U S Environmental Protection Agency
Office of Research and Development
Electric Power Research
EPA-600/7-78-0368
Industrial Environmental Research Cr A~oOO/7~ I
Laboratory + f\^n
Research Triangle Park North Carolina 27711 MSCCR 1978
hossil Fuel Powi••
Plants Department
Palo Alto. California 94303
CONTROL OF UTILITY BOILER
AND GAS TURBINE POLLUTANT
EMISSIONS BY COMBUSTION
MODIFICATION - Phase I
Interagency
Energy-Environment
Research and Development
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 poHutants 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
REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication Approval does not signify that the contents necessarily reflect the
views and policies of the Government. nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161
-------
EPA-600/7-78-0363
March 1978
CONTROL OF UTILITY BOILER
AND GAS TURBINE POLLUTANT
EMISSIONS BY COMBUSTION
MODIFICATION - Phase I
by
A.R Crawford. E H Manny, and W Bartok
Exxon Research and Engineering Company
P 0 Box 8
Linden, New Jersey 07036
Contract No 68-02-1415
EPA Program Element No. EHE624A
EPRI Project No. 200
EPA Project Officer: Robert E. Hall
EPRI Project Manager. Robert C Carr
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, N.C. 27711
Prepared for
Fossil Fuel Power Plants Department Office of Research and Development
Electric Power Research Institute and U S. Environmental Protection Agency
Palo Alto, California 94303 Washington, D.C. 20460
-------
TABLE OF CONTENTS
Page
SUN .ARY
1. INTRODUCTION . . . 3
2. FIELD STUDY PLANNING AND PROCEDURES 5
Selection of Power Generation Combustion Equipment.
Test Program Strategy
Gaseous Sampling and Analysis
Particulate Sampling
Furnace Waterwall Fireside Corrosion Probe Measurements
Boiler Efficiency
S0 2 /S0 3 Gaseous Sampling
Stack Plume Opacity
3. RESULTS AND DISCUSSION 19
3.1 Gaseous Emission Results . • 19
3.1.1 Widows Creek, Boiler No. 5
(Tennessee Valley Authority)
3.1.2 Ernest C. Gaston, Boiler No. 1
(Southern Electric Generating Company)
3.1.3 Navajo Station, Boiler No. 2,
Salt River Project
3.1.4 Comanche Station, Boiler No. 1,
Public Service Co. of Colorado, Pueblo,
3.1.5 Barry Station, Boiler No. 2
(Alabama Power Company)
3.1.6 Morgantown, Boiler No. 1
(Potomac Electric Power Company) . .
3.1.7 Mercer Station, Boiler No. 1,
Public Service Electric and Gas Company
(New Jersey)
3.1.8 Morgantown Gas Turbine No. 3
(Potomac Electric Company)
Particulate Emission Results
Furnace Waterwall Corrosion Probing
Effect of Combustion Modifications on Boiler Performance.
Summary of Sulfur Oxides Measurements
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
5
6
10
13
14
15
18
18
Colorado
20
26
32
40
45
53
60
66
68
81
88
92
3.2
3.3
3.4
3.5
i
-------
TABLE OF CONTENTS (Continued)
Page
4. CONCLUSIONS 98
4.1 Gaseous Emission Measurements 98
4.2 Side Effects of Combustion Modifications 105
5. REFERENCES 107
ACKNOWLEDGMENTS 108
APPENDIX A — Cross Sectional Drawings of
Typical Utility Boilers 109
APPENDIX B — Conversion Factors 113
APPENDIX C — Operating and Gaseous
EmissionsSummary 114
APPENDIX D — Coal Analyses 123
APPENDIX E - E ocon Mobile Van Analytical
Sampling System 131
ii
-------
LIST OF FIGURES
Number Page
2—1 Exxon Research transportable sampling and analytical
system 11
2—2 Corrosion probe, detail of 2½” extension pipe and end
plate (outside of furnace) 16
2—3 Corrosion probe, detail of corrosion coupon assembly
(inside of furnace) 17
3—1 NO emissions vs. staged firing patterns for Widows
CreekNo. 5 unit 21
3—2 PPM NOx vs. % stoichiometric air to active burners for
Widows Creek, boiler No. 5 • • 22
3—3 Babcock and Wilcox dual register pulverized coal burner. . 27
3—4 Mill—burner configuration of E. C. Gaston boilers No. 1
and 2 28
3_5 NOx emissions vs. % stoichiometric air for Gaston
boiler No. 1 31
3—6 PPM NO emissions vs. % oxygen for Navajo No. 2 unit
under full load, normal firing operation 34
37 PPM NOx emissions vs. % oxygen for Navajo No. 2 unit
operating with overf ire air dampers 100% open 35
3—8 PPM NO emissions vs. % oxygen for Navajo No. 2 unit
under staged firing operation 37
3—9 Effect of opening overf ire air dampers on NO emissions
from Navajo No. 2 unit at full load 38
3—10 Effect of burner nozzle tilt on NO emissions from
ComancheNo. lunit 42
3—11 Effect of overfire air damper setting on Comanche No. 1
unitN0 emissions 43
3—12 N emissions vs. flue gas oxygen measurements on
ComancheNo. lunit 44
3—13 Side elevation drawing of Barry No. 2 unit 46
3—14 N emissions vs % in flue gas for 100% coal fired test
runs on BarryNo. 2 unit 50
3—15 NQ emissions vs % coal in coal—gas mixed—fuel firing
operation on Barry No. 2 unit 51
iii
-------
LIST OF FIGURES (Continued)
Number Page
3—16 NOx emissions vs % coal in oil—coal mixed—fuel, normal
firing operation on Morgantown No. 1 unit
3—17 NOx emissions vs % coal in coal/oil mixed fuel, staged
firing operation on Morgantown No. 1 unit 56
3—18 PPM NOx emissions vs % coal in coal/oil mixed—fuel
firing operation at reduced load on Morgantown
No. 1 unit
3—19 NOx emissions vs % 02 in flue gas on Mercer No. 1 unit . . 64
3—20 NOx emissions vs gross load — Morgantown gas turbine
No. 3 67
3—21 Particulate emissions from utility boilers 71
3—22 Particle size distribution, Southern Electric Generating
Company, F. C. Gaston, boilers No. 1 and No. 2
3—23 Particle size distribution, Salt River project, Navajo
station, boilerNo. 2 78
3—24 Particle size distribution, Public Service Company of
Colorado, Comanche station, boiler No. 1 7 9
3—25 Particle size distribution, Public Service Electric and
Gas Co., Mercer station, boiler No. 1 80
4—1 PPM N0 vs % 02 in flue gas from coal—fired boilers under
normal firing operation 101
4—2 Effect of excess air level on NOx emissions for coal
fired boilers under normal firing operations 103
4—3 Effect of excess air on NO emissions of coal fired
boilers under modified firing conditions 104
iv
-------
Number
LIST OF TABLES
Page
2—1 Summary of Coal and Mixed—Fuel Fired
Boilers Tested
2—2 Summary of Coal—Fired Boilers Tested
Exxon Field Test Programs
in a Previous
7
8
12
23
29
33
39
48
54
61
62
63
66
70
73
9
2—3 Summary of Oil—Fired Boilers Tested
in a Previous Exxon Field Program
2—4 Continuous Analytical Instruments in
ER&EVan
3-1 Test Program Experimental Design - PPM NO and % 02
Widows Creek, Boiler No. 5
3—2 Test Program Experimental Design —
(Ernest C. Gaston, Boiler No. 1)
3—3 Test Program Experimental Design — Run No., % 02,
PPM NO (Navajo No. 2 Unit — Full Load (795—808 MWe)
3—4 Reduced Load Test Data —
Navajo No. 2 Unit
3-5 Test Program Experimental Design - Run No., % 02,
PPM NO (Barry No. 2 Boiler)
3—6 Test Program Experimental Design —
Morgantown Boiler No. 1
3—7 Summary of Operating and Emission Data
(Mercer No. 1 Unit)
3—8 Summary of Operating and Emission Data
Mercer No. 1 Unit, Reduced Load Test
3—9 Test Program Experimental Design — Run No., %
PPM NOx (Mercer No. 1 Boiler)
3—10 Summary of Operating and Emission Data —
Morgantown Station, Gas Turbine No. 3
3—11 Particulate Emission Test Results. . .
3—12 Particle Size Distribution, Comanche Station,
Boiler No. 1
3—13 Particle Size Distribution, Navajo Station,
Boiler No. 2 74
3—14 Particle Size Distribution, E.C. Gaston Station,
Boilers No. 1 and No. 2 75
3—15 Particle Size Distribution, Mercer Station,
Boiler No. 1 76
V
-------
LIST OF TABLES (Continued)
Number
3—16 Tennessee Valley Authority — Widows Creek Station —
Corrosion Rate Data
Page
82
3—17
Southern Electric Generating Company —
E. C. Gaston Station, Corrosion Rate Data. . . .
83
3-18
Salt River Project — Navajo Station,
CorrosionRate Data
84
3—19
Public Service Company of Colorado,
Comanche Station, Corrosion Rate Data
85
3—20
Public Service Electric and Gas Company,
Mercer Station, Corrosion Rate Data
86
3—21
ASME Test Form for Abbreviated Efficiency Test
89
3—22
ASME Test Form for Abbreviated Efficiency Test
90
3—23
Summary of Boiler Performance Calculations
91
3—24
Public Service Electric and Gas Company,
Mercer Station, S0 2 /S0 3 Analyses
93
3—25
Public Service Company of Coloradb,
Comanche Station, Pueblo, Colorado, S0 2 /S0 3 Analyses . .
94
3—26
Salt River Project, Navajo Generating Station,
S0 2 /S0 3 Analyses
95
3—27
Southern Electric Generating Company,
E. C. Gaston Station, Wilsonville, Alabama,
S0 2 /S0 3 Analyses
96
3—28
Potomac Electric Power Company
Morgantown Station, Morgantown, Maryland,
502/503 Analyses
97
4—1
Sunnary of NO Emissions from Coal Fired Boilers . . . .
99
4—2
Navajo No. 2 — Baseline Operation Emission Data
100
vi
-------
SUMMARY
Exxon Research and Engineering Company (ER&E) has been performing
a continuing field study on the application of combustion modification to
control pollutant emissions from power generation combustion sources.
This program, jointly sponsored by EPA (Contract No. 68—02—1415) and EPRI
(Project No. 200), was conducted with the cooperation of equipment
manufacturers and equipment operators. Measurements of gaseous emissions,
particulate mass, particulate size distribution, accelerated corrosion, and
slagging tendency, and boiler efficiency were made using ER&E’s mobile
sampling/analytical van.
The major emphasis in this program was placed upon combustion
modification of coal—fired boilers. In addition, field testing of mixed—
fuel fired boilers, waste fuel—fired boilers, gas turbines and stationary
internal combustion engines is also included within the scope of this program.
Seven coal—fired boilers, including two with mixed—fuel firing
(coal—oil, coal—gas) capabilities and a gas turbine were tested during the
period of performance between July 1974 and December 1975 under this contract.
Among the units tested, four coal—fired boilers were equipped with special
NOx emission reduction equipment. Three of the Combustion Engineering
designed boilers tested have overf ire air ports and a Babcock and Wilcox
designed boiler has been retrofitted with that company’s “low N0, ”, dual
register, compartimentalized windbox coal burners. A Foster Wheeler boiler
designed to operate with a wet bottom at low loads was also tested for
possible side effects under “low NOx” operation. The three—phase test
programs conducted on most of these units consisted of, first, a statistically
designed short period test run to determine baseline and minimum NOx emission
capability of the unit, second, a one or two day sustained run under “low NOx”
operation to determine if potential operating problems arise and third, a
300—hour sustained run under baseline and tilow NO x” conditions to check on
other potential adverse side effects such as increased corrosion and slagging
problems, loss in boiler efficiency and changes in particulate emissions.
Several key findings were determined from an analysis of the
gaseous emission data. The B&W boiler retrofitted with the new burners
emitted 36% less N0, under normal firing and over 50% less NOx under full
load, staged firing operation than was emitted from an identical boiler
equipped with conventional burners. Tangentially fired boilers equipped
with overf ire air ports emitted about 20% less NOx than similar boilers
when tested under normal firing operation. Modified firing using overfire
air ports and low excess air resulted in an average of over 40% additional
reduction in NOx emissions at full load, for a reduction of over 50 percent
over boilers not equipped with overf ire air ports. Mixed fuel—fired boiler
NOx emission levels increased as the % of coal in the coal—oil or coal—gas
mixture increased, but not linearly. NO emissions from a 50 MWe electrical
output, gas turbine were about 219 ng/J t375 ppm on a 3% 02 basis) at full
load, 234 ng/J (400 ppm) at peak load, with reductions to about 190 ng/J and
146 ng/J (325 and 250 ppm) at 50% and 20% of full load, respectively.
(Note that all ppm emission values given in this report are corrected
to a 02, dry basis and all megawatt values are on an electrical output
basis, M% e).
1
-------
Only relatively minor differences have been observed in
particulate mass loadings between “low NOR” and baseline firing conditions.
Unburned carbon in the fly—ash, which was observed to increase in previous
studies (especially on front wall and horizontally opposed fired boilers)
was found to decrease on some of the boilers tested in this program.
Particle size distribution, which could affect electrostatic precipitator
collection efficiency, was found to be very similar for “low NOR” operation
and for baseline conditions. As in previous studies, no major changes in
boiler efficiency have been observed as a result of “low NOR” operation.
In 300—hour sustained runs, no major differences in corrosion
rates of corrosion coupons have been observed for low NO firing compared
to normal operation. However, long term tests measuring actual furnace
wall tube wastage are needed to settle this important question.
2
-------
1. INTRODUCTION
Since 1970, Exxon Research and Engineering Company (ER&E) has
been conducting field studies under EPA sponsorship on the application
of combustion modification techniques to control pollutant emissions from
utility boilers. The emphasis in these studies has been on controlling
NOx emissions without causing adverse side—effects.
Under EPA Contract No. CPA 70—90 (i), significant reductions of
NO were achieved for gas and oil—fired boilers using combustion modification
techniques in field testing of limited duration, without attempting to
optimize the technology. The principal modifications investigated consisted
of minimizing excess air, staged introduction of the combustion air, flue
gas recixculation, varying boiler load, and varying air preheat temperature.
Also, as part of this study, it was possible to achieve significant reductions
in NOx emissions for two of the seven coal—fired boilers tested, through the
combination of low excess air with staged firing.
Because of the difficulty of controlling NOx emissions from coal—
fired boilers, in the subsequent EPA—sponsored ER&E study (Contract No.
68—02—0227) ( ) the emphasis shifted to a more detailed investigation of
emission control for coal—fired utility boilers, again in cooperation with
boiler owner—operators and manufacturers. These field studies, on twelve
coal—fired units representative of the current design practices of the
major U.S. boiler manufacturers (Babcock and Wilcox, Combustion Engineering,
Foster Wheeler, and Riley—Stoker) , have produced very promising results. It
was possible to achieve reductions in N0 emissions ranging between about
30% and 50%, without apparent adverse side—effects. In addition to gaseous
emissions measurements, the studies included particulate mass loading and
unburned combustible measurements, accelerated furnace corrosion probing,
determination of boiler efficiency, and observations on changes in boiler
operability with particular attention to slagging, fouling, and flame
problems.
Based on the encouraging results of the above work, the Environmental
Protection Agency and the Electric Power Research Institute decided to
jointly fund the present ER&E field study. The scope of the work was
broadened to determine the effects of combustion modification techniques on
the control of pollutant emissions and on the performance of fossil fuel
fired power generation equipment. In this field program, the combustion
equipment selected for study included coal—fired, mixed—fuel fired and
waste fuel—fired boilers to be studied, in addition to short—term tests
on stationary gas turbine and I.C. engine equipment. The principal
continuing emphasis in this work remained on coal—fired utility boilers.
All but one of the coal—fired boilers tested in this study had dry—bottom
furnaces.
In the field studies detailed in this report, the combustion
equipment selected was tested in cooperation with the owner—operators and
the manufacturers. The effectiveness of equipment modifications designed
to provide N0 emission control, such as coal—fired boilers constructed
with overf ire air ports, and the use of low N0 emitting improved burner
design were explored. Potential adverse side—effects of combustion
3
-------
modifications were studied in more detail than previously. Thus, in
addition to the gaseous emission measurements (including pollutants,
stable combustion products and unburned combustibles), particulate mass
and size distribution measurements were made under both normal and low N0
modes of boiler operation.
As in the earlier field studies, the effect of combustion
modifications on boiler efficiency and operability was determined. Special
attention was paid to the demonstration of potential furnace fireside tube
wall corrosion problems that may result from staged firing with coal, i.e.,
with parts of the boiler operated under fuel rich, reducing conditions.
For this purpose, 300—hour corrosion probe runs with air—cooled coupons
were made under both low NO and baseline firing conditions in the boilers
studied.
Under Contract No. 68—02—1415, this final report on Phase I
will be followed by a final report on Phase II of the program. The final
report on Phase II will include summary tables from this report plus test
data on additional pulverized coal—fired boilers (Crist No. 7, Sewaren
No. 5, Cooper No. 2 and Comanche No. 2), additional gas turbines (Wharton
No. 42, and Wharton No. 43). A waste fuel supplemented coal fired boiler
and a pulverized coal—fired boiler utilizing a slag reducing additive may
also be included. Trace species pollutant data determined from the newly
developed Source Assessment Sampling System (SASS) train will also be
covered in the final report on Phase II. A separate, Data Supplement
containing detailed raw pollutant emission and operating data covering
all units tested will be issued following completion of this program.
In addition, two Guideline publications will be issued: one for utility
operators, and a second for utility boiler designers. Furthermore, a
report will be issued which will describe results of a comprehensive,
long—term fireside tube wall corrosion study while operating under baseline
and low NO conditions.
-------
2. FIELD STUDY PLANNING AN]) PROCEDURES
This section discusses the major steps involved in field study
planning and the testing methods used to obtain emission, corrosion and
performance measurements. Field study planning steps included selection
of representative power generation equipment (with participation of EPA,
EPRI, equipment manufacturers and equipment operators) and designing an
effective test program strategy. Methods of gaseous emission testing
were similar to those developed and applied in ER&E’s “Systematic Field
Study” ( ). Particulate sampling, corrosion probing and performance
measurements were also based on methods used in ER&E’s previous field
studies (2). Boilers firing coal, coal—oil and coal—gas mixed fuel were
tested, as was an oil—fired power generating gas turbine.
2.1 SELECTION OF POWER GENERATION COMBUSTION EQUIPMENT
This program provides for the testing of utility boilers (coal—
fired, mixed fossil—fuel fired and waste—fuel fired) utility sized gas
turbines and a large, stationary IC engine. This report covers the field tests
completed on five coal—fired boilers, two mixed—fuel (coal—oil and coal—gas)
fired boilers and an oil—fired gas turbine. No waste—fuel fired utility
boiler could be located for testing during the period of performance covered
by this report. Also, testing of a stationary I.C. engine used for power
generation was deferred to a later date to permit the completion of the
boiler tests.
The successful selection of boilers representing current design
practices was the result of a cooperative planning effort by Exxon Research,
EPA, EPRI and boiler manufacturers (Babcock and Wilcox, Combustion Engi-
neering, Foster Wheeler and Riley Stoker). The process of developing
boiler selection criteria, reviewing boiler manufacturers’ candidate
boilers meeting the criteria, selecting a tentative list of boilers for
detailed field meetings and final selection and scheduling of test programs
was described in a previous report (2).
Design factors were the prime consideration in selecting boilers
in the current field test program after other criteria such as size,
operating flexibility, boiler measurement and control capability had been
met. Boilers representing the current design practices of all four utility
boiler manufacturers were sought for this study. Coal—fired boilers equipped
with overf ire air ports or with new burners installed for NO emission
control were of special importance to compare the emission control performance
of such units to those operated under modified conditions but without
making the NOx control capability an integral feature of the equipment.
The cooperation of boiler operators was secured to use three tangentially
fired (2 coal, 1 coal—gas mixed fuel) boilers with overfire air ports and
a wall fired boiler retrofitted with special low NO burners. Field testing
of an operating front—wall, coal—fired utility boiler equipped with overfire
air ports was attempted but could not be arranged with its owner—operator.
5
-------
The design and operating features of the seven coal or mixed—
fuel fired boilers tested in the current program are summarized in
Table 2—1. Four of these boilers are tangentially fired (ranging in size
from 130 NWe to 800 MWe) and three of them are wall fired (a rear wall fired
125 MWe unit, a front wall fired, 290 MWe unit, and a horizontally opposed
fired 270 MWe capacity unit). As discussed earlier, these boilers have
been selected for field studies at the recommendation of their respective
manufacturers as representative of current design practices and/or because
of specially designed N0 reduction capabilities. Tables 2—2 and 2—3 present
summaries of coal and oil fired boilers tested in a previous program (2)
for comparison purposes.
2.2 TEST PROGRAM STRATEGY
The up—to—date, comprehensive information obtained in field
meetings provided the necessary data to develop detailed, run—by—run
proposed test program plans for review by the sponsors and the cooperat-
ing parties. Each test program, tailored to take full advantage of the
particular combustion control flexibility of each boiler, was comprised
of three phases: (1) short test—period runs, (2) a 1—3 day sustained
“low N0 ” run, and (3) 300—hour sustained “low N0 ” and normal operation
runs. Thus, the strategy used for field testing coal—fired boilers
consisted of first, defining the optimum operating conditions for N0
emission control, without apparent adverse side—effects in short—term
statistically designed test programs. Second, the boiler was operated
for 1—3 days under the “low NOR” conditions determined during the
optimization phase, for assessing boiler operability problems. Finally,
where possible, sustained 300—hour runs were made under both baseline
and modified (“low NOR”) operating conditions. During this period, air—
cooled carbon steel coupons were exposed on corrosion probes in the
vicinity of furnace water tubes, to determine through accelerated cor-
rosion tests whether operating parts of the boiler under the reducing
conditions associated with staged firing results in increased furnace
tube wall corrosion rates. Particulate samples were obtained under
both baseline and “low NOX” conditions. Engineering information on
boiler operability, such as flame instability, impingement and slagging
problems, and data on boiler performance were obtained.
Statistical principles (as discussed in more detail in a previous
report (1)) provided practical guidance in planning the Phase I test programs,
e.g., how many, and which test runs to conduct, as well as the proper
sequence in which they should be run. These procedures allow valid con-
clusions to be drawn from analysis of data on only a small fraction of
the total possible number of different test runs that could be made in
principle, resulting in increased productivity of the research work.
Our earlier studies (2) describe in detail an example of these principles
applied to field testing of a wall—fired boiler.
Statistical principles were also used extensively in the
analysis of emission data recorded from each test run. For example,
least squares regression analysis was used to obtain valid estimates of
the relationship between NO emission levels and operating variables,
such as excess air level. These techniques provide objective, uniform
methods of analysis that can be easily verified, provide the basis for
efficient assessment of outlier data, and limit the risk of drawing
false conclusions.
6
-------
TABLE 2—1
SUMMARY OF COAL AND MIXED FUEL FIRED BOILERS TESTED
No. of NO 5 Emissions
Station and Boiler Type of Fuel(c) NCR No. of Test Test Baseline Low NO B NO (h)
Boiler No. Mfr.(a) Firing(b) Burned Burners Variables Runs ppm (lb/b 6 BTU) ppm (lb/lOt B1 ’U) Redu tion
(I) U )
1. Tennessee Valley Widows Creek — 5 B6W RW C 125 16 4 31(d) 597 (0.81) 468 (0.64) 22
Authority
2. Southern Electric E. C. Gaston — 1 B&W HO(e) C 270 18 5 37(d) 389 (0.53) 278 (0.38) 29
Generating Company
3. Alabama Power Company Barry — 2 CE T(f) CG 130 16 6 38 341 (0.46) 189 (0.26) 45
4. Potomac Electric Morgantown — 1 CE T CO 575 40 5 27 552 (0.75) 403 (0.55) 27
Power Company
5. Salt River Project Navajo — 2 CE T(f) C 800 56 4 36(d) 492 (0.67) 282 (0.38) 43
6. Public Service Company Comanche — 1 CE T(f) C 350 20 4 30(d) 417 (0.57) 261 (0.35) 37
of Colorado
i 7. Public Service Electric Mercer — 1 FW FW(g) C 270 24 4 33(d) 1383 (1.88) 876 (1.19) 37
and Gas Company — __________ __________ —
Average of Coal Fired Boilers 33 656 (0.89) 433 (0.59) 34
(a) 36W — Babcock and Eilcox, CE — Combustion Engineering, FW — Foster Wheeler
(b) RW — rear wall, HO — horizontally opposed, T — tangential, VW — front wall
(c) C — coal, C—G — coal—gas mixed, C—O — coal—oil mixed
(d) Particulate and corrosion probe tests performed on these boilers
(e) Special low NO emission burners
(f) Overfire air ports
(g) Wet bottom furnace
(h) B NO reduction at full or near full load
(i) PPM NO — 3% O . dry basis
-------
(a) B&W — Babcock and Wilcox
CE — Combustion Engineering
F—W — Foster Wheeler Energy
RS — Riley Stoker
TABLE 2-2
SUMMARY OF COAL-FIRED BOILERS TESTED
TN A PREVIOUS EXXON FIELD TEST PROCRAII (2 )
(b) RW — Rear Wall
FW — Front Wall
HO — Horizontally Opposed
T — Tangential
Turbo - Turbo-Furnace
Low NO 5 I NOxte)
ppm (lb/106 BTU) Reduction
(F)
379 (0.50) 40
550 (0.73) 34
463 (0.62) 35
488 (0.65) 48
219 (0.29) 59
273 (0.36) 34
310 (0.41) 24
398 (0.53) 34
359 (0.48) 49
375 (0.50) 33
214 (0.28) 53
384 (0.51 ) 12
368 (0.49) 39
Boiler Operator
Station and
Boiler No.
Boiler
Mfr.(a)
B&W
Type of
Firing(b)
RW
NCR
Q j
125
No. of
Burners
16
Test
Variables
4
No. of
Test
Runs
41 (c)
pp
C !)
Baseline
(lb/ b 6 BTU)
634 (0.84)
1.
Tennessee Valley Authority
Widows Creek - 6
2.
Gulf Power
Crist — 6
F—W
FW
320
16
4
22(c)
832 (1.11)
3.
Georgia Power
Harilee Branch —
3
B&W
HO
480
40
4
51(c)(d)
711 (0.95)
4.
Arizona Public Service
Four Corners — 4
B&W
HO
800
54
5
26(c)(d)
935 (1.24)
S.
Utah Power and Light
Naughton — 3
CE
T
330
20
6
26(c)(d)
531 (0.71)
6.
Alabama Power
Barry — 4
CE
T
350
20
7
46(c)(d)
415 (0.55)
7.
Alabama Power
Barry — 3
CE
T
250
48
4
8
410 (0.55)
8.
Tampa Electric
Big Bend — 2
ES
Turbo
350
24
4
14
600 (0.80)
9.
Central Illinois Light
E. D. Edwards —
2
RS
FW
256
16
4
19
703 (0.93)
10.
Basin Electric
Leland Olds — 1
B&W
HO
218
20
3
13
569 (0.76)
11.
Pacific Power and Light
Dave Johnston —
2
B&W
FW
105
18
3
14
454 (0.60)
12.
Pacific Power and Light Dave Johnston —
Average of Coal Fired Boilers
4
RW
T
348
28
7
6
24
434 (0.53)
602 (0.80)
(c) Particulate tests performed on these boilers
(d) Corrosion probe tests performed on these units
(e) Z NO reduction at full or near full load
(f) PPM NO — 32 02, dry basis
-------
(a) B&W — Babcock and Wilcox, CE — Combustion Engineering
‘a
(b) 1W — front wall, T — tangential, Cy — cyclone
(c) PPM NO — 32 O2 dry basis
Cd) 2 NO reduction at full or near full load
TABLE 2-3
SUMMARY OF OIL—FIRED BOILERS TESTED
IN A PREVIOUS EXXON FIELD PROGRAM (2 )
Station and
Boiler No.
Boiler
M l r.(a)
B&W
Type of
Firing(b)
1W
NCR
55
No. of
Burners
6
Test
Variables
2
No. of
Test
Runs
8
NO Emissions
ppm
(C)
Baseline
(lb/106 STU)
142 (0.19)
ppm
(c)
Low NO
Clbf 106 STU)
2 NO, (d)
Reduction
1..
Atlantic
City
Electric
Deepvater — 3
118 (0.16)
17
2.
Atlantic
City
Electric
Deepvater — 5
36W
1W
55
6
2
4
221 (0.29)
175 (0.23)
21
3.
Atlantic
City
Electric
Deepwater — 8
36W
1W
83
16
3
25
246 (0.33)
124 (0.16)
50
4.
Atlantic
City
Electric
Deepwater — 9
CE
T
23
6
2
7
286 (0.38)
101 (0.13)
65
5.
Atlantic
City
Electric
B. L. England — 1
86W
Cy
136
3
2
7
441 (0.59)
313 (0.42)
29
6.
Atlantic City Electric B. L. England — 2
Average of Oil—Fired Boilers
B&W
Cy
168
4
1
2
9
361 (0.48)
283 (0.37)
303 (0.42
189 (0.25)
16
33
-------
2.3 GASEOUS SAMPLING AND ANALYSIS
The mobile sampling and analytical system used in this study to
obtain reliable gaseous emission data from field tests was described in
detail in previous reports (1,2) and, for convenience, is included in
Appendix E. A schematic diagrim of the revised sampling system used in
this program is provided in Figure 2-1. The gaseous species analyzed are
also shown on this diagram. Table 2-2 tabulates the instruments contained
in the ER&E Sampling and Analytical Van according to manufacturer, operating
technique employed, and measurement range capabilities. This system has been
used in all of our preceding field testing to obtain reliable data with a
minimum of operating difficulties.
A major consideration in obtaining reliable gaseous emission
data is to insure that the sample gas is virtually moisture free. Moisture
in the sample gas can influence the readings obtained by some of the ana-
lytical monitoring instrumentation, for example, moisture is an unwanted
interference with NO, C02 and CO nondispersive infrared analyzers. This
problem is avoided effectively in the sampling system used by passing
the sample gases through a refrigerated water trap coil. Recently,
permeation—type drying tubes* (Perinatubes) have become available for re-
moving moisture in gas sampling systems without changing the concentrations
of other gaseous species. For this reason, Permatube driers were adapted to
our sampling train as an addition to the refrigerated traps to insure a
maximum drying capability for the gaseous emission sampling system. This
system has proven to be effective in obtaining dry sample gases.
Concurrent with the present program, Exxon Research and
Engineering Company conducted an investigation under contract to EPA
(Contract No. 68—02—1722) to “Determine the Magnitude of the Stratifica-
tion of Gases in the Ducts of Fossil Fuel Fired Power Plants t ’ (3). The
results of this investigation showed that stratification of the gases
in the ducts indeed does exist and that reliable “average” flue gas con-
centrations can be obtained only by multiple probe sampling of the inner
50% of the duct. Also, average emission data obtained in this manner
are within plus or minus 10 percent of the actual concentration values
in adjacent ducts on the same boiler, with the possible exception of
oxygen, because the concentration in the ducting is influenced by air
infiltration.
The findings of our boiler duct stratification study verify
the design principles utilized by the ER&E sampling—analytical system
for obtaining representative gaseous emission data. In the ER&E system,
samples are taken from zones of “equal areas” in the flue gas ducts. At
least two probes are installed in each flue gas duct, or a minimum of
four are used when there is only one large flue gas duct on the boiler.
Each of the probes consists of three stainless steel sampling tubes
(short, medium and long) reaching to the mid—point of zones of equal
area through the depth of the duct. Thus, a minimum of six sampling
points per duct, or 12 per boiler are provided, assuring that representa-
tive gas samples are obtained.
* Manufactured by Perma Pure, Inc., Oceanport, N. J.
10
-------
BC I LER
DUCT
800°F
CO
Co 2
NO
NO 2
SO 2
02
HYDROCARBONS
SNV1PLI NG
VAN
I
VENT
5 PSI RELIEF
VALVE
PROBE (4 EAcH)
PITOT TUBE
500°F
PARTICULATE FILTERS (HEATED)
ROTAMETERS
‘.J
I- .
cO&I4TER
FLOA
PURGE
LINES
H 2 0
200
SOLENOID
VALVE
NOx & NO
Figure 2—1.
Exxon Research transportable sampling and analytical system.
-------
TABLE 2-4
CONTINUOUS ANALYTICAL
INSTRUMENTS IN ER&E VAN
Beckman Measuring
Instruments Technique Range
NO Non—dispersive Infrared 0—400 ppm
0—2000 ppm
NO 2 Non—dispersive ultraviolet 0—100 ppm
0—400 ppm
02 Polarographic 0—5%
0—25%
Co 2 Non—dispersive infrared 0—20%
CO Non—dispersive Infrared 0—200 ppm
0—1000 ppm
0—23,600 ppm
so 2 Non—dispersive Infrared 0—600 ppm
0—3000 ppm
Hydrocarbons Flame ionization detection 0—10 ppm
0—100 ppm
0—1000 ppm
Thermo Electron
NO/NO Chemiluminescence 0—2.5 ppm
x 0—10.0 ppm
0—25 ppm
0—100 ppm
0—250 ppm
0—1000 ppm
0—2500 ppm
0—10,000 ppm
12
-------
A complete range of calibration gas cylinders in appropriate
concentrations with N2 purge and zero gas for each analyzer is installed
in the system. The instruments are calibrated daily before each test,
and also in—between tests if necessary, to assure reliable and accurate
results.
