and
64
-------
stainless steel tubing. From information presented by Slowik
and Sansone (16) , and Byers and Davis (17) , errors due to the
stainless steel portions would be less than 1% while errors due
to the filter paper itself could conceivably be as much as 10%,
depending on the relative humidity of the flue gas.
A further look at the efficiency of this unit was made by
calculating the cost of yearly operation. It was assumed that
the unit (boiler No. 2) would be run for one year at an average
load of 6000 Ib/hr (2730 Kg/hr) of steam. This steam production
is equivalent to the generation of 52,560,000 Ib (23,914,800 Kg)
of steam for the year. This boiler, referring to Figure 19,
assuming all coal operation produced 7.7 Ib of steam/lb of coal.
To generate 52,560,000 Ib of steam would require 6,826,000 Ib of
coal using the 7.7 Ib/lb index. This was 3410 tons of coal.
At $37.53/ton delivered the total coal cost was $128,000.
If the same boiler was run for the year at 6000 Ib/hr with
a blend of 1 wood to 1 coal, the steam required is still
52,560,000 Ib of steam. The boiler production from Figure 19
was 7.5 Ib of steam/lb of fuel. This production and load
required 7,008,000 Ib of fuel or 3504 tons of fuel. The cost
to provide this fuel at $12.50/ton for the wood and $37.53/ton
for coal was $106,950.00. This cost provided a savings of
$21,050 over all coal operations and would have saved 887 tons
of coal.
The amount of fuel for 2:1 was 3576 tons, which was 1645
wood and 1931 coal. The cost was $72,470.00 for coal and
$20,560.5 for wood, a total of $93,032.5 or a potential savings
of $35,000.60 and 1453 tons of coal. The dollar savings for
this 6000 Ib/hr load was plotted in Figure 20 for different
wood weight ratios.
The completion of the testing was an attempt to use bark.
A dump truck load of the bark was brought to tne power plant and
blended with the coal on a 1:1 basis. There were immediate
problems. The fuel segregated badly as it was elevated into the
plant. The segregation was worse in the bunker and the boiler
fuel was constantly varying from all coal to almost all fines
to wood and coal. It was difficult to pass the combustion air
through the blend and holes again appeared. An attempt was made
to fire the boiler like it was fired with the wet fuel, but it
just wasn't successful. The conclusion drawn is that the bark
has to be hogged, but it would still contain excessively high
concentrations of fines. The pelletizing of the bark was
attempted and appears very promising, but wasn't tested. The
HHV density and ash of the pellets were measured, and indicated
they could possibly be fired alone with no coal. This
possibility is very encouraging.
65
-------
120
40|
to
<30|
o
o
C20|
CO
e>
z
colOl
UJ
100
co 80
£T
<
O
O
060
o
o
CO
840
-I
UJ
20
FUEL COST
O 2x1 wood
I xl wood
10
20
30
40
50
WEIGHT PERCENT WOOD
Figure 20. Economic Results from the Blended Fuel Use
66
-------
REFERENCES
1. Lowe, R.A., "Energy Recovery from Waste", 2nd Interim
Report, U.S. Environmental Protection Agency, (SE-360.ii),
1973.
2. Even, J.C., et. al., "Evaluation of the Ames Solid Waste
Recovery System, Part I", Environmental Protection Agency
Technical Series, EPA-60012-77-205, Nov. 1977.
3. Massengale, Robert, "Sawdust, Slab and Edging Weights from
Mixed Oak Logs of the Missouri Ozarks", the National Logger
and Timber Processor, April 1971.
4. Amos, J.A. and Massengale, R. , "The Wood Residues of
Missouri's Eastern Ozarks Region", Report to the Economic
Development Administration, Oct. 1976.
5. "Timber Resource of Missouri Eastern Ozark Region", 1972,
USDA, Forest Service Resource Bulletin, NC-19, Printed 1974.
6. Biotherm Energy Systems, Publication of American Wood Fiber
Institute, Winn., Mich., 48896.
7. Hall, H.H., et. al., "Comparison of Fossil and Wood Fuels",
Environmental Protection Agency Technology Series, EPA-
60012-76-056, March 1976.
8. Jones, S.G. and Massengale, R., "Missouri Sawmill Improve-
ment Program", a Publication of the Missouri Conservation
Department, March 1975.
