ABMA
American
Boiler Manufacturers
Association
1500 Wilson Boulevard
Arlington VA 22209
DoE
United States
Department,
of Energy
Division of Power Systems
Energy Technology Branch
Washington DC 20545
EPA
U.S. Environmental Protection Agency
Office of Research and Development
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600/7-79-237a
November 1979
Field Tests of Industrial
Stoker Coal-fired Boilers
for Emissions Control
and Efficiency
Improvement - Site D
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA 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-600/7-79-237a
November 1979
Field Tests of Industrial Stoker Coal-fired
Boilers for Emissions Control and
Efficiency Improvement - Site D
by
J.E. Gabrielson, P.L Langsjoen, and T.C. Kosvic
KVB, Inc.
6176 Olson Memorial Highway
Minneapolis, Minnesota 55422
IAG/Contract Nos. IAG-D7-E681 (EPA), EF-77-C-01-2609 (DoE)
Program Element No. EHE624
Project Officers: Robert E. Hall (EPA) and William T. Harvey, Jr. (DoE)
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
U.S. DEPARTMENT OF ENERGY
Division of Power Systems/Energy Technology Branch
Washington, DC 20545
and
AMERICAN BOILER MANUFACTURERS ASSOCIATION
1500 Wilson Boulevard
Arlington, VA 22209
-------
ACKNOWLEDGEMENTS
The authors wish to express their appreciation for the assistance
and direction given the program by project monitors W. T. (Bill) Harvey of
the United States Department of Energy (DOE) and R. E. (Bob) Hall of the
United States Environmental Protection Agency (EPA). Thanks are due to
their agencies, DOE and EPA, for co-funding the program.
We would also like to thank the American Boiler Manufacturers
Association, ABMA Executive Director, W. H. (Bill) Axtman, ABMA Assistant
Executive Director, R. N. (Russ) Mosher, ABMA's Project Manager, B. C.
(Ben) Severs, and the members of the ABMA Stoker Technical Committee chaired
by W. B. (Willard) McBurney of the McBurney Corporation for providing support
through their time and travel to manage and review the program. The
participating committee members listed alphabetically are as follows:
R. D. Bessette Island Creek Coal Company
T. Davis Combustion Engineering
N. H. Johnson Detroit Stoker
K. Luuri Riley Stoker
D. McCoy E. Keeler Company
J. Mullan National Coal Association
E. A. Nelson Zurn Industries
E. Poitras The McBurney Corporation
P. E. Ralston Babcock and Wilcox
D. C. Reschley Detroit Stoker
R. A. Santos Zurn Industries
We would also like to recognize the KVB engineers and technicians who
spent much time in the field, often under adverse conditions, testing the boilers
and gathering data for this program. Those involved at Site D were Jim
Burlingame, Russ Parker, Mike Jackson, and Jim Demont.
Finally, our gratitude goes to the host boiler facilities which in-
vited us to test their boiler. At their request, the facilities will remain
anonymous to protect their own interests. Without their cooperation and
assistance this program would not have been possible.
11
-------
TABLE OF CONTENTS
Section Page
ACKNOWLEDGEMENTS ii
LIST OF FIGURES v
LIST OF TABLES vi
1.0 INTRODUCTION 1
2.0 EXECUTIVE SUMMARY 3
3.0 DESCRIPTION OF FACILITY TESTED AND COALS FIRED 9
3.1 Boiler D Description 9
3.2 Overfire Air System 9
3.3 Particulate Collection Equipment 12
3.4 Test Port Locations 12
3.5 Coals Utilized 12
4.0 TEST EQUIPMENT AND PROCEDURES 17
4.1 Gaseous Emissions Measurements 17
4.1.1 Analytical Instruments and Related Equipment ... 17
4.1.2 Recording Instruments 21
4.1.3 Gas Sampling and Conditioning System 21
4.1.4 Gaseous Emission Sampling Techniques 21
4.2 Sulfur Oxides (SOx) Measurement and Procedures 23
4.3 Particulate Measurement and Procedures 25
4.4 Particle Size Distribution Measurement and Procedure . . 25
4.5 Coal Sampling and Analysis Procedure 29
4.6 Ash Collection and Analysis for Combustibles 30
4.7 Boiler Efficiency Evaluation 31
4.8 Modified Smoke Spot Number 31
4.9 Trace Species Measurement 33
5.0 TEST RESULTS AND OBSERVATIONS 35
5.1 Overfire Air 35
5.1.1 Particulate Loadings vs Overfire Air 37
5.1.2 Nitric Oxide vs Overfire Air 42
5.1.3 Carbon Monoxide vs Overfire Air 45
5.1.4 Boiler Efficiency vs Overfire Air 45
5.2 Excess Oxygen and Grate Heat Release 49
5.2.1 Excess Oxygen Operating Levels 49
5.2.2 Particulate Loading vs Oxygen and Grate Heat
Release 51
5.2.3 Nitric Oxide vs Oxygen and Grate Heat Release . . 54
5.2.4 Carbon Monoxide vs Oxygen and Grate Heat Release 63
5.2.5 Combustibles vs Oxygen and Grate Heat Release . . 63
5.2.6 Boiler Efficiency vs Oxygen and Grate Heat Release 68
111
-------
TABLE OF CONTENTS
Section
5. 3 Coal Properties 71
5.3.1 Chemical Composition of the Coals 71
5.3.2 Coal Size Consistency 77
5.3.3 Sulfur Balance 77
5.4 Particle Size Distribution of Flyash 77
5.5 Efficiency of Multiclone Dust Collector 88
5.6 Modified Smoke Spot Number 88
5.7 Source Assessment Sampling System 94
5. R Data Tables 94
APPENDICES 101
IV
-------
LIST OF FIGURES
Figure
No- Page
3-1 Boiler D Schematic 13
3-2 Boiler D Sample Plane Geometry 14
4-1 Flow Schematic of Mobile Flue Gas Monitoring Laboratory ... 22
4-2 SOx Sample Probe Construction 24
4-3 Sulfur Oxides Sampling Train 24
4-4 EPA Method 5 Particulate Sampling Train 26
4-5 Brink Cascade Impactor Sampling Train Schematic 28
4-6 Field Service Type Smoke Tester 32
4-7 Source Assessment Sampling (SASS) Flow Diagram 34
5-1 Overfir3 Air Configuration 36
5-2 Particulate Loading Breakdown for Perfect 8 Coal as a Function
of OFA 38
5-3 Particulate Loading Breakdown for Century Coal as a Function
of OFA 39
5-4 Nitric Oxide vs Overfire Air 43
5-5 Carbon Monoxide vs Overfire Air 46
5-6 Oxygen vs Grate Heat Release 50
5-7 Oxygen vs Grate Heat Release 52
5-8 Boiler Out Part, vs Grate Heat Release 53
5-9 Multiclone Out Part, vs Grate Heat Release 55
5-10 Nitric Oxide vs Grate Heat Release 56
5-11 Nitric Oxide vs Oxygen 58
5-12 Nitric Oxide vs Oxygen '. 59
5-13 Nitric Oxide vs Oxygen 60
5-14 Nitric Oxide vs Oxygen 61
5-15 Trend in Nitric Oxide Emissions as a Function of Oxygen ... 62
5-16 Carbon Monoxide vs Grate Heat Release 64
5-17 Carbon Monoxide vs Oxygen 65
5-18 Carbon Monoxide vs Oxygen 66
5-19 Boiler Out Comb vs Grate Heat Release 67
5-20 Bottom Ash Comb, vs Grate Heat Release 69
5-21 Boiler Efficiency vs Grate Heat Release 70
5-22 Size Consistency of "As Fired" Perfect 8 Coal vs ABMA Recom-
mended Sizing for Overfeed Stokers 79
5-23 Size Consistency of "As Fired" Century Coal vs ABMA Recom-
mended Sizing for Overfeed Stokers 80
5-24 Size Consistency of "As Fired" Victoria Coal vs ABMA Recom-
mended Sizing for Overfeed Stokers 81
5-25 Particle Size Distribution of the Boiler Outlet Flyash from
Bahco Classifier and Sieve Analysis 85
5-26 Particle Size Distribution at the Boiler Outlet from Brink
Cascade Impactor 86
5-27 Particle Size Distribution at the Boiler Outlet from SASS
Gravimetrics 87
5-28 Multiclone Efficiency vs Grate Heat Release 90
5-29 Smoke Spot Number vs Opacity 92
5-30 Smoke Spot Number vs Multiclone Outlet Part 93
v
-------
LIST _OF__TARLJ;:G
Tab IP
Page
2-1 Emission Data Summary
3-1 Design Data , 10
3-2 Predicted Performance and Acceptance Test 13
3-3 Average Coal Analysis 15
5-1 Effect of Overfire Air on Emissions and Efficiency - Perfect 8 . 40
5-2 Effect of Overfire Air on Emissions and Efficiency - Century . . 41
5-3 Nitric Oxide Emissions vs Overfire Air 44
5-4 Carbon Monoxide Emissions vs Overfire Air 47
5-5 Boiler Efficiency vs Overfire Air 48
5-6 Average Heat Losses by Coal Type 71
5-7 Fuel Analysis - Perfect 8 Coal 72
5-8 Fuel Analysis - Century Coal 73
5-9 Fuel Analysis - Victoria Coal 74
5-10 Mineral Analysis of Coal Ash 75
5-11 Coal Properties Corrected to a Constant 106Btu Basis 76
5-12 As Fired Coal Size Consistency 78
5-13 Sulfur Balance 82
5-14 Description of Particle Size Distribution Tests at the Boiler
Outlet 83
5-15 Results of Particle Size Distribution Tests at the Boiler Outlet 84
5-16 Efficiency of Multiclone Dust Collector 89
5-17 Modified Smoke Spot Data 91
5-18 Polynuclear Aromatic Hydrocarbons Analyzed in the Site D SASS
Sample 94
5-19 Particulate Emissions 95
5-20 Heat Losses and.Efficiencies 96
5-21 Percent Combustibles in Refuse 97
5-22 As Fired Coal Size Consistency 98
5-23 Steam Flows and Heat Release Rates 99
-------
l.U INTRODUCTION
The principal objective of the test program described in this report,
one of several reports in a series, is to produce information which will in-
crease the ability of boiler manufacturers to design and fabricate stoker boilers
that are an economical and environmentally satisfactory alternative to oil-
fired units. Further objectives of the program are to: provide information to
stoker boiler operators concerning the efficient operation of their boilers;
provide assistance to stoker boiler operators in planning their coal supply
contracts; refine application of existing pollution control equipment with
special emphasis on performance; and contribute to the design of new pollution
control equipment.