Two Joy Manufacturing Company, and two Aerotherm Acurex High
Volume EPA type particulate sampling trains designed according to EPA
Method No. 5 (4) were used to obtain particulate mass loading data.
These trains were modified to incorporate Brink multi—stage cascade
impactors in the heated sampling box for the determination of particle
size distribution. This arrangement permits particle size distribution
determinations on the outside of the boiler duct under isokinetic sampling
conditions.
2.4 PARTICULATE SAMPLING
To control NOx emissions, combustion modification techniques
are used at less intense combustion conditions than those corresponding
to conventional firing methods. Combustion modifications such as lowering
and optimizing excess air levels, staging the combustion process and
re—adjustment of burner dampers to decrease the air supplied to the flame
zone may lead to carbon burnout problems reflected in changes in particulate
emissions. Lowering excess air tends to limit the amount of oxygen
available for completion of the combustion process. This factor is
expected directionally to increase the probability of burnout problems.
Staging the combustion pattern, where the majority of the burners are
operated at sub—stoichiometric air supply conditions for NO emission
control, and the remaining air is introduced either through inactive burners
or overf ire air ports to complete the combustion process, is expected to
have similar effects. Since with staged combustion the amount
of available oxygen is limited in the burner zone, the flames are lengthened,
and turbulence may be reduced, resulting in slower diffusive mixing of air
and fuel. Thus, staging the combustion pattern may also increase
unburned combustibles. Beyond the issue of combustibles (soot particulates
and unburned carbon), modifications to the combustion process for NOx
emission control may also affect the quantity, size distribution and
composition of the particulate matter emitted from the boiler.
To obtain information on the effect of combustion modifications
on particulate emissions, the field studies included the measurement of
particulate mass loading and flyash particle size distribution upstream
of the particulate collector device of pulverized coal fired utility
boilers. Measurements of total mass loading and particle size distribution
were made under baseline and optimized low NO operating conditions to
determine the extent of adverse effects of staged firing and other NO,
emission reduction techniques. Flyash was analyzed for unburned carbon
content for use in these comparisons. Such information is needed to evaluate
the impact of potential changes in particulates on electrostatic precipitator
performance and to determine the effect of combustion modifications on
trace species emissions.
13
-------
2.5 FURNACE FIRESIDE TUBE WALL
CORROSION PROBE MEASUREMENTS
Under certain conditions, pulverized coal—fired boilers are
subject to wastage of the furnace tubes. Normal corrosion occurs in
oxidizing atmospheres due to the corrosive effect of iron alkali sulfate
attack on the tube metal surface. However, under fuel—rich reducing
conditions, corrosion of the furnace tubes may be accelerated (particularly
with high sulfur, high iron content coals) due to Increased slagging on
the tube surface and penetration of iron sulfide into metal surfaces.
Under normal firing conditions this type of corrosion is most likely to
occur in areas where a localized reducing environment might exist adjacent
to the midpoint of furnace sidewalls near burner elevations where flame
Impingement might occur. To counteract such effects, normal boiler
operating practice is to increase the excess air level so that an oxidizing
atmosphere prevails at these locations, and to increase the fineness of
pulverization, so that the oxidation of the pyrites in the coal is completed
before these species can come into contact with the furnace wall tube surface.
For new boilers, an improved design feature consists of increasing the
separation between the burners and the sidewalls, thus minimizing potential
flame impingement problems.
The most effective combustion modifications to control NO
emissions from coal—fired boilers are staged combustion and low exc ss air
firing, i.e., conditions that are potentially conducive to furnace tube
wall corrosion. The need to investigate the effects of staged combustion
on furnace tube corrosion was recognized and preliminary measurements were
made in earlier studies by Exxon Research and Engineering Company under EPA
Contract No. 68—02—0227. Details and results of these studies were reported
previously (2).
In the earlier program, the approach used for measuring corrosion
rates was to expose corrosion coupons installed on the end of probes. The
probes were inserted into available openings located near “vulnerable” areas
of the furnace under both baseline and low NO firing conditions. Coupons
were fabricated of SA 192 carbon steel, the same material as that currently
used for furnace wall tubes. Exposure of the coupons for 300 hours at
elevated temperatures of 742 K (875°F) (higher than normal furnace tube
wall temperature of about 589 K (600°F) was chosen in order to deliberately
accelerate corrosion so that measurable values could be obtained, Coupons
were also mildly acid pickled to remove the existing oxide coating prior to
exposure to eliminate potential differences caused by surface conditions,
The conclusion of these earlier corrosion probing tests was that no major
differences in corrosion rates could be found between coupons exposed to
low NO firing conditions compared to coupons exposed under normal boiler
operating conditions. Coupon corrosion rates were, however, considerably
higher under both baseline and low NO conditions than those corresponding
to normal furnace tube wastage rates because of the accelerated nature of
these corrosion probing tests.
14
-------
The approach used for the measurement of furnace tube corrosion
rates in the current program was similar to that of the earlier work, but
with several significant differences. First, corrosion coupons were no
longer mildly acid pickled (which removed the protective iron oxide film)
but instead, were dipped in acetone and air dried prior to weighing to
remove any oil deposited during machining. Second, coupon temperatures
were maintained during exposure in the boiler at approximately the metal
temperatures of the furnace wall tubes in an effort to more closely
approximate actual furnace conditions. Thus, instead of being operated
at 742 K, as in the prior work, the coupons were kept at 589 K. Thirds
three coupons were installed on each probe, to increase the amount of data
obtained compared with only two coupons per probe in the prior program.
Time of exposure (300 hours) was held the same as in the previous studies
so that the results of the present corrosion probing runs could be compared
to the earlier work.
Figures 2—2 and 2—3 show details of the corrosion probes used
in the present and previous studies. The design of these corrosion probes
was based on information supplied by Combustion Engineering, with appropriate
modifications for this work. Essentially, the design consists of a “pipe
within a pipe”, where the cooling air from the plant air supply is admitted
to the ring—shaped coupons exposed to furnace atmospheres at one end of the
probe, through a 19 mm (3/4—inch) stainless steel tube roughly centered
inside of the coupons. The amount of cooling air is automatically
controlled to maintain the desired set—point temperature of 603—672 K
(625—750°F) for the coupons. The cooling air supply tube is axially
adjustable with respect to the corrosion coupons, so that temperatures of
coupons may be balanced. To simplify the pictorial presentation, the
thermocouples mounted in each coupon are not shown in Figures 2—2 and 2—3.
Normally, one thermocouple is used for controlling and the other one is
used for recording temperatures. Also, as noted above, a third coupon is
used on the probe (between the coupons with the thermocouples) which is
not shown in Figure 2—3. The cooling air travels backwards along the 63.5 mm
(2—1/2—inch) extension pipe and discharges outside of the furnace. Thus,
the cooling air and the furnace atmosphere do not mix at the coupon location.
2.6 BOILER EFFICIENCY
The application of staged combustion for N0 emission control
is based on fuel rich operation of a number of burners. This results in
less intense combustion conditions which may increase carbon burnout
problems. Thus, because this mode of boiler operation may increase the
amount of unburned combustibles, there may be an adverse impact on boiler
efficiency. However, potential efficiency losses may be compensated by
the increase in efficiency resulting from reduced stack losses under such
conditions. To determine the effect of combustion modification techniques
applied for N0 control on boiler efficiency, particulate mass loading
tests were run for each boiler under baseline and low NO operating condi-
tions and the fly ash samples were analyzed f or unburned carbon. Boiler
efficiency was then calculated, using the ASME Abbreviated Efficiency Test
based on the heat loss method, and the results were compared to evaluate
the effects of low NO firing on efficiency. The results obtained are
presented and discussed in Section 3.4.
15
-------
DRILLED AND TAPPED FOR 1/ IPT THREAD
(SWAGELOK FITTINGS - FOR THERMOCOUPLES)
HOLE FOR 1/4” SS END PLATE
GAS SAMPLING TUBE
SWAGELOK FITTING DRILLED FOR
SUPPLY TUBE (THREADS CUT OFF
WELDED OR SILVER SOLDERED TO
0 ’
END PLATE WELD
2-1/2” I.P.S. PIPE
EXTENSION
(2)
______ AIR SUPPLY
(3/4” SS TUBING)
1/2” ss Am
AND FITTING
END PLATE
1—1/4”
LING
Figure 2—2.
Corrosion probe, detail of 2½” extension pipe and end plate (outside of furnace).
-------
FACE OF FURNACE WALL TUBES
END CAP
2-1/2” PIPE EXTENSION
- .4
CORROSION
THERMOCOUP LE SOCKETS
SAMPLING TUBE L ii’. L4m_h/ 4t ’ PL
Figure 2—3. Corrosion probe, detail of corrosion coupon assembly (inside of furnace)
-------
2.7 SO Q. GASEOUS SAMPLING
Wet chemical analyses for the determination of sulfur dioxide
and sulfur trioxide content of boiler flue gases were made using an Exxon
Research and Engineering Company modification of EPA Method No. 6.
Flue gas samples were extracted from the boiler ducts through a heated
probe and passed through a series of two adsorbers, the first one
containing isopropanol to absorb the sulfur trioxide and the second one
a hydrogen peroxide solution in isopropanol to absorb the sulfur dioxide.
The amounts absorbed were determined titrimetrically using barium perchiorate
as the titrant and thorin as the indicator. Results are presented and
discussed in section 3.5.
2.8 STACK PLUME OPACITY
EPA Method 9 procedures were used to take readings of stack
plume opacity. The readings were made by two field study team members who
had been certified by a U.S. Environmental Protection Agency approved
smoke school. Stack plume observations following prescribed procedures
were made and recorded for each individual boiler test. Stack plume opacity
readings are listed in the tables in Appendix C of this report summarizing
the major operating variables and flue gas measurements for each boiler for
each test run.
18
-------
3. RESULTS AND DISCUSSION
The field test results obtained on the individual boilers investi-
gated under a variety of operating conditions in this study are presented in
four parts. These parts are, respectively, gaseous emission measurements,
flue gas particulate mass loadings and size distributions measureu upstream
of particulate collector equipment, corrosion probing data obtained in
accelerated external furnace tube corrosion tests, and boiler efficiency
calculated by the heat loss method. Gaseous emission data and most of the
particulate emission data were obtained under normal, as well as modified
firing operation. As discussed before, particulate loadings of the flue
gas were determined only under conditions corresponding to baseline and
low NO operation, for purposes of comparison of the relative effect of
modified combustion operation on flue gas particulate mass loadings and
size distribution in coal combustion. Similarly, the sustained 300—hour
corrosion probing tests were conducted to determine relative differences
in corrosion rates under baseline and modified combustion modes. It was
recognized that short term corrosion probing tests cannot substitute for
long term tests in which actual furnace tube wastage is determined. Thus,
the measurements in the current studies were made to detect major
differences with the anticipation that confirmation of these findings will
have to be based on more extensive long term corrosion tests.
The gaseous emission data obtained under baseline and modified
firing conditions at various load levels are presented first. Throughout
this report, NO concentrations are expressed as ppm, adjusted to three
percent 02 in t e flue gas, on a dry basis.
In addition to the results obtained in testing coal—fired boilers,
this section also presents the gaseous emission data on two mixed fuel (coal—
gas and coal—oil) fired units, and an oil—fired gas turbine used for power
generation. The mixed fuel tests were run to determine whether NO emissions
and their response to combustion modifications vary linearly with fuel
composition. The gas turbine test, as discussed earlier, was run to broaden
the available data base on emissions from power generation combustion equip-
ment.
3.1 GASEOUS EMISSION RESULTS
Test programs were conducted on seven coal—fired boilers consisting
of a rear—wall fired, a front—wall fired, an opposed—wall fired and four
tangentially fired boilers. Two of the tangentially fired boilers were
equipped to fire mixed fuels, one firing mixtures of coal and gas, and the
other one firing coal and oil mixtures. Typical cross—sectional diagrams
for these types of boilers are shown in Appendix A. Table 2—1 lists each
boiler by station and number, boiler manufacturer, type of firing, fuels
burned, full load MWe rating, and number of burners. In addition, the
number of operating test variables included in each test program and the
number of test runs completed are also shown in Table 2—1,
19
-------
In presenting the results and discussion on the field tests for
each of these power generation combustion equipment types, the unit will
be briefly described. First, the reasons for its selection will be dis-
cussed; second, the key operating variables tested (load, excess air level,
staged firing patterns, etc.) will be summarized and the resulting NOx and
other emission data will be given for each test run according to the experi-
mental test design used; and third, these results will be presented in
graphical form. Major conclusions and overall findings of this program will
be presented in Section 4 of this report. Summary tables containing operating
and emission data from each test run for each boiler tested are presented in
Appendix C. A separate data supplement report will be available for those
interested in detailed raw data.
3.1.1 Widows Creek, Boiler No. 5
( Tennessee Valley Authority )
Tennessee Valley Authority’s Unit No. 5 at the Widows Creek
Station was the first coal—fired unit tested in our current field
test program. The principal reason for including this boiler in the
field test program was to obtain short—term (300 hour) corrosion rate
data using our corrosion probes to be correlated with long—term (six
month) corrosion rate data based both on tube wall thickness measurements
and corrosion probe coupon weight loss measurements to be obtained by TVA
independently. TVA was provided with details of the ER&E methods used for
corrosion probing measurements in order to have a common methodology available
for both programs. In addition, this cooperative test program provided TVA
personnel with the opportunity to compare the results obtained with their new
gaseous and particulate sampling system which was designed based on the
features of the ER&E sampling—analytical van.
Widows Creek Unit No. 5 is a 125 MWe, 16—burner, rear—wall,
pulverized coal—fired Babcock and Wilcox boiler. Although the boiler
originally had a rated generating capacity of 140 MWe, full load is currently
considered to be 125 MWe. It has a single, dry—bottom furnace with a
division wall. The 16 burners are arranged with four burners in each of
four rows. Each row is fed with coal from a separate pulverizer, designated
as “A” mill for the top row, through “D” mill supplying coal to the bottom
row of burners. The burners are numbered 1 through 4 on each row from left
to right when facing the rear wall (see Figure 3—1).
Table C—l of Appendix C contains a sui miary of the key operating
and emission data from the 31 short—period test runs conducted on this
boiler. Table 3—1 presents these results according to the test program
experimental design. The four operating variables included in the test
program were gross load (125 and 100 MWe), excess air level (normal and
low), secondary air register setting (60% and 20% open), and burner firing
pattern (Sl through S9). The emission data presented in Table 3—]. are
average percent 02 and ppm N0 (3% 02, dry basis). Four probes, each
containing short, medium, and long sampling tubes were positioned at the
centers of 12 equal areas in the flue gas ducts upstream of the air preheaters
and downstream of the primary superheater. Composite gas samples from each
probe were analyzed sequentially over four complete cycles, resulting in
16 measurements of each gaseous component corresponding to each test run.
20
-------
LI
I A _____
hAM
Mil1A 0 0 0 0
MI11B 0 0100
Mi11C 0000
MilD 00100
BURNER
CONFIGURATION
rI
FULL
LOAD
(125 MWe)
REDUCED
LOAD
______ (100 MWe)
T1
STAGED FIRING PATTERNS (s 2 THROUGH s 9 )
Figure 3—i.
NO emissions vs. staged firing patterns for Widows Creek No. 5 unit.
C l)
>-
C”
0
C.,
0
600 —
500
400
300 —
200 —
100 —
(I
NOTE: Letter A*s in burner
configuration diagrams indicate
locations of burners on air
only for each staged firing
pattern.
21
-------
‘ I
S — 100 MWe
IT \
Burner Firing Gross
Symbol Pattern (Active! Load
Air Only) MWe
C) S (16/0) 125
s (16/0) 100
S 2 (14/2) 125
S3_5 (14/2) 125
c D 6—8 (12/4) 100
U S 9 (12/4) 100
I I I I I I
90 100 110 120
AVERAGE % STOICHIOMETRIC AIR TO ACTIVE
Figure 3—2. PPM NO vs. % stoichiometric air
burners for Widows Creek, boiler no.
130
BURNERS
to active
5.
140
U)
U)
c..l
0
700 -
600 —
500 -
400 —
300 —
200 —
100 —
08d
fi
S 69 —100 MWe
22
-------
TABLE 3—1
TEST PROGRAN EXPERIMENTAL DESIGN - WIDOWS CREEK NO. 5
(RUN NO., AVERAGE % 02 AND AVERAGE PPM NO EMISSIONS (3% 02, DRY))
Ai—Norma l
Au—Low
Mr
L 1 —Full Load (125 fi4)
L.a—Reduced Load (100 MW)
Ai—Nnrm 1 Air
A —Tr
Air
L 3
gPatte j 4.tL.
60%
Open
20%
Open
60% —
Open
20%
Open
-- --
60%
Open
20%
Open
L
60%
Open
20%
Open
S 1 —16 Firing Coal
6i
2.4%—409
U9 2.5%—44l
3.5%
574
3.2%
502
3.2%
548
5.5%—660
5.4%—603
5.4%
667
3.0%
462
2.1%
381
S 2 —14 Firing Coal
(A 1 A 4 Air Only)
GJ 3.0%
486
3.2%
468
S 3 —14 Firing Coal
(C 1 C 4 Air Only)
4.8%
639
4.1%
474
S 4 —14 Firing Coal
(B 1 B 4 Air Only)
4.0%
521
3.6%
517
S 5 —14 Firing Coal
(D 1 D 4 Air Only)
3.4%
598
3.6%
580
S 6 —l2 Firing Coal
(A 1 A 4 D 1 D 4 Air Only)
3.2%
365
4.3%
421
S 7 —12 Firing Coal
(A 1 A 4 B 2 B 3 Air Only)
3.0%
374
3 3.3%
302
S —12 Firing Coal
(A 1 A 4 B 1 B 4 Air Only)
4.4%
368
3.3%
316
S9_l2 Firing Coal
(A 1 A 2 A 3 A 4 Air Only)
4.5%
329
5.3%
305
-------
Figure 3—2 presents the N0 emissions measured, plotted as the
function of the average percent stoichiometric air to active burners.
Least squares lines have been fitted to the data points for normal firing
at 125 MWe, normal firing at 100 MWe and staged firing at 125 MWe. Test
run numbers are shown within the symbols which represent different operating
conditions. The test runs made with the firing patterns that produced the
lowest NO emission levels with two burners on air only, and four burners
on air only, were S 2 and S 9 , respectively. These patterns are indicated
by diamonds and squares, respectively, in Figure 3—2.
Baseline NO emissions calculated from the least squares
regression lines for 120% stoichiometric air were 332 and 296 ng/J (567
ppm and 506 ppm) at 125 and 100 MWe, respectively. These results are
about 58 ng/J (100 ppm) lower than the results obtained in our previous
field test program on Widows Creek “sister” Boiler No. 6 (2). This
difference is probably due to the differences in the coals fired and the
fact that the inside walls of No. 5 furnace were much cleaner (less ash
deposit) than the furnace walls of No. 6 boiler when it was previously
tested.
Operation with low excess air firing reduced NO emissions
as shown by the least squares line of Figure 3—2. A reduction of 10% in
the stoichiometric air supplied to the burners (i.e., 120% to 110%) reduced
NOx emissions by 21% both at full load, from 332 ng/J to 261 ng/J (from
567 to 447 ppm) and at 100 MWe load, from 288 to 227 ng/J i from 493 to
388 ppm) with normal, 16 burner firing operation.
Decreasing the gross load by 20% from 125 MWe to 100 NWe with
normal excess air, 16 burner firing would result in an estimated lowering
of NO emissions by 13%, from 332 ng/J (567 ppm) to 288 ng/J (493 ppm)
at 120% stoichiometric air. Since the actual excess air level was increased
during reduced load operation, actual NO emissions were higher during
reduced load, “normal” air operation than under normal air, full load
operation.
Staged firing test runs were conducted only at low excess air
levels. This procedure was used so that additional staged firing patterns
could be tested within the same test period, and to concentrate the effect
on measurements at low excess air operation. The minimum level of excess
air was established based on what was considered acceptable CO emissions
[ less than 117 ng/J (200 ppm)] and on normal stack plume opacity.
Four staged firing patterns were tested at 125 MWe load with
two burners on air only (14 active burners). The lowest average NO
emissions were measured with the top row outer burners on air only
(477 ppm using S 2 ). The next lowest average NOx emissions were measured
with next to top row wing burners on air only at 304 ng/J (519 ppm) using
S 4 while S 3 and S 5 staged firing patterns with the next to bottom and the
bottom row wing burners on air only produced an average of 325 and 344 ng/J
(557 and 589 ppm) NO emissions, respectively. Thus, there was a consistent
pattern of decreasing NO emissions as the “air only” burners were shifted
from “underfire” air through the bottom row to “overfire” air through the
top row of burners, as shown by the top line of Figure 3—1. Changing the
secondary air register settings from 60% open to 20% open on the active
24
-------
burners consistently resulted in lowered NO emissions for all four firing
patterns, with an average reduction in NO of 9%. The best operating
combination of burner register setting and staged firing pattern (test run
No. 12) produced an average NOx emission level of 274 ng/J (468 ppm) which
was 22% lower than the 349 ng/J (597 ppm) under actual base level operation
(test run No. lÀ).
Four staged firing patterns were also tested at 100 MWe load
with four burners on air only (12 active burners). The lowest average N0
emission level of 185 ng/J (317 ppm), was measured with the top row of
burners fed from A mill on air only (S 9 ). The next to lowest N0 level
of 198 ng/J (338 ppm) was measured when operating with A 1 , B 2 , B 3 , and A 4
burners on air only (S 7 ). Staged firing pattern S 8 (A 1 , A 4 , B 1 and B 4
burners on air only) produced N0 emissions slightly higher than firing
pattern S7, while firing pattern S 6 (A 1 , A 4 , D 1 and D 4 burners on air only)
produced the highest average NO emission level of 230 ng/J (393 ppm) among
the 4 staged burner firing patterns. Reducing secondary air register
settings from the 60% open to the 20% open position on active burners
reduced NOx emissions by an average of 6% for these four firing patterns.
The combination of staged firing pattern (Sg) and air register setting
(test run No. 20) produced the lowest NO emission level of 178 ng/J
(305 ppm), or about 49% below the level for full load, baseline operation.
The average NO emission levels measured for staged firing patterns 6 through 9
are plotted in Figure 3 —1, where the data points are connected by the lower
solid line.
Two factors appear to be responsible for lower NOx emissions
when operating with secondary air register settings of 20% open on the
active burners. The first factor is that the overall level of excess air
can be lowered (without exceeding the maximum acceptable CO emission
levels), and the second one is that operating in this manner allows lower
substoichiometric air to be supplied to the active burners.
The test results obtained from Widows Creek boiler No. 5 were
in agreement with expectations based on the results obtained on the
previously tested No. 6 boiler and other front—wall, coal—fired boilers
tested in previous studies Q ,2). Each of the 4 operating variables, load,
excess air level, secondary air register setting and firing pattern has a
significant effect on NO emission levels. From a full load, base level
NO emission level of 332 ng/J (567 ppm) at 120% stoichiometric air, low
excess air operation reduced N0 emissions by about 12% to 294 ng/J (502 ppm).
Reducing load by 20% to 100 NWe reduced NO emissions by 13%. Low excess
air, staged firing at full load reduced NO emissions to as low as 274 ng/J
(468 ppm) when secondary air registers were closed down to “20% open” setting.
Low excess air, staged firing at reduced load (100 MWe) reduced NO emissions
to as low as 178 ng/J (305 ppm) with the burners supplied by A mill on
air only and with closed down secondary air registers. The most effective
staged firing patterns are those that maximize the amount of “overfire”
air across the burner area.
25
-------
3.1.2 Ernest C. Gaston, Boiler No. 1
( Southern Electric Generating Company )
Southern Electric Generating Company’s Boiler No. 1 at the
Ernest C. Gaston Station, was the second boiler tested in our current
field test program. The major reason for selecting this boiler was
that it had been retro—fitted with newly designed, low NOx burners by
its manufacturer, Babcock and Wilcox. These dual register compartmen-
talized wind box burners produce a limited turbulence, controlled diffu-
sion flame. They are designed to minimize the amount of fuel and air
mixed at the burner to that required to obtain ignition and to sustain
combustion of the fuel. Figure 3—3 presents a line diagram of the dual
register pulverized coal burners installed in Boiler No. 1.
Gaston Units No. 1 and 2 are 270 MWe, 18—burner, horizontally
opposed, pulverized coal fired Babcock and Wilcox boilers. Two division
walls divide each furnace into 3 equal compartments, each having six
burners arranged 3 high in both the front and rear walls of the furnace.
The furnace is 18.3 m (60 feet) wide, with volume of 3950 in 3 (139,500
cubic feet) and a total wall area of 2391 m (25,732 square feet). Six
pulverizers each with 16,300 kg (36,000 pounds) capacity feed coal to 3
burners each as shown in Figure 3—4. The maximum continuous rating of each
boiler is 771,000 kg (1,700,000 pounds) of steam per hour. Steam design
pressure at superheat outlet is 303 kPa (2,075 PSIG) at 811 K (1000°F). Steam
design pressure at reheater outlet is 72 kPa (495 PSIG) at 811 K (1000°F).
Steam temperature controls include flue gas recirculation into the hopper
area of the furnace and spray attemperators for primary superheat and reheat.
Table 2 of Appendix C contains a summary of the key operating and
emissions data collected on this boiler. Table 3—2 presents these results
according to the test program experimental design. The five operating vari-
ables (and experimental levels tested) included in the test program were:
1. Gross load ful1 load (270 MWe), 250 MWe, and 190 MWe to 205 MWeI.
2. Excess air level (high, normal and low).
3. Secondary air register setting (30, 50, 70,and 100% open).
4. Tertiary air register setting (50 and 100% open).
5. Firing pattern (all burners firing, 1 top mill off, 2 top
mills off).
The emission data shown in Table 3—2 and Table 2 of Appendix C are average
% oxygen and ppm N0 (3% 02, dry basis). Four probes, each containing
short, medium and long sampling tubes were positioned near the centers of
12 equal areas in the flue gas ducts between the economizer and the air
preheaters.
26
-------
Adjustableair Watercooled
Retractable lighter outer
vanes and registers
and auxiliary burner assy
I _ U oat
ii:
i plu , ______________________________________
Primary
P
bservat ion door
air/coal venturi .and burner
flame detector Air cooled /
inner throat
Winclbox
Figure 3—3. Babcock and Wilcox dual register pulverized coal burner, (Courtesy, Babcock and Wilcox Company)
-------
Division Walls
B
Figure 3—4, Mill—burner configuration of E. C, Gaston boilers No, 1 and 2
Mill
Mill
E
F
C
A
D
Front—Wall Burners
28
-------
TABLE 3-2
TEST PROGRAM EXPERIMENTAL DESIGN - ERNEST C. GASTON BOILER NO. 1
(RUN NO., AVERAGE Z 02 AND AVERAGE ppm NO — 3% 02, DRY BASIS)
Li (268 — 273 MWe)
L2 (236— 250 MWe)
L3 (187 — 208 MWe) 6 (148 MWe)
A 2 —Normai
Air
A 3 -LOw
Air
A2-Nor.
Air
A 3 —Low
Air
A 1 -Maximum
Air
A 2 —Normal
Air
A 3 -Low
Air
A 2 -Nor.
Air
0
Firing
a tern
Secon.
Tert>-..4ir
Air
9A 1 —
30—50%
SA 2 —
70—100%
SA 2 —
30—50%
SA 2 —
70—100%
SA 2 —
70%
2 —
70%
SA 1 —
30%
SA 2 —
70%
SA 1 —
30%
SA 2 —
70%
SA 2 —
70%
SA 2 —
70%
S 1 — All
Burners
Firing
Coal
Ti — L00%
Open
3.7% ( ) 4.3%—389
377 3.9 —356
t ) 3.6% 4) 3.7 —349
409 3.8 —390
3.9 —349
3.9 —316
4.4%—372
2.4%
289
(7 ) 2.6%
329
2.3%
318
2.4%
278
)2.4%
336
3.9%
322
i 6.8%
451
7.0%
454
3.9%
297
4.9%
325
T 3 — 50%
Open
(7) 2.5%
362
2.2%
312
9 2%
296
— B
or E
Miii.
On Air
Only
T 1 — 100%
Open
I 3.6%
289
2.6%
257
1.9%
240
3.8%
301
S 3 — B
and E
fills
Air
Only
T 1 — 100%
Open
3.6%
240
1.6%
182
6.9%
148
T 2 — 70%
Open
1.8%
200
-------
Composite gas samples from each probe were analyzed and recorded in turn
over four complete cycles, resulting in 16 measurements of each gaseous
component in each test run. The level of excess air on “low” excess air
test runs was established as the minimum excess air level operation that
would not produce over 115 ng/J (200 ppm) CO emissions.
Figure 3—5 is a plot of ppm NOx (3% 02, dry basis) vs. average
% stoichiometric air to active burners. The numbers within the plotted
symbols identify the run number from which the data were obtained. Least
squares, linear regression lines were calculated from the data points re-
presenting normal firing (Si) at 270 MWe, normal firing at 205 MWe, staged
firing (S 2 — top mill burners on air only on front or rear wall) at 250
MWe and staged firing operation (S 3 — top mill burners of front and rear
wall on air only) at 190 MWe.
Baseline NO emissions calculated from the least squares regres-
sion lines shown in F gure 3—5 for 120% stoichiometric air were 204 and 160
ng/J (356 ppm and 279 ppm) at 270 and 205 MWe, respectively on Boiler No. 1
using the new low NO burners. Thus, a load reduction of 24% resulted in a
22% reduction in NOx emissions. From a full load, base level NO emission
level of 220 ng/J (377 ppm) [ 124% stoichiometric air], low excess air opera-
ting reduced NOx emissions by about 20% to 178 ngfJ (305 ppm). Two staged
firing patterns were tested: B or E mill burners on air only (S2) and E
and B mill burners on air only (S3). Low excess air, staged firing at 250
MWe reduced NOx emissions to as low as 140 ng/J (240 ppm) with B mill burners
on air only (S2). Low excess air, staged firing at a lower loads of 190 and
148 MWe reduced NOx emissions to as low as 106 and 87 ng/J (182 and 148 ppm),
respectively, with both E and B mill burners on air only (S 3 ). Analysis of
secondary air register setting vs. N0 shows that the lowest NOx level was
generally reached when the register setting was approximately 70% open with
lower and higher settings producing higher NOx levels.
NOx measurements on Boiler No. 2 when operated at full load of
270 MWe, averaged 346 ng/J (591 ppm) at 24% excess air. Boiler No. 1 when
operated at full load of 270 MWe, averaged 220 ng/J (377 ppm) at 24% excess
air. Therefore, the new dual register burners used in Boiler No. 1 reduced
NOx emissions significantly, i.e., by 36% compared to conventional Babcock
and Wilcox burners.
A short—term program of experimental runs was conducted to deter-
mine the effect of reduced pulverizer air temperature on NO emissions. At
the normal operating temperature of 350 K (l70”F), average NO emission
measured 211 ng/J (361 ppm), while at a reduced temperature of 333 K (140°F),
average NO emissions measured 183 ng/J (313 ppm). Thus, a 12% reduction
in NOx emissions resulted from the use of lowered pulverizer air temperature.
Further experimentation would be necessary to determine if the reduced
temperature operation would result in mill operating problems and to
validate these results over a longer time period with different coals and
pulverizer settings.
30
-------
p S 1 —270 MWe
(Boiler No. 2
equipped with
conventional burners)
500
(I )
400
300
z
200
100
70
S 2 —250
I I I
S 1 —205 N
600 -
NWe
80
S 3 —148 MWe
90 100 110 120 130 140
Average % Stoichiometric Air to Active Burners
Figure 3—5. NO emissions vs. % stoichiometric air for Gaston boiler No. 1.
I I
150
160
-------
3.1.3 Navajo Station, Boiler No. 2
Salt River Project
Navajo Unit No. 2 is a twin furnace, 800 MWe, Combustion 3
Engineering Boiler. The two—cell furnace has a volume of 13,182 m
(465,500 Cu. ft.), width of 25.5 in (83 ft. 6—3/4 in.) and front to rear
length of 12.45 in (40 ft. 10—1/4 in.). Maximum continuous rated primary
steam flow is 2,450,000 kg/hr. (5,400,000 lb./hr.), at 814 K (1005°F),
with a reheat steam flow of 2,200,000 kg/hr. (‘4,850,000 lb./hr.) at 813 K
(1003°F). The normal fuel fired is Black Mesa sub—bituminous coal with a
higher heating value of 24950 kJfkg (10,725 BTU/lb.), and containing 10.4%
ash, 10.3% moisture, 38% volatile material and 41.4% carbon. At maximum
continuous rating, 296,000 kg/hr. (652,000 lb./hr.) of coal are fired and
the design furnace efficiency is 88.77%. Seven pulverizers feed 56 burners
arranged to fire at seven different levels. Overf ire tilting air ports
are located above the top row of burners. Main steam pressure is 523
kPa (3590 lbs./sq. in.) at the superheater outlet.
Table 3, Appendix C, contains a summary of the major operating
variables and flue gas measurements for each of the 36 test runs com-
pleted on Navajo No. 2 Unit. Operating variables included in the experi-
mental program were gross load, excess air level, burner tilt, and firing
pattern. Table 3-3 presents a summary of emission data (run no., % 02,
ppin NOx) arranged according to the planned experimental design. Figures
3—6 through 3—9 display the test data in graphical form.