9. Massengale, R., "Missouri Forest Products Industries",
Missouri Department of Conservation, 1975.
10. Environmental Assessment of Waste to Energy Processes,
"Stationary Source Testing at Power Plant of University of
Missouri at Rolla", EPA Contract No. 68-02-2166, MRI Project
No. 4290-L, September 1977.
11. Badzioch, S., "Correction for an Isokinetic Sampling of
Gas-Borne Dust Particles", Journal of the Institute of
Fuel, March 1960.
67
-------
12. Fryling, C.R., ed., Combustion Engineering/ 2nd edition,
Combustion Engr. Co., Riverside, Cambridge, Mass., 1966.
13. Smith, M.C. and Stinson, K.W. , Fuels and Combustion, Firs.t
Edition, McGraw-Hill, 1952.
14. Potter, P.J., Power Plant Theory and Design, 2nd Edition,
Ronald Press Co., 1959.
15. Anonymous, Steam, 38th Edition, Babcock and Wilcox, New
York, N.Y., 1972.
16. Slowik, A.A. and Sansone, E.B., "Diffusion Losses of Sulfur
Dioxide in Sampling Manifolds", Journal of Applied Pollution
Control Association, Vol. 24, No. 3, March 1974, pp. 245- •
247.
17. Byers, R.L. and Davis, J.W., "Sulfur Dioxide Absorption and
Desorption on Various Filter Media", Journal of Applied
Pollution Control Association, Vol. 20, No. 4, April 1970,
pp. 236-238.
68
-------
APPENDIX A
FACILITIES STATEMENT
The Power Plant to be used in the investigation has four
boilers as listed below:
No. 1 Babcock and Wilcox, Integral Furnace and
Boiler, 1944; 18,000 Ib/hr at 150 psig
saturated
No. 2 Babcock and Wilcox, Integral Furnace and
Boiler; 18,00 Ib/hr at 150 psig saturated
No. 3 Union Iron Works, 1957
35,000 Ib/hr at 150 psig 405°
No. 4 Erie City Iron Works
50,000 Ib/hr at 150 psig, 405°
All the boilers have Laclede chain-grate stokers. The
power plant has two turbines as listed below:
Worthington Turbine, 150 psig steam with a General
Electric Generator, 500 kw at 36000 RPM
Elliott Turbine Generator, 150 psig, steam 36000
RPM, 1000 kw.
The power plant's primary load is the heating and cooling
of the facilities on the campus. The turbines are not presently
being used.
69
-------
APPENDIX B
DATA COLLECTED AND CALCULATED
Date
6/14
6/15
6/16
6/21
6/22
6/22
7/5
7/6
7/7
7/8
7/12
7/13
7/14
7/19
7/20
7/21
7/26
7/27
7/27
7/28
Blend by
Volume
Blend
Iw: Ic
Iwrlc
coal
2w:lc
c
2w:lc
Iw: Ic
Iw: Ic
2w:lc
coal
2w:lc
Iw: Ic
Iw: Ic
2w:lc
2w:lc
3w:2c
coal
coal
coal
coal
Firing
Rate
#/hr
886.7
758.5
831.25
1026.2
954.3
902.1
934.8
921.7
947.0
759.0
990.4
928
950
1162
1048
956
884
835
885
1019
TABLE
DATA
% Sulfur
Ash
% SA
0.72
0.5
1.5
1.43
0.53
1.72
.27
.98
1.08
1.04
1.43
.05
.058
.055
.055
.055
.05
.05
1.3
1.2
B-l
% Carbon
Ash
% CA
21
21
33
22
18.3
20.7
20.2
20.6
21.6
27.9
21
21
19.8
22.6
24.7
27.5
25.4
28
30
31
Weight
Coal Ib/lb
WF
c
0.72
0.72
1
.5967
1
0.5926
0.7565
.7473
.6027
1
.5694
0.72
0.72
.563
.563
.6316
1
1
1
1
Sulfur
in Coal
S
2.44
2.44
2.44
2.44
3.12
2.444
3.12
3.12
3.12
3.12
3.12
1.911
1.911
1.911
1.911
1.911
1.911
3.12
3.12
70
-------
TABLE B-2
DATA PLOTTED
Sulfur
2.44
2.44
2.44
2.44
2.44
3.12
3.12
3.12
3.12
3.12
3.12
3.12
1.91
1.91
1.91
1.91
1.91
3.29
Weight %
Coal
.72
.72
100
.5967
.5926
100
.7576
.7473
.6027
100
.5694
100
.563
.563
.6316
100
100
100
SO 2
lb/106Btu
3.985
3.29
2.84
2.54
3.06
5.05
4.94
3.58
4.51
3.99
3.18
3.9
1.49
2.07
2.6
3.03
2.5
4.54
%
Deviation
126
98.9
75.7
94.8
107
104
115.7
89.8
124.9
86
101
86
72
99.6
107
100
83
.100
NOX