In order to meet these objectives, it is necessary to define stoker
boiler designs which will provide efficient operation and minimum gaseous and
particulate emissions, and define what those emissions are in order to facili-
tate preparation of attainable national emission standards for industrial size,
coal-fired boilers. To do this, boiler emissions and efficiency must be
measured as a function of coal analysis and sizing, rate of flyash reinjection,
overfire air admission, ash handling, grate size, and other variables for different
boiler,- furnace, and stoker designs.
A field test program designed to address the objectives out!ined above
was awarded to the American Boiler Manufacturers Association (ABMA), sponsored
by the United States Department of Energy (DOE) under contract number
EF-77-C-01-2609, and co-sponsored by the United States Environmental Protection
Agency (EPA) under inter^agency agreement number IAG-D7-E681. The program is
directed by an ABMA Stoker Technical Committee which, in turn, ha.-i subcontracted
the field test portion to KVB, Inc., of Minneapolis, Minnesota.
This report is the Final Technical Report for the fourth of eleven
boilers to be tested under the ABMA program. It contains a description of the
facility tested, the coals fired, the test equipment and procedures, and the
results and observations of testing. There is also a data supplement to this
report containing the '"raw" data sheets from the tests conducted. The data
KVB 15900-529
-------
supplement has the same EPA report number as this report except that it iy
followed by "b'' rather than "a" . As a compilation of all data obtained at
this test site, the supplement acts as a research tool for further data
reduction and analysis as new areas of interest are uncovered in subsequent
testing.
At the completion of this program, a Final Technical Report will com-
bine and correlate the test results from all sites tested. This final report
will provide the technical basis for the ABMA publication on "Design and
Operating Guidelines for Industrial Stoker Firing," and will be available
to interested parties through the ABMA, EPA, or DOE. A separate report covering
trace species data will also be written at the completion of this program.
It, too, will be available to interested parties through the ABMA, EPA, or
DOE.
Although it is EPA policy to use S.I. units in all EPA sponsored
reports, an exception has been made herein because English units have been
conventionally used to describe boiler design and operation. Conversion
tables are provided in the Appendix for those who prefer S.I. units.
To protect the interests of the host boiler facilities, each test
site in this program has been given a letter designation. As the fourth
site tested, this is the Final Technical Report for Test Site D under the
program entitled "A Testing Program to Update Equipment Specifications and
Design Criteria for Stoker Fired Boilers."
KVB 15900-529
-------
2.0 EXECUTIVE SUMMARY
A mass fed vibrating grate stoker rated at 90,000 Ibs steam/hour
was extensively tested for emissions and efficiency between July 28 and
September 15, 1978. Steam loading on this unit was limited to 88% of its
design rating due to draft losses associated with a retrofitted multiclone
dust collectori For consistency with other reports in this series, steam
loading will be presented as percent of design steam capacity. It should
be understood, however, that the unit was tested at its maximum derated load
whenever the reported load is in the range of 86 to 89% of design capacity -
This section summarizes the results of all tests performed at Site D and
provides references to supporting Figures, Tables and commentary found in
the main text of the report.
UNIT TESTED: Described in Section 3.0, pages 9-12
0 Babcock & Wilcox Boiler
Built 1964
Type FP — two-drum Stirling
90,000 Ib/hr rated capacity
200 psig operating steam pressure
Saturated steam
Economizer
0 Detroit Vibragrate Stoker
Mass fed
Vibrating inclined grate
No flyash reinjection
Two rows OFA on front water wall
COALS TESTED; Individual coal analysis results given in Tables 5-7, 5-8,
5-9 and 5-10, pages 72-75. Commentary in Section 3.0,
pages 12-15.
0 Perfect 8 Coal
13,003 Btu/lb
8.71% Ash
2.60% Sulfur
3.23% Moisture
2050°F initial ash deformation temperature
KVB 15900-529
-------
0 Century Coal
13,629 Btu/lb
5.86% Ash
1.23% Sulfur
2.68% Moisture
2680°F initial ash deformation temperature
0 Victoria Coal
13,405 Btu/lb
6.72% Ash
1.11% Sulfur
3.11% Moisture
2590°F initial ash deformation temperature
OVERFIRE AIR TEST RESULTS: Overfire air pressure was varied over its operating
range of 5" to 15" H2O, at low, medium and high
boiler loadings (Section 5.1, pages 35-48).
0 Particulate Loading
Mixed results were obtained. On Perfect 8 coal, particulate
loading increased with increasing OFA pressure. On Century
coal, particulate loading decreased with increasing overfire
a'ir pressure. There was insufficient data on Victoria coal to
draw any conclusions. (Section 5.1.1, page 37; Figures 5-2 and
5-3, pages 38, 39; Tables 5-1 and 5-2, pages 40, 41.)
0 Nitric Oxide•
Nitric oxide concentrations increased an average ten percent
when overfire air pressure was increased from 5" to 15" f^O
(Section 5.1.2, page 42; Table 5-3, page 44; Figure 5-4,
page 43).
0 Carbon Monoxide
At high loads, 13" to 15" H2O were necessary to control carbon
monoxide. At low loads, carbon monoxide concentrations remained
insignificant over the full operating range of the overfire air
system (Section 5.1.3, page 45; Table 5-4, page 47; Figure 5-5,
page 46).
0 Boiler Efficiency
The data indicate that maximum overfire air pressure will
minimize the heat loss due to combustibles in the flyash. Com-
bustible heat loss reductions of up to 0.47% were measured
(Section 5.1.4, page 45; Table 5-5, page 48).
KVB .15900-529
-------
BOILER EMISSION PROFILES: Boiler emissions were measured over the load range
45-89% of design capacity which corresponds to a
grate heat release range of 141,000 to 359,000
Btu/hr-ft2- Measured oxygen levels ranged from
7.3 to 12.5% (Section 5.2, pages 49-71).
0 Excess Oxygen Operating Levels
The excess oxygen operating level was highly dependent on
boiler load at Site D. As load increased, 02 decreased.
At maximum obtainable load (about 88% of design capacity) the
lowest obtainable 02 was 7.3%.* Extrapolation of the data
trend to 100% capacity would give an O2 of 6.4%. The boiler's
design operating level is 31% excess air, or about 5% O2
(Section 5.2.1, pages 49, 51; Figures 5-6 and 5-7, pages 50, 52)
*This lower O2 limit was a function of control limitations
and/or operator's reluctance to lower O2 further. De-
finitive 02 limits were not established on this boiler.
0 Particulate Loading
At maximum load the particulate loading averaged 1.03i.l9
lb/106Btu at the boiler outlet and 0.55±.14 lb/106Btu after
the multiclone dust collector. The particulate loadings
dropped more than 20% at loads below a grate heat release of
300,000 Btu/hr-ft2 (Section 5.2.2, pages 51, 54; Figures 5-8,
and 5-9, pages 51, 53).
0 Nitric Oxide
At maximum load the nitric oxide emissions averaged 0.28 lb/10^
Btu. NO emissions increased sharply as load decreased due to
the resulting increase in O2. Maximum measured NO was 0.46 lb/
lO^Btu at an operating level of 12.5% O2 (Section 5.2.3, pages
54, 57; Figures 5-10 through 5-15, pages 56, 58-62).
0 Carbon Monoxide
Carbon monoxide was found in significant concentrations only at
loads which resulted in a grate heat release above 300,000 Btu/
hr-ft2. At these loads it could be controlled by supplying
overfire air pressures greater than 13" H2O (Section 5.2.4,
page 63; Figures 5-16 through 5-18, pages 64-66).
0 Combustibles in Ash
Combustibles in the boiler outlet flyash were greatest under
high load, low 02 conditions where they ranged in value from 22
to 40% by weight. At lower loads they ranged from 14 to 35% by
weight. Combustibles in the bottom ash ranged in value from 8
to 36% by weight (Section 5.2.5, pages 63, 68; Figures 5-19 and
5-20, pages 67, 69).