Figure 3—6 is a plot of ppm NO (3% 02, dry basis) vs. average
% oxygen measured in the flue gas for all test runs conducted at full
load with normal firing CS 1 ). Lines have been drawn to show th& effect
of excess air level on NO emission levels for +250, +10 to +15 , 00 and
—10 to —15° burner tilts. The effect of overfire cooling air (about 10%
open) has no beneficial NO reduction effect when the burners are tilted
up 250. However, as the burners are tilted towards the horizontal posi-
tion, the distance of separation between the overf ire cooling air and
the bulk flame zone increases and increased NO reductions are apparent.
Operating with burner tilted at —10 to —15° produced higher NOx emission
levels than 00 tilt operation due to the concentration of heat release
in the bottom of the furnace. The least squares regression line calcu-
lated from the seven test runs made at +10 to +150 burner tilt (ppm NO =
118 + 51% 02), showed an average reduction of 51 ppm NO for each per-
cent reduction in flue gas oxygen content.
Figure 3—7 is a plot of ppm NOx (3% 02, dry basis) vs. % oxygen
for the test runs made at full load with overf ire air dampers 100% open
CS 7 ). The least squares regression line calculated from the eight test
runs conducted with +10 to +15° burner tilt (ppm NO = 90 + 54% 02) showed
an average reduction of 32 ng/J (54 ppm) NOx for each per cent reduction
in flue gas oxygen content. The average ppm NO emission level for the
five S 7 test runs conducted at the lowest excess air level (3.5 to 3.8%
02) was 165 ng/J (282 ppm) or 17% below the 199 ng/J (341 ppm) NO level
for comparable excess air S 1 test runs.
32
-------
TABLE 3—3
TEST PROGRAM EXPERIMENTAL DESIGN — RUN NO. • % 0 2,, ‘ - 1 - x
(NAVAJO NO. 2 UNIT — FULL LOAD (795—808 (NWe)
Tilt Pattern
S 1 —Normal
Firing
Row Air
S 2 —Top
25% OFA
S 4 -
50% OFA
S 5 -
75% OFA
S 6 *
100% OFA
S 7 *
—10° to —15°
3,9% 332
5.2% 346
Horia.
4.9% 343
5) 5.4% 329
+100 to 150
5.4%
5 • 4%
4 • 9%
4 • 8%
5 • 2%
5.2%
416
385
366
346
378
383
5.0% 318 J 4.0% 332
3.9% 310
4.8% 492
;3: 3.8% 292
5.2%
4.4%
4.6%
4.6%
354
349
346
349
3.8% 314
4 • 2%
309
335
298
4.0%
3.6%
3.8% 411
3.6% 259 d, J 3.6% 328 3.6% 326 j 4.0% 289
set at 25% open for S 4 , 50% open for S 5 75% open for S 6 and 100% open for S 7 .
3.5% 271
3.8% 289
3.6%
4.2%
3.6%
3.7%
282
291
288
280
A 1 -
Normal
Excess
Air
+100 to +15°
A 2 -
Low
Excess
Air
T 3
Ta
+250
* Overfire air registers were
-------
Figure 3—6.
PPM NO emissions vs. % oxygen for Navajo No, 2 unit
under full load, normal firing operation.
•1•1
c.’1
0
C- ,
0
z
x
3.0 3.5 4.0 4.5 5.0 5.5 6.0
Average % Oxygen Measured in Flue Gas
34
-------
40C —
+10 to +15° Burner Tilt
(—io to —15°
Burner Tiltj
j I I I
3.0 3.5 4.0 4.5 5.0
Average % Oxygen Measured in Flue Gas
Figure 3-’7 ,
PPM N0 emissions vs. % oxygen for Navajo No, 2 unit
operating with overf Ire air dampers 100% open.
r1
a
c’J
0
0
z
35( —
300 —
250
00 Burner Tilt
5.5
6.0
35
-------
Figure 3—8 is a plot of ppm NO vs. % oxygen for the S 2 , S 4 ,
S 5 and S 6 staged firing test runs. The least squares line labeled S 7
from Figure 3—7 was drawn on Figure 3—8 for purposes of comparison.
Short lines parallel to the S 7 line were drawn throi’gh the averaged
data for S 4 , S 5 and S 6 test runs. The lowest NO emissions resulted
from test runs No. 7 and No. 8 conducted with the top tier of burners
on air only (S 2 ) and cooling air through the overf ire air registers.
The beneficial effect of Increasing overfire air register openings from
25% CS 4 ) to 50% (S 5 ) and then to 75% (S 6 ) is apparent from examination
of Figure 3—8. Furthermore, Figure 3—9 was constructed to present the
effect of changing overf ire air register settings more directly. Only
test runs conducted with approximately equal excess air levels (3.6 to
4.0% 02) and burner tilts (+100 to +15°) are shown on Figure 3—9 so
that the effect of overfire air damper settings on NO emission levels
can be seen directly. Average NO levels decreased rom 193 ngfJ (330 ppm)
to 186 ngfJ (318 ppm), 170 ng/J ( 9O ppm) and 165 ng/J (283 ppm) as the
overf ire air damper openings were increased from 25% to 50%, 75% and 100%,
respectively.
The effect of boiler load on NO emission levels can be
estimated from the data shown in Table 3—4. The first six columns pro-
vide data on the three test runs conducted at reduced loads of 565 to
305 MWe. The seventh and eighth columns list NOx emission levels meas-
ured for full load operation at comparable excess air levels and burner
tilts based on the data of Figure 3—6. The last column indicates the
% reduction in NO levels at the reduced loads.
Thus, a 29% reduction in load resulted in a 10 to 20% reduction
in N0 emission levels and a 52% reduction in load resulted in a 26% N0
emission reduction.
36
-------
400
350
° 300
0
z
250
3.0
Average % Oxygen Measured in Flue Gas
Figure 3—8. PPM N0 emissions vs. % oxygen for Navajo No. 2 unit
under staged firing operation.
3.5 4.0 4.5 5.0 5.5
37
-------
350
300
250 —
. 1 -I
a
0
m
0
z
E
0 .
0.
0
Figure 3—9, Effect of opening overf ire air dampers on N0 emissions from
Navajo No, 2 unit at full 1oad
40
60
.6% 0xyg
.6% 0xyg
3.7% Ox
en
20
I I
Overfire Air Dampers — % Open
80
100
38
-------
TABLE 3—4
REDUCED LOAD TEST DATA
(Navajo No. 2 Unit)
NO EIfliSSIOflS
Run Gross Burner % Reduced Load Full Load % NOx
No. Load (MWe) Tilt 02 ng/J ng/J Reduction
23 565 +100 5.3% 205 350 226 388 10
24 565 +300 5.0% 236 404 296 506 20
18D 305 +25° 4.6% 204 349 277 474 26
39
-------
3.1.4 Comanche Station, Boiler No. 1
Public Service Co. of Colorado, Pueblo, Colorado
Comanche Unit No. 1 is a single furnace, tangentially fired,
Combustion Engineering boiler. It was selected for this program because
it is a new unit equipped with overf ire air ports (OFA) over the top
level of burners for NO emissions control. The furnace has a volume
of 6739 m 3 (238,000 cu. ft.), width of 13.8 m (45 ft. 4 in.) and front to
rear length of 12.2 m (40 ft.). Maximum continuous rated primary steam flow
at 365 kPa (2500 lb./in 2 ) is 1,149,000 kg/rir. (2,534,000 lb./hr.) at 814 K
(1005°F) with a reheat steam flow of 997,500 kg/hr. (2,155,000 lb./hr.) at
814 K (1005°F). A Wyoming sub—bituminous coal is fired with nominal values
of 19,190 kJ/kg (8,250 BTU/lb.), 5.2% ash, 0.57% sulfur, 29% moisture, i2 ,4%
fixed carbon and 33.4% volatile matter, and an ash fusion temperature of
1450 K (2150°F). At maximum continuous rating (350 MWe), 194,000 kg/hr.
(428,000 lb./hr.) coal is fired and a furnace efficiency of 84.b5% is predicted,
Five pulverizers feed 20 burners firing at 5 levels (designated “A” for the top
and “E” for the bottom levels).
This unit was the first pulverized coal fired boiler designed by
Combustion Engineering to have overf ire air ports (through extended wind—
boxes) for reducing nitrogen oxides emissions. The boiler is equipped with
a hot electrostatic precipitator for particulate collection.
Table 4 of Appendix C contains a summary of the major operating
variables and flue gas measurements for each of the 30 test runs completed on
Comanche Unit No. 1. Operating variables included In the experimental pro-
gram were overf ire air damper settings (closed to 100% open and horizontal
to —15° tilt), burner nozzle tilt (horizontal to —26°), secondary air regis-
ter settings (22—40% open auxiliary and 100% open coal registers) and boiler
load (316 to 340 MWe).
Several operating limitations were experienced due to load, weather
and boiler conditions. July and August are peak load demand months and thus
only normal load variations were treated. Only normal excess air operation
was used due to maximum ID fan capacity (economizer section blockage) and plant
operating management’s desire to avoid possible slagging problems (previously
encountered) that might be caused by low excess air operation. Burners are
normally operated tilted at —20° from the horizontal to avoid high steam out-
let temperature and to reduce excess air level variations to a minimum between
flue gas economizer outlet ducts. Thus, the burners were not tilted above
horizontal. Secondary air register settings at full load are varied auto-
matically to maintain 18 to 20 inPa (4.5 to 5.0 In. H 7 0) furnace to wind—
box differential by varying auxiliary register settings, while the coal air
registers are maintained 100% open. Thus, if the overf ire air registers are
opened from 5% (closed) to 50% open, the auxiliary air registers automatically
close to 22 to 26% open from about 50% open. Maximum gross load during our
test period was 320 to 340 MWe due to limited ID fan capacity mentioned above
and poorer than normal coal (up to 7% ash instead of 5% ash content). How-
ever, in spite of these limitations most of the objectives of the test program
could be accomplished satisfactorily due to the excellent cooperation of the
plant personnel.
40
-------
Table 4, Appendix A, presents a summary of the operating and
emission data recorded for each test run. Flue gas samples were taken
from the centers of six equal areas from each of the two ducts (“A” and
“B”) between the economizer and the hot—side precipitator. Sixteen
measurements were made of each gaseous component during each 32 minute
“steady state” test run.
Figure 3—10 is a plot of ppm NOx (3% 02, dry basis) vs burner
nozzle tilt (degrees from horizontal). The number within the symbols
indicate the test run. Straight lines were drawn through the data points
obtained from similar firing patterns to indicate the relationship
between N0 emission and degree of burner tilt. Baseline operations
(normal firing pattern with OFA ports closed) resulted in 235 ng/J (410
ppm) N0 when operating normally at _150 to _200 burner tilts. Hori-
zontal burner operation reduced NO emissions by 18% to 196 ng/J (336 ppm),
(Run 5) while lowering burner nozzle tilts to _260 increased NO emissions
to 262 ng/J (448 ppm) (Run 3), or 9% above baseline operation.
As expected, operation of the overfire air registers had a large
effect on N0 emission levels. The three lines on Figure 3—10 drawn
through data from S 1 operation (closed OFA ports), S 3 (25% open OFA ports)
and S 4 (50% open OFA ports), respectively, indicate the reduction of NOx
emission levels as OFA port registers are opened. Figure 3—11 is a plot
of ppm NO emissions vs the percent open OFA registers for the data from
test runs conducted at the normal operating range of —15 to _200 burner
nozzle tilts. The solid line drawn through these data points indicates
that NO emissions are reduced sharply from 238 ng/J (407 ppm) with over—
f ire air ports closed (actually, 5% open for cooling purposes), to 209
ng/J (358 ppm) with OFA ports 25% open to 169 ng/J (289 ppm) with 50% open
OFA ports to a level of 154 ng/J (264 ppm) NO emissions when operating
with 75 or 100% open OFA ports. Thus, a 36% reduction in N0 emissions
levels from baseline operations was obtained through the use of overf ire
air ports. Additional improvements would be expected with low excess air
operation and raising the burner nozzle tilt angles closer to a horizontal
position.
Excess air levels were maintained within “normal” levels during
the entire test period at Colorado Public Service Company’s request to
avoid possible slagging problems. Consequently, test run average flue
gas 2 measurements varied within the narrow range of 3.5% to 4.4% making
meaningful correlations of NO vs 02 difficult. However, as commonly found
in tangentially fired boilers, average 02 and NO measurements from one
duct (A) were consistently lower than those for the other duct (B) on each
test run. This situation provided a means for correlating NO levels with
02 levels within the test runs. Figure 3—12 is a plot of ppm NO vs 02
data by flue gas duct for test runs 22 through 31. 02 measurements in
Duct B averaged 2.2% higher than in Duct A, while NOx measurements in Duct
B were 15.8 ng/J (27 ppm) higher than in Duct A. Although these data do
not represent the full N0 reduction potential of low excess air operation,
they indicate directionally the improvements that might be obtained.
41
-------
450 —
400 —
350 —
300 -
c y 15 ç y
c?< 250 c E .— 5 — OFA 75% OPEN
S 6 — OFA ].00%OPEN
(NLrVBERS IN SYIVBOLS INDICATE RW NO.)
200 l C) CLOSED (5% OPEN)
S 2 10% OPEN
- S 3 25% OPEN
150_
S 5 75% OPEN
- S 6 l00% OPEN
10( I I I I
—30° —25° —20° —15° —10° 5° 0° +5°
BURNER NOZZLE TILTS — DEGREES FROM HORIZONTAL
Figure 3—10. Effect of burner nozzle tilt on NO emissions from Comanche
x
No. 1 unit.
S 1 - NORMAL FIRING — OFA PORTS CLOSE]
42
-------
450
400
350
300
250
200
150
OVERFIRE AIR REGISTERS — % OPEN
Figure 3—11. Effect of overf ire air damper setting on Comanche No. 1
unit NO emissions.
x
I - ’
U)
I -
U)
C 1
0
ox
0
(NUMBERS WIThIN SYMBOLS INDICATE RUN NO.)
0 20 40 60 80 100
43
-------
I I
(N1J 1BERS IN CIRCLES INDICATE RW NO.)
I I
I I
I I
1 2 3 4 5 6 7
AVERAGE % OXYGEN IN FLUE GAS
Figure 3—12. NO emissions vs. flue gas oxygen measurements on Comanche
No. 1 unit.
400
350
300
U,
C%1
0
cn
ox
200
0
150
1 AA
250
DUCT “A”
DUCT “B”
8
44
-------
3.1.5 Barry Station, Boiler No. 2
( Alabama Power Company )
Alabama Powar Company’s Boiler No. 2 at the Barry Station was
the third boiler to be tested in our current field test program. There
were two major reasons for selecting this boiler. First, it is one of
very few boilers that has been retrofitted with overf ire air ports for
NOx emission control. The second reason for selecting this boiler was
that it is capable of firing gas (up to 40% of full load), coal and
mixed fuel firing. No other boiler in the United States is known to have
these combined capabilities. Figure 3—13 is a side elevation view of
this boiler.
Barry Station, Unit No. 2 is a natural circulation, balanced
draft boiler which fires coal through four elevations of tilting tangential
fuel nozzles. Each elevation of burners is fired by a pulverizer. The
steam capacity at maximum continuous rating is 408,000 kg/hr (900,000 lbs/hour)
main stream flow with a superheat outlet temperature and pressure of 811 K
(l 0 00°F) and 272 kPa (1875 PSIG), respectively. Superheat and reheat tempera-
tures are controlled by burner tilt and water spray desuperheating. The fur-
nace is 11.63 m (38 feet 2 inches) wide and 8.57 m (28 feet 1—1/4 inches) in
depth. Two wind boxes (upper and lower) feed secondary air through 10 com-
partments (4 coal and 6 auxiliary) located at the four corners of the furnace.
Vertical burner spacing is 1.5 m (4’ 11”) from center line to center line be-
tween elevations 1 to 2 and 3 to 4 with 2.29 m (7’ 6”) spacing between eleva-
tions 2 and 3. Overf ire air ports were designed to supply about 20% of total
air through new ducts located at each corner at 2.44 m (8 feet) above the upper
fuel zone as well as through the top two compartments of the upper windbox.
The experimental program was planned to produce required
emission and operating information on this unusually flexible boiler
from a minimum number of test runs. The program was divided logically
into three experimental blocks, one at full load and two at reduced loads.
The prime objectives of the full load (130 MWe) block were to measure N0
emissions under baseline and modified combustion operation while firing 100%
coal (Alabama and Illinois) as well as mixtures of gas and coal up to
the maximum gas firing capability of the boiler under both normal and
overf ire air operation. Test runs to evaluate the effect of excess air
level, burner nozzle tilt and secondary air register settings were also
included in the full load experimental block. The major objective of the
reduced load block at 95 MWe was to compare the NO reduction effectiveness
of overfire air port operation with modified staged firing (top row of
burners on air only). Ninety—five (95) MWe is the maximum load capability
of the boiler with the top mill off. To measure N0 emissions while firing
100% gas, the test runs of block 3 were conducted at the maximum load possible
(55 MWe) while firing gas only.
45
-------
Figure 3—13. Side elevation drawing of Barry No. 2 unit.
46
-------
Table 5, Appendix C contains a summary of the key operating and
emission data for the 37 short—period test runs completed on this unit.
Table 3—5 presents the average ppm N0 emissions (3% 02, dry basis) and
percent oxygen measurements for each test run according to the test
program experimental design. The six operating variables (and experimental
levels tested) included in the test program were:
1. Gross load [ 130 MWe (full load), 95 MWe, and 55 MWe]
2. Fuel fired (Alabama coal, Illinois coal, mixed fuel and
natural gas)
3. Excess air level (normal and low)
4. Firing patterns (all coal levels firing with NO ports
closed, 50% open and 100% open; top row coal off with
NO ports closed and 50% open)
5. Burner tilt (horizontal, 200 up and 200 down)
6. Secondary air register setting (coal/auxiliary air at
30/100 and 100/50 % opening)
Full Load Test Data Results — Barry No. 2 Unit
Analysis of the NO emission data from the full load test block
reveals several significant hndings:
1. Full load, baseline operation (normal firing of all 16 burners,
burner nozzles tilted down or horizontal, and normal secondary air
register settings) produced an average NO, emission level of 199
ng/J (341 ppm) [ 3% 02, dry basis] while firing 100% coal.
2. Staged firing using the special overfire air ports reduced N0
emissions by 10 to 20% from baseline operation at the same excess
air levels. 100% open overfire air ports produced slightly lower
NO emissions than 50% open operation (227 ppm vs 238 ppm).
3. Excess air level was the most significant single operating variable
affecting NO emission levels. N0 emissions were reduced through
low excess ah operation by 43% to 115 ng/J (196 ppm) under normal
firing operation, by 39% under staged firing operation, S , (50%
open overfire air ports) and by 29% under staged firing operation
S (100% open overfire air ports). The solid regression lines on
Figure 3—14 indicate the relationship between ppm NO and the average
oxygen content in the flue gas for these three firing patterns while
firing 100% coal and operating with horizontal or down burner tilts
and normal secondary air register settings.
4. Normal secondary air operation (30% open coal air and 100% open
auxiliary air registers) produced an average of about 10% lower
NO emissions than reversed secondary air operation (100% coal air
an 50% open auxiliary air registers) when firing 100% coal. -
47
-------
TABLE 3-5
ThST PROGRAM EXPERIMENTAL DESIGN —
BARRY NO. 2 UNIT
(RUN NO., AVERAGE % 02 AND AVERAGE PPM NO _3% O2 DRY BASIS)
* Test runs 2 and 5: burner tilt —20°
** Test run 37: D 2 100 coal and 50 auxiliary air damper settings
% Open
A 1 -
Nor.
A 2 —
Low
A 1 —
Nor.
A 2 -
Low
L 1 —Full Load—13O MWe L 2 —95 MWe
F 1 —Coal— F —Coal— F —.8 Coal ‘ F —.6 Coal FtCoal—
Alabama I 1inois d.2 Gas 1 .4 Gas Ill i nois
L 3 —55 MWe
F 2 —Coal— F 5 —Nat’ l.
Kentucky
A 1
Nor.
A 2 -
Low
Al_IA 2 -
Nor. I
Tow
A 1 —
Nor.
At
Low
A 1 —
Nor.
A 2 -
Low
A-
Nèr.
A 2 -
Low
—Normal
Ports
Burners
T 1 —Hor.
Burner
Tilt
D 1 —3OCoal
100% Aux
.4
345
314
206
168
i
452
284 254
639.2
122
J7.3%
97
D 2 —lOOCoal
50 Aux
299
I
T 2 -+20°
D 1 —3OCoal
100 Aux
6 z.o
320
I
I
T 3 —20°
D 1 —3OCoal
100 Aux
365
68.1%
196
273
P.
Open
T 1 —Hor.
D 1 —3OCoal
lOOAux
‘ 6.8%
294
*
53.5%
182
*
@5.1% I
186
I
5.O%
184 182
6%
184
(.1%
411
.8%
253
Ports
Open
Burners
T 1 —Hor.
D,—3OCoal
ibO Aux
243
.9% 4.9%
182
143
I
D —lOOCoal
O Aux
6j .ox
233
I
I
T 2 +20°
D 2 —100 Coal
50 Aux
5j4.l%
274
I
i
T 3 ——20°
D 1 —3OCoal
100 Aux
.7%
265
4).8%
189
c 7 4.O%
218
**
i
I
Row Air
Burn.—Coa
1/2 Op
Row Air
Burn.-Coa
T 1 —Hor.
n
T 1 —Hor.
D 1 —3OCoal
100 Aux
D —3OCoal
l 0 Aux
-
,_ _ ._
.._
_
P.l%
330
1J 6.8%
272
fj 4.6%
212
1)4.3%
192
-------
5. Burner nozzle tilt over the range of _200 to horizontal produced
no consistent change in NO, emission levels. However, raising
burner nozzle tilts to +20v resulted in significantly higher N0
emission levels. This effect can be explained by the fact that
the overfire air directional dampers were fired at —11° and thus
the effectiveness of staged combustion was reduced with raised
burner tilts due to reduced separation of the overfire air from
the main combustion zone.
6. Mixed fuel firing at full load reduced NO emissions significantly
but not linearly. Figure 3—15 is a plot of ppm NO (3% 07, dry) vs
percent coal in the fuel fired (on a heat release asis) For comparable
full load test runs under normal firing, S 1 , and staged firing (S
and S 3 ) operations. Firing with 80% of the heat release from coa
reduced NO emissions by an average of 30%, while firing with 60% coal
resulted in an average of 32% NO reduction from 100% coal firing over
the three firing patterns. Thus, within the range of full load,
mixed fuel firing capability of this boiler, the data indicates that
replacing coal with gas fuel lowers NOx in the direction of full gas
firing but the relationship is not linear. As discussed below, at
reduced load (55 M ie), firing with 100% gas reduced NO emissions
from 157 to 64 mg/J (269 to 110 ppm) (60%) compared to 100% coal
firing. A comparable linear relationship at full load would thus be
a 12% reduction in N0 emissions when firing 80% coal—20% gas mixture
and a 24% reduction when firing a mixture of 60% coal—40% gas.
Reduced Load Tests
Earlier, we discussed the necessity for conducting two test
program blocks at reduced load. The first block (comprising seven test
runs under 100% coal firing) was conducted in order to obtain information
on the NOx emission reduction effectiveness of staged combustion using
overfire air ducts compared to modified staged combustion operation where
the top tier of burners are operated on air only. Since the maximum capacity
of the boiler with the top mill inactive is 95 MWe, all of these runs were
made at this load for purposes of comparison. The second block (four test
runs) was conducted at the maximum load capacity of this boiler (55 MWe)
when firing 100% gas.
Reduced load test runs at 95 MWe were conducted under normal,
all burners firing operation and at three different staged firing patterns:
S 2 , overfire air port dampers set at 50% open and all four levels of burners
firing; S 4 , overfire air port dampers closed (cooling air only) with top
tier of burners on air only; and S 5 . overf ire air dampers 50% open and top
tier of burners on air only.
The dashed lines on Figure 3—14, labeled S 2 , S 4 and S 5 connect the
normal excess air and low excess air N0 emission data points for each
of these staged firing patterns. These results can be summarized as follows:
49
-------
®sl
100% OPEN
AIR ONLY
AIR ()JLY
50% OPEN
1 I I I I I
3 4 5 6 7 8
A\’ERA( % OXYGEN IN FLUE GAS
Figure 3—14. NO emissions i/s % 02 in flue gas for
100% coal fired test runs on Barry No. 2 unit.
9
‘S 2
0
C l )
a
c’.1
0
z
a-
a-
REDUCED
‘LOAD
TEST RLt 1S
AT 95 4qe
400 —
300 —
200 —
100 —
2
S 1 -55 F4 te RUNS
i7
“V
0 s 1 OFA CLOSED
V
S 2 OFA 50% OPEN
S 3 OFA
II II
A
I, IS
OFA
50
-------
400
C ’,
‘1
200 —
‘cn
ox
z
z
0
a-
() N0RM( L FIRING — S 1
50% OPEN OFA PORTS — S 2
0 100% OPEN OFA PORTS — S 3
I I I I I
20 40 60 80 100
% COAL IN COAL—GAS MIXED FUEL FIRING
Figure 3—15. NO emissions vs % coal in coal—gas
mixed—fuel firing operation on Barry No. 2 unit.
301 —
—4.5%
/
—5.1%
100 —
51
-------
1. Baseline NO emissions were 264 ng/J (452 ppm) at the higher excess
level required under reduced load operation.
2. Low excess air operation reduced NO emissions sharply in all three
staged firing configurations compared to normal excess air levels as
shown by the dashed lines of FIgure 3—14.
3. As expected, all three staged firing operations resulted in significantly
lower NO emission levels, with the combined use of overfire air and
top row of burners on air only producing the best results. Compared
to the 95 MWe, baseline N0 emission level of 264 ng/J (452 ppm), low
excess air, S 2 operation (overf ire air ports ½ open) reduced NOx emis-
sions by 44% to 148 ng/J (253 ppm); low excess air, S 4 (top burners on
air only) operation reduced N0 emissions by 53% 124 ng/J (212 ppm) and
low excess air, S 5 operation (50% open overf ire air ports plus top row
of burners on air only) reduced NO emissions by 58% to 112 ng/J (192 ppm).
The second set of reduced load test iuns (numbered 13 through 16)
were designed to supply information on 100% gaseous fuel firing compared to
100% coal firing under both normal and low excess air operation. Under nor-
mal operation at 55 MWe load, coal was fired through the bottom two rows of
burners with relatively high excess air levels. Baseline NOx emissions of
166 ng/J (284 ppm) were reduced to 148 ng/J (254 ppm) with low excess air
operation. 100% gas firing operation produced 71 ng/J (122 ppm) NOx under
normal excess air levels and 57 ng/J (97 ppm) when using low excess air opera-
tion. Thus, at this low load, gas firing resulted in a 57% NO emission
reduction under normal air operation and a 62% NO emission reduction under
low excess air operation, compared to 100% coal firing.
52
-------
3.1.6 Morgantown, Boiler No. 1
( Potomac Electric Power Company )
Potomac Electric Power Company’s, Boiler No. 1 at Morgantown,
Md. was the fourth boiler tested in the current field test program.
This boiler was selected, with the assistance of Combustion Engineering
Company, because it represents modern design practices and has the
capability of firing oil, pulverized coal or mixtures of these two fuels.
Morgantown Boiler No. 1 is a 575 MWe, Combustion Engineering tangen-
tially fired, twin furnace boiler. Five pulverizers feed 40 burners arranged
at five levels (from A pulverizer feeding the top 8 burners to E pulverizer
feeding the bottom 8 burners). Thirty—two oil burners are positioned at the
four levels (8 per level) between the five pulverized coal burner levels, I.e.,
between A and B, B and C, C and D and D and E. The furnace has a volume of 8269
m 3 (292,600 cubic feet), a width of 19.25 in (63.15 feet) and a length of 10.8 m
(34.45 feet). Full load steam rate is 1,928,000 kg/hr (4,250,000 pounds/hour)
at 814 K (1000°F) superheat temperature and 555 kPa (3810 PSIC) pressure.
Table 6 of Appendix C contains a suimnary of the operating and emis-
sion data from the 27 test runs completed on this boiler. Table 3—6 presents
the results arranged according to the test program experimental design. The
five operating variables (and experiment levels) included in the test
program were:
1. Gross load (full load — 575 MWe, and reduced load — 300 MWe)
2. Excess air level (normal, low, and high)
3. Fuel mixture fired (100% coal to 100% oil)
4. Burner tilt (horizontal, +15°, —15°)
5. Firing pattern (normal firing — S 1 , and staged firing — S 2 )
Flue gas measurements of average 02 and N0 concentrations are Included in
Table 6 of Appendix C, as well as the calculated percent stoichiometric air to
the active burners. The flue gas sampling system consisted of four probes,
each containing short, medium and long sampling tubes positioned at the
centers of 12 equal areas in the ducts between the economizer and the air
heaters. Composite gas samples from each set of probes were analyzed and
recorder in turn over four complete cycles, resulting in 16 measurements
of each gaseous component in each test run.
The N0 measurements made are plotted in three separate figures
using identical coordinate scales to show the results obtained under dif-
ferent local conditions and firing patterns. Figure 3—16 is a plot of ppm
NO (3% 02, dry basis) vs. % coal (on a heat release basis) for coal—oil
conditions (no overf ire air). Figure 3—17 Is a plot of test runs conducted
under full load, staged firing (air only at top pulverized coal burner
level) conditions. Figure 3—18 is a plot of test runs conducted at reduced
load (about 300 MWe) under both normal and staged firing conditions. The
53
-------
4 — Full Load (575 MWe)
L 2 — Reduced Load (300 MWe)
F — 93% C
*1 7%
F — 73% C
2 27% 0
F — 45% C
55% 0
F — 22% C
78% 0
F 1 — 100% C
0% 0
F 2 — 63% C
37% 0
F — 50% C
50% 0
F 4 — 202 C
80% 0
F - 0% C
100% 0
S 1 —
Normal
Firing
Pattern
A 1 —
Normal
Excess
Air
T 1 —
Nor. Tilt
4.9%
552
4.4%
593
525
Ci) 4.4z
458
6.5%
544
6.8%
460
6.9%
408
5.9%
247
T 2 —
+15 Tilt
4.7%
553
—
—:15° Tilt
4.7%
568
—Low
Air
—
Hor. Tilt
3.8%
549
3.4%
511
2.8%
502
2.1%
386
6.3%
515
—
Staged
Firing
A 1 — Nor.
Air
T 1 —
Hot. Tilt
4.7%
403
4.5%
465
4.6%
375
5.4%
300
5.1%
165
A 2 — Low
Excess
Air
T 1 —
Nor. Tilt
3.1%
423
3.3%
310
4.2%
149
T 2 —
+15° Tilt
GJ
334
—
—15° Tilt
3.1%
314
coal and % oil on heat release basis.
TABLE 3-6
TEST PROGRAM EXPERIMENTAL DESIGN - MORGANTOWN BOILER NO. 1
(Run No., Average 2 02 and Average PPM NO — 32 0 , Dry Basis)
* Fuel fired —
-------
% Coal in Oil—Coal Mixed Fuel Firing
Figure 3—16. NO emissions vs % coal in oil—coal
mixed—fuel, normal firing operation on Morgantown No. 1 unit.
Low Excess
Air
C l )
•rl
Cl )
Cu
) .1
a
C .’1
0
0
z
300
200
100
0 20 40 60 80 100
55
-------
600
500 —
Normal Excess Air
. 400
(n
Low Excess Air
a
c 1
o 300—
0
z
z
200 —
100 —
0 I I I I
0 20 40 60 80 100
% Coal in Coal/Oil Mixed Fuel Firing
Figure 3—17. NO emissions vs % coal in coal/oil mixed fuel,
staged fifing operation on Morgantown No. 1 unit.
56
-------
% Coal in Coal/Oil Mixed Fuel Firing
Figure 3—18. PPM NOx emissions vs Z coal in coal/oil
mixed—fuel firing operation at reduced load on Morgantown No. 1 unit.
C a
-ri
U )
a ’
11
C J
0
en
Normal Firing Pattern
0
z
x
p4
p ..
Normal Excess Air
200
Staged Firing Pattern
Excess Air
100
20
40
60
80
100
57
-------
numbers shown within the circles (normal excess air operation) or squares
(low excess air operation) indicate test run indentification numbers.
The conclusions reached from analysis of the data are as follows.
Low excess air operation (established by the maximum permissible
level of 200 ppm CO emission for any single probe) consistently reduced NOx
emissions compared to normal excess air operation. Under full load (575 MWe)
operation, the average reduction in NO emissions was 10% under normal firing
operation and 18% under staged firing operation. At reduced load (300 MWe),
reductions of 5% and 10% were experienced under normal firing and staged
firing conditions respectively. Lines connect the normal excess air and low
excess air test run data points in Figures 3—16 and 3—17 to show this com-
parison.
Staged firing operation (top mill burners on air only) produced
substantially lower NO emission levels than normal firing operation (top
mill burners firing coal) under both full load and reduced load operation
over a wide range of fuel mixture compositions. Reductions of 18%, 19%
and 32% were experienced at full load when firing with mixed fuels containing
22% coal, 49% coal and 73% coal, respectively, under normal excess air, staged
firing operation compared to normal excess air, normal firing operation. The
combination of low excess air and staged firing resulted in further improve-
ments. At reduced load (300 NWe), NO emission reductions of 26 to 40% were
obtained through staged firing operation compared to normal firing operation.