lb/106Btu
.3843
.306
.442
.405
.091
.05
.166
.54
.25
.24
.68
1.3
.46
.43
.51
.138
.36
.37
71
-------
TABLE B-2
DATA PLOTTED
Coal
Sulfur
1.91
1.91
1.91
1.91
1.91
1.91
3.12
3.12
3.12
3.12
3.12
3.29
2.44
2.44
2.44
2.44
2.44
2.44
Opacity
9.4
7.3
7.8
8.2
8.1
8.2
7.4
7
6.7
6.6
7.6
7.1
6.4
4.5
4.6
4.25
6
6.5
Excess Air
95
77
104
112
116
147
113
123
108
106
153
129
88
58
76
106
122.5
135
Weight %
Coal
72
56.3
56.3
100
56.94
63.16
100
100
1
74.13
60.27
100
59.26
72
59.67
72
100
100
72
-------
APPENDIX C
REPORT OF ANALYSIS
MATERIAL:
One (1) sample of coal identified as Coal Sample 12-30-2
TEST REQUESTED:
Analysis of coal and screen test
RESULTS OF TESTS:
SIEVE SIZE % RETAINED ON SIEVE % PASSED SIEVE
1 in
.750 in
.500 in
.375 in
.250 in
# 4
# 6
% Moisture
% Volatile Matter
% Fixed Carbon
% Ash
% Sulfur
Btu
Btu (Moisture & Ash Free)
19.81
18.78
22.00
17.26
12.18
3.54
3.17
'ree)
AS RECEIVED
7.60
35.83
47.00
9.57
100.00
2.03
11,892
80.19
61.41
38.41
22.15
9.97
6.43
3.26
DRY BASIS
38.78
50.86
10.36
100.00
2.2
12,870
14,357
73
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
\. REPORT NO.
EPA-600/7-80-103
2.
3. RECIPIENT'S ACCESSIOf*NO.
4. TITLE ANDSUBTITLE
Wood Waste as a Power Plant Fuel in The Ozarks
5. REPORT DATE
May 1980
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
V. J. Flanigan
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Mechanical Engineering Department
University of Missouri - Rolla
Rolla, Missouri 65401
10. PROGRAM ELEMENT NO.
EHE 62UB
11. CONTRACT/GRANT NO.
R804270-010
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Industrial Environmental Research Laboratory
Office of Research and Development
U. S. Environmental Protection Agency
Cincinnati. Ohio 45268
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The report discusses the testing program conducted on a chain-grate stoker
boiler with a blended coal and wood waste fuel. The boiler was designed to produce
18,000 Ib/hr (8,200 kg/hr) of saturated steam at 150 psig (103.4 N/cm2). The
objective of the tests was to determine the difference, if any, in the performance
and the emissions of the boiler co-firing wood and coal as compared to firing coal
alone. Four different coals with different sulfur contents were fired with the wood
waste. The wood waste content was varied up to 2/3 by volume.
Results indicate that particulate and nitrogen oxide emissions are not substan-
tially altered by using wood waste as supplemental fuel with coal for the test
conditions adopted in the program. Sulfur dioxide emissions decreased with increased
proportions of wood waste in the coal-wood mixture, whereas total hydrocarbon concen-
tration increased when wood waste content was increased. Cyanide concentrations were
not substantially affected by substitution of wood waste for coal; particulate poly-
cyclic organic emissions were below the detectable limits for all fuels used in the
tests.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. cos AT I Field/Group
Air Pollution
Wood Wastes
Environmental Tests
Boilers
Wood-Fired Boiler
SASS Train
SAM-1A Analysis
Multimedia Effluent
Sampling
18. DISTRIBUTION STATEMENT
Release to the Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OP PAGES
86
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
. US GOVERNMENT MINTING OFFICt 1980-657-146/5680
------- |