KVB 15900-529
-------
BOILER EFFICIENCY: Boiler efficiency was determined for 15 tests using the
ASTM heat loss method. Boiler efficiency averaged 83.3%
and did not vary with boiler load (Section 5.2.6, page 68;
Figure 5-20, page 69; Table 5-6, page 71).
COAL PROPERTIES; Emissions and boiler efficiency were studied to determine
any effects which could be related to differences in the
properties of the three coals fired (Section 5.3, pages 71-
77) .
0 Sulfur Dioxide
Sulfur dioxide emissions were proportional to sulfur content
of the coals. Sulfur retention in the ash averaged two percent
of the fuel sulfur (Section 5.3.3, page 77; Table 5-13, page 82).
0 Combustibles in Bottom Ash
Combustibles in the bottom ash were highest for the two
Victoria coal tests (Figure 5-20, page 69).
0 Boiler Efficiency
Boiler efficiency was highest while burning Century coal. This
was due in part to a low dry gas loss and a low combustible
loss. Victoria coal showed the lowest boiler efficiency's due
to high combustible heat losses (Table 5-6, page 71; Figure
5-21, page 70).
0 Multiclone Collection Efficiency
Multiclone collection efficiency was highest while burning
Century coal, and was low for the single Victoria coal test.
Particle size distribution data helps explain this difference.
Century coal had 39% of its flyash below 20 micrometers in
diameter. This high percentage of fine particulate matter
helps explain the low collection efficiencies (Section 5.5,
page 88; Table 5-16, page 89; Figure 5-28, page 90).
0 Other Emissions
Boiler outlet particulate loading, nitric oxide, carbon monoxide
and flyash combustible levels were all unaffected by the dif-
ferences in the three coals tested.
KVB 15900-529
-------
SMOKE SPOT: A Bacharach smoke spot tester was used at the multi-done outlet
and one, two or three pumps .were taken instead of the specified
ten pumps used in oil fired units.
0 Results
The results did not correlate well with stack opacity or particu-
late loading (Section 5.6, page 88; Table 5-17, page 91;
Figures 5-29, 5-30, pages 92, 93).
SOURCE'ASSESSMENT SAMPLING SYSTEM: Flue gas was sampled for polynuclear
aromatic hydrocarbons and trace elements
during one test on Victoria coal. Data
will be presented in a separate report
at completion of test program (Section
5.7, page 94; Table 5-18, page 94).
The emissions data are summarized in Table 2-1 on the following page.
Other data tables are included at the end of Section 5.0, Test Results and
Observations. For reference, a Data Supplement containing all the unreduced
data obtained at Site D is available under separate .cover but with the same
title followed by the words "Data Supplement," and having the same EPA docu-
ment number followed by the letter "b" rather than "a". Copies of this
report and the Data Supplement are available through EPA and NTIS.
KVB 15900-529
-------
co
TABLE 2-1
EMISSION DATA SUMMARY - TEST SITE D
Test
No.
1
2
3
4
5
6
7
8
9
10
HA
11B
lie
11D
HE
11F
12
13A
13B
13C
13D
14
15
16
17
18++
19
20
21
22++
23
% Design
Date Capacity Coal Test Description
7/27/78
7/28/78
8/03/78
8/15/78
8/16/78
8/18/78
8/22/78
8/23/78
8/29/78
8/31/78
8/31/78
8/31/78
8/31/78
8/31/78
8/31/78
8/31/78
9/06/78
9/08/78
9/08/78
9/08/78
9/08/78
9/12/78
9/12/78
9/13/78
9/14/78
9/15/78
9/15/78
9/15/78
9/15/78
9/21/78
9/22/78
67
73
69
64
86
65
63
66
45
63
52
52
52
52
52
52
87
89
89
89
89
87
88
64
87
79
59
61
P
P
P
P
P
P
P
P
P
P
t
P
P
P
P
P
P
P
P
P
P
C
C
c
c
c
c
c
c
V
V
Med Load
Med Load
Med Load
Med Load
High Load
Med Load
Med Load
Med Load
Low Load
Med Load
Low Load
Low Load
Low Load
Low Load
Low Load
Low Load
High Load
High Load
High Load
High Load
High Load
High Load
High Load
Med Load
High Load
SOx - Normal
NOx (PDS Flask)
NOx (PDS Flask)
NOx (PDS Flask)
SASS s SOx - Normal
Med Load
- 6"
- 6"
- 6"
- 6"
- 10"
- 5"
- 15"
- 10"
-' 10"
- 10"
- 10"
- 5"
- 15"
- 10"
- 5"
- 15"
- 10"
- 5"
7"
- 15"
- 10"
7"
- 15"
- 10"
- 10"
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
OFA
Condi tions
- 12"
- 9"
- 15"
OFA
OFA
OFA
Conditions
- 13"
OFA
°2
dry
8
9
10
9
7
8
10
9
11
9
11
10
12
11
10
12
8
7
7
7
7
7
7
8
7
8
9
10
11
10
10
5
0
1
3
9
•7
2
1
5
9
,
5
5
0
8
0
4
8
4
6
6
5
8
7
3
1
0
5
5
2
4
C02
dry
10.9
10.2
9.3
9.8
11.2
10.9
9.9
10.0
8.5
10.3
8.5
9.2
7.8
8.8
9.0
8.2
11.4
11.9
12.1
12.0
12.0
12.2
11.6
10.8
12.0
11.2
10.4
9.5
8.5
10.0
9.9
CO
ppm
dry
112
108
24
89
87
29
57
56
61
65
126
85
72
129
80
321
2345
640
73
378
618
39
25
885
34
32
25
24
29
19
Excess
Air
63
69
75
74
55
65
88
80
114
83
112
96
143
106
102
129
61
54
49
52
52
51
54
65
48
58
70
94
115
89
92
N0
ppm NO NO2
dry lb/106Btu lb/106Btu
183
206
225
235
222
184
255
226
262
233
278
241
341
298
282
332
188
172
200
202
199
226
217
234
199
230
154
147
155
243
260
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
249
280
306
321
302
250 0.001
346 -0.008
306 -0.023
353 0.000
317
378
328
464
405
383
451
253 0.000
232
269 0.003
276
272
307 -0.001
294 0.000
314 0.000
269 0.000
312 -0.010
313
343
400
327
352
Part.
Blr Out
lb/106Btu
0
0
0
0
0
0
0
1
0
1
1
0
1
0
0
.664
.832
.694
.793
.788
.732
.707
.097
.837
.336
.000
.798
.072
.969
.762
Part.
D.C. Out
lb/106Btu
0
0
0
0.
0
0
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
355
557
574
372
488
328
456
354
667
490
754
424
319
393
421
595
Stack*'
Opacity
30
23
12
12
12
7
12
19
29
35
19
12
32
19
11
* The maximum obtainable load was 89% of the design capacity of 90,000 Ib stm/hr due to
draft losses from a retrofit dust collector.
** p - Perfect 8 Coal; C - Century Coal; V - Victoria Coal
*** Expressed as NC>2
t Stack opacity data based on transmissometer readings
ft SOx data was as follows: Test 18 SO2 - 2.402, 803 - 0.000 lb/106Btu
Test 22 SO2 - 2.425, SO3 - 0.019 lb/106Btu
KVB 15900-529
-------
3.0 DESCRIPTION OF FACILITY TESTED
AND COALS FIRED
This section discusses the general physical layout and operational
characteristics of the boiler tested at Test Site D. The coals utilized
in this test series are also discussed.
3.1 BOILER D DESCRIPTION
Boiler D is a Babcock & Wilcox type FP-18 No. 52 Integral Furnace
Boiler Unit, designed for 700 psig, and capable of a maximum continuous
capacity of 90,000 pounds of steam per hour at 200 psig and saturated tempera-
ture using feedwater at 250°F. The unit has a Detroit Stoker Company Vibra-
grate stoker, with water-cooled vibrating continuous ash discharge grate.
Coal is mass fed to the upper end of the vibrating grate and is shaken toward
the ash discharge end by the grate's vibrating action. There is no suspension
burning. Undergrate air can be controlled in five zones. Design data on the
boiler and stoker are presented in Table 3-1. Predicted performance data and
performance data from a 1964 acceptance test are given in Table 3-2.
The boiler is equipped with a OOP multiclone dust collector. There
is no flyash reinjection.
3.2 OVERFIRE AIR SYSTEM
The overfire air system on Boiler D consists of two rows of air jets
on the front wall. The nozzles are 1-1/4" in diameter and there are sixteen
nozzles in each row. The overfire air was found to be operating at about
10" H2O. At maximum flow the pressure is about 15" H2O. The overfire air
is supplied by an independent fan, having a 4000 CFM capacity.