Changing the tilt angle of the coal injected nozzles had a minor
effect on NO emissions when firing mixed fuel, presumably because the oil
burners were fired in a horizontal position. Horizontal burner tilts resulted
in NO emissions that were about 6% lower than with 15° up or down tilting
burners.
Reduced load operation (300 NWe vs. 575 MWe) resulted in lowered
NOx emissions. However, the magnitude of the reduction was highly dependent
upon the amount of oil in the coal—oil mixture fired. While firing mostly
coal (93% at full load and 100% at reduced laod), NO emissions were reduced
on the average by only four percent. However, reducing load while firing 22
to 55% oil in oil/coal mixtures, reduced N0 emissions by 26% under normal
firing and 35% with staged firing.
N0 emission levels generally increased with increasing amounts
of coal in mixed fuel firing. Figure 3—18 shows this trend over almost the
full range of fuel mixtures under normal firing operation at reduced load.
A straight line through the N0 data point from test run 25 (100% coal firing)
and test run 2lA (5% coal firing) does not pass through the NOx data points for
intermediate mixtures of coal—oil fuel. Thus, test run 22A (43% coal — 57%
oil fired) measured 13% higher NO emissions and test run 24 (62% coal —
38% oil fired) produced 9% higher NO emissions than would be expected from
a linear relationship calculated from 100% coal and 100% oil firing. Figure
3—16 shows a pronounced departure from a linear relationship for mixed fuel
firing over the range from 22% coal to 93% coal. Increasing the percentage of
58
-------
coal in the coal—oil mixture from 22% to 45% or 49% resulted in significantly
increased NO emissions (by 58 ng/J, or 93.5 ppm) but little change in N0
emissions resulted from further increases in the percentage of coal fired.
During the test period this boiler had limited fuel burning capability, and
therefore, full load test runs on 100% coal or 100% oil could not be run.
Only 21 of the 32 oil burners were operational due to mechanical problems,
and therefore, 22% of the heat release had to be supplied from coal. At the
same time, coal supplies were short and some coal handling equipment could
not be operated. Consequently, plant management had to limit the length of
time that high coal mixtures could be burned as well as the maximum coal con-
tent of the mixed fuel.
In sunmiary, it was found that most of the operation variables
included in the experimental program had significant effects on NO emis-
sion levels. Low excess air operation reduced NO emissions by about 10%
for a wide range of coal/oil mixtures at full and reduced load. Staged
firing operation resulted in 20 to 30% NO emission reductions over a wide
range of coal/oil fuel mixtures. Combined low excess air and staged firing
operation reduced NO by 30 to 40%. Tilting burners over the range of
+15° to —15° produced minor changes in NO levels with horizontal levels
producing the lowest NO emission levels. Increasing the proportion of
coal In simultaneous oil/coal firing increased NO emission levels
nonlinearly under both normal and staged firing operation. The N0
emission level increase was higher in changing from 0% to 50% coal, than
the NO emission increase when changing from 50% to 100% coal fuel. It
is likely that such behavior is exhibited because of the changing rela-
tive contributions of thermal NO and fuel NOx to the total emissions.
59
-------
3.1.7 Mercer Station, Boiler No. 1
Public Service Electric and Gas Company (New Jersey )
Mercer Unit No. 1 is a twin furnace, front—wall fired, wet—
bottom Foster Wheeler boiler. This unit was selected for testing because
of its flexibility for combustion modification, and to determine the
side—effects of such modifications on a wet—bottom unit. The boiler has
3579 m 2 (38,526 ft 2 ) of furnace heating surface, a furnace volume of
5333 m 3 (188,332 cubic feet), with each furnace measuring 11.95 m (39 feet
2 1/2 inches) in width and 7.99 m (26 feet 2 1/2 inches) in depth. Maximum
continuous rated steam flow is 934,000 kg/hr (2,060,000 lb/hr) at 348 kPa
(2400 PSIG) pressure and 867 K (1100°F) superheat steam temperature and
839 K (1050°F) reheat steam temperature. Three ball—type pulverizers feed
the 24 burners arranged In three rows of four burners in each of the two
twin furnace front walls. The pressurized furnaces are equipped with flat
floors and slag—taps.
Table 3—7 contains a summary of the operating and emission data for
each of the 12 test runs conducted at full load during the Phase I test program.
Gperating variables listed are gross load (MWe), excess air level (normal and
low), firing pattern (Sl—norma]. and S 2 —biased firing, I.e. top row of burners
fuel—lean with middle and bottom burner rows fuel—rich), and secondary air
register settings (80% open and 50% open). Gaseous measurements of % 0
and ppm NO (3% 02, dry basis) are listed for the reheat furnace out1et
the superheater furnace outlet, and for boiler average. Table 7, Appendix C,
contains a summary of the operating and emission data for all 36 test runs
completed on this unit.
Table 3—8 contains a summary of the operating and emissions data
for each of the 21 reduced load test runs completed in the Phase I test
program. Operating variables were gross load (MWe), excess air level
(normal and low), firing pattern (S 1 —normal, S 3 —staged, and S 4 —top mill
air and coal off), and secondary air registers (off and opened to: approxi-
mately 15%, 50% and 80% open). Normal secondary air register setting is
about 80% open while 15% open is used at initial firing and closing down
to about 40 to 50% open is an intermediate setting.
Table 3—9 presents the test data (run number, % oxygen, and ppm
NO ) according to the partially replicated, factorial experimental design
conducted for the Phase I test program. Repeat test runs were conducted
corresponding to the conditions of runs No. 10, 23, 25, and 26, to provide
the data needed for the determination of experimental error under these
key test conditions.
Figure 3—19 is a plot of ppm NO (3% 02, dry basis) vs. average
% 02 measured in the flue gas for each test run condition. Lines have been
drawn through the data for the normal and low excess air firing for each
operating condition (boiler load, firing pattern, and secondary air register
setting).
Analysis of these test results indicate that all of the operating
variables included in the experimental program had a significant effect on
NOx emission levels. At full load (290 MWe) baseline operation resulted in
809 ng/J (1383 ppm) NO emissions. This high NO level is caused by the
unusual furnace design of this boiler in which pulverized coal can be burned
60
-------
TABLE 3—7
SUMMARY OF OPERATING AND EMISSION DATA
(MERCER NO. 1 UNIT)
Date—
Run
No.
Gross
Load
(MWe)
Excess
Air
Level
Firing
Pattern
(1)
Secondary
Air egisters (2)
Gaseous Measurements
Reheat_________
Super1 ’ ’at
Boiler
Top
Mid
Bot
% O _
ng/J
PPM NO _
% O
ng/J
%02
ng/J
PPM NO,
12/3—1
2
3
4
12/4—5
6
7
8
1A
12/5—lB
1C
1D
296
290
290
290
288
285
283
284
290
280
291
292
Normal
Low
Normal
Low
Normal
Low
Normal
Low
Normal
Normal
Normal
Normal
S 1 —Nornial
S 1 —Normal
S 2 —Biased
S.,—Biased
S —Normal
S 1 —Normal
S 1 —Biased
S 2 —Biased
S —Normal
S —Norma1
S —Norma1
S —Norma1
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
5O%*
5O%*
50%*
50%*
80%
80%
80%
80%
80%
80%
80%
80%
50%*
5O%*
50%*
5O%*
80%
80%
80%
80%
4.08
2.05
3.21
1.45
4.06
2.04
4.10
1.89
3.60
5.02
3.86
3.81
751
637
688
540
839
605
608
495
725
677
729
715
1284
1098
1176
924
1434
1034
1040
846
1240
1157
1246
1223
3.70
1.66
3.25
1.25
3.78
1.85
3.68
1.62
4.60
4.78
3.95
4.15
867
699
754
538
844
609
753
530
818
792
816
807
1482
1196
1289
920
1444
1042
1288
906
1399
1355
1396
1380
3.89
1.86
3.24
1.35
3.92
1.95
3.89
1.76
4.10
4.90
3.91
3.98
809
670
723
539
842
607
681
512
771
734
772
761
1383
1147
1237
922
1439
1038
1164
876
1319
1256
1321
1302
(1) S 2 —Biased Firing: top row of burners fuel—lean with middle and bottom burners fuel—rich.
(2) Secondary air register settings are 80% open at full load and closed down to 40 to 50% open for low load firing.
I - ’
* Reheat furnace 50% open and superheat furnace 80% open.
-------
TABLE 3—8
SUMMARY OF OPERATING AND EMISSION DATA
PSE&G CO. (N.J.), MERCER NO. 1 UNIT
(Reduced Load Test Runs)
Date—
Run
No.
Gross
Load
(MWe)
Excess
Air
Level
Firing
Pattern
(1)
Secondary
Air egiscers (2)
Gaseous Measurements
Reheat
Superheat
Boiler
Top
Mid
Bot
0
ng/J PPM NO _
% O
ngTJ
PPM NO
°2 ng
f3’ PPM NO
1218— 23 150 Low S 4 —Top Off 0% 80% 80% 4.91 439 750 5.05 471 805 4.98 455 778
24 151 Normal S 3 —Top Off 0% 80% 80% 6.42 524 897 6.78 535 915 6.60 530 906
26 152 Low S 3 —Staged 80% 80% 80% 4.00 193 330 4.56 206 352 4.28 199 341
25 152 Normal S 3 —Staged 80% 80% 80% 7.95 409 700 8.46 492 841 8.21 451 771
29 155 Normal S 3 —Staged 50% 80% 80% 7.92 480 820 8.08 504 861 8.00 491 840
30 150 Low S 3 —Staged 50% 80% 80% 4.16 288 492 4.82 275 471 4.49 282 482
12/9— 10 219 Low S 3 —Staged 80% 80% 80% 3.51 216 370 3.70 236 403 3.60 226 386
9 218 Normal S 3 —Staged 80% 80% 80% 7.15 473 808 6.41 427 730 6.78 449 769
27 216 Normal S 3 —Staged 50% 80% 80% 6.91 524 896 6.34 477 816 6.62 501 856
28 212 Low S 3 —Staged 50% 80% 80% 3.20 230 394 2.99 195 334 3.10 213 364
20 210 Low S 3 -Staged 15% 80% 80% 2.60 405 692 2.02 311 532 2.31 358 612
19 211 Normal S 3 —Staged 15% 80% 80% 6.32 610 1043 6.25 578 989 6.28 594 1016
12/10—11 238 Normal S 1 —Normal 80% 80% 80% 5.15 641 1096 6.56 688 1176 5.86 664 1136
12 232 Low S 1 —Normal 80% 80% 80% 3.34 592 1012 2.97 600 1026 3.15 596 1019
22 226 Low S 4 —Top Off 0% 80% 80% 3.60 552 944 2.83 542 927 3.22 547 936
l2/ll—23A 165 Low S 4 —Top Off 0% 80% 80% 4.30 413 707 3.70 422 722 4.00 417 714
26A 161 Low S 3 —Staged 80% 80% 80% 3.53 205 350 4.78 219 374 4.16 212 362
25A 162 Normal S 3 —Staged 80% 80% 80% 6.75 392 670 7.30 414 708 7.02 403 689
12/12—17 236 Normal S 1 —Normal 80% 50% 50% 5.47 643 1100 6.00 687 1174 5.73 665 1137
18 232 Low S 1 —Normal 80% 50% 50% 3.24 602 1029 4.24 720 1.232 3.47 661 1130
1OA 211 Low S 3 —Staged 80% 80% 80% 2.80 160 275 3.82 222 380 3.31 191 327
-------
TABLE 3-9
TEST PROGRAM DESIGN — MERCER NO. 1 UNIT
(TEST RUN NO., % OXYGEN AND PPM NO (3% 0 , DRY BASIS)
*SecoMary Air Registers: Top row of burners set at 50% open, bottom and middle registers set at 80% open.
**Top row of burners set at 15% open, bottom and middle row burners set at 80% open.
L 1 —(290 MWe)
L 2 —(220 MWe)
(Max. Load—i Miii Of f)
L 3 —(155 MWe)
A 1 -Normal
Excess Air
A 2 —Low
Excess Air
A 1 —Norma l
Excess Air
A 2 -Low
Excess Air
A 1 -Normal
Excess Air
A 2 -Low
Excess Air
a.,
S 1 —Base
(Normal
Firing)
R 1 —Normal
2nd Air Rag.
R 2 ’-Closed
Down 2nd
® 3.9%
1383
® 3,9%
1439
® 1.9%
1147
® 1.95%
1038
© 5.9%
1136
5.7%
1137
3.15%
1019
35%
1130
S 2 —Biased
Firing
(Top Burners
Lean Fuel)
R 1 —Normai
2nd Air Reg
RrC losed
Down 2nd
(33 3.2%
1237
3.9%
1164
1.35%
922
® 1.8%
876
© X
3 )
8 27 771
70% RQ
) 8.0%
840
Firing
(Top Burners
S 3 —Staged
Air Only)
S 4 —Base
(Top Burners
Air and
Coal Of f)
2nd Air Reg.
R 1 —Normal
RrC losed*
Down to 50% opei
,,,, ,,,,//
769
(3. 6.8%
6.6%
856
3.6% 386
3 % 327
3.1%
364
4
L9!
341
6 4.5%
482
R 2 _C losed**
R 1 —Normai
Down to 0% open
2nd Air
R —C1osed
Down 2nd Air
6.3%
( )
1016
2.3%
J) 3.2%
612
936
I11’m
6 % 906
u5
778
4.0% 71L
-
-------
l40
Full Load (283—296 MWe)
(232—237 MWe)
(210 MWe)
Oi 2
3
4
5 6 7 8
Average % Oxygen in Flue Gas
Figure 3—19. NO emissions vs % 02
in flue gas on ercer No. 1 unit.
Co
Co
Ii
C..’
o 800
0
z
600
400
64
-------
at low loads with a wet bottom furnace. The flat furnace floor is
relatively close to the bottom row of burners so that unusually high gas
temperatures are maintained in the bottom of the furnace in order to maintain
the slag in a molten state. Within the limited operating flexibility
under full load operation, low excess air was the most important variable,
reducing NOx emissions by an average of 24%. Biased fired (top row burners
fuel—lean with bottom and middle row burners fuel—rich) reduced NOx emissions
by an average of 16%. Reducing the secondary air register setting from
about 80% open to about 40% to 50% open on the reheat furnace increased NO
emission levels by about 4% under normal firing operation (S 1 ) and reduced C
NOx emissions by an average of 8% under biased firing operation (S 2 ). At
full load, the lowest NOx emission levels were obtained under test run No. 8
operating conditions of biased firing, low excess air, and closed down
secondary air registers. This test condition produced 876 ppm NOR, a
reduction of 36% from 809 ng/J (1383 ppm) produced under baseline operation.
Table 3—9 presents the emission results from the 12 test runs
conducted at approximately 220 MWe, the maximum load achievable under staged
firing (1 mill on air only) operation. Under normal firing operation (Si),
NO emissions were reduced by an average of 12% due to the load reduction
from 290 to 220 MWe (24% reduction). Low excess air operation reduced NOx
emissions by an average of 5% under normal firing (S 1 ) and by 50% under
staged firing (S 3 ). Staged firing (top row of burners on air only) was
carried out with the secondary air registers of the top row of burners set
at 80% open, partially closed down to 40 to 50% open, or almost completely
closed down. As expected, the greatest reduction in NOx emissions occurred
when the top row secondary air registers were set at the maximum opening
of about 80% open. Thus, the average reduction from the NO emissions for
normal firing of 630 ng/J (1078 ppm NOR) at 200 MWe were 24 , 43%, and 48%
under staged firing Cs 3 ) conditions when the secondary registers were set
at 0%, 50% and 80% open, respectively. Test run No. 10 operating conditions
of staged firing, low excess air and normal secondary air register settings
(80% open) produced an average level of 208 ng/J (356 ppm) NO, or 69% below
the 664 ng/J (1136 ppm) level experienced under baseline operation at about
the same load.
Nine test runs were conducted at approximately 155 MWe which is
the normal night—time low load conditions for this boiler. Under baseline
operations at this load (top burners with air and coal off), NO emissions
were 34% less than under full load operation. Low excess air and staged
firing operation at this load again resulted in large NO emission reduc-
tions. Thus, low excess air combined with staged firing (test run 26
conditions) lowered NO emissions by 61% from baseline operation at low load.
In summary, the wet bottom furnace boiler demonstrated a high degree
of NOx emission reduction capabilities through combustion modification from
the high baseline operating level of 809 ng/J (1383 ppm) at full load.
Excess air level, firing pattern, and secondary air register settings were
all important combustion control variables. The optimum operations at
full load (290 MWe), intermediate load (220 MWe), and low load (155 MWe)
produced NO emission levels and percent reductions of 512 ng/J (876 ppm)
or 27%, 2O8Xng/J (356 ppm) or 74%, and 205 ng/J (351 ppm) or 75%, respectively.
65
-------
3.1.8 Morgantown Gas Turbine No. 3
( Potomac Electric Company )
Potomac Electric Company’s Gas Turbine No. 3 was the first
stationary power generating gas turbine tested in the current field test
program. This General Electric model MS 7001 B gas turbine has a maximum
continuous rating of 50 MWe and a peak output of 54 MWe. It fires No. 2
distillate oil. This unit is not equipped for water injection, so that
‘ 1 wet” control of gas turbine NOx emissions could not be explored.
Flue gas samples were taken from the centers of four equal
rectangular areas from each of two stacks and from the centers of each
of the stacks in order that possible stratification could be measured and
to obtain representative flue gas samples. Twelve to 18 measurements were
recorded during the 15 to 20 minute duration of each test run.
TABLE 3—10
SUMMARY OF OPERATING AND EMISSION DATA -
MORGANTOWN STATION, GAS TURBINE NO. 3
Run No.
Operating_Conditions
- Flue Gas Measurements
Gross Load
(MWe)
Fuel Flow
1/tn (GPMj
Exhaust
Temp.
K °F
% O _
7 Enii sions (Dry Basis)
15% 07
ng/J PPM
3% 02
ng/J PPM
1
2
3
4
10
25
48
54
125 33
178 47
280 74
307 81
601 621
611 640
758 905
800 980
17.9
17.3
15.2
14.7
48 82
63 108
73 125
78 133
144 247
188 322
220 376
233 398
Table 3—10 contains a summary of the major operating and emission
data obtained from this test program. Figure 3—20 presents the ppm NO
measured vs gross load (MWe), with the NO values adjusted to both 3% 02
and 15% 02 bases. As expected, the NO emissions increase with load, but
not very sharply up to the peak load of 54 MWe.
Measured gaseous concentration stratification was very low in
the stacks of this gas turbine. The reasons for this lack of stratification
were the high velocity of the flue gas through the long path of ductwork
(about 130 feet) in the switchback silencer and the location of the sample
tubes at four diameters from the nearest upstream disturbance.
66
-------
Figure 3—20. NO emissions vs gross load —
Morgantown gas turbine No. 3.
Co
r 1
Co
Li
0
z
0 10 20 30 40 50
Gross Load (NWe)
60
67
-------
3.2 PARTICULATE EMISSION RESULTS
As discussed in section 2.4 of this report, the use of combustion
modifications to limit “NOX” emissions may result in a tendency to increase
burnout problems. Potentially, carbon burnout problems might increase
particulate emissions, and the increase in unburned carbon in the fly ash
could decrease precipitator collection efficiency. A further question
to answer Is whether particulate size distribution may also be affected
by low NO operation of coal fired boilers, in addition to potential changes
In particulate mass loading. Therefore, part of the overall program plan
was to obtain sufficient data to afford a comparison of the effects of
“low NOR” firing techniques on particulate emissions by comparing measure-
ments of particulate mass loading and size distribution, and In the fly ash
with similar data obtained under baseline operation.
The particulate emission tests were conducted using sampling
trains complying with EPA Method 5 requirements. The Method 5 type trains
were modified to incorporate Brink, multi—stage cascade impactors for
particle size determination in the heated oven between the cyclone and
filter of the normal EPA train. This arrangement permitted obtaining mass
loading and particle size distribution data simultaneously under isokinetlc
conditions with the Brink impactor located outside of the boiler. All tests
were made in the boiler flue gas ducts downstream of the airheaters but
upstream of any collecting devices, such as mechanical dust collectors or
electrostatic precipitators. Consequently, the dust loadings are higher
than at the outlet of the precipitator, but the results are not affected by
the existing collection devices. Comparison of the test results taken
under baseline operating conditions with data obtained under “low NO “
operation, therefore, provides a direct determination of potential s de
effects of staged firing on particulates.
Particulate emissions in pulverized coal fired utility boilers
are affected by the ash content of the coal fired, coal fineness, burner
design, type of firing and by burnout problems (increased carbon on
particulates) that may result from staged combustion and low excess
firing implemented to lower NO emissions. Since particulate tests were
conducted on each boiler with no change in pulverizer mill fineness
adjustment, coal fineness does not appear to be a valid contributing
factor. Carbon burnout problems, however, as evidenced by increased
particulate carbon content, could be a contributor to increased particulate
emissions.
En earlier ER&E field studies (2), some “side effects” attributed
to “low NOR” firing were observed based on the fact that mass loadings
in some boilers tended to Increase but these effects, if any, appeared to
be minor. The flue gas particulate mass loading data obtained in this
program are summarized in Table 3—li. Referring to Table 3—li, it appears
that the emission data appear to be reliable and consistent within
the accuracy of this type of testing. Comparison of emission results in
the table and data on the required collection efficiency of electrostatic
precipitators or other collection devices to meet the NSPS of 0.1 lb/b 6
BTU, for baseline vs. “low NOR” operation in some cases shows a tendency
toward increased particulate emissions with “low firing. Data on
the Salt River Project, Navajo, Boiler No. 2 (tangentially fired) and the
Public Service Electric and Gas Company, Mercer, Boiler No. 1 (front wall
fired) are examples. However, emissions from the boilers at the
Tennessee Valley Authority, Widows Creek, Boiler No. 5 (rear wall fired),
68
-------
Southern Electric Generating Company, E. C. Gaston, Boiler Nos. 1 and 2
(horizontally opposed fired) and Public Service Electric Company of
Colorado, Comanche, Boiler No. 1 (tangentially fired) are either the same
or lower for “low NOx” operation. It is concluded from the particulate
emission data obtained in these tests, similar to the conclusion of our
earlier field study (2), that any increases or decreases in dust loadings
caused by “low NOx” firing conditions do not appear to be significant.
The particulate mass loading data of Table 3—11, are plotted in
Figure 3—21 as a function of unit load. It is seen that particulate
emissions are not correlated with boiler size or load. The highest particulate
mass loadings measured in this program were 3178 ng/J (7.39 lbs/10 6 BTU)
on a 270 MWe boiler operating at 268 MWe. In contrast, an 800 MWe boiler
produced particulate loadings of 2946 n /J (6.85 lbs/l0 6 BTEJ) at 808 MWe,
and a loading of 2292 ng/J (5.33 ibs/lO BTIJ) was measured on a 125 MWe
rated boiler operating at near full load. In the latter test (on the
125 MWe boiler), however, coal ash content was 18.27 percent compared to
9.84 and 6.62 percent, respectively, on the 270 and 800 MWe boiler tests.
Mass loading emissions might also be influenced slightly by the small
differences in the heating value of the fuels burned in these tests.
Another potential side effect of low NO operation may be an
increase in carbon on fly ash or other unburned combustibles. As discussed
above and in section 2.4, staging the combustion process and decreasing
excess air levels produces longer, “lazier” flames with increased tendencies
toward incojnplete combustion. Any increase in unburned carbon as a
consequence of low NO combustion modifications, would have a corresponding
adverse effect on boiler efficiency and might tend to increase particulate
loading emitted from the boiler. An assessment of the magnitude of this
potential side effect may be made based on the data of Table 3—11 by
comparing the percent carbon on the particulate measured under baseline
and low NO operations, respectively. In comparing these data it may be
noted that only in two cases, i.e., the Southern Electric Generating Company,
E. C. Gaston, Boiler No. 1 (horizontally opposed firing) and the
Public Service Electric and Gas Company, Mercer, Boiler No. 1 (front wall
firing) the carbon losses are higher for low N0 operation. In all other
cases the low N0 carbon loss data are the same or lower than for baseline
operation. There is also an indication that particulate carbon content may
decrease slightly under “low N0 ” firing conditions in tangentially fired
boilers. This may be seen for the data for the Public Service Company of
Colorado, Comanche, No. 1 and Salt River Project, Navajo. No. 2 units. These
differences, however, are not statistically significant. Again, as in
earlier studies (2), carbon losses in boilers fired with Western coals,
i.e., Salt River Project, Navajo No. 2 and Public Service of Colorado,
Camanche, No. 2 have been noted to be very low indicating the easy burnability
of these coals. The effects of changes in unburned carbon resulting from
low N0 firing on boiler efficiency are discussed in section 3.4 and are
shown to be minor.
Potential changes in particle size distribution that could result
from low N0 combustion modifications were also assessed in this study.
Changes in size distribution of flue particulates could affect electro-
static precipitator collection efficiency. These data were obtained upstream
of the electrostatic precipitators (or mechanical dust collectors) using a
69
-------
TABLE 3—11
PARTICULATE EMISSION TEST RESULTS (a)
PULVERIZED COAL FIRING
(a) Sampling at air heater
(b) Boiler No. 2.
(c) Boiler No. I..
outlet upstreanrof collection devices.
-J
0
Utility Date
Test No.
Firing
Condition
Load,
MWe
Emissions
Reqd.
Efficiency
to Meet
0.1 lb.!
106 BTU
F
Carbon on
Particulate
Coal
Ash
Wet, F
11EV,
cal/g
Wet
BTIJ/lb.
mg/rn 3
Gr/SCF
lb./lOb
BTU
10/7/74
TVA, Widows 10/8/74
Creek, Boiler 10/15/74
No. 5 10/15/74
( lB
1 1C
‘\ 12C
L12D
Base
Base
Low NO
Low NO
128
124
125
125
1.02
1.02
.87
.81
4.46
4.45
3.80
3.55
2288
2292
1961
1832
5.32
5.33
4.56
4.26
98.12
98.12
97.81
97.65
13.04
9.11
6.21
8.08
14.64
18.27
12.30
12.30
6267
5699
5901
5901
11,281
10,258
10,622
10,622
1/22/75
1/22/75
So. Elec. 1/23/75
Ceo. Co., 1/29/75
E. C. Gaston 1/17/75
Boilers No. 1 & 1/20/75
No. 2 1/20/75
28A (b)
28B (b)
29 (b)
30 (b)
26 (c)
27A (c)
27B (c)
Base
Base
Base
Base
Low NO
Low NO
Low NO
272
272
268
270
270
268
268
1.13
1.04
.87
.87
1.02
1.07
1.38
4.93
4.55
3.80
3.78
4.46
4.68
6.02
2657
2451
2030
2043
2356
2473
3178
6.18
5.70
4.72
4.75
5.48
5.75
7.39
98.36
98.25
97.88
97.89
98.18
98.26
98.65
2.69
1.09
1.69
2.01
5.65
3.37
4.08
15.00
15.00
16.06
13.28
12.12
9.84
9.84
6494
6494
6367
6332
6529
6462
6462
11,690
11,690
11,461
11,398
11,752
11,632
11,632
6/18/75
6/24/75
Salt River 6/26/75
Project, Navajo 6/19/75
Boiler No. 2 6/20/75
6/26/75
( 3C
I 2A
/ 3D
18B
18C
t 18E
Base
Base
Base
Low NO
Low NOx
Low NOx
799
792
785
808
808
798
.73
.80
.44
1.13
1.22
.54
3.17
3.48
1.93
4.95
5.35
2.36
1836
1836
1075
2692
2946
1234
4.27
4.27
2.50
6.26
6.85
2.87
97.66
97.66
96.00
98.40
98.52
96.52
0.92
1.79
1.98
0.80
1.16
1.53
11.49
5.94
6.04
11.49
6.62
6.04
5672
6149
6214
5672
5989
6214
10,210
11,068
11,186
10,210
10,780
11,186
7/29/75
7/30/75
Pub. Ser. Co. 7/23/75
of Colorado, 7/23/75
Comanche, 8/1/75
Boiler No. 1 8/1/75
1/20/76
1/21/76
Pub. Ser. Elec. 1/22/76
& Gas Co.. Mercer,1120/76
Boiler No. 1 1/21/76
1/22/76
(30
33
) 27A
\ 27B
34A
34B
( 4A
I 4B
1 4C
4A
1 4B
4C
Base
Base
Low NOx
Low NO
Low NO
Low NO
Base
Base
Base
Low N0
Low NO
Low N0
325
323
321
321
330
330
265
268
269
265
268
269
.59
.59
.42
.55
.45
.46
.52
.56
.27
.57
.77
.26
2.56
2.59
1.85
2.41
1.96
2.02
2.25
2.44
1.17
2.47
3.38
1.12
1337
1367
972
1264
1006
1036
1329
1290
671
1286
1793
606
3.11
3.18
2.26
2.94
2.34
2.41
3.09
3.20
1.56
2.99
4.17
1.41
96.79
96.86
95.58
96.60
95.73
95.85
96.76
96.88
96.44
98.14
98.66
95.69
0.66
0.55
0.46
0.18
0.50
0.59
2.1
1.3
2.2
3.0
1.7
5.8
5.36
5.36
5.43
5.43
5.62
5.62
9.57
9.23
11.00
9.57
9.23
11.00
4497
4497
4513
4513
4508
4508
7567
7615
7323
7567
7615
7323
8,094
8,094
8,123
8,123
8,115
8,115
13,620
13,708
13,181
13,620
13,708
13,181
-------
Pulverized Coal Firing
Figure 3—21.
Particulate emissions from utility boilers.
n
8
6
C
4
4000-
3000 -
2000 -
0
‘-4
‘I
6
I ,
a—. .c
Joo 0
a
.-4 —
.0
U
‘.4
4J
a
1000- —
500 -
3
7
0
——______
0
0
0
0
©
•
2
U
4
I
U
A
•
•
o Tennessee Valley Authority
Widows Creek, No. 5
o Southern Electric Generating
Company, E. C. Gaston, No. 1
&No.2
A Salt River Project, Navajo,
No. 2
• Public Service Company of
Colorado, Comanche, No. 1
• Public Service Electric and
Gas Company, Mercer, No. 1
I I
2
1
100
300 400 500 600 700 800
Load, Mwe
71
-------
Brink multi—stage cascade impactor incorporated in an EPA type sampling
train between the cyclone and the filter as discussed above and in section
2.4. A separate pump and flow rate measuring device was used to pull
samples through the Brink impactor for particle size distribution measure-
ments. This arrangement permitted simultaneous sampling for mass loading
and particle size distribution determination at the required rates in iso—
kinetic traverses of the boiler flue gas ducts, with the Brink equipment
located outside of the boiler. Thus, the proper recommended flow rates
were obtained in the Brink impactor to obtain the specified aerodynamic
diameter cut points. The Brink impactor consists of five stages, each
having cut points at rated conditions at diameters of 2.5, 1.5, 1.0, 0.5
and 0.25 pm followed by a final filter.
Particle size distribution measurement results obtained in this
program are summarized in Tables 3—12, 3—13, 3—14 and 3—15. The data
include all of the material collected in the probe and the cyclone for
the size range greater than 2.5 microns. Comparing the data in Tables
3-13, 3-14 and 3—15, it is seen that there are no significant differ-
ences in particle size distribution between low NO and baseline opera-
tion for the three boilers tested. The data in Table 3—12 however,
show major increases for the smaller size fractions with low NO firing
that could adversely affect precipitator collection efficiency. The
significance of the latter data, however, is open to question, first, due
to the relatively small size of the sample obtained in these tests in the
Brink impactor and, second, because of the problems with leaks in the
sampling equipment during this particular set of measurements.
The sampling arrangement used for obtaining particulate particle
size distribution information employed the cyclone of the EPA train as the
first cut point ahead of the Brink impactor. It is generally accepted
that this device collects all material greater than 7 pm in diameter. The
fractionation size (D 50 ) for each of the Brink impactor stages is the
particle size (aerodynamic diameter in pm) collected with 50% efficiency
by the impactor stage. For the Brink impactor the aerodynamic diameter
(D 50 ) for each stage is as noted above when operated at the prescribed
flow rates. Figures 3—22, 3—23, 3—24 and 3—25 present the size distribu-
tion data vs aerodynamic diameter for each boiler tested as a function of
solids loading in cumulative percent below the stated size. It should be
noted that all the measurements were made at or near full boiler load under
both baseline and low N0 operating conditions. Referring to the results
shown in Figures 3-23 and 3—24, data plotted exhibit consistent trends for
both baseline and low NO operations indicating no adverse side effects
resulting from low NO firing modifications for these two boilers firing
Western coal. On the other hand, some scatter is indicated in Figure 3—25
for the lower size fractions in tests conducted on the wet bottom furnace
firing Eastern coal at the Public Service Electric and Gas Company’s
Mercer Station Boiler No. 1. However, there is no indication of detrimental
side effects due to low NO operation. The data in Figure 3—22 cannot be
reconciled with those in Figures 3-23, 3-24 and 3-25 which is not surprising
in view of the small samples obtained and some leaks in the sampling equip-
ment experienced during the tests on the Southern Electric Generating
Company’s boilers.
72
-------
TABLE 3—12
PARTICLE SIZE DISTRIBUTION, WT.%
E. C. GASTON BOILER NOS. 1 & 2
SOUTHERN ELECTRIC GENERATING COMPANY
Base — Boiler No. 2 Low NO., — Boiler No. 1
Size Range 29 30 Avg. 34 37 Avg .