KVB 15900-529
-------
TABU: 3-1
DESIGN DATA
TEST SITE D
BOILER: Manufacturer Babcock & Wilcox Company
Type Two-drum FP-18 Integral Furnace Boiler
Boiler Heating Surface 7706 ft2
Design Pressure 700 psig
Tube Diameter 2"
ECONOMIZER:
Type
Heating Surface
Design Pressure
Tube Diameter
Continuous Tube
4275 ft2
750 psig
2"
FURNACE: Volume
Projected Area of Water Cooled Furnace Walls
3520 ft3
737 ft2
STOKER: Manufacturer
Type
Width
Length
Effective Grate Area
Detroit Stoker Company
Vibra-Grate
15'6"
16'6.5"
256.3 ft2
HEAT RATES: Steam Flow
Input to Furnace
Furnace Width Heat Release
Grate Heat Release
Furnace Liberation
90,000 lbs/hr
102.8xl06Btu/hr
6.63xl06Btu/ft furnace width-hr
401x10 %tu/ft2-hr
29.7xl03Btu/ft3-hr
KVB 15900-529
10
-------
TABLE 3-2
PREDICTED PERFORMANCE AND ACCEPTANCE TEST
TEST SITE D
Steam Flow
Fuel
Excess Air Leaving Boiler
Coal Flow
Flue Gas Leaving Boiler
Air Leaving F.D. Fan
Steam Pressure at Boiler Outlet
Economizer to Drum Pressure Drop
Temp. Flue Gas Leaving Boiler
Temp. Flue Gas Leaving Economizer
Temp. Water Entering Economizer
Temp. Water Leaving Economizer
Temp. Air Entering F.D. Fan
Furnace Draft Loss
Boiler Draft Loss
Economizer Draft Loss
Damper and Flue Draft Loss
Total Draft Loss
Dry Gas Heat Loss
H2O and H2 in Fuel Heat Loss
Moisture in Air Heat Loss
Unburned Combustible Heat Loss
Radiation Heat Loss
Unaccounted for and Manufacturers Margin
Total Heat Loss
Efficiency of Unit
Predicted
90,000 Ibs/hr
Western Kentucky Coal
31%
8,090 Ibs/hr
112,500 Ibs/hr
94,200 Ibs/hr
200 psig
10 psig
673°F
370°F
250°F
342°F
80°F
0.1"H2O
1.5"H20
1.0"H20
1.1"H20
3.7"H20
7.0%
4.6%
0.2%
1.6%
0.5%
1.5%
15.4%
84.6%
Acceptance Test
2/20/64
88,389
31%
7,060 Ibs/hr
218 psig
677°F
398°F
262°F
342°F
67°F
0.08
1.04
1.58
86.7%
KVB 15900-529
11
-------
3.3 PARTICULATE COLLECTION EQUIPMENT
The boiler is equipped with a UOP multiclone dust collector. Some of
the tubes have been blanked off in order to increase the velocity through the
remaining cyclones and so improve the efficiency of the collector.
3.4 TEST PORT LOCATIONS
Emissions measurements were made at two locations -- at the boiler
outlet (before the multiclone dust collector) and at the stack outlet. The
locations of these sample sites are shown in Figure 3-1. Their geometry is
shown in Figure 3-2.
Whenever particulate loading was measured it was measured simultaneously
at both locations using 24-point sample traverses. Gaseous measurements of 02,
CO2f CO, and NO were obtained by pulling samples individually and compositely
from six probes distributed along the width of the boiler outlet duct. SOx
measurements and SASS samples for organic and trace element determinations
were each obtained from single points within the boiler outlet duct. For these
measurements a heated sample line was attached to one of the middle gaseous
probes at the boiler outlet. Its purpose was to eliminate losses due to con-
densation when measuring NO2 and unburned hydrocarbons.
3.5 COALS UTILIZED
Three coals were test fired at Test Site D. These were Perfect 8
coal, Century coal, and Victoria coal. Coal samples were taken for each test
involving particulate or SASS sampling. Most of the test data were obtained
on Perfect 8 coal, the primary coal at this site. Eight tests were run on
Century coal and only two tests on Victoria coal. Certain sections in this
report do not discuss Victoria coal because the data was insufficient to draw
meaningful conclusions. The average analyses obtained from these samples are
presented in Table 3-3. They show significant differences in moisture, sulfur
and Btu content. Individual analyses for each coal sample are presented in
Section 5.0, Test Results and Observations, Tables 5-7 through 5-11.
KVB 15900-529
12
-------
STACK SAMPLING
PLANE
BOILER OUTLET
SAMPLING PLANE
FIGURE 3-1. BOILER D SCHEMATIC
KVB 15900-529
13
-------
Boiler Outlet Sampling Plane
Cross Sectional Area = 64.82 ft2
-K,
13V
-*-:
14V
o.
D 0
O
o
15'10.5"
O
O
19',
100-7/8"
Particulate Sampling Points
Gaseous Sampling Points
SOx
SASS
Stack Sampling Plane
Cross Sectional Area =55.5 ft2
FIGURE 3-2. BOILER D SAMPLE PLANE GEOMETRY
KVB 15900-529
14
-------
TABLE 3-3
AVERAGE COAL ANALYSIS
TEST SITE D
PROXIMATE (As Rec'd)
% Moisture
% Ash
% Volatile
% Fixed Carbon
Century
Coal
2.68
5.86
38.44
53.02
Perfect 8
Coal
3.23
8.71
39.17
48.89
Victoria
Coal
3.11
6.72
37.25
52.94
Btu/lb
% Sulfur
13629
1.23
13003
2.60
13405
1.11
ULTIMATE (As Rec'd)
% Moisture
% Carbon
% Hydrogen
% Nitrogen
% Chlorine
% Sulfur
% Ash
% Oxygen (Diff)
2.68
76.08
5.15
1.33
0.13
1.23
5.86
7.54
3.23
72.07
5.01
1.23
0.12
2.60
8.71
7.03
3.11
74.96
5.00
1.30
0.07
1.11
6.72
7.76
KVB 15900-529
15
-------
4.0 TEST EQUIPMENT AND PROCEDURES
This section details how specific emissions were measured and the
sampling procedures followed to assure that accurate, reliable data were
collected.
4.1 GASEOUS EMISSIONS MEASUREMENTS (NOx, CO, CO?., O2 / HC)
A description is given below of the analytical instrumentation, re-
lated equipment, and the gas sampling and conditioning system, all of which
are located in a mobile testing van owned and operated by KVB. The systems
have been developed as a result of testing since 1970, and are operational
and fully checked out.
4.1.1 Analytical Instruments and Related Equipment
The analytical system consists of five instruments and associated
equipment for simultaneously measuring the constituents of flue gas. The
analyzers, recorders, valves, controls, and manifolds are mounted on a panel
in the vehicle. The analyzers are shock mounted to prevent vibration damage.
The flue gas constituents which are measured are oxides of nitrogen (NO, NOx),
carbon monoxide (CO), carbon dioxide (CO2), oxygen (02), and gaseous hydro-
carbons (HC) .
Listed below are the measurement parameters, the analyzer model
furnished, and the range and accuracy of each parameter for the system. A
detailed discussion of each analyzer follows:
Constituent: Nitric Oxide/Total Oxides of Nitrogen (NO/NOx)
Analyzer: Thermo Electron Model 10 Chemiluminescent Analyzer
Range: 0-2.5, 10, 25, 100, 250, 1000, 2500, 10,000 ppm NO
Accuracy: +1% of full scale
Constituent: Carbon Monoxide
Analyzer: Beckman Model 315B NDIR Analyzer
Range: 0-500 and 0-2000 ppm CO
Accuracy: il% of full scale
KVB 15900-529
17
-------
Constituent: Carbon Dioxide
Analyzer: Beckman Model 864 NDIR Analyzer
Range: 0-5% and 0-20% CC>2
Accuracy: -1% of full scale
Constituent: Oxygen
Analyzer: Teledyne Model 326A Fuel Cell Analyzer
Range: 0-5, 10 and 25% O2 full scale
Accuracy: il% of full scale
Constituent: Hydrocarbons
Analyzer: Beckman Model 402 Flame lonization Analyzer
Range: 5 ppm full scale to 10% full scale
Accuracy: ±1% of full scale
Oxides of nitrogen. The instrument used to monitor oxides of nitrogen
is a Thermo Electron chemiluminescent nitric oxide analyzer. The instrument
operates by measuring the chemiluminescent reaction of NO and 03 to form NO2.
Light is emitted when electronically excited NO2 molecules revert to their
ground state. The resulting chemiluminescence is monitored through an optical
filter by a high sensitivity photomultiplier, the output of which is linearly
proportional to the NO concentration.
Air for the ozonator is drawn from ambient air through a dryer and
a ten micrometer filter element. Flow control for the instrument is accomplished
by means of a small bellows pump mounted on the vent of the instrument down-
stream of a separator that prevents water from collecting in the pump.
The basic analyzer is sensitive only to NO molecules. To measure NOx
(i.e., NO+NO2), the NO2 is first converted to NO. This is accomplished by a
converter which is included with the analyzer. The conversion occurs as the
gas passes through a thermally insulated, resistance heated, stainless steel
coil. With the application of heat, NO2 molecules in the sample gas are reduced
to NO molecules, and the analyzer now reads NOx. NO2 is obtained by the dif-
ference in readings obtained with and without the converter in operation.
Specifications: Accuracy 1% of full scale
Span stability ±1% of full scale in 24 hours
Zero stability ±1 ppm in 24 hours
Power requirements 115±10V, 60 Hz, 1000 watts
Response 90% of full scale in 1 sec. (NOx mode),
0.7 sec. NO mode
Output 4-20 ma
KVB 15900-529
18
-------
Sensitivity 0.5 ppm
Linearity il% of full scale
Vacuum detector operation
Range: 2.5, 10, 25, 100, 250, 1000, 2500, 10,000 ppm
full scale
Carbon monoxide. Carbon monoxide concentration is measured by a
Beckman 315B non-dispersive infrared analyzer. This instrument measures the
differential in infrared energy absorbed from energy beams passed through a
reference cell (containing a gas selected to have minimal absorption of infra-
red energy in the wavelength absorbed by the gas component of interest) and a
sample cell through which the sample gas flows continuously. The differential
absorption appears as a reading on a scale from 0 to 100 and is then related
to the concentration of the specie of interest by calibration curves supplied
with the instrument. The operating ranges for the CO analyzer are 0-500 ppm
and 0-2000 ppm.
Specifications: Span stability ±1% of full scale in 24 hours
Zero stability -1% of full scale in 24 hours
Ambient temperature range 32°F to 120°F
Line voltage 115JT15V rms
Response 90% of full scale in 0.5 or 2.5 sec.
Precision -1% c
Output 4-20 ma
Precision -1% of full scale
Carbon dioxide. Carbon dioxide concentration is measured by a Beckman
Model 864 short path-length, non-dispersive infrared analyzer. This instrument
measures the differential in infrared energy absorbed from energy beams passed
through a reference cell (containing a gas selected to have minimal absorption
of infrared energy in the wavelength absorbed by the gas component of interest)
and a sample cell through which the sample gas flows continuously- The dif-
ferential absorption appears as a reading on a scale from 0 to 100 and is then
related to the concentration of the specie of interest by calibration curves
supplied with the instrument. The operating ranges for the CO2 analyzer are
0-5% and 0-20%.
Specifications: Span stability -1% of full scale in 24 hours
Zero stability -1% of full scale in 24 hours
Ambient temperature range 32°F to 120°F
Line voltage 115-15V rms
Response 90% of full scale in 0.5 or 2.5 sec.
Precision -1% of full scale
Output 4-20 ma
19 KVB 15900-529
-------
Oxygen. The oxygen content of the flue gas sample is automatically
and continuously determined with a Teledyne Model 326A Oxygen analyzer.
Oxygen in the flue gas diffuses through a Teflon membrane and is reduced
on the surface of the cathode. A corresponding oxidation occurs at the anode
internally and an electric current is produced that is proportional to the
concentration of oxygen. This current is measured and conditioned by the
instrument's electronic circuitry to give a final output in percent 02 by
volume for operating ranges of 0% to 5%, 0% to 10%, or 0% to 25%.
Specifications: Precision -1% of full scale
Response 90% in less than 40 sec.
Sensitivity 1% of low range
Linearity ±1% of full scale
Ambient temperature range 32-125°F
Fuel cell life expectancy 40,000%-hours
Power requirement 115 VAC, 50-60 Hz, 100 watts
Output 4-20 ma
Hydrocarbons. Hydrocarbons are measured using a Beckman Model 402
hydrocarbon analyzer which utilizes the flame ionization method of detection.
The sample is drawn to the analyzer through a heated line to prevent the loss
of higher molecular weight hydrocarbons. It is then filtered and supplied -to
the burner by means of a pump and flow control system. The sensor, which is
the burner, has its flame.sustained by regulated flows of fuel (40% hydrogen
plus 60% helium) and air. In the flame, the hydrocarbon components of the
sample undergo a complete ionization that produces electrons and positive ions.
Polarized electrodes collect these ions, causing a small current to flow through
a circuit. This ionization current is proportional to the concentration of
hydrocarbon atoms which enter the burner. The instrument is available with
range selection from 5 ppm to 10% full scale as CH^.
Specifications: Full scale sensitivity, adjustable from 5 ppm CH4 to
10% CH4
Ranges: Range multiplier switch has 8 positions: XI,
X5, X10, X50, X100, X500, XlOOO, and X5000. In
addition, span control provides continuously variable
adjustment within a dynamic range of 10:1
Response time 90% full scale in 0.5 sec.
Precision ±1% of full scale
Electronic stability -1% of full scale for successive
identical samples
KVB 15900-529
20
-------
Reproducibility ±1% of full scale for successive
identical samples
Analysis temperature: ambient
Ambient temperature 32 °F to 110°F
Output 4-20 ma
Air requirements 350 to 400 cc/min of clean, hydro-
carbon-free air, supplied at 30 to 200 psig
Fuel gas requirements 75 to 80 cc/min of pre-mixed
fuel consisting of 40% hydrogen and 60% nitrogen
or helium, supplied at 30 to 200 psig
Electrical power requirements 120V, 60 Hz
Automatic flame-out indication and fuel shut-off valve
4.1.2 Recording Instruments
The output of the four analyzers is displayed on front panel meters
and are simultaneously recorded on a Texas Instrument Model FLO4W6D four-pen
strip chart recorder. The recorder specifications are as follows:
Chart size 9-3/4 inch
Accuracy ±0.25%
Linearity <0.1%
Line voltage 120V±10% at 60 Hz
Span step response: one second
4.1.3 Gas Sampling and Conditioning System
The gas sampling and conditioning system consists of probes, sample
lines, valves, pumps, filters and other components necessary to deliver a
representative, conditioned sample gas to the analytical instrumentation. The
following sections describe the system and its components. The entire gas
sampling and conditioning system shown schematically in Figure 4-1 is contained
in the emission test vehicle.
4.1.4 Gaseous Emission Sampling Techniques
Boiler access points for gaseous sampling are selected in the same sample
plane as are particulate sample points. Each probe consists of one-half inch
316 stainless steel heavy wall tubing. A 100 micrometer Mott Metallurgical
Corporation sintered stainless steel filter is attached to each probe for
removal of particulate material.
KVB 15900-529
.21
-------
HO- NO*
sr-<4™
w
,„
"°,"~ h^
Th"" "
-- :""")
£ ; .;.;.;
= ] 1 .r.p.n.rj
...I
Hralfl
rilror
Jl
n1
i
FIGURE 4-1. FLOW SCHEMATIC OF MOBILE FLUE GAS MONITORING LABORATORY
KVB 15900-529
-------
Gas samples to be analyzed for 02, CO2» CO and NO are conveyed to
the KVB mobile laboratory through 3/8 inch nylon sample lines. After
passing through bubblers for flow control, the samples pass through a dia-
phragm pump and a refrigerated dryer to reduce the sample dew point temper-
ature to 35°F. After the dryer, the sample gas is split between the various
continuous gas monitors for analysis. Flow through each continuous monitor
is accurately controlled with rotometers. Excess flow is vented to the out-
side. Gas samples are drawn both individually and compositely from all
probes during each test. The average emission values are reported in this
report.
4.2 SULFUR OXIDES (SOx) MEASUREMENT AND PROCEDURES
Measurement of SO2 and 803 concentrations is made by wet chemical
analysis using the "Shell-Emeryville" method. In this technique the gas
sample is drawn from the stack through a glass probe (Figure 4-2), containing
a quartz wool filter to remove particulate matter, into a system of three
sintered glass plate absorbers (Figure 4-3). The first two absorbers contain
aqueous isopropyl alcohol and remove the sulfur trioxide; the third contains
aqueous hydrogen peroxide solution which absorbs the sulfur dioxide. Most
of the sulfur trioxide is removed by the first absorber. The remainder,
which passes through the absorber, is completely removed by the secondary
absorber mounted above the first. After the gas sample has passed through the
absorbers, the gas train is purged with nitrogen to transfer sulfur dioxide,
which has dissolved in the first two absorbers, to the third absorber to
complete the separation of the two components. The isopropyl alcohol is used
to inhibit the oxidation of sulfur dioxide to sulfur trioxide before it gets
to the third absorber.
The isopropyl alcohol absorber solutions are combined and the sulfate
resulting from the sulfur trioxide absorption is titrated with standard lead
perchlorate solution using Sulfonazo III indicator- In a similar manner, the
hydrogen peroxide solution is titrated for the sulfate resulting from the
sulfur dioxide absorption.
The gas sample is drawn from the flue by a single probe made of
quartz glass inserted into the duct approximately one-third to one-half way.
23 KVB 15900-529
-------
Flue Wall
Asbestos Plug
Ball Joint
Vycor
Samole Probe
Pryometer
and
Thermocouple
FIGURE 4-2. SOX SAMPLE PROBE CONSTRUCTION
Spray Trap
Dial Thermometer
Pressure Gauge.
Volume Ir.dica-. \
.Lee
Vapor Trap Diaphragm
Pumo
Dry Test Meter
FIGURE 4-3. SULFUR OXIDES SAMPLING TRAIN
KVB 15900-529
24
-------
The inlet end of the probe holds a quartz wool filter to remove particulate
matter. It is important that the entire probe temperature be kept above
the dew point of sulfuric acid during sampling (minimum temperature of
260°C). This is accomplished by wrapping the probe with a heating tape.
Three repetitions of SOx sampling are made at each test point.