>2.511* 93.67 91.44 92.6 59.50 59.23 59.4
2.5ji 2.53 2.65 2.6 11.74 9.79 10.8
0.63 0.60 0.6 4.05 4.10 4.0
l.Op 0.89 1.03 1.0 5.67 6.10 5.9
O.5 1.26 1.63 1.4 8.91 10.19 9.5
<0.511 1.02 2.65 1.8 10.12 10.60 10.4
* Includes probe and cyclone
-------
TABLE 3-13
PARTICLE SIZE DISTRIBUTION, WT.%
SALT RIVER PROJECT BOILER NO. 2
NAVAJO STATION — PAGE, ARIZONA
Baseline Firing _____ Low NO Firing
Size Range 2A (Box 1) 2A (Box 2) 3D Avg. 18C (Box 1) 18C (Box 2) 18E Avg .
>2.5p* 95.80 92.17 90.29 92.7 95.75 93.47 92.61 93.9
2.5 t 1.18 2.97 4.76 3.0 0.83 1.96 2.88 1.9
0.42 0.71 0.97 0.7 0.36 0.52 0.89 0.6
1.Op 0.67 1.02 1.23 1.0 0.68 0.72 1.18 0.9
O.5p 1.03 1.49 1.12 1.2 1.15 1.24 0.92 1.1
0.90 1.64 1.60 1.4 1.23 2.09 1.52 1.6
* Includes probe and cyclone.
-------
TABLE 3-14
PARTICLE SIZE DISTRIBUTION, WT. %
COMANCHE BOILER NO. 1
PUBLIC SERVICE COMPANY OF COLORADO
Norn a1 Firing j Low NO Firing
Size Range 30 33 Avg. 27A 27B 34 35
>2.5p* 82.18 81.38 81.8 77.4 83.84 82.41 79.29 80.7
8.73 9.5 8.9 11.31 9.53 7.04 7.76 8.9
1.96 2.05 2.0 2.71 1.62 2.29 2.48 2.3
l.Op 2.49 2.79 2.6 3.17 1.86 2.9 3.73 2.9
2.94 2.89 2.9 2.94 2.1 3.52 4.45 3.3
-------
-4
0 ’
TABLE 3—15
PARTICLE SIZE DISTRIBUTION, WT. %
MERCER STATION BOILER NO. 1
PUBLIC SERVICE ELECTRIC AI D GAS COMPANY
Base
Furnace No. 12
_____________________________ Low NO
______________________ _______ Furnace No. 11
Size Range 4A 4B 4C Avg. 4A 4B 4C Avg .
>2.5p* 84.8 88.3 81.9 85.0 88.6 90.7 80.0 86.4
2.5p 2.7 4.9 2.9 3.5 3.2 2.9 9.7 5.3
1.51.1 2.9 2.4 1.5 2.3 2.2 0.5 2.7 1.8
1.011 3.8 2.0 0.9 2.2 1.3 2.2 2.4 2.0
1.9 1.5 0.1 1.2 2.2 1.0 0.6 1.2
3.9 0.9 12.7 5.8 2.5 2.7 4.6 3.3
* Includes probe and cyclone.
-------
Aerodynamic Diameter, pm
Figure 3—22. Particle size distribution, Southern Electric
Generating Company, E. C. Gaston, boilers No. 1 and No. 2.
100. . ,
50. ..
I .
Co
Co
a)
N
0)
. c
C’ IJ
o u
ra
.
•d E-o
I-I
o U)
Co
a)
0)
•r•I
Cu
I
l0.u
5_
l.u
FirJ]g__
0
— ——
.-____.
— 0
. --- --, .
.— —
— —
—
____
-C’
I E EH
- — - ---..
:::z
ii:::::
-.-
——
——
F
—
—
—
- . .
—
—
—
pd
,0
‘
.
— — — - . .
74_
/9,----..
(
- 0 TI
— —U-: .. .. I
•Te t N . 0, 1 i e ne
DTetN. 9: lie ne
ATe: t N • 7: 1 x
OTetN 4i
I
0. .
0.1
0.1
0.5 1.0 5.0
10.0
77
-------
100.0
— -
‘ 1 r rized Co
1 F l
i
— -
Aerodynamic Diameter, iim
Figure 3—23. Particle size distribution,
Salt River project, Navajo station, boiler No. 2.
5O. .,
1,
(
U,
C i,
C U
x
U
N
U
•rI 1-i
, CU
CU
o
U) CU
0
•rl El
‘-I
o 0 )
U )
U
U
CU
L)
5.C
-
, -‘!!
— .-
::::: i
- .
0
2( ’__
:::::
——
—----..
:::
o Test N
DTestN.
. 2A
2A
Bo
Bo
1
2
] a
lae
e
I
——
— -
A Test N. 3C
£ Test N. 18
OTestN.18,Bx
- •TestN. 18
Ba el
, L w
jx
m
‘0,
,Lw 0
Lw 0
1.1’
0.
0.
0.1 0.5 1.0 5.0 10.0
78
-------
100.
U,
0)
U )
a)
•t-I 1.
, c
U, 1.1
o U)
(0 (
,
..-l (-1
H
o ca
(I) U)
a)
0)
.1-I
U,
H
0
Aerodynamic Diameter, urn
Figure 3—24. Particle size distribution, Public Service
Company of Colorado, Comanche station, boiler No. 1.
- 1. rized Co
1 Fii —
50
10.0
5.0
1.0
0.5
—
—
- -
H
—
1:
/
;fflr’
—A,
I yø
zL::::
——- -..
.
••
::::::
-----.._
OTest N
ATest N
. 30
. 33
Bas
as
ii
11
e
e
O.loi_
OTest N
27
Lo
N
•Test N
27:
Lo’
N
£Test N
OTest N
34,
34:
.ow
ow
NO
NO
0.5 1.0
x
x
5.0 10.0
79
-------
Aerodynamic Diameter, urn
Figure 3—25. Particle size distribution, Public Service
Electric and Gas Co., Mercer station, boiler No. 1.
1
I—
v
ed Coal
irini
C a
Ca
a,
N
r1
a,
1-i
o 11)
U, ca
r4 E-l
o w
ci ca
a,
r 1
c 1
I -I
0
.
r
—I
—
.1
-
- - -
l00.
50.
10.
5.u
l.c
0.1
0.1
—
— r
—;
i
; ___—_ 3
-3 — — - -
l’
rw
f—
I
‘
;IJ --; ;
..
—----.._
-----.._
IEEE
EEEE
OTest N
OTest N
•4A
. 4B
ase
ase
in
in
ATest N
OTestN
ATest N
. 4C ase
.4Aow
. 4B ow
•TestN.4C q
in
0
0
0.5
1.0
5.0
10.0
80
-------
3.3 FURNACE WATERWALL CORROSION PROBING
In the current program, significant changes were made in the
conditions for measuring corrosion rates by exposed coupons in order to
better relate rates obtained on corrosion probes to actual furnace wall
tube corrosion rates. In prior ER&E studies (2), coupon corrosion rates
obtained in 300—hour tests averaged approximately 50 mils per year with
considerable scatter between high and low values. These high rates were
the result of the higher than normal exposure temperature of 742 K (875°F)
and the mild acid pickling of the coupons prior to exposure which resulted
in high initial corrosion rates. In the present program, coupon temperatures
were controlled at temperatures approximating those of the furnace wall tubes,
603—658 K, (625—725°F) and the acid pickling procedure was eliminated. As
a result, the corrosion rates measured were generally lower and more consistent
for each boiler than the previous data. Other test conditions were kept the
same as in the previous program, i.e., probes were inserted through openings
in the furnace wall as close as possible to vulnerable furnace areas and the
exposure was maintained at approximately 300 hours. The analytical procedures
used were also the same in both programs. Thus, each coupon was visually
inspected after exposure and was photographed to record its appearance.
Scale was then removed from the outside diameter surfaces by dry honing with
glass beads under low pressure. The coupon was then weighed to obtain the
weight loss from the outside surface. This was followed by dry honing of
the inside diameter surface scale after which the coupon was reweighed to
determine the weight loss from the inside surfaces. Corrosion rates were
then calculated as the loss in niils per year (ni/yr) using the weight loss
data, the combined exposed coupon areas, the metal and scale densities, and
the exposure time.
Coupon corrosion rates were determined for 60 coupons mounted on
20 probes (3 coupons/probe) exposed in 8 boilers firing pulverized coal.
The data obtained are tabulated in Tables 3—16, 3—17, 3—18, 3—19 and 3—20.
Referring to the tables, it may be seen that the corrosion rates obtained
are quite consistent, ranging from values of 6 muls/year to around 40 mils/year
and averaging about 19 mils/year (excluding the data in Table 3—18 and for
probe No. 3 in Table 3—19 for reasons that will be discussed later). These
rates are considerably lower than the 50 mils/year average obtained in our
previous study under accelerated conditions. The lower and more consistent
coupon corrosion rates measured in this program reflect the changes made in
test procedures to more closely approximate actual furnace wall tube conditions.
However, these rates are still an order of magnitude greater than the 1 to 3
muls per year corrosion rates that are expected for the wastage of actual
furnace waterwall tubes under normal firing conditions. Therefore, the
corrosion probing results should be still viewed as only the relative measure
of corrosion tendency under baseline and low NOx firing conditions.
Referring to Table 3—18, it is seen that the coupons on probes No.
1 and No. 3 had undergone higher corrosion rates than those on probes No. 2
and No. 4, respectively. The higher rates are due to some of the metal having
been inadvertently removed in disassembling the coupons caused by galling of
the threads between coupons. The indicated rates of corrosion on probes No.
8 ].
-------
TABLE 3—16
CORROSION RATE DATA
WIDOWS CREEK STATION
BOILERS NO. 5 AND NO. 6
TENNESSEE VALLEY AUThORITY
Coupon
Corrosion Rate,
Boiler Probe No . (a) Firing Condition Mils/Year Average
( 14.6
Boiler No. 6 (North Side) 4 Base 12.0 12.3
I.. 10.2 (b)
( 11.1
Boiler No. 6 (South Side) 3 Base 11.6 10.8
L 9.7(b)
( 18.8
Boiler No. 5 (North Side) 2 Low NO 13.4 14.9
X L 12.5 (b)
( 15.1
Boiler No. 5 (South Side) 1 Low NO ( 13.4 13.4
x ¼. 11.7 (b)
(a) Probes located between A B burners, top & next to top rows, respectively.
(b) Coupons farthest into furnace.
-------
TABLE 3—17
CORROSION RATE DATA
E. C. GASTON STATION
SOUThERN ELECTRIC GENERATING COMPANY, WILSONVILLE, ALABAMA
Coupon
Corrosion Rate,
Boiler (a) Probe No. Firing Condition Mils/Year Average
Boiler No. 2 @ (IR No. 1) 3 Base ( 10.3 9.3
7.3
L 10.2 (b)
(8.0
Boiler No. 2 @ (IR No. 10) 4 Base 6.3 6.8
L 6.0(b)
( 20.0
Boiler No. 1 @ (IR No. 10) 2 Low NO 12.3 14.2
X I . 10.4 (b)
(11.4
Boiler No. 1 @ (IR No. 1) 1 Low NO 4 7.1 9.5
X ( 10.1 (b)
(a) Probes inserted through slag blower ports in side walls about 8 feet above top
burners and 12 feet from rear furnace corner.
(b) Coupons farthest into furnace.
-------
TABLE 3—18
CORROSION RATE DATA
NA VAJO STATION
SALT RIVER PROJECT, PAGE, ARIZONA
Coupon
Corrosion Rate,
Boiler (a) Probe No. Firing Condition Mils/Year Average
(62.8
No. 1 (North Side) 1 (b) Base 32.0 38.7
1. 21.3 (c)
(15.1
No. 1 (South Side) 2 Base 4 10.6 13.0
.. 13.2 (c)
(43.8
No. 2 (North Side) 3 (b) Low NO 29.2 30.9
x .19.7 Cc)
(16.2
No. 2 (South Side) 4 Low NO s 15,1 15.6
X 15.5 (c)
(a) Probes inserted through observation doors in side walls between B & C burner rows.
(b) Probes with damaged coupons.
(c) Coupons farthest into furnace.
-------
TABLE 3—19
PUBLIC SERVICE COMPANY OF COLORADO
COMANCHE STATION - BOILER NO. 1
CORROSION RATE DATA
Coupon Corrosion
Probe No. Location (a) Firing Cond. Rate 0 Mils/Yr. Average
S.E. Corner Between (81.9
3 (b) B&C Burners Base ( 49.1 48.9
( 15.8 (c)
N.W. Corner Between ( 34.0
4 B&C Burners Base 24.0 27.1
L 23.4 (c)
S.E. Corner Between (41.6
1 B&C Burners Low NO 25.6 27.6
X t. 15.6 (c)
W.V. Corner Between ( 15.5
2 B&C Burners Low NO . 13.5 41.7
X (, 96.2 (c)
(a) 5th floor elevation.
(b) Probe with damaged coupons.
(c) Coupons farthest into furnace.
-------
TABLE 3—20
CORROSION RATE DATA
MERCER STATION
PUBLIC SERVICE ELECTRIC AND GAS COMPANY
Coupon Corrosion
Boiler/Furnace Location Probe No. Firing Cond. Rate, 1411s/Yr. Average
Boiler No. 2 Elev No. 3
10.1
3 Base 73.1 66.8
(Furnace No. 1) Between Furnaces 117.3 (a)
( 6.1
Boiler No. 2 South Wall, 4 Base 9.7 17.3
(Furnace No. 2) Elev No. 3 1 38.2 (a)
f 5.2
Boiler No. 1 South Wall,
1 Low NO 17.1 61 8
(Furnace No. 1) Elev No. 3 X L. 163.2 (a)
( 12.4
Boiler No. 1 Elev No. 3
2 Low NO 41.9 59 2
(Furnace No. 1) Between Furnaces X 123.4 (a)
(a) Coupons farthest m W furnace.
-------
1 and No. 3, therefore, are not reliable. Comparing the data from probes
No. 2 and No. 4 with the results in Tables 3—16 and 3—17, It may be seen
that corrosion rates tend to vary between 10—15 mIls/year with the maximum
at about 20 mIls/year. Also, although these results suggest somewhat higher
corrosion rates for the probes exposed under low N0 conditions, the magnitude
of the increase is relatively minor.
The validity of the corrosion rates determined for probe No. 3
In Table 3—19 is also suspect due to inadvertent metal removal caused by
thread damage. The results from the remaining probes, however, appear to
be quite reliable even though they are higher than for the other tests.
Comparing the corrosion rates for probes No. 1 and No. 2 to those for probe
No. 4, there is a suggestion of a lower rate of corrosion under “low N0 ”
firing conditions. However, within the accuracy of corrosion rate measure-
ments with probe coupons, it appears that the effect of “low N0 ” is minor.
The corrosion rates measured on the Mercer Station boilers of the
Public Service Electric and Gas Company, as shown in Table 3—20, are quite
different from those determined in the other tests. Due to the close spacing
of the furnace tubes and the construction of the access doors in which the
probes were mounted on the furnace, extensions were required on the probes
to hold the coupons. Three “dummy” coupons were inserted between the probe
proper and the coupons exposed for test (6 coupons in series). As indicated
by the high apparent corrosion rates measured, the extensions evidently
affected the air cooling available for the exposed coupons. This is sup-
ported by the fact that the highest rates were measured for each probe for
the coupon inserted furthest into the furnace, with successively lower rates
measured for the adjacent coupons closer to the furnace walls. Using the
coupon rates measured only for the coupons inserted closer to the tube walls,
no significant difference can be detected between the values obtained for
baseline and low N0 firing conditions, respectively.
Directionally it appears that under some conditions “low N0 ”
firing conditions may result in somewhat higher corrosion rates based on
the results of coupon corrosion probe measurements. Thus, although the
conclusion of the corrosion probing results is that no major differences
in corrosion can be observed for coupons exposed under low N0 operating
conditions compared to those subjected to baseline boiler operating con-
ditions, only long term corrosion measurements of actual furnace tube
wastage can answer the question of the magnitude of corrosion rate increase
caused by staged firing of boilers with coal. The corrosion probing data
obtained so far provides confidence that catastrophic corrosion problems
will not occur as a result of stage firing of coal, and therefore, long
term corrosion testing under low N0 conditions is a reasonable approach.
It is planned that further detailed corrosion probing results will be
correlated with actual tube wastage measurements obtained in long term
corrosion tests.
87
-------
3.4 EFFECT OF COMBUSTION MODIFICATIONS ON BOILER PERFORNANCE
Prior field study results (2) indicated a tendency for particulate
carbon content to increase under low NO operating conditions, especially
for wall fired boilers. Potentially, an increase in unburned carbon should
result in lower boiler efficiency but this adverse side effect did not mate-
rialize in the previous study because of the offsetting effect of lower
excess air operation under low NO operating conditions. In the present
study the effect of modified combustion operation on boiler performance was
investigated and evaluated for each boiler tested where particulate data
were obtained under baseline and optimum “low NOR” conditions. Control
room board data and other pertinent information representative of each test
run were recorded and boiler efficiency was calculated using the heat loss
method (ASME Steam Generating Units, Power Test Codes, Abbreviated Efficiency
Test). Calculations were based on the assumption that fly ash and bottom ash
combustible content were the same and unmeasured losses were 0.5 percent. An
example of typical performance data and the calculations made are shown :in
ASME test forms in Tables 3—21 and 3—22.
Table 3—23 lists boiler efficiencies calculated for each boiler
tested along with other pertinent boiler performance information. Differences
in calculated boiler efficiencies between baseline and low NOx test conditions
provide a comparison of any efficiency debit or credit accruing to low NO
operating conditions. Boiler efficiency, in general, increases with load and
decreases with increasing coal ash or unburned combustible content of the par—
ticulates.
It can be seen from the data in Table 3—23 that low NO
modified combustion operation does not decrease boiler efficiency signifi-
cantly. In fact, some apparent increases are calculated in some cases. For
example, the Tennessee Valley Authority, Boiler No. 5, the Salt River Project,
Boiler No. 2, and the Public Service Company of Colorado, Boiler No. 1, test
data show small increases in efficiency under low NOx firing conditions due to
decreases in uncombustible losses, lower coal ash, and excess air levels.
Differences In boiler efficiency are also Inconsequential with low NOx opera-
tion where carbon losses increased substantially, as in the case of the
Southern Electric Company, Boilers No. 1 aix! No. 2. Here the efficiency
debits due to increases in unburned carbon for the low N0 tests are offset
by the increases in efficiency resulting from operation at lower excess air
and the lower ash content of the coal burned during the low NOx tests. In
general, staging the combustion operation and reducing excess air levels for a
boiler to reduce NO emissions tends to increase somewhat the unburned com-
bustibles. However, the debit in efficiency resulting from such incomplete
utilization of the fuel is offset by the improved efficiency resulting from
lower excess air operation. Therefore, the conclusion reached based on these
results is that “low NOR” firing, as used in this study, had no major effect
on boiler efficiency.
88
-------
TABLE 3—21
ASME TEST FORM
SUMMARYSHEET FOR ABBREVIATED EFFICIENCY TEST
TEST NO ].A BOILER NO. 6 DATE4— 18—72
TV& LOCATION Widows Creek
Esso Research & Engineering Co.OBJECTIVE OF TEST Boiler PerformanceuRAllo..14 Hrs.
B&W Radiant RATED CAPACITY 125 MW —
Type E BURNER. TYPE & SIZE
Coal MINE COUNTY STATE SIZE AS FIRED
TEMPERATURES FUEL. DAT
—
A
BOILER DRUM
S. H. OUTLET po.o
COAL AS FIRED
°ROX ANALYSIS
w,
OIL
37
MOISTURE
SI
FLASH POINT F ’ —
R H. INLET pita
38
VOL MATTER
52
St Gravity Deq.API ’
H. OUTLET
39
FIXEDCARBON
53
VISCOSITY AT SSU
BURNER SSF
AT S H. OUTLET
40
ASH
44
TOTAL. HYDROGEN
5 wt
AT R H. INLET F
AT R H. OUTLET F
ENTERING(ECON )(BOILER) F
— TOTAL
411 Bt . p., lb AS FIRED
I ASH SOFT TEMP.’
42j ASTMMETHOD
/i1/..2
41 Bti, par lb
GAS
%VOL
MOISTURE OR P P M —
COAL OR OIL AS FIRED
— ULTIMATE ANALYSIS
54
CO
F
43
CARBON
35
Cl i . METHANE
COMBUSTION
Ten.per.ti.re) I F
23
44
HYDROGEN
4L 2
—
C ,H, ACETYLENE
FUEL F
45
OXYGEN
57
C,H, ETHYLENE
(Bo.I.r) ( (con.) (Atr H,,.)
72 ,
46
NITROGEN
C,H ETHANE
AH (If condition, to be
F
SULPHUR
O
H, S
QUANTITIES
40
ASH
60
CO 3
LIQUID (TOTAL HEAT) Btu/Ib
37
MOISTURE
61
H 2 HYDPOGF.N
OF(SATURATED)(SUPERHEATED)
Btu/Ib
TOTAL
TOTAL
FEED TO (BOILER) Btu/Ib
COAL PULVERIZATION
48 GRINDABILITY
62
TOTAL HYDROGEN
DENSITY 68 F
REHEATED STEAM R.H. INLET Bt 5 ,/Ib
INDEX
ATM. PRESS
REHEATED STEAM R. H
Bt ./Ib
49
FINENESS %THRU
—___________
63
BiUPERCUFT
STEAM (ITEM 16—ITEM 17) Btu/Ib
50
FINENESS % THRU
200
41
Btu PER LB
STEAM(ITEM 19—ITEM 18) Biu/Ib
64
INPUT.OUTPUT ITEM 31 • 100
EFFICIENCY OF UNIT S ITEM 29
FLY ASH) PER LB
lb/lb
J 7
HEAT LOSS EFFICIENCY
Bt ./Ib
A. F. FUEL
S of A F
FUEL
(WEIGHTED AVERAGE) Bt ./Ib
LB AS FIRED FUEL lb/lb
jj J
65
66
HEAT LOSS DUE TO DRY GAS
HEATLOSSDUETOMOISTURE IN FUEL
HEAT LOSS DUE TO H 3 O FROM COMB OPH
fr. O
,.51’
FIRED FUEL BURNED lb/lb
/J,
67
QUANTITIES
68
HEAT LOSS DUE TO COMBUST IN REFUSE
lb/h,
69
HEATLOSSDUETORADIATION
lb/h,
(AS FIRED wi) lb/hr
0
71
UNMEASURED LOSSES
TOTAL
a
73.Ib
(Li n... 28 x lien. 41) kB/h,
1000
72
EFFICIENCY (100 — lien, 71)
f 1
‘
FLUE GAS ANAL (BOILER)(ECON) (AIR HTR) OUTLET
32
CO,
%VOL
L 1 J
33
35
0,
-_____
N, (BYDIFFERENCE)
%VOL
c.vOL
%VOL
j3
_i), .o2
36
EXCESS AIR
plc 4.1-0 (1964
‘0
HEAT OUTPUT IN BLOW-DOWN WATER
TOTAL
HEAT ( lien. 26th. . ’. 20)G(Itet ’ . 27 Ite, 21 ) + lien, 30
OUTPUT 1000
• Not Required for Efficiency Testing
For Point of Measuren.enl See Pa, 7 2 8 I-FTC 4 1.1964
89
-------
For rigorous ia •rmina?,on ol auc.ss air see Appetid.o 9 2 — Plc 4 1-1964
• If buss ate --r -reosured, one ABMA Standard Rod.ohon Loss Char’. F . g 8, PTC 1 1- 1961
Unm..su,sd losses listed in Plc 4 1 b.t not tob Ioied above racy by proaided for by aligning a mutually
agreed upon e.ue For Item 70
TABLE 3-2: ’
ASME TEST FORM
CALCULATIONSHEET FOR ABBREVIATED EFFICIENCY TEST
PlC 4.1-b ( 964)
Revised September, 7965
OWNEROFPLANT TVA TESTNO 1A BOILERNO 6 DATE4_18_72
30
ITEM IS ITEM 17
HEAT OUTPUT IN BOILER BLOW-DOWN WATER °LB OF WATER BLOW-DOWN PER HR x I - — - - .]= hE/h:
L 1000
24
If ,mproCttCoI to weigh reFuse, this I
item can be estimated as Follows
% ASH IN AS FIRED COAL
DRY REFUSE PER LB OF AS FIRED FUEL u NOTE- IF FLUE DUST & ASH
100 — % COMB. IN REFUSE SAMPLE I
PIT REFUSE DIFFER MATERIALLY
1
ITEM 43 ITTEM 22 ITEM 231 IN COMBUSTIBLE CONTENT. THEY
SHOULD BE ESTiMATED
CARBON BURNED 7,2.7 [ o i.i 7 ( SEPARATELY SEE SECTION?.
PER LB AS FIRED a — COMPUTATIONS
100 14.500
FUEL
25
DRY GAS PER LB IICO, + 8O + 7(N 1 • CO)
AS FIRED FUEL a X (LB CARBON BURNED PER LB AS FIRED FUEL ÷
3(CO 3 • CO) 1
BURN ED
lix /4i.9- Bx 3.3 ‘( z i ô.o -) xi I-/i
ITEM 32 ITEM 33 ITEM 35 ITEM 34 rITEM 24 ITEM 471
- ( 1TEM32 ITEM34’ [
267 J --
3 x t -*. +
36
EXCESS
AIR?
CO
01 — —
2
o 100 x
2682N 1 — (01 —
2
a 100 x
ITEM 34
ITEM 33 — ________
2
.2632 (ITEM 35) — (ITEM 33 — ITEM 34
2
HEAT LOSS EFFICIENCY
Bt 0 /bb
AS FIRED
FUEL
LOSS x
• R
LOSS
%
HEAT LOSS DUE LB DRY GAS ITEM 25 (ITEM13) —(ITEM II )
65 TO DRY GAS PER LEAS xCx (‘lag — ‘our) a xo 24 372 y - a7 O
FIRED FUEL Unit
79O
65
— X ?OOa
41
6. c
66 HEAT LOSS DUE TO LB 1130 PER LB [ (ENTHALPY OF VAPj T 1 PSIA & T GAS LV
MOISTURE IN FUEL AS FIRED FUEL
/2 l7.
— (ENTHALPY OF LIQUIDAT T AIR)) ITEM 37 XRENTHALPY OF POR
— .56 100
AT 1 PSIA & T ITEM 13) —(ENTHALPY OF LIQUID AT I ITEM 1 i,I 3
x 100 a
41
67 HEAT LOSS DUE TO H 1 O FROM COMB OF H, = 9H , x [ ENTHALPY OF VAPOR AT 1 PSIA & T GAS
2j7 / z 7 5? LVG) — (ENTHALPY OF LIOI I T T AIR))
x I EM 41 x [ (ENTIfALPY OF VAPOR AT_). PSIA & T ITEM 13) — (EIITHALPY OF LIQUID AT
100 1 ITEM I I) ] a -
5
6
x 100 =
3.’/?
68 NEAT LOSS DUE TO ITEM 22 ITEM 23
COMBUSTIBLE IN REFUSE o 4/5’7 x t/, 2 / a / & L .ä’
/ tL
68
— x l OO a
i
/ 24
6 HEAT LOSS DUE TO - TOTAL BTU RADIATION LOSS PER HR
-
RADIATIOM LB AS FIRED FUEL — ITEM SE
- O2—
69
41
7 , j)o2,
70 UNMEASURED LOSSES
-
70 100
41
—
-
71 TOTAL
...
í3./o
72 EFFICIENCY u (100 — ITEM 71)
Y6.8
90
-------
TABLE 3—23
SUMMARY OF BOILER PERFORMANCE CALCULATIONS
Coal
NO Emissions at Ash, %
Company, Boiler Firing Test Load, 3% O (Wet % Carbon on Boiler
Station No. Mode No. MWe %02 PPM lb./106 BTU Basis) Particulate Efficiency. %
Tenn. 5 Base lB 128 2.4 409 .56 239 14.64 13.0 86.6
Valley Base 1C 124 2.5 441 .60 258 18.27 9.1 86.2
Authority,
Widows 5 Low NO 12C 125 2.5 412 .56 241 12.30 6.2 87.7
Creek Low NO 12D 125 2.5 412 .56 241 12.30 8.1 87.3
Station
So. Electric 2 Base 28A 272 4.2 584 .79 343. 15.00 2.7 89.8
Cenerating Base 28B 272 4.2 584 .79 341 15.00 1.1 90.1
Company, Base 29 268 4.1 572 .78 335 16.06 1.7 89.9
E. C. Gaston Base 30 270 4.3 606 .82 354 13.28 2.1 89.2
Station
1 Low NOx 26 270 3.9 349 .47 204 12.12 5.7 89.2
Low NOx 27A 268 3.9 316 .43 185 9.84 3.4 89.6
‘O Low NO 27B 268 3.9 316 .43 185 9.84 4.1 89.8
Salt 2 Base 3B 799 4.8 492 .67 288 11.49 0.9 89.0
River Base 2A 792 3.9 332 .45 194 5.94 1.8 89.6
Project, Base 30 785 4.8 346 .47 202 6.06 2.0 89.6
Navajo
Station 2 Low NO 18B 808 4.4 349 .47 204 11.49 0.8 89.2
Low NO 18C 808 4.6 346 .47 202 6.62 1.2 89.5
Low N0 18E 798 3.7 280 .38 163 6.04 1.5 89.7
Pub. Ser. Base 30 325 3.7 380 .52 222 5.36 .66 80.8
Co. f Colorado, Base 33 323 3.9 389 .53 227 5.36 .55 80.2
Comanche 1
Station Low NOx 27A 321 3.9 246 .33 144 5 43 .46 80.7
Low NOx 27B 321 3.9 246 .33 144 5.43 .18 80.7
Low NOx 34 330 3.4 205 .28 120 5.62 .50 80.3
Low NOx 35 330 3.4 205 .28 120 5.62 .59 80.3
Base 2 265 3.8 1095 1.49 640 9.57 2.1 90.0
Pub. Ser. E&G Base 4 268 3.1 1175 1.60 687 9.23 1.3 90.4
Co., Mercer Base 6 269 3.3 1070 1.46 626 11.00 2.2 90.1
Station
Low NOx 1 265 1.5 940 1.28 550 9.57 3.0 90.1
Low NO 3 268 1.9 827 1.12 484 9.23 1.7 90.3
Low NO 5 269 2.2 924 1.26 540 11.00 5.8 89.8
-------
3.5 SUMMARY OF SULFUR OXIDES MEASUREMENTS
Flue gas samples for sulfur oxides (SO 2 and SO 3 ) determination
were taken from the boiler ducts at the same location as the samples
extracted for continuous analysis by the instrumental train of the ER&E
analytical van discussed in section 2.7. Normally this location is at
the boiler outlet downstream of the economizer but upstream of the
regenerative airheaters. Wet chemical SO 2 /S0 3 determinations followed
procedures outlined in section 2.7.
Tables 3—24, 3—25, 3—26. 3—27 and 3—28 summarize the results
of sulfur oxides measurements made in this program. Sulfur oxides emis-
sion concentrations calculated on the assumption of 100 percent conversion
of the sulfur content of the fuels burned during the test period, given
in the tables, can be compared with the separately determined values of
SO 2 and SO 3 . It can be seen that calculated SOx values are higher than
measured sulfur oxides concentrations. This is not unexpected for pul-
verized coal fired boilers. In dry bottom pulverized coal fired boiler
installations, about 15 to 20 percent of the ash is extracted as bottom
ash and 80 to 85 percent is carried with the flue gases to electrostatic
precipitators (or other collection devices) which collect up to 98/99+
percent of the flyash. Another portion of the fuel ash remains in the
boiler as slag on the furnace walls or as deposits in the convection
sections and hoppers. In each case, a portion of the fuel sulfur content
may remain with these ash accumulations to decrease the amount emitted as
SO 2 or SO 3 in the flue gases. Although firm sulfur balance data are not
available, from 5 to 25 percent of the fuel sulfur may be retained in a
boiler, especially where a boiler is fired with high calcium content coal.
Based on a comparison of calculated S0 emissions to the experimental data,
it appears that the results obtained in this program are in line with
expectations for the boilers tested.
Sulfur trioxide determinations, which range between 8 and 16
ppm as shown on Tables 3—24, 3—25 and 3—26, are in accord with results
obtained from other pulverized coal fired installations. The SO 3 values
are within the anticipated range of up to five percent of total SO in
the flue gas except for the data presented in Table 3—27 which appear
to be too high, and are therefore suspect. For mixed coal—oil fuel
firing, the sulfur dioxide emission data on Table 3—28 compare favorably
with calculated SO emissions based on the sulfur content of the coal
used at this plant. With oil firing, most or nearly all of the fuel
sulfur appears in the flue gases so the values shown in the table
may be proportionately higher due to the small percentage of oil being
fired during these tests.
92
-------
Table 3—24
Public Service Electric and Gas Company
Mercer Station
Wet Chemistry S02/S0 3 Analyses
Of Flue Gases
Pulverized Coal Firing
Wet Chemistry Analysis
Run Boiler Firing Load 02, Calculated SO 2 _____ S03 S0 3 /S02
Date (76 ) No. No. Condition MWe SO, PPM ng/J PPM ng/J PPM Ratio, %
1—29 201 2 Base 160 4.5 487 611 8 8
1.3
3.0 924 532 667 9 9
1—29 202 2 Base 180 4.5 J 487 611 9 9
1.5
532 667 10 10
i O 2—10 203 1 Low NO 290 2.4 ) 567 711 10 10
x I 1.4
990 550 689 10 10
2—10 204 1 Low NO 290 2.4 1 603 756 11 11 1.5
585 733 11 11
2—11 205 1 Base 210 4.4 759 952 9 9 9
1039 821 1030 10 10
2—11 206 1 Base 210 4.4 ) 736 923 11 11 1.2
796 998 12 12
(l)ppm Ratio
-------
Table 3—25
Public Service Company of Colorado
Comanche Station, Pueblo, Colorado
Wet Chemistry S0 2 /S0 3 Analyses
Of Flue Gases
Pulverized Coal Firing
Wet Chemistry Analysis
Boiler Firing Load 02, Calculated SO 2 _____ SOi S0 3 /S0 2 ( 1 )
Date No. Condition MWe ____ SO , 1 , PPM ng/J PPM ng/J PPM Ratio,
7—16—75 1 Base 331 4.0 \ 215 270 15 15
5.6
3.0 499 228 286 16 16
7—16—75 1 Base 331 3.8 J 235 295 15 15
5.1
3.0 246 308 16 16
7—18—75 1 Base 332 — — 376 225 282
—— ) 219 275 9 9 3.4
7—30—75 1 Low NO 324 3.9 224 281 9 9
x I 3.1
3.0 (2) 236 296 9 9
7—30—75 1 Low NO 324 3.9 1 225 282 10 10
x I 3.5
3.0 ) 238 298 10 10
(1)ppm I atio
2 Coa1 Data Not Available
-------
Table 3—26
Salt River Project
Navajo Generating Station
Wet Chemistry S0 2 /S0 3 Analyses
Of Flue Gases
Pulverized Coal Firing
Wet Chemistry Analyses 3
Boiler Firing 02, Calculated SO 2 SO S0 3 /S02
Date No. Condition S0, , PPM ng/J PPM ng/J PPM Ratio, %
6—12—75 2 Base 6.1 218 273 5 5
393 1.8
3.0) 263 330 6 6
6—9—75 2 Base 5.2 230 288 — — — —
3.0 450 261 327(1) — — ——
6—9—75 2 Base 6.0 ( 210 263 —— ——
3.0 J 259 315(2) — — ——
6—23—75 2 Low NO 3.8 158 198 9 9
X (4) 4:7
3.0 ) 165 207 10 10
At Economizer Outlet
360’ Stack Elevation
3 ppm Ratio
4 Coa1 Data Not Available
-------
Table 3—27
Southern Electric Generating Company
E. C. Gaston Station, Wilsonville, Alabama
Wet Chemistry S02/S03 Analyses
Of Flue Gases
Pulverized Coal Firing
Wet Chemistry Analysis (1)
Boiler Firing Calculated SO 2 _______ S03 S0 3 /S02 (2)
Date No. Condition SOT, PPM ng/J PPM ng/J PPM Ratio, %
1—31—75 2 Base 1072 726 910 129 129 14.2
1—31—75 2 Base 1193 552 692 127 127 18.4
1—30—75 2 Base 1415 883 1107 130 130 11.7
w
0 ’ (1)
Uncorrected Values
(2)
ppm Ratio
-------
Table 3—28
Potomac Electric Power Company
Morgantown Station, Morgantown, Maryland
Wet Chemistry SO 2 Analyses
Of Flue Cases
Coal/Oil Combination Firing
Wet Chemistry Analyses
Test Boiler Sampling Firing 50 2(1 ) Calculated
No. No. Location Condition ng/J PPM SOS, PPM( 2 )
1 1 At Exit Gas Duct Base 1247 1563 1685
2 1 At Remote Pumping Base 1246 1562 1849
Unit
3 1 At End of Sampling Base 1253 1571 1751
Line At Van
Uncorrec ted Values
2 Calculated from Coal Analysis
-------
4. CONCLUSIONS
In this section of the report, the overall conclusions of the
field study results will be sumin rlzed. The effects of combustion inodi—
fications used for controlling NO and other pollutant emissions or
gaseous and particulate emissions, furnace water—wall corrosion and
boiler performance will be discussed based on the results obtained.
4.1 GASEOUS EMISSION MEASUREMENTS
In section 3.1 of this report, the results obtained in field
studies on eight power generation combustion sources were presented for
each individual unit tested. In this section, the conclusions based on
emission results for the individual units will be presented, including
overall correlations based on the results of the seven coal fired boilers
tested.
Table 4—1 presents a sunnuary of the NO emissions for coal
fired boilers under baseline and “low NOR” operation for the seven boilers
tested during this contractual period. Widows Creek boiler No. 5 is a
rear wall fired unit; Ernest C. Gaston boiler No. 1 is a horizontally
opposed fired unit (equipped with the new B & W dual register low NO
burners); Mercer boiler No. 1 is a twin furnace, front wall fired, wet
bottom unit; while the remaining four boilers are of the tangential
design type (three of these tangentially fired boilers are equipped
with overfire air ports). Baseline emission data were also obtained on
E. C. Gaston boiler No. 2 (a “sister” unit of Gaston No. 1) equipped
with conventional pulverized wall burners. Gaseous measurements of 02,
NO and CO are listed for baseline operation, “low NOx—I” (modified
firing operation at full load) and “low NOx—Il” (modified firing opera-
tion at reduced load). Baseline NO emission levels are listed at both
actual excess air levels and calculated for 20% excese air so that compari-
son between boilers can be made on the same operating basis. Note that
correcting emissions to a 3% 02 basis provide data on a constant dilution
basis regardless of operating conditions.
A numerical example will help explain why correction of actual
NO emissions to a 3% 02 dry basis for dilutiiin . and the normalization of
baseline operation emission levels are useful. Table 4—2 contains baseline
operation emission data from Navajo No. 2 Unit for three excess air level
operations: 116%, 120% and 129% stoichiometric air. N0 emissions test data
are listed on the bases: ppm NOx (as measured), ppm NO (3% 02 dry basis)
and lbs. NO /lO 6 BTU. The % change (A 1 ) in emissions are listed for the
change in operation from 116% to 120% and 120% to 129% stoichiometric air.
Note that the A for ppm NO (3% 02, dry) and lb. N0 /1O 6 BTU are identical
since they are both on a mass emission basis while the ppm NOR, as measured,
are lower because of a dilution factor. Thus, a correction for dilution
is necessary to put the data on a comparable mass emission basis. It should also
be noted that if the NO data were not normalized to 120% (or some other
arbitrary fixed level), Navajo No. 2 Unit operating at 29% excess air would
give a 23% higher NO emission level than an identical emitting boiler which
operated at 20% excess air under baseline operation. Thus, normalizing
baseline operation to a constant excess air level removes one extraneous
pperating variable so that the effect of other variables on NO emissions
such as type of firing, size, etc. can be more easily analyzed.
-------
TABLE 4-1
SUMMARY OF NO EMISSIONS FOR COAL FIRED BOILERS
NO Emissions
Operating Mode ppm Lb . ppm CO
Boiler ( Gross Load — MWe ) °2 ( 3% 02) 106 BTU ng/J ( 3% 02 )
1. Widows Creek No. 5 (RW)(a) Base — 125 4.0 597(567)(b) 0.81 349 29
“Low NOx — I” — 125 3.2 468 0.64 274 88
“Low 1 Ox — II” — 100 5.3 305 0.41 178 285
2. Ernest C. Gaston No. 1 (HO)(a)Base (NO.2 Unit)- 270 4.3 591(491)(b) 0.53 227 26
(Low NO Burners) “Low NO — I” — 270 2.4 278 0.38 163 65
“Low NO, — II” — 190 1.6 182 0.25 106 329
3. Navajo No. 2 (T)(a) Base — 800 4.8 492(400)(b) 0.67 288 64
(Overfire Air Ports) “Low NOx — I” — 802 3.6 282 0.38 165 330
4. Comanche No. 1 (T)(a) Base — 323 4.0 417(370)(b) 0.57 244 27
(Overfire Air Ports) “Low NOx — I” — 333 3.7 261 0.35 153 35
5. Barry No. 2 (T)(a) Base — 130 49 341(250)(b) 0.46 199 22
(Overfire Air Ports) “Low NO — I” — 129 3.3 189 0.26 111 49
6. Morgantown No. 1 (T)(a) Base(c) — 570 4.9 552(534)(b) 0.75 323 21
“Low NO — I” — 572 4.7 403 0.55 236 20
(d)
7. Mercer No. 1 (FW)(a) Base — 290 3.9 1383(1354)(b) 188 809 22
“Low N0 — I” — 220 3.4 356 0.48 208 218
“Low NOx — II” — 160 4.2 351 0.48 205 455
(a) Types of Firing: RW, rear wall; HO, horizontally opposed; T, tangential; FW, front wall.
(b) Baseline NO emissions calculated for 20% excess air operation.
x
(c) Fuel burned 93% coal, 7% oil.
(d) Fuel burned 74% coal, 26% oil.
-------
TABLE 4.2
NAVAJO NO. 2 - BASELINE OPERATION EMISSION DATA
1 2 3
—1—
% Stoichionietric Air 116 (3.0% 02) +3.4% 120 (3.6% 02) +7.5% 129 (4.8% 02)
PPM NO (As Measured) 360 +7.5% 387 +14.2% 442
PPM NO (3% 02, Dry) 360 +11.1% 400 +23.0% 492
Lbs. NO /106 BTU 0. 490 +11.1% 0. 544 +23.0% 0.669
The wall fired boilers (Widows Creek No. 5, Gaston No. 2 and
Mercer No. 1) under baseline operating conditions produced NO emissions
above the 301 ng/J (0.7 lb./lQ BTU) level of the NSPS for coal—fired steam
generators. However, under modified combustion operation the dry bottom,
wall fired boilers, Widows Creek No. 5 (low excess air, staged firing) and
Gaston No. 1 (low N0 Babcock & Wilcox burners) emitted N0 below the level
of NSPS under full load operation, while the wet bottom Mercer boiler No. 1
required a 20% load reduction and low excess air, staged firing operation
to reduce NO emissions below 0.7 lb./10 6 BTU. At reduced load (by 20%)
Widows Creek boiler No. 5 produced NO emissions as low as 179 ng/J (305 ppm)
in short term tests, i.e., 49% below the baseline NO level at full load.
Gaston boiler No. 1 equipped with newly designed dual register burners,
produced 36% lower NOx emissions than Gaston boiler No. 2 (a sister boiler
equipped with conventional burners) under baseline conditions. Modified
firing operation of Gaston No. 1 unit at full load and at reduced load
lowered NO emissions by 59% and 69%, respectively, compared to baseline
operation with the conventional burners. For the wet bottom Mercer boiler
No. 1 operated at 20% reduced load, staged combustion with low excess air
reduced NOx emissions by 74% from the baseline emission level.
Under full load, baseline operation (cooling air only), the
three tangentially fired boilers that are equipped with overfire air ports
met the 301 ng/J (0.7 lb./l0 6 BTU) NSPS for NOR. Morgantown boiler No. 1
emitted NO slightly above this level under baseline firing conditions
with coal fuel. Modified firing using the overfire air ports at low over-
all excess air resulted in reductions of 37 to 45% in NOx emissions. Staged
firing of Morgantown boiler No. 1 reduced NO emissions by 27% for this
particular unit. Comparison of baseline NOx emissions normalized to 20%
excess air for the three boilers equipped with OFA to those measured on
four tangentially fired boilers in previous ER&E field test studies show
an 18% NO reduction due to cooling air.
The ranges of NO emissions measured under normal firing operat-
ing conditions as a function of excess air level (% 02 in the flue gas)
are shown in Figure 4—1. The line representing the wet bottom Mercer No. 1
boiler (labeled ME) is plotted using the right hand N0 scale while all of
the other boilers are plotted using the left hand ordinate values.
100
-------
OPPOSED WALL FIRED
G — GASTON NO. 1
TANGENTIALLY FIRED
M - DRGANTOWN NO. 1
C — COMANCHE NO. 1
N - NAVAJO NO. 2
B — BARRY NO. 2
0 1 2 3 4 5 6
1400
6 ’
U,
I -
U)
1300 >-
0
1200
1100
AVERAGE % OXYGEN IN FLUE C,AS
Figure 4—1. PPM NO vs % 02 in flue gas from
coal—fired boilers un er normal firing operation.
FUEL FIRED
C’)
a
c 1
0
m
0
a-
600
500
400
300
200
100
0
0
,
WALL FIRED
ME — MERCER NO. 1
WC — WIDOWS CREEK NO, 5
101
-------
As discussed in section 3.1, excess air had a significant effect
On NOx emissions for each boiler under normal firing operation. The
relationship of N0 vs 02 is shown in FIgure 4—1. With the exception of
Comanche boiler No. 1, (which had limited operating flexibility for changing
the excess air level) the slopes, calculated by least squares, of these
lines are fairly consistent. However, the actual N0 levels vary because
of differences in boiler size, type of firing, type and composition of coal
fired, and other factors.
Figures 4—2 and 4—3 show the overall relationship between N0
emissions and excess air level (% 02 in flue gas), on a consistent basis
for normal firing and modified firing operation for the seven coal fired
boilers tested in this program.
Figure 4—2 is a plot of “normalized” N0 emissions expressed as
the fraction of baseline NO emissions (full load and 20% excess air) as a
function of the average 02 concentration measured In the flue gas for normal
firing conditions. The solid lines shown for each boiler are based on
the least—squares, linear regression analysis of all full load test runs
made under normal firing operation (all burners firing coal and with
closed overf ire air ports). With the exception of the Comanche boiler
No. 1 mentioned above and Barry No. 2, all of the lines fit within a
relatively narrow band.
Figure 4—3 is a plot of “normalized” NO emissions (expressed
as the fraction of baseline NO emissions at full load and 20% excess
air) vs the average 02 concentration measured in the flue gas under modi-
fied combustion operating conditions. Thus, the ordinates are identical
in Figures 4—2 and 4—3. However, the least squares regression lines of
Figure 4—3 do not necessarily pass through the normalized NO value of
100% at 3.6% 02, as they must, by definition, in Figure 4—2.
Figure 4—3 indicates the strong effect of lowering the excess
air on NO emissions when boilers are operated in the staged combustion
mode. The slopes of these lines are very similar except for Morgantown
boiler No. 1 (which was fired with a mixture of coal and oil) and Barry
No. 2 unit (which has overfire air ports located relatively far from the
top row of active burners). It should be noted that all of the boilers
tested had baseline NO emission rates higher than those normalized to
20% excess air (used for constructing Figure 4—3), and that optimized,
modified firing operation generally produced lower NO operation than
shown by the least squares lines. Thus, the improvement resulting from
modified firing operation compared to actual baseline N0 emission levels
averaged 38% instead of 26% based on the data in Figure 4—3. Mercer
boiler No. 1 (lIE), the only wet bottom boiler tested in this program,
had unusually high baseline N0 emission levels. This is reflected by
the large reduction obtained under 80% load, modified firing operating
conditions. An additional point to recall again is that three of the
tangentially fired boilers are equipped with overf ire air ports, which
res ilted in an average 18% reduction normalized baseline NO emissions
compared to the results for the tangentially fired boilers tested pre-
viously (2). This reduction in baseline NO emissions is due to the use
of cooling air through the overfire air ports.
102
-------
140
1 .
C ,)
U)
w
w
0
c..1
I -I W
0
‘.3
120 —
100
80 —
60
40
E
FRONT OR REAR WALL FIRE)
OPPOSE) WALL FIRED
TANGENTIALLY FIRED
I I I• I
2 3 4 5
AVERAGE % OXYGEN IN FLUE GAS
Figure 4—2. Effect of excess air leve]. on N0 emissions
for coal fired boilers under normal firing operations.
0
-------
120
100
em
.
d i
C l )
80-
.r-l
aj 60
C l )
S
Cl 1
0
40 -
I -----
2
3
5
6
Opposed Wall Fired
/ \ Tangential Fired
(a) 45% Coal — 55% Oil Mixture
(b) Wet Bottom Boiler — 80% Load
Average % Oxygen in Flue Gas
Figure 4—3. Effect of excess air on NO emissions of
coal fired boilers under modified firing conditions.
(b) ( J Front or Rear Wall Fired
4
I I
7
-------
The results obtained on mixed fuel fired boilers indicate that
NO emission levels increased as the fraction of coal in coal—oil and
coal—gas mixtures increases. However, the relationship between NO ends—
sions and fuel coal content is not linear. Additional test data Je re-
quired to establish a more reliable correlation for predicting NO emis-
sions from mixed fuel fired units than the single assumption of l!near
dependence.
A large 50 MWe General Electric gas turbine was tested under
normal operating conditions while firing fuel oil. NO emissions were
about 219 ng/J (375 ppm) at full load, 234 ng/J (400 p m) at peak load
(54 ?tWe) with reductions to about 190 and 146 ng/J (325 and 250 ppm) at
50% and 20% of full load operation, respectively. Thus, in the absence
of NO control modifications, load reduction was found to exhibit a
strong influence on NO emissions.
‘ C
This section can be summerized as follows:
• Conventional wall and tangentially fired boilers that were not designed
with special NO reduction features did not meet the NSPS standard of
0.7 lb/b 6 Bit tor NOx emissions under full load, baseline operation.
• Wall and tangentially fired boilers designed with NOx reduction features
such as low turbulence, dual register burners or overf ire air ports were
capable of meeting the NSPS for NOx emissions under baseline operations.
• A wall—fired boiler (Gaston No. 1) retrofitted with Babcock and Wilcox,
dual register burners emitted substantially less N0 (377 ppm vs 591 ppm
or —36%) than a similar boiler equipped with conventional burners.
• Three tangentially fired boilers equipped with overfire air ports were
capable of reducing N0 emissions by 37 to 45% under low excess air,
staged firing operation (282 ppm vs 492 ppm, 261 ppm vs 417 ppm and
198 ppm from 341 ppm).
• Low excess air, staged firing operation on conventional wall fired units
reduced NO emissions significantly on both dry and wet bottom units.
The dry bottom unit met the NSPS for NO emissions at full load under
this operation while the wet bottom unit (with extremely igh baseline
emissions) was forced to reduce load to meet the 0. llb/lO BTU NO emission
standard for new boilers.
• NO emissions increased from mixed—fuel fired boilers as the fraction of
coh in coal—oil and coal—gas mixtures increased. However, the relation-
ship between NO emissions and fuel coal content was not linear in the
two units teste to date.
• The firing pattern used in low excess air, staged firing operation in
boilers without overfire air ports, had a significant influence on the
level of NO reduction obtained. Firing patterns utilizing top burners
on air on1y were nore effective than firing patterns with middle or
lower row burners on air only. In addition, the use of maximum second-
ary air register openings on air only burners while simultaneously re-
ducing air register openings on operating burners had a beneficial effect
on the level of NO emissions.
x
105
-------
4.2 SIDE EFFECTS OF COMBUSTION MODIFICATIONS
As discussed in Section 3, the modified combustion techniques
studied for controlling NO emissions did not produce major, and for the
most part, did not produce significant adverse side—effects in short—term,
300—hour test runs. Thus, particulate mass loading and size distribution,
carbon on fly ash, and boiler efficiency have not changed significantly when
comparing low NO modified combustion operation with baseline conditions.
Long—term testing to determine furnace tubewall corrosion rates by actual
tube wall thickness measurements is required to establish whether fuel rich,
staged firing of different coal types used In different boiler design types
results in undue tubewall corrosion rates.
106
-------
5. REFERENCES
1. W. Bartok, A. R. Crawford and C. J. Piegarl, “Systematic Field
Study of N0 Emission Control Methods for Utility Boilers,” Esso
Research and Engineering Company Final Report No. CRU.4G.NOS.71,
NTIS No. PB 210—739, December 1971.
2. A. R. Crawford, E. H. Manny and W. Bartok, “Field Testing:
Application of Combustion Modifications to Control NO Emissions
from Utility Boilers,” Exxon Research and Engineering Company,
EPA Report No. EPA—650/2—74—066, NTIS No. PB 237—344/AS, June 1974.
3. A. R. Crawford, M. W. Gregory, E. H. Manny, and W. Bartok, “Determination
of the Magnitude of SO 2 , NO, CO 2 and 02 Stratification in the Ducting
of Fossil Fuel Power Plants,” Exxon Research and Engineering Company,
EPA Report No. EPA—600/2—75—053, September 1975, NTIS No. PB 252—565/AS.
4. EnvIronmental Protection Agency, Standards of Performance for New
Stationary Sources,” Method 5, Published in the Federal Register,
December 23, 1971, Vol. 36, Number 247, p. 24888.
107
-------
ACKNOWLEDGMENTS
The authors wish to acknowledge the constructive participation
of Mr. R. E. Hall, EPA Project Officer and Mr. R. C. Carr of the
Electric Power Research Institute, co—sponsors of this program, In
planning the field test programs and providing coordination with boiler
operators and manufacturers. The assistance and cooperation of the
General Electric Company and Westinghouse Electric Company personnel in
helping the selection of gas turbines for testing are also gratefully
acknowledged. The helpful cooperation, participation and advice of Babcock
and Wilcox, Combustion Engineering and Foster Wheeler were essential in
selecting representative boilers for field testing and conducting the
program. The voluntary participation of electric utility boiler operators
in making their boilers available is gratefully acknowledged. These
boiler operators include the Southern E lectric Generating Company, the
Alabama Power Company, the Tennessee Valley Authority, the Potomac Electric
Company, the Salt River Project and the Public Service Company of Colorado.
The authors also express their appreciation for the extensive coal analyses
services provided by Exxon Research’s Coal Analysis Laboratory at Baytown,
Texas and to Messrs. A. A. Ubbens and E. C. Winegartner for their contri-
butions and advice on coal related matters. The valuable assistance of
Messrs. L. W. Blanken, R. W. Schroeder, W. Petuchovas, and Mrs. M. V.
Thompson in performing these field studies Is also acknowledged.
LEGAL NOTICE
This report was prepared by Exxon Research and Engineering
Company as an account of work sponsored by th Electric Power Research
Institute, Inc. (EPRI), and the United States Environmental Protection
Agency (EPA). Neither EPRI, members of EPRI, EPA nor Exxon Research and
Engineering Company, nor any person acting on behalf of them:
a. Makes any warranty or representation, express or implied,
with respect to the accuracy, completeness, or usefulness of the
Information contained in this report, or that the use of any information,
apparatus, method, or process disclosed in this report may not infringe
privately owned rights; or
b. Assumes any liabilities with respect to the use of, or
for damages resulting from the use of, any information, apparatus,
method or process disclosed in this report.
108
-------
APPENDIX A
CROSS SECTIONAL DRAWINGS OF TYPICAL UTILITY BOILERS
Typical utility boiler designs representative of the types
of boilers tested in this program are shown in the cross sectional
drawings in Figures 1 through 3 of Appendix A. Typical front wall,
horjzontally opposed and tangentially fired boilers are shown in
Figures 1, 2 and 3, respectively.
109
-------
APPENDIX A
FIGURE 1
TYPICAL FRONT WALL FIRED BOILER
DRAWING FURNISHED THROUGH THE COURTESY OF
THE BABCOCK AND WILCOX COMPANY
110
10 P tCSPtTATOm
* 1 10
iwouao-o UT N
._—2?.O” - 32.O”
•32’.O” L 30’-O” -j
-------
APPENDIX A
FIGURE 2
TYPICAL HORIZONTALLY OPPOSED FIRED BOILER
DRAWING FURNISHED THROUGH THE COURTESY OF
THE BABCOCK AND WILCOX COMPANY
111
—32 ’-0” 4 - 3-0” 35-0’ 29-0” 4 - 25-0” - I 30-0” I 25-0” 4 - 39-6”
-------
APPENDIX A
FIGURE 3
TYPICAL TANGENTIALLY FIRED BOILER
DRAWING FURNISHED THROUGH THE COURTESY OF
COMBUSTION ENGINEERING, INC.
112
-------
APPENDIX B
CONVERSION FACTORS
ENGLISH TO SI UNITS
To Convert Prom To Multiply By*
Btu Gigajoule, GJ 1.055 056 E—O6
Btu/h Megawatt, MW 6.330 336 E—02
Btu/lb Gigajoule/kilogram, GJ/kg 2.326 000 E—06
Btu/lb Calories/gram, cal/g 5.559 273 E—Ol
Pounds/square inch Megapascal, MPa 6.894 757 E—03
Kilopascal, kPa 6.894 757 E+O0
Pounds/hour, lb/h Kilogram/hour, kg/h 4.535 924 E—01
Pounds/hour, lb/h Metric tons/hour, t/h 4.535 924 E—04
Tons (short) Metric tons, t 9.071 847 E—0l
Pounds, lb Kilogram, kg 4.535 924 E—01
Feet, ft Metre, m 3.048 000 E—O1
Square Feet, ft 2 Metres 2 , m 2 9.290 304 E—O2
Inches, in Centimetres, cm 2.540 000 E+OO
Inches, in Metre, m 2.540 000 E—O2
Grains/cubic foot Mil higrams/metre 3 ,mg/m 3 2.288 351 E—Ol
Cubic feet, ft 3 Metre 3 , m 3 2.831 685 E—02
Decimetre 3 , dm 3 2.831 685 E+0l
Gallons Metres 3 , m 3 3.785 412 E—O3
Decimetre 3 , dm 3 3.785 412 E+OO
*Numbers expressed in power—of—ten notation using B ± xx to indicate
the power of ten. For example:
1.055 056 E—06 = 1.055 056 x l0_6 = .000001055056
113
-------
APPENDIX B
CONVERSION FACTORS
English and Metric Units to SI Units
Multiply Concentration
To Obtain From Multiply By To Obtain ng/J of in ppm at 3% O by
ngIJ lb/MBtu 430 Natural Gas Fuel
ng/J g/MCa1 239 CO 0.310
GJ.hrkm 2 MBH/ft 2 11.356 MC 0.177
GJ.hr 1 .m 3 NE l l/ft 3 37.257 NO or NO (as equivalent NO 2 ) 0.510
GJ/hr lO lb/hr* 1.055 SO or SO 0.709
or iO 6 Btu/hr 2 x
Oil Fuel
in ft 0.3048
CO 0.341
cm in 2.54
2 2 HC 0.195
i n ft 0.0929
3 3 NO or NO (as equivalent NO 2 ) 0.561
i n ft 0.02832 ‘ C
SO or 50 0.780
Kg lb 0.4536 2 x
Coal Fuel
Celsius Fahrenheit t = 5/9 (t —32)
C CO 0.372
Kelvin tK = 519 (tF_ 3 2) + 273
BC 0.213
Pa psig P = (P + 14.7) (6.895X10 )
pa g NO or NO (as equivalent NO 2 ) 0.584
Pa psia P = (P )(6.895Xl0 ) ‘ C
pa pssa SO or SO 0.850
Pa iwg (39.2°F) pa iwg 2 4 2 x
*lb/hr of equivalent saturated steam
-------
APPENDIX B
CONVERSION FACTORS
SI Units to Metric or English Units
To Obtain ppm Multiply Concentration
To Obtain From Multiply By at 3% 0 of in ng/J by
g/Mcal ng/J 0.004186 Natural Gas Fuel
1o 6 Btu GJ 0.948 CO 3.23
MBH/ft 2 GJ.hr 1 .m 2 0.08806 HC 5.65
MBH/ft 3 GJ .hr 1 .m 3 0.02684 NO or NO 1.96
Btu gin cal 3.9685 x l0 SO 2 or so 1.41
lO lb/hr* or MBH GJ/hr 0.948 Oil Fuel
lb/MBtu ng/J 0.00233 Co 2.93
ft m 3.281 NC 5.13
in cm 0.3937 NO or NO 1.78
ft 2 in 2 10.764 50 or SO 1.28
ft 3 in 3 35.314 Coal Fuel
lb Kg 2.205 CO 2.69
Fahrenheit Celsius tF = 9/5(t)+32 NC 4.69
Kelvin t = 1.8K — 460 NO or NO 1.71
psig Pa psig = (Ppa)(l•450XlO 4 )_14•7 SO 2 or SO 1.18
psia Pa psia = (Ppa)(1•450XlO 4 )
iwg (39.2°F) Pa 1 ’iwg = pa40l4X1 3 )
*lb/hr of equivalent saturated steam
-------
APPENDIX C
OPERATING AND GASEOUS
EMISSION DATA SUMMARIES
This section of the report contains 7 tables summarizing the
operating and gaseous emission data by test run for each of the 7 coal
fired boilers and a gas turbine tested.
Table Boiler
1 Widows Creek No. 5
2 Ernest C. Gaston No. 1 and No. 2
3 Navajo No. 2
4 Comanche No. 1
5 Barry No. 2
6 Morgantown No. 1
7 Mercer No. 1
8 Morgantown Gas Turbine No. 3
Percent oxygen and % carbon dioxide are listed on an as
measured basis. Carbon monoxide and gaseous hydrocarbons are listed
on a ppm (3Z 02, dry) basis. NO emissions are listed on a ppm (3%
02, dry) basis, pounds per i06 Btu and nan ograms per joule bases. The
conversion factor used for adjusting ppm (3% 0 , dry) to pounds per
106 Btu is 0.00136. The conversion factor for converting pounds NO
per 106 Btu to ng/J is 430.
116
-------
APPENDIX C
TABLE 1
SU}8IAET OF OPERATING AND EMISSION DATA —
WIDOWS CREEK. BOILER NO. S
(125 MWe, Front Wall, Pulverized Coal Fired)
Date and
Run No.
(1975)
Operating_Conditions
Gross Firing Pattern- Sec. Air Excess 2 Stoich. Air
Losd (Burners on Registers Air to Active
S!! f L Air Only)(a) (2 Open) l.evel Burners (b)
2z_
NO
Gas
Measurements
.Sz_
CO
Temp.
2
PPM
(32 02,
Dry)
Pounds
Per 106
BTIJ
Nanogram
Per
Joule
2
PPM
(32 02,
Dry)
Kelvin
9/25—1
2
3
4
105
118
120
121
5 1 —None
5 1 —None
5 1 —None
5 1 —None
60
60
20
20
Normal
Low
Normal
Low
134
117
119
117
5.4
3.2
3.5
3.2
603
502
574
548
0.82
0.68
D.78
0.75
353
294
336
320
12.4
14.6
14.1
14.2
29
403
54
62
661
630
651
743
9/261A
5
6
7
8
121
121
117
119
120
SrNone
5 2 —(AjA 4 )
5 3 —(C 1 C 4 )
54—(B1B4)
5 5 —(D 1 D 4 )
60
60
60
20
20
Normal
Low
Low
Low
Low
123
101
113
105
105
4.0
3.0
4.8
3.6
3.6
597
486
639
517
580
0.81
0.66
0.87
0.70
0.79
349
284
374
302
339
14.0
14.0
14.0
14.7
14.3
43
124
35
317
31
648
644
753
646
647
9/27—9
10
11
12
115
110
112
120
5 5 —(D 1 D 4 )
5 3 —(C 1 C 4 )
5 2 —(A 1 A 4 )
60
60
20
20
Low
Low
Low
Low
104
108
108
102
3.4
4.0
4.1
3.2
598
521
474
468
0.81
0.71
0.64
0.64
350
305
277
274
14.5
13.5
13.2
14.6
52
74
37
88
647
644
647
645
10/2—13
14
15
16
102
98
98
97
5 1 —None
5 1 —None
5 1 —None
5 1 —None
60
60
20
20
Normal
Low
Normal
law
135
116
134
110
5.5
3.0
5.4
2.1
660
462
667
381
0.90
0.63
0.91
0.52
386
270
390
223
12.8
14.5
13.0
15.6
33
63
18
563
650
627
643
621
10/3—17
lB
19
20
99
100
99
100
5 6 —(A 1 A 4 D 1 D 4 )
5 7 —(A 1 A 4 B 2 B 3 )
5 8 —(A A 4 B 1 B 4 )
Sg—(4A 2 A 3 A 4 )
60
60
20
20
law
Low
Low
Low
88
87
88
100
3.2
3.0
3.3
5.3
365
374
316
305
0.50
0.51
0.43
0.41
213
219
184
178
14.3
14.4
14.5
13.2
283
208
106
503
628
627
633
615
10/4—21
22
23
24
99
100
100
101
5 9 —(A 1 A 4 A 3 A 2 )
5 8 —(A 1 A 4 B B 4 )
S 6 —(A 1 A 4 D1D 4 )
5 7 —(A 1 A 4 B 2 B 3 )
60
60
20
20
Low
Low
Low
law
94
94
93
88
4.5
4.4
4.3
3.3
329
368
421
302
0.65
0.50
0.57
0.41
192
215
246
176
13.2
13.3
13.6
13.8
155
82
63
103
632
628
635
628
10/7—lB
125
5 1 —None
60
Low
112
2.4
409
0.56
239
15.5
710
636
10/8—ic
125
5 1 —None
60
Low
113
2.5
441
0.60
258
14.6
618
640
10/14—128
124
S2—(Al4)
20
Low
97
2.2
465
0.63
272
16.0
B7
640
l0/15—12C
125
5 2 —(A 1 A. 4 )
20
Low
99
2.5
412
0.56
241
15.5
214
642
l0/17—12E
108
5 2 —(A 1 A, 4 )
20
Low
105
3.6
561
0.76
328
14.4
78
635
10,22—12P
132
S 2 —(A 1 A 4 )
20
Low
97
2.2
504
0.69
298
15.5
429
675
(a) Four rows of four burners each; A row on top to 0 row at bottom.
(b) Based on aaaumption that equal amounts of air flow through all active and inactive burners.