4.3 PARTICULATE MEASUREMENT AND PROCEDURES
Particulate samples are taken at the same sample ports as the gaseous
emission samples using a Joy Manufacturing Company portable effluent sampler
(Figure 4-4). This system, which meets the EPA design specifications for
Test Method 5, Determination of Particulate Emissions from Stationary Sources
(Federal Register, Volume 36, No. 27, page 24888, December 23, 1971), is used
to perform both the initial velocity traverse and the particulate sample
collection. Dry particulates are collected in a heated case using first a
cyclone to separate particles'larger than five micrometers and a 100 mm glass
fiber filter for retention of particles down to 0.3 micrometers. Condensible
particulates are collected in a train of four Greenburg-Smith impingers in an
ice water bath. The control unit includes a total gas meter and thermocouple
indicator. A pitot tube system is provided for setting sample flows to obtain
isokinetic sampling conditions.
All peripheral equipment is carried in the instrument van. This
includes a scale (accurate to io.l mg), hot plate, drying oven (212°F), high
temperature oven, desiccator, and related glassware. A particulate analysis
laboratory is set up in the vicinity of the boiler in a vibration-free area.
Here filters are prepared, tare weighed and weighed again after particulate
collection. Also, probe washes are evaporated and weighed in the lab.
4.4 PARTICLE SIZE DISTRIBUTION MEASUREMENT AND PROCEDURE
Particle size distribution is measured using several methods. These
include the Brink Cascade Impactor, SASS cyclones, and the Bahco Classifier.
Each of these particle sizing methods has its advantages and disadvantages.
KVB 15900-529
25
-------
PROBE
THERMOMETER
PROBE
HEATED AREA
STACK
WALL
THERMOMETER
FILTER HOLDER
STACK
THERMOMETER-
REVERSE-TYPE
PITOT TUBE
VELOCITY
PRESSURE
GAUGE
\
IMPINGERS ICE BATH
THERMOMETERS FINE CONTROL VALVE
ORIFICE
GAUGE
CHECK VALVE
VACUUM LINE
VACUUM
GAUGE
COARSE CONTROL VALVE
DRY TEST METER
AIR-TIGHT
PUMP
FIGURE 4-4. EPA METHOD 5 PARTICIPATE SAMPLING TRAIN
KVB 15900-529
-------
Brink. The Brink cascade impactor is an in situ particle sizing de-
vice which separates the particles into six size classifications. It has the
advantage of collecting the entire sample. That is, everything down to the
collection efficiency of the final filter is included in the analysis. It
has, however, some disadvantages. If the particulate matter is spatially
stratified within the duct, the single-point Brink sampler will yield
erroneous results. Unfortunately.- the particles at the outlets of stoker
boilers may be considerably stratified. Another disadvantage is the instru-
ment's small classification range (0.3 to 3.0 micrometers) and its small sample
nozzle (1.5 to 2.0 mm maximum diameter). Both are inadequate for the job at
hand. The particles being collected at the boiler outlet are often as large
as the sample nozzle.
The sampling procedure is straight forward. First, the gas velocity
at the sample point is determined using a calibrated S-type pitot tube. For
this purpose a hand held particulate probe, inclined manometer, thermocouple
and indicator are used. Second, a nozzle size is selected which will main-
tain isokinetic flow rates within the recommended .02-.07 ft3/min rate at
stack conditions. Having selected a nozzle and determined the required flow
rate for isokinetics, the operating pressure drop across the impactor is
determined from a calibration curve. This pressure drop is corrected for
temperature, pressure and molecular weight of the gas to be sampled.
A sample is drawn at the predetermined AP for a time period which is
dictated by mass loading and size distribution. To minimize weighing errors,
it is desirable to collect several milligrams on each stage. However, to
minimize reentrainment, a rule of thumb is that no stage should be loaded
above 10 mg. A schematic of the Brink sampling train is shown in Figure 4-5.
Bahco. The Bahco classifier is described in Power Test Code 28. It
is an acceptable particle sizing method in the power industry and is often
used in specifying mechanical dust collector guarantees. Its main disadvantage
is that it is only as accurate as the sample collected. Most Bahco samples are
collected by cyclone separation; thus, particles below the cut point of the
cyclone are lost. The Bahco samples collected at Test Site D came from the
cyclone in the EPA Method 5 particulate train. These samples are spatially
representative because they are taken from a 24-point sample matrix. However,
KVB 15900-529
27
-------
PRESSURE TAP
FOR 0-20"
MAGNAHELIX
CYCLOME
STAGE 1
STAGE 2
STAGE 3
STAGE 4
STAGE 5
FINAL FILTER
ELECTRICALLY HEATED PROBE
DRY GAS
METER
FLOW CONTROL
VALVE
DRYING
COLUMN
FIGURE 4-5. BRINK CASCADE IMPACTOR SAMPLING TRAIN SCHEMATIC
KVB 15900-529
28
-------
much of the sample below about seven micrometers is lost to the filter. The
Bahco test data are presented in combination with sieve analysis of the same
sample. An attempt was made to correct for the lost portion of the sample.
SASS. The Source Assessment Sampling System (SASS) was not designed
principally as a particle sizer but it includes three calibrated cyclones
which can be used as such. The SASS train is a single point in-situ sampler.
Thus, it is on a par with cascade impactors. Because it is a high volume
sampler and samples are drawn through large nozzles (0.25 to 1.0 in.), it
has an advantage over the Brink cascade impactor where large particles are
involved. The cut points of the three cyclones are 10, 3 and 1 micrometers.
A detailed description of the SASS train is presented in Section 4.9.
4.5 COAL SAMPLING AND ANALYSIS PROCEDURE
Coal samples at Test Site D were taken during each test from the
unit's two coal scales. The samples were processed and analyzed for both
size consistency and chemical composition. The use of the coal scale as
a sampling station has two advantages. It is close enough to the furnace
that the coal sampled simultaneously with testing is representative of the
coal fired during the testing. Also, because of the construction of the
coal scale, it is possible to collect- a complete cut of coal off the scales'
apron feeder thus insuring a representative size consistency.
In order to collect representative coal samples, a sampling tray
having a twenty pound capacity was custom built. The tray has the same width
as the apron feeder belt and can be moved directly under the belt's discharge
end to catch all of the coal over a short increment of time (approximately
five seconds).
The sampling procedure is as follows. At the start of testing one
increment of sample is collected from each feeder. This is repeated twice more
during the test (three to five hours duration) so that a six increment sample
is obtained. The sample is then riffled using a Gilson Model SP-2 Porta
Splitter until two representative twenty pound samples are obtained.
KVB 15900-529
29
-------
The sample to be used for sieve analysis is weighed, dried in an
oven at 200°F for about four hours, and re-weighed. Drying of the coal is
necessary for good separation of fines. If the coal is wet, fines cling to
the larger pieces of coal and to each other. Once dry, the coal is sized
using a six tray Gilson Model PS-3 Porta Screen. Screen sizes used are 1",
1/2", 1/4", #8 and #16 mesh. Screen area per tray is 14"xl4". The coal in
each tray is weighed on a triple beam balance to the nearest 0.1 gram.
The coal sample for chemical analysis is reduced to 2-3 pounds by
further riffling and sealed in a plastic bag. All coal samples are sent to
Commercial Testing and Engineering Company, South Holland, Illinois. Each
sample associated with a particulate loading or particle sizing test is
given a proximate analysis. In addition, composite samples consisting of
one increment of coal for each test for each coal type receive ultimate
analysis, ash fusion temperature, mineral analysis, Hardgrove grindability
and free swelling index measurements.
4.6 ASH COLLECTION AND ANALYSIS FOR COMBUSTIBLES
The combustible content of flyash is determined in the field by KVB
in accordance with ASTM D3173, "Moisture in the Analysis Sample of Coal and
Coke" and ASTM D3174, "Ash in the Analysis Sample of Coal and Coke."
The flyash sample is collected by the EPA Method 5 particulate sample
train while sampling for particulates. The cyclone catch is placed in a desic-
cated and tare-weighed ceramic crucible. The crucible with sample is heated
in an oven at 230°F to remove its moisture. It is then desiccated to room
temperature and weighed. The crucible with sample is then placed in an
electric muffle furnace maintained at a temperature of 1400°F until ignition
is complete and the sample has reached a constant weight. It is cooled in a
desiccator over desiccant and weighed. Combustible content is calculated as
the percent weight loss of the sample based on its post 230°F weight.
KVB 15900-529
30
-------
At Test Site D the bottom ash samples were collected in several in-
crements from the stoker ash pit at completion of testing. These samples
were mixed, quartered, and sent to Commercial Testing and Engineering
Company for combustible determination. Multiclone ash samples were taken
from ports near the base of the multiclone hopper. This sample, approximately
two quarts in size, was sent to Commercial Testing and Engineering Company
for combustible determination.
4.7 BOILER EFFICIENCY EVALUATION
Boiler efficiency is calculated using the ASME Test Form for Abbre-
viated Efficiency Test, Revised, September, 1965. The general approach to
efficiency evaluation is based on the assessment of combustion losses. These
losses can be grouped into three major categories: stack gas losses, com-
bustible losses, and radiation losses. The first two groups of losses are
measured directly. The third is estimated from the ABMA Standard Radiation
Loss Chart.
Unlike the ASME test in which combustible losses are lumped into one
category, combustible losses are calculated and reported separately for com-
bustibles in the bottom ash, combustibles in the mechanically collected ash
which is not reinjected, and combustibles in the flyash leaving the mechanical
collector.