(c) NC were not measured due to instrument problems.
ii i
-------
APPENDIX C
TABLE 2
SUMMARY OF OPERATING AND EMISSION DATA -
ERNEST C. GASTON, BOILERS NO. 1 AND 2
(270 MWe, Horizontal Opposed, Pulverized Coal Fired)
Date and
Run No.
(1975—
1976)
— ODerating Conditions
Air Register
Gross Firing Pattern — S ’
Load (Burners on Vanes
Air Only) Sec. Tert ! L
Flue Gas Measurements
Excess
Air
Level
B Stoich. Air
to Active
Burners(d)
NO,,
9L
CO
SC
Temp.
B
PPM
(3% 07,
Dry)
Pounds
Per 106
8Th
Nanogram
Per
Joule
B
PPM
(3% 02,
Dry)
-
PPM
(3% 02,
Dry)( )
Kolvin
11/14—1
2
3
270
270
270
5 1 —None
S 1 —None
S 1 —None
70
70
50
100
100
100
90
90
100
Normal
Low
Low
125
112
112
4.3
2.4
2.4
389
278
289
0.53
0.38
0.39
227
163
169
13.9
14.9
15.5
26
65
46
——
——
——
688
686
684
11/15-4
5
6
7
270
270
270
270
S 1 —None
5 —None
S 1 —None
S 1 —None
30
100
50
50
100
100
100
50
100
100
90
90
Low
Low
Low
Low
114
112
112
113
2.6
2.4
2.3
2.5
329
336
318
362
0.45
0.46
0.43
0.49
192
196
186
212
14.7
15.1
15.5
15.0
55
63
79
36
——
——
——
——
699
683
684
683
11/18—8
9
270
270
S 1 —None
S 1 —None
30
100
50
50
90
90
Low
Low
111
110
2.2
2.0
312
296
0.42
0.40
188
173
14.8
15.3
44
59
——
642
684
11/20—10(a)
270
S 1 —None
Normal
125
4.3
600
0.82
351
14.2
32
——
810
11/22—11
12
13
14
15
16
230
250
195
190
187
236
S2(E Mill)
S 2 -(E Miii)
S 3 —(E,B Mills)
S 3 —(E,B Mills)
S 3 —(E.B Mills)
S 4 —(B Mill)
70
70
70
70
70
70
100
100
100
100
70
100
90
90
90
90
90
90
Normal
Low
Normal
Low
Low
Low
100
95
80
72
72
91
3.6
2.6
3.6
1.6
1.8
1.9
289
257
240
182
200
240
0.39
0.35
0.33
0.25
0.27
0.33
169
150
140
106
117
140
14.0
14.5
13.9
14.8
14.8
15.3
45
178
39
329
245
162
——
——
——
——
——
——
679
676
668
658
658
669
12/9—17(b)
18(b)
19(b)
20(b)
204
206
206
204
S 1 —None
S 1 —None
S 1 —None
S 1 —None
70
70
30
30
100
100
100
100
90
90
90
90
Normal
Maximum
Maximum
Normal
129
149
147
122
4.9
7.0
6.8
3.9
325
454
451
295
0.44
0.62
0.61
0.40
190
265
264
172
14.0
11.9
11.1
13.6
21
21
25
28
——
——
——
——
——
——
——
——
1/15—21
22
23
24
273
270
270
270
S 1 —None
S 1 —None
S 1 —None
S 1 —None
30
50
70
100
100
100
100
100
90
90
90
90
Normal
Normal
Normal
Normal
120
120
122
120
3.7
3.6
3.9
3.7
377
409
356
349
0.51
0.56
0.68
0.47
220
239
208
206
14.1
14.8
14.5
14.8
36
32
31
33
0
—
——
2
——
——
——
——
1/16—25
268
S 1 —None
70
100
90
Normal
121
3.8
390
0.53
228
15.8
——
2
——
1/17—26
270
S 1 —None
70
100
90
Normal
122
3.9
349
0.47
204
14.0
39
—
663
1/20—27
268
S 1 —None
70
100
90
Normal
122
39
316
0.43
185
14.3
34
2
639
1/22—28(a)
272
S 1 —None
——
——
——
Normal
124
4.2
584
0.79
341
15.2
——
17
——
1/23—29(a)
268
S 1 —None
——
——
——
Normal
123
4.1
572
0.78
334
15.2
20
7
——
1/29—30(a)
270
S 1 —None
——
——
——
Normal
124
4.3
606
0.82
354
14.8
21
2
——
2/3—34
2/4—35
148
269
208
243
S 3 —(E,8 Mills)
S 1 —None
70
70
100
100
90
90
Normal
Normal
97
125
6.9
4.4
148
372
0.20
0.51
87
217
12.6
13.5
43
22
3
1
626
631
2/5—36
37
S4—(B Mill)
S 1 —None
70
70
100
100
90
90
Normal
Normal
101
122
3.8
3.9
301
324
0.41
0.44
176
189
15.1
15.9
27
49
2
2
621
629
(a) Boiler No. 2 data
(b) Load limited to 205 MWe due to excess turbine vibration.
(c) 50% — No spin, 100% — maximum counterclockwise spin of secondary air.
(d) Based on assumption that equal amounts of air flow through all active and inactive burners.
(e) Hydrocarbons were measured on each teat but values were negligible except where indicated.
118
-------
APPENDIX C
TABLE 3
SUMMARY OF OPERATING AND ENISSION DATA — NAVAJO, BOILER NO. 2
(800 MWe, Tangential. Pulverized Coal Fired)
Date end
Run No.
(1975)
Operating Conditione
Groee Exceee Firing Pattern
Load Air OverE ’ ’ Air No. Active Burner
Level Symbol( 2 Open Tilt Burners Tilt
:
2
NO 5
Flue Gas
Heasurem
ante
CO
MC
Temp.
PPM
(32 02,
Dry)
Pounds
Per 106
BTU
Nanogram
Per
Joule
B
PPM
(3% 02,
Dry)
PPM
(3% 02,
Dry)
Kelvin
5/29—3
803
Normal
S 1
10
Mont.
56
+10’
4.9
366
0.50
214
13.5
34
1.
666
5/30—6
4
5
6A
3A
1
802
801
797
800
799
798
Low
Low
Low
Low
Normal
Normal
5
S 1
S
S
S
S
10
10
10
10
10
10
)Ioniz.
Roriz.
Horiz.
Maria.
Horiz.
Morlz.
56
56
56
56
56
56
+10’
Horiz.
—15’
+25’
+25’
Horiz.
4.0
4.2
3.8
3.8
4.8
4.9
335
309
314
411
492
343
0.46
0.42
0.43
0.56
0.67
0.47
196
181
184
240
288
201
14.0
13.5
14.5
14.8
14.0
13.7
70
94
137
288
64
54
3
4
3
3
3
4
658
656
658
665
666
666
6/2—11
13
15
17
795
795
795
795
Normal
Normal
Normal
Normal
S 4
S 5
6
S 7
25
50
75
100
+1.0°
+10’
+10’
+10°
56
56
5
56
+10’
+10’
+10’
+10°
4.0
3 9
3.8
3.6
332
310
293
288
0.45
0.42
0.40
0.39
194
181
171
168
13.4
13.3
1.3.3
13.2
94
83
92
112
4
3
3
2
652
652
652
649
6/3—18
16
14
12
802
798
799
798
Low
Low
Low
1.0w
S 7
6
55
S 4
100
75
50
25
+10°
+10°
+10°
+10°
56
56
56
56
+10°
+10°
+10°
+10°
3.6
4.0
3.6
3.6
282
289
326
328
0.38
0.39
0.44
0.45
165
169
191
192
13.4
12.7
14.2
15.0
339
261
309
253
—
—
—
—
655
655
653
658
6/4—lA
800
Normal
S
10
+10’
56
+10°
5.2
378
0.51
221
12.9
108
—
674
6/5—8
8A
7
73
21
19
20
22
800
800
800
800
798
795
798
795
Low
Low
Normal
Normal
Normal
Normal
Low
Low
S 2
l
S
l
57
37
S 7
S 7
10
10
10
10
100
100
100
100
+10°
+10°
+10°
+10°
Mona.
Mona.
Horiz.
Horiz.
48
48
48
56
56
56
56
56
+10°
+10°
+10’
+10°
Moniz.
—15°
—15
Horiz.
3.6
3.6
5.0
5.2
5.4
5.2
3.5
3.8
259
298
318
383
329
346
271
289
0.35
0.41
0.63
0.52
0.45
0.47
0.37
0.39
151
174
186
224
192
202
158
169
15.2
15.7
13.8
14.0
13.5
13.8
15.7
15.4
549
97
124
83
96
85
160
256
—
—
—
—
—
—
—
—
652
649
650
664
669
663
658
658
6/6—33
hA
18A
802
806
802
Normal
Normal
Low
S 1
S 7
57
10
100
100
+10°
+10°
+10°
56
56
56
+10’
+10°
+10°
5.4
5.2
4.2
416
354
291
0.57
0.48
0.40
243
207
170
13.5
13.6
15.1
111
109
183
—
-
—
663
663
652
6/18—3C
803
Normal
l
10
+10.
56
+10’
5.4
385
0.52
225
13.6
14
—
663
6/19—188
808
Normal
S 7
100
+10°
56
+12°
4.4
349
0.47
204
14.2
36
—
659
6/20—18C
808
Normal
S 7
100
+10°
56
+11°
4.6
346
0.47
202
14.0
21
—
663
6/24—J.8D
2A
305
792
Low
Normal
S 7
l
100
10
+10°
+10°
32
56
+25°
—10°
46
3.9
349
332
0.47
0.45
204
194
14.9
15.6
12
30
—
—
569
661
6/26—3D
185
23
785
798
565
Normal
Low
Normal
S 1
S 7
l
10
100
10
+10°
+10°
+10’
56
56
48
+10°
+10°
+10°
4.8
3.7
5.3
346
280
350
0.47
0.38
0.48
202
164
205
14.4
15.3
14.3
22
52
10
—
666
659
646
6/27—24 565 Normal S 1 10 +10° 40 +30° 5.0 404 0.55 236 13.1 17 623
(a) Firing Patterns — Normal Firing
— Staged FinIng — Top Row Burners on Air Only
54 — Overf ire Air Ports 23% Open
S — Overf ire Air Porte 50% Open
S 6 — Overfire Air Ports 75% Open
S 7 — Overfire Air Ports 100% Open
119
-------
APPENDIX C
TABLE 4
SUMMARY OF OPERATING AND ISSION DATA — COMANCHE, BOILER NO. 1
(350 MWe, Tangential, Pulverized Coal Fired)
Date and
Run No.
(1975)
Operating Conditions
— Firing Pattern — Secondary
Gross Excess No. of Burners Burner Air Reg.
Load Air Firing Air Over Fire Air Nozzle Aux/Coal
Level Coal 2 Open Tilt (2 Open)
— Flue Gas Measurements
NO
CO
HC
2
PPM
(3% 02.
Dry)
Pounds
Per 10
BTU
Isnograt
Per
Joule
B
PPM
(3% 02,
Dry)
PPM
(3% 02.
Dry)
Kelvin
7/14—1
3
5
7
340
334
340
340
Normal
Normal
Normal
Normal
S 1 —20
S 1 —20
S 1 —20
S 2 —20
None
None
None
None
Closed
Closed
Closed
102
Horiz.
Horiz.
Horiz.
Horiz.
—14’
—26’
Horiz.
—16’
40/100
40/100
40/100
37/100
3 9
4.0
3.7
3.8
391
448
336
373
0.53
0.61
0.46
0.51
229
262
196
218
15.6
15.1
15.7
15.8
22
19
33
28
2
1
1
1
796
796
800
800
7/15—9
11
13
15
335
335
333
332
Normal
Normal
Normal
Normal
S 3 —20
S 4 —20
S 5 —20
S 6 —20
None
None
None
None
252
502
752
100%
—15’
—15’
—15’
—15’
—17’
—17°
—17’
—18°
35/100
30/100
27.5/100
26.5/100
4.1
4.2
3.7
3.5
355
306
261
266
0.48
0.42
0.35
0.36
208
179
153
156
15.4
15.4
15.8
15.8
31
28
37
48
1
1
1
1
——
——
7/16—2
4
17
18
332
331
331
332
Normsl
Normal
Normal
Normal
S 1 —20
S 1 —20
S 3 —20
S 3 —20
None
None
None
None
Closed
Closed
25%
252
—15’
—15’
—15’
—15’
—16’
—18’
—26’
Horiz.
38/100
40/100
36/100
36/100
3.8
4.0
4.0
4 0
404
417
383
308
0.55
0.57
0.52
0.42
236
244
224
180
15.9
15.8
15.7
15.5
31
27
38
98
0
0
0
0
796
805
808
——
7/17—6
335
Normal
S 1 —20
None
Closed
—15’
—16’
38/100
3.7
405
0.55
237
15.6
35
0
800
7/18—20
8
21
19
322
323
321
322
Normal
Normal
Normal
Normal
S 3 —20
S 1 —2O
S 3 —20
S 3 —20
None
None
None
None
252
Closed
25%
25%
Horiz.
Horiz.
Horiz.
Horiz.
—17’
—20’
—23’
—4’
29/100
36/100
33/100
32/100
3.9
3.8
3.9
4.0
362
428
364
332
0.49
0.58
0.50
0 45
212
250
213
194
15.5
15.6
15.2
15.4
55
52
50
53
0
0
0
0
795
802
803
798
7/21—22
23
24
322
321
316
Normal
Normal
Normal
S 4 —20
S 4 —20
S 4 —20
None
None
Wane
50%
50%
50%
Horiz.
Hong.
Non,.
Hong.
—15’
—25’
26/100
22/100
22/100
4.0
4.2
4.4
280
316
326
0.38
0.43
0.44
164
185
191
15.5
15.2
15.0
169
40
42
0
0
0
785
785
773
7/22—25
26
322
321
Normal
Normal
S 4 —20
S 4 —20
None
None
50%
50%
Honiz.
Honiz.
—15°
—18’
22/100
24/100
4.2
4.1
268
284
0.36
0.39
137
166
15.1
15.5
35
25
0
0
778
785
7/23—27
321
Normal
S 4 —20
None
50%
Honiz.
—20’
23/100
3.8
245
0.33
143
15.1
36
0
791
7/24—28
322
Normal
S 4 —20
None
50%
Honiz.
—21’
22/100
3.8
259
0.35
151
15.6
38
0
794
7/25—29
321
Normal
S 4 —20
None
50%
Horiz
—19’
24/100
4.4
289
0.39
169
15.2
57
0
800
7/29—30
31
325
322
Normal
Normal
S 1 —20
S 4 —20
None
None
Closed
50%
Horiz.
Won,
—20’
—20’
37/100
27/100
3.7
3.9
380
271
0.52
0.37
222
158
15.5
15.8
85
112
2
2
809
808
7/30—32A
328
33
323
324
323
Normal
Normal
Normal
S 4 —20
S 4 —20
S 1 —20
None
None
None
502
50%
Closed
Hong.
Hong.
Horiz.
—15’
—15’
—15’
26/100
26/100
37/100
3.9
3.9
3.9
273
275
389
0.37
0.37
0.53
160
161
227
15.2
15.7
15.8
96
70
77
2
2
2
800
——
——
8/1—34
330
Normal
S 4 —20
None
50%
Hone.
—7°
24/100
3.4
200
0.27
117
16.0
39
1
789
120
-------
APPENDIX C
TABLE 5
SU)0(ARY OF OPERATING AND ENISSION DATA — BARRY. BOILER NO. 2
(130 MWe, Tangential, Pulverized Coal Fired)
Date and
Run No.
(1975)
Operating Conditions
Flue Gas Men
nuremen
1
Gross
Load
Excess
Air
Level
— Firing Pattern
Burners
Mills No. on No. on OFA Ports
Oj. L Coal Air B Open(b’
Burner
Tilt
Secondary
Air Reg.
B Open
Conl/Aux.
Fuel Fired
NOw 1C02
CO
NC
i o
Opacity
B
PPM
(3% 02,
• _
Pounds Nanogram
Per 106 Per
BTU joule 2
PPM
(3% 02,
Dry)
PPM
(3% 02.
lialvir
2/12—1
2
3
4
5
6
131
130
128
129
129
126
Normal
Normal
Normal
Low
Low
Low
S 1 —None
S 2 —None
S 3 —jlone
S 3 —None
S 2 —None
S 1 —None
16
16
16
16
16
16
0
0
0
0
0
0
5
50
100
100
50
5
—17
—17
—17
—17
—17
—17
301100
30/100
30/100
30/100
30/100
30/100
Aleb. Coal
Alab. Coal
Alab. Coal
Alab. Coal
Alab. Coal
Alab. Coal
4.9
4.8
4.7
3.8
3.5
3.1
365
296
265
189
182
196
0.50
0.40
0.36
0.26
0 25
0.27
213
172
155
111
106
115
15.0
14.7
14.7
16.0
16.4
16.9
24
25
22
49
227
783
1
1
1
3
2
1
576
583
583
579
574
571
——
——
——
55—60
50—70
55—70
2/13—25
7
8
11
12
9
10
26
95
93
93
93
95
94
95
130
Normal
Normal
Low
Normal
Low
Normal
Low
Normal
S 1 —None
S 2 —None
S 2 —None
S 5 —A’
S 5 —’A”
S 4 —”A ”
S 4 —’A”
S 1 —None
16
16
16
12
12
12
12
16
0
0
0
4
4
4
4
0
5
50
50
50
50
5
5
5
—2
—2
—2
—2
—2
—2
—2
—2
30/100
30/100
30/100
30/100
30/100
30/100
30/100
30/100
Ill. Coal
Ill. Coal
111. Coal
Ill. Coal
Ill. Coal
Ill. Coal
111. Coal
Ill. Coal
7.3
7.1
4.8
6.8
4.3
7.1
4.6
5.4
452
411
253
272
192
330
212
345
0.61
0.56
0.34
0.37
0.26
0.45
0.29
0.47
264
240
148
159
112
193
124
202
11.7
11.7
13.5
11.9
16.6
12.6
15.5
15.0
22
21
60
22
48
18
203
22
2
6
3
4
3
7
3
2
569
569
556
566
555
566
555
582
50—60
50—60
60—75
60
60—70
50—60
60—70
60—75
2/19—21
22
23
24
27
28
29
30
130
128
132
130
133
130
130
128
Normal
Normal
Normal
Low
Normal
Normal
Low
Normal
S 1 —’A”
S 2 —’A”
S 3 —’A”
5 3 — ’A”
5 1 —C’
S 2 — ’C ”
S 3 —C
5 3 —”C”
12
12
12
12
12
12
12
12
0
0
0
0
0
0
0
0
5
50
100
100
5
50
100
100
—2
—2
+1
+1
+1
+1
+1
+1
30/100
30/100
30/100
30/100
30/1.00
30/100
30/100
30/100
Kty.+.4 Gas
Kty.+.4 Gas
Kty.+.4 Gas
Kty.+.4 Gas
Kty.+.4 Gas
Kty.+.4 Gas
Kty.+.4 Gas
Kty.+.4 Gas
3.9
4.6
4.9
3.7
4.5
5.1
5.0
4.9
168
184
182
143
206
186
184
190
0.23
0.25
0.25
0.19
0.28
0.25
0.25
0.26
98
108
106
84
120
109
108
111
13.2
12.5
12.0
13.2
13.2
12.5
12.6
12.6
743
30
18
389
210
65
78
47
3
2
3
2
6
4
3
3
585
586
586
584
586
586
585
587
60
55—70
55—65
50—60
50—60
60
60
60
2/20—17
18
19
20
15
16
13
14
131
128
132
134
58
56
54
55
Low
Low
Low
Low
Normal
Low
Normal
Low
S 1 —None
S 1 —None
S 2 —None
S 3 —None
S 1 —ABCD
S 1 —AZCD
S 1 —A,B
S 1 —A,B
16
16
16
16
0
0
8
8
0
0
0
0
0
0
0
0
5
5
50
100
5
5
5
5
+1
+1
+1
+1
+9
+6
+9
+9
30/100
30/100
30/100
30/100
30/100
30/100
20/50
20/50
Kty.+.2 Gas
Kty.+.2 Gas
Xty.+.2 Gas
Kty.+.2 Gas
Gas
Gas
Kty. Coal
Kty. Coal
4.9
6.2
4.8
4.6
9.2
7.3
8.8
6.9
218
198
212
191
122
97
284
254
0.30
0.27
0.29
0.26
0.17
0.13
0.39
0.35
127
116
124
112
71
57
166
149
13.9
14.3
13.9
14.0
7.2
8.3
10.8
11.7
41
547
137
143
10
12
28
26
6
2
2
2
10
5
6
7
587
580
586
585
538
533
537
535
70
75
70
——
——
——
80—100
65—80
2/21—31
32
33
34
35
37
36
38
133
131
135
134
132
128
130
126
Normal
Normal
Normal
Normal
Low
Low
Low
Normal
S 1 —None
5 1 —None
S 1 —None
5 1 —None
S 2 —None
S 2 —None
S 2 —None
S2—None
16
16
16
16
16
16
16
16
0
0
0
0
0
0
0
0
5
5
5
5
50
50
50
50
+1
+20
—20
0
0
—20
+20
0
30/100
30/100
30/100
100/50
100/50
100/50
100/50
30/100
Kty. Coal
Kty. Coal
ICty. Coal
Kty. Coal
Kty. Coal
Kty. Coal
KEy. Coal
Kty. Coal
4.3
4.0
3.9
3.8
4.0
4.0
4.1
4.7
314
320
273
299
233
218
274
243
0.42
0.44
0.37
0.41
0.32
0.30
0.37
0.33
184
187
160
175
136
127
160
142
14.1
13.0
12.8
12.8
13.3
13.5
13.3
12.7
20
70
40
47
58
32
103
28
——
——
——
——
——
——
4
3
588
587
585
586
549
587
590
588
——
——
——
——
55—70
——
——
——
(a) Mills are designated ‘A’ for the mill feeding the top row of burners through “0”, the mill feeding the bottom row of burners.
(b) Overfire air “NO” ports are about 5% open for cooling, when in a “closed” position.
121
-------
APPENDIX C
TABLE 6
OF OPERATING AND EMISSION DATA - MORGANTOWN STATION, BOILER NO. 1
(575 MWe, Pulverized Coal and/or Oil Tangentially Fired)
Date and
Run No.
(1975)
— Operating Conditions —
Gross Excess — Firing Patterr t a) E Stoich. Air
Load Air Mills on No. of Oil Burner Fuel Fired — to Active
Q ) Level Svmb9 Coal Air Burners Tilt 2 Cosl/X Oil Burners(b)
Flue Gas Measu ’ ”senta
Z
PPM
(32 02,
NO x 1C02
Pounds Nanogram
Per 10 Per
HTU Joule 2
. ._.
PPM
(3% 0 ,
PPM
(32 02,
Celvin
3/14—1
2
3
4
5
6
576
576
576
575
576
584
Normal
Low
Normal
Low
Low
Low
S 1
S 1
S 2
S 2
2
2
A,C
A,C
C,E
C,E
C,E
C,E
None
None
A
A
A
A
21
21
21
21
21
21
—2°
—2°
0°
0°
+13°
—12°
22/78
22/78
22/78
22/78
22/78
22/78
125
110
102
94
94
96
4.4
2.1
4.6
3.3
3.4
3.1
458
386
375
310
334
314
0.62
0.52
0.51
0.42
0.45
0.43
268
226
219
181
195
184
13.8
15.5
13.3
13.8
13.1
13.0
21
53
16
39
27
18
2
1
1
1
1
1
——
——
——
——
3/18—7
8
9
576
576
576
Normal
Normal
Normal
S
S 1
S
A,B,C,D
A,B,C,D
A,B,C,D
None
None
None
18
18
18
0°
+15°
—15°
45155
45/55
45/55
126
128
128
4.5
4.7
4.7
525
553
568
0.71
0.75
0.77
307
323
332
13.8
13.2
13.0
15
17
16
2
1
1
——
——
——
3/19—7A
10
11
12
576
576
576
576
Normal
Low
Normal
Low
S
S
S 2
Sz
A,B,C,D
A,B,C,D
B,C,D,E
B,C,D,E
None
None
A
A
17
17
17
16
0°
0°
0°
+1°
49/51
49/51
49/51
49/51
128
115
101
94
4.7
2.8
4.5
3.1
577
502
465
423
0.78
0.68
0.63
0.58
337
293
272
247
13.9
15.4
13.8
15.7
13
18
14
18
0
0
0
0
—
——
3/20—13
378
Normal
S 1
A,B,C,D,E
None
10
+1°
73/27
125
4.4
593
0.81
347
14.9
160
14
3/21—14
15
17
18
572
572
570
572
1.0w
Normal
Normal
Low
S 1
S 2
S
S 1
A,B,C,D,E
B,C,D,E
A,B,C,D,E
A,B,C,D,E
None
A
None
None
10
10
2
2
+2°
+2°
+2°
+2°
72/28
74/26
93/7
9317
119
102
129
121
3.4
4.7
4.9
3 8
511
403
552
549
0.69
0.55
0.75
0.75
299
236
323
321
13.9
14.4
14.0
14.8
25
20
21
30
2
1
1
1
——
623
624
622
3/25—19
20
21
22
25
26
303
300
300
306
301
314
Low
High
High
High
High
Low
S 2
S 2
2
2
S 1
S 1
None
None
a
B.C
,B,C,D
•,B,C,D
A
A
A
A
None
None
16
16
14
8
0
0
0°
0°
0°
0°
0°
0°
0/100
0/100
20/80
50/50
100/0
100/0
105
99
107
107
144
142
5.1
4.2
5.4
5.4
6.5
6.3
165
149
246
300
544
515
0.22
0.20
0.33
0.40
0.74
0 70
96
87
144
175
318
301
11.8
13.1
13.0
13.8
13.0
13.3
32
63
19
57
20
20
23
34
16
12
12
9
566
——
569
569
569
574
3/26—21g
22A
24
298
300
300
Normal
Normal
Normal
S 1
1
i
,D,E
,D,E
None
None
None
16
8
6
0°
0°
0°
5/95
43/57
62/38
138
148
147
5.9
6.9
6.8
247
408
460
0.34
0.55
0.63
144
239
269
12.2
11.8
12.0
19
20
20
27
24
17
569
630
580
(a) Firing Psttern S — Normal Firing, S 2 — Overfire Air Staged Firing, 5 Mills (A at top to E on botton).
(b) Based on simplifying assumption that air flows to each burner are equal.
122
-------
APPENDIX C
TABLE 7
SU 0(ARY OF OPERATING AND ENISSION DATA — MERCER. BOILER NO. 1
(315 MWe, Horizontally Opposed, Pulverized Coal Fired)
Date and
Run No
(1975—
1976)
Op sting_Cond!ione
Flue
Cas Measurements_______
NOx
anogrs
Per
Joule
CO
NC
Opacity
Gross
Load
Excess
Air
Level
Firing
Pattern(s)
Secondsry
Air
2
PPM
(3% 02,
Dry)
Pounds
Per io6
8Th
2
PPM
(3% 02,
Dry)
PPM
(3% 02,
Dry)
Ke1vi
2
Mid
Rot
12/3—1
2
3
4
296
290
290
290
Normal
Low
Normsl
Low
SI—Normal
Sj—Normal
S2—Bissed
S 2 —Bissed
80%
802
80%
802
802
80%
802
802
802
80%
80%
80%
3.9
1.9
3.2
1.4
1383
1147
1237
922
1.88
1.56
1.68
1.25
808
671
723
539
14.6
17.3
16.0
17.7
22
32
21
131
1
3
5
3
705
689
705
700
——
22
15
15
12/4—5
6
7
8
1A
288
285
283
284
290
Normal
Low
Normal
Low
Normsl
S 1 —Normal
S 1 —Normsl
S 2 —Biased
S 2 —Biased
S 1 —Normal
80%
80%
80%
80%
80%
502(c)
502(c)
SO2(c)
502(c)
802
502(c)
502(c)
502(c)
502(c)
3.9
2.0
3.9
1.8
4.1
1439
1038
1164
876
1319
1 96
1.41
1.58
1.19
1.79
841
607
681
512
771
15.4
17.0
15.4
17.1
14.6
21
25
24
61
15
5
3
3
2
4
71.1
711
700
691
705
30
20
30
20
35
12/5—lB
1C
1D
280
291.
292
Normal
Normal
Normal
S 1 —Normal
S 1 —Normal
S 1 —Normal
80%
802
802
80%
80%
802
802
802
80%
4 9
3.9
4.0
1256
1321
1302
1.71
1.80
1 77
735
773
761
14.0
14.4
15.0
17
19
18
2
3
3
705
700
703
30
30
——
12/8—23
24
26
25
29
30
150
151
152
152
155
150
Low
Normal
Low
Normal
Normal.
Low
5 4 —Top Off
S 3 —Top Off
S 3 —Staged
S 3 —Staged
S 3 —Staged
S 3 —Staged
02
02
80%
802
502
502
80%
802
802
802
802
80%
80%
80%
80%
80%
80%
802
5.0
6.6
4.3
8.2
8.0
4.5
778
906
341
771
840
482
1.06
1.23
0.46
1.05
1.14
0.66
455
530
199
451
491
282
14.1
12.1
14.0
10.6
10.8
14.2
1.6
15
27
22
16
18
3
3
1
2
2
1
661
661
641
664
658
658
——
——
——
——
12/9—10
9
27
28
20
19
219
218
216
212
210
211
Low
Normal
Normal
Low
Low
Normal
S 3 —Stsged
S 3 —Staged
S 3 —Staged
S 3 —Staged
S 3 —Stnged
S 3 —Staged
802
802
502
50%
152
152
802
80%
802
80%
80%
802
80%
802
802
802
80%
80%
3.6
6.8
6.6
3.1
2.3
6.3
386
769
856
364
612
1016
0.52
1.05
1.16
0.50
0.83
1.38
226
449
501
213
358
594
15.0
11.6
11.6
15 0
15.8
11.8
218
20
17
338
69
26
1
2
2
1
0
2
658
673
675
658
650
675
——
——
——
——
——
——
12/10—11
12
22
238
232
226
Normal
Low
Low
S 1 —Normal
S 1 —Normal
S 4 —Top 0ff
802
802
0%
80%
802
802
802
802
802
5.9
3.2
3.2
1136
1019
936
1.54
1.39
1.27
664
596
547
12.6
15.0
14.4
16
19
21
4
3
4
700
678
680
——
——
——
12/ll—23A
26A
25A
165
161
162
Low
Low
Normal
S4—Top Off
S 3 —Staged
S 3 —Staged
02
80%
80%
802
802
802
802
802
80%
4.0
4.2
7.0
714
362
689
0.97
0.49
0.94
417
212
403
15.3
14.3
11.3
17
43
33
1
1
3
630
633
642
——
——
——
12/12—17
18
1OA
236
232
211
Normal
Low
Low
S 1 —Normal
S 1 —Normsl
S 3 —Staged
80%
80%
802
50%
50%
802
502
502
802
5.7
3.5
3.3
1137
1130
327
1.55
1.54
0.44
665
661
191
12.3
14.4
15.6
12
12
264
4
2
1
683
672
664
——
——
——
1/20—4A(d)
265
Low/Normal
S 2 /S 1
802
802
80%
2.6
1018
1.38
595
15.8
48
2
682
15
1/21—48(d)
268
Low/Normal
S 2 /S 1
802
80%
80%
2.4
1001
1.36
585
16.8
18
3
686
——
1/22—4C(d)
269
Low/Normal
S 2 /S 1
802
80%
802
2.8
997
1.36
583
16.3
24
2
684
—
(a) S 2 —Biased Firing top row of burners fuel—lean with middle and bottom burners fuel—rich; S 3 —Staged Firing top row of burners
on air only, S 4 —Baseline at Low Load top row of burners air and coal off.
(b) Secondary air register settings are Fire 2, 80% open, Fire 1, 502 open for low load firing, Fire Ignite (Fign), 15% open
(c) Reheat furnace — 50% and superheat furnace 80% open, respectfully.
(d) Reheat furnace operated under low excess air, biased firing (S 2 ) and superheat furnace operated under baseline operations (S 1 ).
123
-------
APPENDIX C
TABLE 8
SUMMARY OF OPERATING AND EMISSION DATA -
MORGANTOWN STATION, GAS TURBINE NO. 3
(50 MWe Oil Fired)
Date and
Run No.
1975
Operating Conditior9
Flue Gas Measurements
Gross
Load
MWe
Fuel
Flow
1/sec.
Ambient
Air Temp.
Kelvin
Compressor
Inlet Temp.
Kelvin
..... .22_
N2 _______
C0
CO
HC
Temp.
%
PPM
(15% 02
Dry)
PPM
(3% 02
Dry)
Pounds
Per 106
BTU
Nanogr.
Per
Joule
¼
PPM
(3% 02
Dry)
PPM
(3% 02
Dry)
Kelvin
4/1—1
2
3
4
10
25
48
54
2.08
2.96
4.67
5.11
290
290
290
290
290
290
291
293
17.9
17.3
15.2
14.7
82
108
125
133
247
322
376
398
0.33
0.43
0.50
0.53
144
188
220
233
2.1
2.5
4.0
4.3
244
89
12
11
73
33
17
16
601
1l
758
800
-------
APPENDIX D
Coal Analyses
Representative coal samples were taken for each major test
under baseline and “low NOR” operating conditions. The samples were
submitted to the Exxon Research and Engineering Company’s Coal Analysis
Laboratory at Baytown, Texas for analysis. Ultimate analysis determinations,
which were of most importance to the study, were made on all samples as
indicated in the following tables for each boiler tested. Proximate
analyses information are also tabulated, where available. Ash fusion
temperature determinations under reducing and oxidizing conditions and
analyses for critical coal ash elements were obtained on coal samples
taken during certain important tests in an attempt to shed more light on
potential slagging or fouling side effects of “low NOR” firing techniques.