4.8 MODIFIED SMOKE SPOT NUMBER
Modified Bacharach smoke spot numbers are determined using a Bacharach
field service type smoke tester. ASTM procedures for this measurement apply
only to oil fired units. Therefore, KVB defined its own set of procedures,
which differ from ASTM D2156-65 procedure in the number of strokes taken with
the hand pump. At this test site, one, two and three strokes were taken at
the multiclone outlet.
Smoke spot measurements are obtained by pulling a fixed volume of flue
gas through a standard filter paper. The color (or shade) of the spot that is
KVB 15900-529
31
-------
produced is matched visually with a standard smoke spot scale. The result
is a "smoke number" which is used to characterize the density of smoke in
the flue gas.
The sampling device is a hand pump similar to the one shown in
Figure 4-6. It is a commercially available item that with ten strokes can
pass 2,250 -100 cubic inches of gas at 60°F and one atmosphere pressure
through an enclosed filter paper for each 1.0 square inch effective surface
area of the filter paper.
Sampling Tube
Filter Paper
Handle
Plunger
FIGURE 4-6. FIELD SERVICE TYPE SMOKE TESTER
KVB 15900-529
32
-------
4.9 TRACE SPECIES MEASUREMENT
The EPA (IERL-RTP) has developed the Source Assessment Sampling
System (SASS) train for the collection of particulate and volatile matter
in addition to gaseous samples (Figure 4-7). The "catch" from the SASS
train is analyzed for polynuclear aromatic hydrocarbons (PAH) and inorganic
trace elements.
In this system, a stainless steel heated probe is connected to an
oven module containing three cyclones and a filter. Size fractionation is
accomplished in the series cyclone portion of the SASS train, which incor-
porates the cyclones in series to provide large quantities of particulate
matter which are classified by size into three ranges:
A) >10 urn B) 3 ym to 10 ym C) 1 urn to 3 urn
Together with a filter, a fourth cut (<1 ym) is obtained. Volatile organic
material is collected in an XAD-2 sorbent trap. The XAD-2 trap is an integral
part of the gas treatment system which follows the oven containing the cyclone
system. The gas treatment system is composed of four primary components:
the gas conditioner, the XAD-2 organic sorbent trap, the aqueous condensate
collector, and a temperature controller. The XAD-2 sorbent is a porous polymer
resin with the capability of absorbing a broad range of organic species.
Some trapping of volatile inorganic species is also anticipated as a result
of simple impaction. Volatile inorganic elements are collected in a series
of impingers. The pumping capacity is supplied by two 10 cfm high volume
vacuum pumps, while required pressure, temperature, power and flow conditions
are obtained from a main controller.
KVB 15900-529
33
-------
Stock T.C.
Convection
OVPR
Filter
Gas cooler
Stack velocity (AP)
magnehelic gauges
1 '>'
w
u
11
1
1
1
i
1
1
-J
«*
—
«
•
Gas
tempera
/ T.C.
-3 1
Orifice AH^
magnehelic gauqe
T.C.
Sorbent
cartrldge-
Condensate
~collector
Imp/cooler
trace element
col lector
Gas
meter
T.C.
Coarse adjustment
Fine valve
X adjustment
[< valve Q' \ (\;
Vacuum pumps
Dry test meter
Impinger
T.C.
Vacuum
gage
FIGURE 4-7. SOURCE ASSESSMENT SAMPLING (SASS) FLOW DIAGRAM
KVB 15900-529
-------
5.0 TEST RESULTS AND OBSERVATIONS
This section presents the results of the tests performed on Boiler D.
Observations are made regarding the influence on efficiency and gaseous and
particulate emissions as the control parameters were varied. Twenty-
three tests were conducted in a defined test matrix to develop this data.
Data Tables 5-19 through 5-23 are included at the end of this section for
reference.
5.1 OVERFIRE AIR
Boiler D had two rows of overfire air jets on the front water wall
configured as shown in Figure 5-1. Several test series were run in which
overfire air pressure (and thus overfire air flow) was the independent
variable. Emissions and boiler efficiency were measured as the overfire air
pressure was varied over its operating range of 5 to 15" I^O static pressure.
The purpose of these tests was to determine which overfire air pressure
settings were optimum in terms of emissions and efficiency. Tests were run
with both Perfect 8 coal and Century coal.
Perfect 8 coal exhibited increases in particulate loading and in nitric
oxide concentrations as overfire air pressure was increased. Carbon carryover
decreased only slightly with increasing overfire air, and boiler efficiency
changes were negligible. At loads of 88% capacity, 13 to 15" H2O pressure
were required to control carbon monoxide levels. However, at lower loads —
in the range of 50 to 65% of capacity — carbon monoxide emissions were negli-
gible at all overfire air levels. The tests showed that 7 to 10" H2O pressure
would be best at low and medium loads because it would help minimize particulate
and nitric oxide concentrations.
The test results for Century coal were largely the reverse of those
for Perfect 8. Both particulate loading and nitric oxide concentrations de-
creased with increasing overfire air pressure at 88% of steam capacity. The
heat loss due to combustibles in the flyash decreased nearly 0.5% as overfire
35 KVB 15900-529
-------
24" 9" 14' 14 9 9" 14 9 9 9 14 9 9 14 14 9 24
-©—® © (j) 0 CD—CD (j) (j) (3—CD CD CD - -© CD (J)
OVERFIRE AIR NOZZLES
(APPROXIMATE MEASUREMENTS)
OVERFIRE AIR
NOZZLES
FIGURE 5-1. OVERFIRE AIR CONFIGURATION
KVB 15900-529
36
-------
air was increased from 7 to 15 "H2O pressure. Overfire air pressures of 13
to 15 "H2O were again necessary to control carbon monoxide emissions at 88%
steam capacity. Lower steam loadings were not tested.
These test results are discussed in greater detail in the following
paragraphs.
5.1.1 Particulate Loadings vs Overfire Air
Five tests were run on Perfect 8 coal and four on Century coal to
determine the effect of overfire air rates on particulate emissions. The
results are shown in Figures 5-2 and 5-3 and in Tables 5-1 and 5-2.
For Perfect 8 coal the particulate loading at the boiler outlet
tended to increase slightly with increasing overfire air flow. Most of this
increase was in the inorganic fraction. Percent combustibles in the boiler
outlet flyash decreased slightly with increasing overfire air flow.
For Century coal the particulate loading trend was just the reverse.
For this coal the particulate loading at the boiler outlet tended to decrease
with increasing overfire air flow. Most of this decrease was in the combus-
tible fraction. Increasing overfire air pressure from 7 to 10 "H2O pressure
was very effective in reducing the combustibles from Century coal. This may
be partly due to the fact that there was a significant concentration (40% in
Test 14) of combustibles in the Century coal flyash at 7" overfire air pressure
which could'be reduced by additional overfire air. By comparison, the equivalent
firing conditions on Perfect 8 coal (Test 13B) produced only 28% combustibles
in the flyash.' Overfire air increases above 10 "f^O pressure had only a small
effect on Century coal particulate loadings.
In general, the multiclone outlet particulate emissions followed the
trends of the boiler outlet emissions. However, the multiclone outlet emissions
were also affected by the multiclone collection efficiency, which varied
between 38 and 63% during these tests.
In summary, the impact of overfire air on particulate emissions varied
with the coal being fired. For both coals, combustible loadings decreased
slightly with increasing overfire air flow. On the other hand, inorganic ash
loading trends were in opposite directions for the two coals and were more
37 KVB 15900-529
-------
9
s
a
hi
o
g
D
o
H
O
ffl
1.6 .
1.4
1.2
1.0
O.S
0.6
0.4
0.2
COMBUSTIBLES
INORGANIC ASH
TEST NO. 6 8 7
OFA PRES. "H2O 5 9.8 15.5
OXYGEN (%) 8.7 9.7 10.2
LOAD (%) 65 66 63
13B
7
7.5
89
12
10
8.4
87
FIGURE 5-2. PARTICULATE LOADING BREAKDOWN FOR PERFECT 8 COAL
AS A FUNCTION OF OVERFIRE AIR
KVB 15900-529
38
-------
D
E-i
pq
9
is
H
Q
U
EH
«
<
CM
EH
1.6
1.4
1.2
1.0
0.8
0.6
0.4 .
0.2
H| COMBUSTIBLES
|[ INORGANIC ASH
TEST NO.
OFA PRES. "H20
OXYGEN (%)
LOAD (%)
14
7
7.5
87
17
10
7.3
87
18
13
3.1
79
15
15.5
7.8
88
FIGURE 5-3.
PARTICULATE LOADING BREAKDOWN FOR CENTURY COAL
AS A FUNCTION OF OVERFIRE AIR
KVB 15900-529
39
-------
TABLE 5-1
EFFECT OF OVERFIKE AIR ON EMISSIONS & EFFICIENCY
PERFECT 8 COAL - TEST SITE D
TEST NO.
Description
OVERFIRE AIR CONDITIONS
Front Upper, "H2O
Front Lower, "HjO
FIRING CONDITIONS
Load, % of Design Capacity*
Grate Heat Release, 103Btu/ft2/hr
Coal Sizing, % Passing 1/4"
Excess Air, %
BOILER OUTLET EMISSIONS
Particulate Loading, lb/10 Btu
Combustible Loading, lb/106Btu
Inorganic Ash Loading, lb/10^Btu
Combustibles in Flyash, %
02, % (dry) .