All coal analyses data, which were used for making various
calculations in this report, are tabulated in Tables 1—7 of Appendix D.
125
-------
APPENDIX D
TABLE 1
COAL ANALYSES
TENNESSEE VALLEY AUTHORITY
WIDOWS CREEK STATION - BOILER NO. 3
Test No. lB 1C 12C-12D
Laboratory No. 96256 96257 96258
Sample Raw Coal Raw Coal Raw Coal
Date 10/7/74 10/8/ 74 10/15/74
Test Conditions Base Base Low NO
x
Proximate Analysis
Moisture (as rec’d.) 8.40 10.18 13.67
Ash 14.64 18.27 12.30
Volatiles 30.18 29.41 32.35
Fixed Carbon 46.79 42.14 41.68
Sulfur 2.77 4.16 3.71
Btu/ lb. 11,281 10,258 10,622
Hardgrove Grindability 65.9 73.1 63.8
Ultimate Analysis-Dry Basis
C - 7, 68.58 63.43 68.0
H - 7. 4.46 4.29 4.63
S - 7, 3.02 4.63 4.30
N — ‘I . 1.21 1.21 1.20
Cl -
0 - 7 . 6.75 6.10 7.63
Ash - 7. 15.98 20.34 14.25
Btu/lb. - 7. 12,316 11,420 12,304
Ash Elements
- 7. 0.40 0.40 0.40
SiO 2 - 7. 49.99 44.67 42.35
Fe 2 0 3 - 7. 16.67 23.08 23.53
Al 03 - 7. 22.52 17.37 20.09
Ti 2 - 7. 0.96 0.80 0.93
CaO - 7. 2.81 5.47 6.08
MgO - 7. 1.02 1.06 0.80
SO 3 - 7. 2.06 3.06 3.22
1 (20 - 7. 2.21 2.34 1.69
Na 2 0 - 7. 0.25 0.29 0.33
Total - 7. 98.89 98.54 99.42
Ash Fusion Temperature
Reducing - I.D. 2100 1950 2010
- H=W 2150 1990 2050
- H=l/2W 2180 2000 2075
- Fluid 2475 2030 2125
Oxidizing - I.D. 2475 2160 2175
- H=W 2485 2350 2400
- H=1/2W 2510 2400 2420
- Fluid 2525 2425 2450
126
-------
APPENDIX D
TABLE 2
COAL ANALYSES
SOUTHERN ELEC RIC GENERAT ING COMPANY
E. C. GASTON STATION - BOILERS No.’s 1 AND 2
Test No. 28A, 28B 29 30 26 27A, 27B
Laboratory No. 96284 96288 96292 96274 96279
Sample Raw Coal Raw Coal Raw Coal Raw Coal Raw Coal
Date 1/22/75 1/23/75 1/29/75 1/17/ 75 1/20/75
Test Conditions Base Base Base Low NO Low NO
x x
Boiler No. 2 2 2 1 1
Proximate Analysis
Moisture (as rec’d.) 6.04 6.61 7.82 6.35 5.29
Ash 15.00 16.06 13.28 12.12 9.84
Volatiles 27.66 29.71 30.55 33.97
Fixed Carbon 51.30 47.62 48.35 47.56
Sulfur 1.38 1.22 1.59 3.11 1.25
Btu/ lb. 11,690 11,461 11,398 11,752 11,632
Hardgrove Grindability ———— 63.0 58.0 59.5 59.5
Ultimate Analysis-Dry Basis
C — 7 . 69.51 69.12 69.89 70.13 69.29
H - 7. 4.57 4.63 4.54 4.86 4.64
S — 7, 1.47 1.31 1.73 3.33 1.32
N — 7, 1.61 1.42 1.42 1.51 1.58
Cl -70 ----
O - ‘I . 15.96 6.33 8.02 7.24 12.77
Ash - 7. ———— 17.20 14.40 12.94 10.39
Btu/lb. 12,272 ———— 12,549 12,282
Ash Elements
P205 - 7. 0.62 0.37 0.42 0.31 0.83
Si0 2 - 7. 50.46 53.00 53.32 47.21 53.65
Fe 2 0 3 - 7. 8.33 10.16 11.77 21.53 8.10
Al 03 - 7. 29.23 26.34 23.35 22.38 27.06
Ti 2 - 7. 1.52 1.20 1.20 1.11 1.29
Ca0 2 - 7. 3.30 2.18 3.10 2.20 2.95
MgO - 7. 1.29 1.50 1.16 1.09 1.36
SO 3 — 7. 1.97 2.12 3.17 2.57 1.68
1 (20 - 7. 1.70 3.18 2.00 1.80 2.55
Na 2 0 - 7. 0.59 0.47 0.35 0.37 0.48
Total - 7. 99.01 100.52 99.83 100.57 99.95
Ash Fusion Temperature
Reducing - I.D. 2425 2275 2075 2000 2475
— H=W 2575 2540 2320 2065 2540
— H=1/2W 2620 2570 2375 2085 2575
- Fluid 2680 2650 2550 2110 2675
Oxidizing — I.D. 2625 2550 2425 2425 2700+
- H=W 2700+ 2625 2575 2530 2700+
- H=l/2W 2700+ 2650 2610 2560 2700+
- Fluid 2700+ 2690 2625 2580 2700+
127
-------
APPENDIX D
TABLE 3
COAL ANALYSES
ALABAMA POWER COMPANY
BARRY STATION - BOILER NO. 2
Test No. 1-6 7-12, 25, 26 21-24, 27-30 13-20 31-38
Laboratory No. 96295 96296 96297 96298 96299
Sample Raw Coal Raw Coal Raw Coal Raw Coal Raw Coal
Source Alabama Ohio/Ky. Rockport, Ky. Rockport, Ky. Rockport, Ky.
Date 2/12/75 2/13/75 2/19/75 2/20/75 2/21/75
Proximate Analysis
Moisture (as rec’d.) 5.51 6.9 ’ 10.11 12.30 10.15
Ash (as recd.) 13.22 10.79 12.03 11.85 11.01
Volatiles (as rec’d.) 28.38 34.62 31.81 30.19 32.49
Fixed Carbon (as rec’d.) 52.89 47.63 46.05 45.66 46.35
Sulfur (as rec’d.) 2.00 3.41 3.11 2.57 2.93
Btu/lb. 12,237 11,802 11,081 11,862 11,389
Hardgrove Grindability ———— 55.2 64.5 57.3 56.6
Ultimate Analysis-Dry Basis
C — 7. 71.32 70.65 68.37 69.13 70.51
H - 7. 4.81 4.95 4.93 4.76 4.79
S - 7. 2.12 3.67 3.46 2.93 3.26
N — 7. 1.59 1.56 1.48 1.55 1.45
0 — 6.17 7.57 8.37 8.12 7.73
Ash — 7. 13.99 11.60 L3.39 13.51 12.26
BtuI lb. 12,950 12,685 12,327 12,328 12,675
Ash Elements
P O 5 7. 0.78 0.30 0.30 0.28 0.23
S o 7. 45.96 43.57 46.49 47.60 45.35
Fe 2 3 7. 16.25 23.46 24.26 24.56 21.67
Al 03 7. 27.45 18.83 20.30 20.57 18.04
Ti 2 7. 1.45 0.86 0.93 0.91 0.87
CeO 7. 1.91 4.76 2.08 1.68 5.13
MgO 7. 1.23 0.87 1.06 1.00 0.99
SO 3 7. 1.02 4.10 1.37 0.74 5.07
1(20 7. 2.51 1.88 2.28 2.96 2.27
Na 2 0 7. 0.40 0.47 0.24 0.17 0.44
Total 98.94 99.10 99.31 100.47 100.07
Ash Fusion Temperatures
Reducing - I.D. 1950 2000 2010 2000 1970
- H=W 2350 2340 2050 2065 2025
- H=l/2W 2400 2410 2065 2090 2040
— Fluid 2570 2450 2100 2110 2065
Oxidizing - I.D. 2660 2360 2500 2510 2350
- H=W 2700+ 2440 2550 2575 2410
- H=1/2W 2700+ 2520 2590 2600 2450
— Fluid 2700+ 2580 2610 2625 2520
128
-------
APPENDIX D
TABLE 4
COAL ANALYSES
POTOMAC ELECTRIC COMPANY
MORGANTOWN STATION - BOILER NO. 1
Test No. 1-6 21A, 22A, 24
Laboratory No. 96300 96301 96302 96303
Sample Raw Coal Raw Coal Raw Coal Raw Coal
Source West Va. Comp. F—334 Comp. F—335
Date 3/14/ 75 3/19—3/25/75 3/26175 3/19-31/75
Proximate Analysis
Moisture (as rec’d.) 7.05 9.14 9.54 8.29
Ash (rec’d.) 20.40 12.92 13.88 15.24
Volatiles (as rec’d.) 20.54 20.77 21.30 21.50
Fixed Carbon (as rec’d.) 52.00 57.17 55.28 54.97
Sulfur (as rec’d.) 1.92 2.05 1.94 1.41
Btu/ lb. 10,792 11,630 11,296 11,428
Hardgrove Grindability 100.3 104.6 96.7 106.0
Ultimate Analysis—Dry Basis
C 7 . 66.17 73.28 71.58 71.59
H 7. 3.99 4.06 4.27 4.18
S 7. 2.06 2.26 2.14 1.53
N 7. 1.20 1.28 1.33 1.33
O 7. 4.63 4.90 5.33 4.75
Ash 7. 21.95 14.22 15.35 16.62
Btu/lb. 11,611 12,800 12,488 12,461
Ash Elements
7. 0.34 0.50 0.47 0.35
sto 7. 53.06 46.65 47.13 50.18
Fe 2 3 7. 12.78 19.20 18.65 14.73
Al 03 7. 25.89 26.41 25.32 27.47
Ti 6 2 7. 1.26 1.32 1.19 1.32
CaO 7. 1.19 1.89 2.18 1.87
MgO 7. 1.08 0.75 0.79 0.89
503 7. 1.05 1.50 1.94 1.10
1(20 7. 2.19 1.65 2.08 2.46
Na 2 0 7. 0.13 0.19 0.16 0.18
Total 98.96 100.05 99.91 100.56
Ash Fusion Temperatures
Reducing - I.D. 2400 2180 2150 2440
- H=W 2475 2350 2340 2510
- H=1/2W 2510 2430 2480 2545
— Fluid 2640 2570 2625 2700+
Oxidizing - I.D. 2625 2570 2535 2660
— H=W 2690 2625 2590 2700+
- 1 1=1 /2W 2700+ 2640 2610 2700+
- Fluid 2700+ 2660 2630 27004-
129
-------
APPENDIX D
TABLE 5
COAL ANALYSES
SALT RIVER PROJECT
NAVAJO STATION - BOILER NO. 2
Test No. 3B, 18B 18B 2A 3D, 18E
Laboratory No. 96304 96305 96306 96307
Date 6/18,19/75 6/20/75 6/24/75 6/26/75
Sample Raw Coal Raw Coal Raw Coal Raw Coal
Conditions Base and Low NO Low NO Base Base and Low NO
x x x
Location of Sample Feeders Feeders Feeders Feeders
Proximate Analysis
Moisture (as rec’d.) 13.27 13.54 12.45 12.81
Ash 11.49 6.62 5.94 6.04
Volatiles 35.65 36.89 38.43 37.96
Fixed Carbon 39.59 42.95 43.18 43.19
Sulfur 0.48 0.48 0.42 0.48
Btu/lb. 10,210 10,780 11,068 10,954
Hardgrove Grindability 40.1 42.3 40.9 39.4
Ultimate Analysis
C - 7. Dry 66.78 71.60 72.42 72.18
H - 7. Dry 4.79 5.06 5.15 5.16
S - 7. Dry 0.56 0.55 0.48 0.55
N - 7. Dry 1.39 1.38 1.43 1.38
o — 7. Dry 13.24 13.75 13.73 13.79
Ash - 7. Dry 13.25 7.65 6.78 6.93
Btu/lb. — 7. Dry 11,772 12,469 12,642 12,563
Ash Elements
P O 5 0.32 0.50 0.61 0.40
sto 52.48 43.43 41.54 47.37
Fe 2 3 3 3.96 5.36 6.62 5.66
A1 2 O 3 23.34 19.46 18.18 17.23
Ti0 2 0.97 0.85 0.95 0.91
CaO 6.24 11.65 11.98 9.34
MgO 1.36 1.83 2.02 1.59
SO 6.82 14.05 14.80 13.37
0.33 0.27 0.17 0.27
Na 0 1.98 3.58 4.29 3.77
To a1 97.0 100.99 101.17 99.91
Ash Fusion Temperatures
Reducing — I.D. 2330 2075 2050 2075
- H=W 2350 2125 2160 2150
— H=1/2W 2390 2180 2190 2180
— Fluid 2550 2225 2210 2225
Oxidizing — I.D. 2370 2200 2175 2175
- H=W 2400 2275 2225 2250
— H=1/2W 2500 2300 2300 2275
— Fluid 2550 2400 2340 2360
130
-------
APPENDIX D
TABLE 6
COAL ANALYSES
PUBLIC SERVICE OF COLORADO
COMANCHE STATION
BOILER NO. 1
PUEBLO, COLORADO
Test No. 30, 33 27A, 27B 34, 35
Laboratory No. 96353 96351 96356
Sample Raw Coal Raw Coal Raw Coal
Date 7/29/75 7/23/75 8/1/75
Conditions Base Low NO Low NO
x
Proximate Analysis
Moisture (as rec’d.) 31.86 31.57 31.54
Ash 5.36 5.43 5.62
Volatiles 29.21 28.94 30.05
Fixed Carbon 33.57 34.06 32.79
Sulfur 0.42 0.32 0.42
Btu/lb. 8,094 8,123 8,115
Hardgrove Grindability 52.3 50.2 49.4
Ultimate Analysis
C - 7. Dry 69.24 69.06 69.08
H - 7. Dry 4.85 4.79 4.79
S - 7. Dry 0.61 0.46 0.61
N - 7. Dry 0.99 0.96 1.06
Cl -7.Dry
O - 7. Dry 16.45 16.80 16.25
Ash - 7. Dry 7.86 7.93 8.21
Btu/lb. — Dry 11,879 11,871 11,854
Ash Elements
1.06 1.38 1.04
s o, 31.67 34.21 33.12
Fe 2 3 4.97 4.61 5.03
A1 9 O 3 18.75 13.87 16.28
Ti0 2 1.20 1.28 1.48
CaO 21.52 20.57 21.36
MgO 3.67 3.57 3.75
SO. 15.15 13.63 15.86
K 2 ö 0.38 0.37 0.36
Na 9 O 1.43 1.51 1.12
Total 99.80 100.00 99.40
Ash Fusion Temperatures
Reducing — I.D. 2125 2085 2075
- H=W 2150 2180 2140
— H=1/2W 2170 2200 2160
— Fluid 2180 2210 2175
Oxidizing - I.D. 2140 2100 2090
- H=W 2165 2160 2120
- H=1/2W 2170 2170 2130
— Fluid 2175 131 2200 2150
-------
APPENDIX D
TABLE 7
coa ANALYSES
PUBLIC SERVICE ELECTRIC AND GAS COMPANY
NERCER STATION, BOILER NO. 1
TestNo. 162 364 566
Laboratory No. 96390 96392
Sample Raw Coal Raw Coal Raw Coal
Date 1/20/76 1/21/76 1/22/76
Test Conditions Base & Low NO >
x
Proximate Analysis
Moisture (as rec’d.) 6.02 6.64 6.31
Ash 9.57 9.23 11.00
Volatiles 29.17 29.74 26.82
Fixed Carbon 55.25 54.38 55.87
Sulfur 1.19 1.14 1.07
Btu/ lb. 12 ,987 12,977 12,526
Hardgrove Grindability 63.8 61.6 67.3
Ultimate Analysis—Dry Basis
C — 77.21 77.47 76.34
H — 4.85 4.89 4.66
S — 1.27 1.21 1.13
N — 1.53 1.54 1.49
Cl - %
0 — 4.95 5.01 4.65
Ash — 10.18 9.88 11.73
Btu/lb. — 13,819 13,900 13,369
Ash Elements
P205 — 0.52 0.54 0.24
Sf02 — 45.99 44.70 50.79
Fe203 — 14.06 15.49 11.54
A1203 — 27.60 27.69 26.62
TiO2 — 1.25 1.19 1.17
CaO — 2.75 2.38 1.77
NgO — 1.03 1.08 1.56
S03 — 2.85 2.63 2.03
K20 — z 1.96 1.97 3.07
Na20 — 0.26 0.24 0.38
Total — z 98.87 97.92 99.16
Ash Fusion Temperature
Reducing — I.D. 2180 2180 2100
— B=W 2440 2450 2415
— H=1/2W 2475 2480 2465
— Fluid 2525 2510 2575
Oxidizing — I.D. 2540 2560 2525
— R=W 2625 2630 2625
— 1 1=1/2W 2640 2640 2660
— Fluid 2685 2675 2700
132
-------
APPENDIX E
E OCON MOBILE ANALYTICAL VAN SAMPLING SYSTEM
Design of Mobile Sampling
and Analytical System
Meeting the objectives of our field program of measuring NO
and related products of combustion emitted from a variety of power boilers
required a versatile, transportable sampling and analytical train. Such
a system had to be self-contained, mobile, and include provisions for wet
chemical analysis of grab samples. Minimum set-up time was another re-
quirement for the sampling and analytical system, which had to be
installed at the actual sampling site to reduce the possibility of changes
in the flue gas composition. Ideally, the instruments should have been
located at the sampling point, but since this location was frequently in-
accessible and was usually unsheltered, some compromise had to be made.
Other requirements for the instruments for the measurement of the concentra-
tion of the flue gas components were easy calibration, maintenance-free and
repeatable operation, and the ability to monitor gas compositions continuously.
The last requirement is of extreme importance in a field program, where
directional effects of operating changes must be assessed itnniediately.
Finally, the instrumental methods had to be compared against
wet chemical methods of analysis, as needed, to validate the accuracy of
the sampling system and continuous monitoring instrumentation.
Sampling System
The objective of obtaining data from coal, as well as oil and
gas fired boilers required the development of an elaborate sampling
system. Consideration of the solubility of NO2 in water, the presence
of oxides of sulfur, and the high concentration of particulates in the
combustion gases were taken into account in the design of the sampling
system. The sampling system was designed with adequate flexibility to
allow gas sampling from different size boilers or other stationary combus-
tion equipment. It could hanile flue gases with heavy particulate loading
from coal fired units, as well as light particulate loading from oil fired
units. The sampling assembly was a dry-type system with appropriate particu-
late filters, pumps, and a refrigeration unit to cool the samples to a
35°F dew point before analysis.
The sampling points for flue gas components were usually located
in the duct-work between the economizer and the air heater. This was done
to provide reasonably homogeneous gas samples at the temperatures to which
the probes could be subjected, and to avoid dilution of the samples by
leakage in the air heaters of the boilers tested. In this part of the
duct-work, tenperatures usually ranged between 550°F and 800°F, and gas
velocities were between 30 and 80 feet per second.
133
-------
The variability of ducting between different boilers required
the design of adjustable sampling probes. These probes were designed
with interchangeable gas sampling tubes. Since we sampled from “equal
areas” in ducts of different sizes, the probes were assembled on location
for the particular duct. A special pitot tube and a thermocouple were
located at the midpoint of each probe with a sampling tube. The re-
maining two gas sampling tubes were then assembled and the entire probe
was ready to be inserted into the duct. Each probe was fitted with a
quick disconnect as a mounting assembly for easy insertion into the
boiler. All pieces of the sampling equipment between the van and the
probes were of the quick-disconnect type for ease of assembly and
assurance of a leak-proof connection at all intermediate points. Figure
E-l shows a schematic diagram of the sampling and analytical system.
In running field tests, the gas samples were withdrawn from
the boiler under vacuum through stainless steel probes to heated paper
filters where the particulate matter was removed. These paper filters
were maintained at 300—500°F. The gases were then passed through
rotameters, which were followed by a packed glass wool column for SO 3
removal. Initially, the gas temperatures were kept as high as possible
to minimize condensation in the particulate filters. After leaving the
packed column at 250—300°F, the gas samples passed at temperatures above
the dew—point through heated Teflon lines to the vacuum/pressure pumps.
The sample was then split with a portion at 120 F sent to the NO 2 instrument
and the balance of the stream refrigerated to a 35°F dew—point before being
sent to the van for analysis. Usually, the van was located 100 to 200 feet
from this point and the gas stream flowed through Teflon lines throughout
this distance.
The sampling system performed well during the test program,
however, some difficulties were encountered with the vacuum pressure pumps.
The pumps originally acquired for our sampling system were stainless steel
bellows pumps. These pumps were manufactured with a clearance volume for
slugging liquid entrainment. After about 40 hours of use the pumps began
to leak and inspection revealed that the bellows became deformed and
perforated with pinholes. The probable cause of failure was condensate
remaining in the pump during the compression stroke deforming the bellows.
The manufacturer (Metal Bellows Corp.) supplied replacement sets of pumps
and we revised the sampling system in an attempt to overcome the problem.
Water knockouts were incorporated before the pumps and the pumps were
mounted upside down to facilitate draining of liquids that condensed dur-
ing shutdown. This procedure did not eliminate the problem and new Teflon
faced neoprene diaphragm pumps (Diapumps) had to be installed. These
proved to be satisfactory in use for the remainder of the field test
program.
Another problem of air leaking into the lines was found to
be due to the flexible lines. While these lines were designed for
high pressures and temperatures, their flexibility was not sufficient
for our purpose. After severe bends, necessitated by probe locations,
leaks would develop when the lines were heated to high temperatures.
We are currently experimenting with a new design which eliminates the
protective wire braid from the line on a replacement basis. Preliminary
evaluation shows that this type of line is superior to the old one. Also,
pressure-testing all lines at each boiler in future work will be required
to correct this problem.
134
-------
Probe (4 each)
CO
Co 2
NO
Hydrocarbons
Sampling
van
vent
5 psi relief
valve
Boiler
Duct
800°F
Particulate filters (he ted)
rotameters
35°F
Remote Instruments
(at boiler duct)
I -I
U I
heated lines
refrigerator
Solenoid
Valve
to NO 2
H 2 0
200
Solenoid
Valve
so 2
02
FIGURE E-1
E O ON RESEARCH TRANSPORTA 8LE SAMPLING AND ANALYTICAL SYSTEM
-------
Analytical Instrument Train
The selection of instruments for the measurement of flue gas
composition was complicated by the relatively short delivery time neces-
sitated by the requirement to begin the field test program in the early
part of the contractual period. The instruments had to be installed and
wired in a console and checked out before the test program could begin.
Beckman Instruments Inc. was chosen as the supplier for these instruments
because of their ability to deliver monitoring instrumentation in a short
time and Exxon Research and Engineering Company’s prior familiarity with
their analyzers in other air pollutant measurement trains. Another reason
was the availability of the field service organization of Beckman which was
felt to be an important asset for field studies.
The instruments selected for monitoring flue gases were those
that had been demonstrated to be accurate and relatively trouble—free
in previously used exhaust gas analytical trains at Exxon. Our van was
equipped with Beckman non—dispersive infrared analyzers to measure NO,
CO, CO 2 and SO 2 , a non—dispersive ultraviolet analyzer for NO 2 measurement,
a polarographic 02 analyzer and a flame ionization detector for hydrocarbon
analysis. The measuring ranges of these continuous monitors are listed in
Table E—l.
ThBLE E-l
Continuous Analytical
Instruments in Exxon Van
Beckman Measuring
Instruments Technique Range
NO Non-dispersive infrared 0-400 ppm
0-2000 ppm
NO 2 Non-dispersive ultraviolet 0-100 ppm
0-400 ppm
Polarographic 0-5%
0-25%
CO 2 Non-dispersive infrared 0-20°f .
CO Non-dispersive infrared 0-200 ppm
0-1000 ppm
SO 2 Non-dispersive infrared 0-600 ppm
0-3000 ppm
Hydrocarbons Flame ionization detection 0-10 ppm
0-100 ppm
0-1000 ppm
136
-------
The instruments were housed in a console which was shock-mounted
inside the van. All connections to the console were made beneath the floor
to prevent a tripping hazard. Separate raceways for piping and electric
wires terminated in the base of the cabinet. Each analyzer was connected
in parallel to the sample and calibration gas lines to insure that each
analyzer would be operated independently of the others.
In the original design of the sampling-analytical train and during
its construction, sufficient flexibility was engineered into our system to
allow additional analyzers to be installed, and for modifications and special
sample handling techniques to be incorporated. In addition to analytical
instrumentation for continuously measuring all of the major flue gas com-
ponents including NO, NO 2 , CO, CO 2 , 02 and hydrocarbons (with an instrument
added later to measure SO 2 ), the temperature and velocity of the gases in
the duct could also be measured. A novel programmable sample timer was
installed to allow any sample cycle tojesimply_dialed into the equipment.
Normally, measurements from four different locations within the ducts
could be made in eight minutes. After steady state conditions in the
boiler had been established, the sampling time of 32 minutes per test
allowed 4 repeats of each location assuring that reproducible data
were obtained. The programmable sample timer proved to be very useful
when monitoring operations at very low excess air levels, because the
most sensitive areas had to be monitored more frequently.
Separate calibration gas cylinders in appropriate concentra-
tions with N 2 carrier gas for each analyzer were installed in the van.
Each cylinder was equipped with a regulator, safety relief valve, ex-
cess flow check valve and other necessary valves and piping. The
cylinders were securely fastened to the body and frame of the vehicle
to insure safe transportation. Each cylinder was piped directly to
the analyzer for ease of operation.
The sample gases were pumped to the van from four separate
probes. While one sample was being monitored, the other three were vented
from the van to insure that a fresh sample would be available when re-
quired. This operation was performed automatically by the sample timer.
The gases were analyzed as received at the 35°F dew point except
for the NO instrument which had twin chemical driers for the removal of
water. The driers were filled with fresh indicating drierite before each
run and were only used until the color change had reached the mid-point of
the tubes.
The hydrocarbon instrument, a Beckman Model 400 flame ionization
detector, measured only the hydrocarbons that reach the instrument. Only
hydrocarbons volatile under the sampling conditions could be measured by this
instrument because of the sample preparation system. The initial filtration
300-500°F removed solid as well as liquid particulates. The glass wool
packed column, maintained at 250°-300°F, might have removed lower boiling
liquids. Gases were then refrigerated to a 35°F dew point before being anal-
yzed for hydrocarbons. The condensate from the refrigeration unit was anal-
yzed for organic carbon in selected test runs, and was found to contain on
the average 20-30 ppm hydrocarbon equivalent in the flue gas.
137
-------
In addition to the recorders in the van, separate trend re-
corders for NO, 02 and CO were provided for remote observation of flue gas
concentrations in the control room of the utility. The effects of changes
in operating variables, therefore, were continuously displayed to provide
information to the operating personnel in the control room.
While in most cases the instruments performed satisfactorily, some
special problems did arise. The CO instrument, which has a long path infra-
red cell, was found to be sensitive to 200 ppm CO full scale. Moisture inter-
ference was a major problem for this long path instrument. Changes were made
in the filter cells, heater circuits, and in the gold coating of the sample
cell. These changes reduced the problem but could not completely eliminate
it. Although this problem may be circumvented by using chemical driers
upstream of the sample cell, a more satisfactory solution is desirable for
future measurements of this type. Narrow band-path optical filters will be
installed to reduce moisture interference with the response of the sample
cell.
The NO 2 instrument, a non-dispersive ultraviolet analyzer, was
designed for 100 ppm NO 2 full scale. However, the accuracy of measuring
extremely small amounts of NO 2 in the flue gases was affected by the noise
level of the instrument. This noise level was substantially increased by
the remote location of the NO 2 instrument, and the varying temperature
environment.
Our experience demonstrated that with the NO 2 analyzer (an in-
strument designed for a laboratory environment) even though it was shock
mounted for vibration, the sensitivity of the mirror adjustment was such
that we lost calibration during runs due to boiler-induced and other
vibrations of very low frequency. Also, because this instrument was used
for measuring hot samples of the corrosive flue gas which had not been
subjected to condensation, the analyzer was fouled easily. Small tempera-
ture variations due to the wind-chill factor at unprotected outside boiler
locations could cause condensation in the analyzer. Then, an elaborate
cleaning procedure was required, which could not be performed during
actual testing. Based on these findings, our future plans are to redesign
this portion of the analytical system.
Integration of Sampling-
Analytical System into Mobile Van
The van used to house the instruments, sampling train, and wet
chemical laboratory is a Winnebago mobile home shell. The basis for the
selection was availability, allowable payload weight and a self-contained
propulsion system to provide maximum mobility. The shell is 27 feet long
by 7 feet 6 inches wide and is mounted on a Dodge truck chasses. The
driver compartment is located in the first 5 feet of the shell. The van
is air conditioned and heated for all-weather operation. A gasoline-
powered electric generator housed in a compartment of the chassis provides
power for lighting and air conditioning during the initial equipment de-
ployment. However, during sampling and data collection the van operated
with electricity provided at the generating station. A floor plan of the
van is presented in Figure E-2.
138
-------
FIGURE E-2 FLOOR PLAN OF SAMPLING-ANALYTICAL VAN
C ,
C
I-
C ,
0
I-A
U ,
tnstrument Cabinet
Overhead Wal
-------
The instruments are housed in a permanent shock mounted instru-
ment console. The calibration gases are permanently installed in the rear
of the van. These cylinders are securely mounted for traveling and each
bottle has its own low pressure safety valve and velocity check in addi-
tion to standard regulators and valves. The rear of the van is designed as
a laboratory bench including a sink. It is used for experimentation and
as a foul weather workbench. A swingaway desk top and a file cabinet
provide an area for data analysis inside the van. A Sony programmable desk
calculator for preliminary data reduction is part of the equipment carried
in the van.
An electrical distribution center for the van includes voltage
regulators ahead of the instrumentation to anticipate any large variations
in line voltage between generating stations. Normal power requirements are
14 KVA for the van and the remote sampling train.
The van holds its own water supply and has a portable winch for
equipment deployment. The external connections to the van are all quick
disconnects in an umbilical area.
140
-------
TECHNICAL REPORT DATA
(Please read !nsj.nlcrzons on the rererse before corn pie tingj
1. REPORT NO.
EPA— 600/7—78—036a 12.
3 RECIPIENTS ACCESSIOF*NO.
4. TITLE AND SUBTITLE Control of Utility Boiler and Gas
Turbine Pollutant Emissions by Combustion
Modification- - Phase I
5. REPORT DATE
March 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
A.R.Crawford, E.H.Manny, and W.Bartok
8. PERFORMING ORGANIZATION REPORT NO.
Project RP 200
9. PERFORMING OROANIZATION NAME AND ADDRESS
Exxon Research and Engineering Company
P.O. Box 8
Linden, New Jersey 07036
10. PROGRAM ELEMENT NO.
EHE624A
11.CONTRACT/GRANTNO.
68-02-1415
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development*
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Phase I; 6/74-6/76
14. SPONSORING AGENCY CODE
EPA/600/13
15.SUPPLEMENTARY NOTES IERL-RTP project officer is Robert E. Hall, Mail Drop 65
919/541-2477. (*)The Electric Power Research Institute cosponsored this work
EPRI project manager Is Robert C. Carr, P.O. Box 10412, Palo Alto, CA 94303.
16.MoaurlM .., The report gives results of a field study to assess the applicability of com-
bustion modification techniques to control NOx and other pollutant emissions from
utility boilers and gas turbines without causing deleterious side effects. Comprehen-
sive, statistically designed tests were used to evaluate the effect of combustion modi-
fications. The most extensively studied combustion modification for utility boiler
applications was staged firing at low excess air, which can reduce NOx emissions by
up to about 50%, based on the results of short term tests. With emphasis on NOx
emission control for coal-fired utility boilers, special attention was paid to the deter-
mination of potentially adverse side effects: increased combustible emissions,
unwanted cbanges in particulate mass loading and size distribution, reduced boiler
efficiency, increased furnace slagging and tube wall corrosion, and flame problems.
Short term tests indicate that staged combustion may be applied to coal-fired utility
boilers. The extent of furnace tube wall corrosion and slagging could not be deter-
mined conclusively, based on the results of 300-hour corrosion probing runs under
low NOx and baseline operating conditions. For this reason, a long term furnace
tube wall corrosion test of at least 6 months duration was initiated on a 500 MWe
front-wall-fired boiler at Gulf Power Company’s Crist Station.
17. KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b IDENTIFIERS/OPEN ENDED TERMS
C COSATI I icld/Group
fr Pollution Electric Power
Coal Generation
Utilities Nitrogen Oxides
Boilers
Gas Turbines
Combustion
Air Pollution Control
Stationary Sources
Combustion Modification
Staged Firing
Low Excess Air
l3B
21D 1OA
07B
l3A
13G
21B
18 DISTRIBUTION STATEMENT
Unlimited
19 SECURITY CLASS (flits Re org)
Unclassified
21 NO OF PAc;I
141
20 SECURITY CLASS (Tins page)
Unclassified
22 PRICE
EPA Form 2220.1 (9.73)
141
------- |