CO, ppm (dry) @ 3% 02
NO, lb/106Btu
MULTICLONE OUTLET EMISSIONS
Particulate Loading, lb/10^Btu
Combustible Loading, Ib/lO^Btu
Inorganic Ash Loading, lb/10^Btu
Combustibles in Flyash, %
Multiclone Collection Efficiency, %
HEAT LOSSES, %
Dry Gas Loss
Moisture in Fuel
H20 from Combustion of H2
Combustibles in Collected Flyash
Combustibles in Emitted Flyash
Combustibles in Bottom Ash
Radiation Loss
Unmeasured Losses
Total Losses
Boiler Efficiency
6
Low OFA
Med Load
5.0
5.3
65
250
9
65
0.694
0.152
0.542
21.9
8.7
87
0.250
0.372
0.051
0.321
13.6
46.4
8.20
0.28
4.03
0.10
0.22
1.74
0.72
1.50
16.79
83.21
8
Med OFA
Med Load
10.0
9.5
66
221
8
80
0.788
0.164
0.624
20.8
9.7
57
0.306
0.328
0.029
0.299
8.90
58.4
9.20
0.27
4.03
0.12
0.24
2.07
0.72
1.50
18.15
81.85
7
High OFA
Med Load
15.5
15.5
63
249
5
88
0.793
0.127
0.666
16.0
10.2
29
0.346
0.488
0.052
0.436
10.7
38.5
8.80
0.30
4.05
0.14
0.18
2.04
0.74
1.50
17.75
82.25
13B
Low OFA
High Load
7.0
7.0
89
344
9
49
0.837
0.236
0.601
28.2
7.5
1225
0.269
0.490
0.043
0.447
8.85
41.5
7.25
0.30
4.01
0.09
0.35
2.63
0.53
1.50
16.66
83.34
12
Med OFA
High Load
10.0
10.0
87
326
9
61
1.097
0.242
0.855
22.1
8.4
321
0.253
0.667
0.059
0.608
8.86
39.2
8.10
0.32
4.02
0.05
0.21
2.00
0.54
1.50
16.74
83.26
* Design capacity of 90,000 Ib stm/hr was not met due to draft
losses resulting from a retrofit dust collector.
KVB 15900-529
40
-------
TABLE 5-2
EFFECT OF OVERFIRE AIR ON EMISSIONS & EFFICIENCY
CENTURY COAL - TEST SITE D
TEST NO.
Description
OVERFIRE AIR CONDITIONS
Front Upper, "H20
Front Lower, "H20
FIRING CONDITIONS
Load, % of Design Capacity*
Grate Heat Release, 103Btu/ft2/hr
Coal Sizing, % Passing 1/4"
Excess Air, %
BOILER OUTLET EMISSIONS
Particulate Loading, lb/106Btu
Combustible Loading, lb/106Btu
Inorganic Ash Loading, lb/10^Btu
Combustibles in Flyash, %
02, % (dry)
CO, ppm (dry) @ 3% O2
NO, ppm (dry) @ 3% O2
HO, lb/106Btu
MULTICLONE OUTLET EMISSIONS
Particulate Loading, lb/106Btu
Combustible Loading, lb/10sBtu
Inorganic Ash Loading, lb/10SBtu
Combustibles in Flyash, %
.Multiclone Collection Efficiency, %
HEAT LOSSES, %
Dry Gas Loss
Moisture in Fuel
H2O From Combustion of H2
Combustibles in Collected Flyash
Combustibles in Emitted Flyash
Combustibles in Bottom Ash
Radiation Loss
Unmeasured Losses -
Total Losses
Boiler Efficiency
14
Low OFA
High Load
7.0
7.0
87
338
13.3
56
1.418
0.569
0.849
40.1
7.5
618
226
0.307
0.759
0.087
0.672
11.4
46.5
7.84
0.27
4.00
0.11
0.81
0.95
0.55
1.50
16.03
83.97
17
Med OFA
High Load
10.0
10.0
87
348
16.4
54
1.084
0.334
0.750
30.8
7.3
885
199
0.269
0.397
—
—
—
63.4
7.63
0.22
3.97
0.02
0.47
0.81
0.54
1.50
15.16
84.84
18
High OFA
High Load
13.0
13.0
79
313
18.1
63
1.033
0.290
0.743
28.1
8.1
34
230
0.312
0.425
0.063
0.362
14.9
58.9
7.76
0.20
3.98
0.27
0.41
1.28
0.60
1.50
16.00
84.00
15
High OFA
High Load
15.0
16.0
88
348
13.7
60
1.010
0.238
0.772
23.6
7.8
39
217
0.294
0.429
0.070
0.359
16.3
57.5
8.11
0.23
4.03
0.09
0.34
0.61
0.54
1.50
15.45
84.55
* Design capacity of 90,000 Ib stm/hr was not met due to
draft losses resulting from a retrofit dust collector.
KVB 15900-529
41
-------
pronounced than the combustible loading trends. On the Perfect 8 coal, low
overfire air flow gave lowest particulate emissions. On Century coal, high
overfire air flow gave lowest particulate emissions.
5.1.2 Nitric Oxide vs Overfire Air
On Perfect 8 coal, increased overfire air flow tended to increase
nitric oxide concentrations. On the single Century coal test series the re-
verse was true. The nitric oxide versus overfire air data is presented in
Figure 5-4 and in Table 5-3.
In conducting these tests, the under grate-air was held constant while
the overfire air was varied. Therefore, the excess air increased with increasing
overfire air flow. This procedure was adopted after unsuccessful attempts to
maintain excess air constant while changing overfire air flows at constant load.
It was characteristic of this boiler that changes in excess air upset the boiler
loading. The length of time required for stabilization of load and excess air
made fuel-air changes at constant load very difficult to perform.
To compensate for the changes in excess air, the nitric oxide concen-
trations were corrected to 9% O2 by applying the established O2-NO relationship:
one percent O2 increase = 0.041 lb/106 Btu NO increase.
Figure 5-4 plots the changes in nitric oxide concentration as a
function of overfire air pressure for several test series. The three low and
medium load test series demonstrate steady increases in NO concentration as over-
fire air pressure increases. The high load test series do not show a uniform
NO concentration increase with OFA pressure. At high load, NO concentrations
decreased as OFA pressure was increased from 7 to 10 "H2O pressure. Above 13
"H20 pressure at high load, NO concentrations increased slightly.
At maximum load (88% design capacity) and 5 "H2O overfire air pressure,
combustion was totally unsatisfactory. Opacity under these conditions averaged
40% and reached as high as 60%. Carbon monoxide concentrations were in excess
of 2000 ppm uncorrected, which is the upper limit on the CO instrument. By in-
creasing the overfire air from 5 to 7 "H2O pressure, the opacity was dropped to
around 20% and the carbon monoxide was dropped to around 500 ppm. Nitric oxide
concentration increased, at the same time, from 0.281 to 0.334 lb/106Btu.
42 KVB 15900-529
-------
CD
O §-
CD
I O
1 f—t _
o
LU 8"
O m.
X
o
CJ
(7 \ I
5.00 7.50
OVERFIRE RIR
I I
10.00 12.50
IN. H20
\
15.00
I HIGH LORD
° MED LORD O S LOW LORD
FIG. 5-4
NITRIC OXIDE
TEST SITE D
VS. OVERFIRE RIR
EACH LINE REPRESENTS A UNIQUE SERIES OF TESTS
PERFORMED IN SEQUENCE
15900-529
43
-------
TABLE 5-3
NITRIC OXIDE EMISSIONS VS OVERFIRE AIR
TEST SITE D
Test
No.
13A
13B
13D
13C
Coal
Perfect
Perfect
Perfect
Perfect
8
8
8
8
% Design
Capacity*
89
89
89
89
°2
%
7.8
7.4
7.6
7.6
OFA Pressure, "f^O
Upper Lower
5.0 5.0
7.0 7.0
10.3 10.0
15.0 14.0
Nitric Oxide, lb/106Btu
Corr Uncorr
14
17
18
15
6
8
7
11B
11A
11C
11E
11D
IIP
Century
Century
Century
Century
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
89
89
89
89
87
87
79
88
65
66
63
52
52
52
52
52
52
7.8
7.4
7.6
7.6
7.5
7.3
8.7
7.8
8.7
9.7
10.2
10.5
11.3
12.5
10.8
11.0
12.0
5.0
7.0
10.3
15.0
7.0
10.0
13.0
15.0
5.0
10.0
15.5
5.0
10.0
15.0
5.0
10.5
15.0
5.0
7.0
10.0
14.0
7.0
10.0
13.0
16.0
5.3
9.5
15.5
5.0
10.0
14.0
5.0
10.0
14.0
0.281
0.334
0.329
0.333
0.368
0.338
0.324
0.343
0.262
0.277
0.297
0.267
0.284
0.321
0.310
0.323
0.329
0.232
0.269
0.272
0.276
0.307
0.269
0.312
0.294
0.250
0.306
0.346
0.328
0.378
0.464
0.383
0.405
0.451
* Maximum obtainable load averaged 88% of design capacity due to draft
losses resulting from a retrofit dust collector.
** Corrected to 9% O2 bY applying the established O2-NO relationship:
1% O2 increase = 0.041 lb/106Btu NO increase
KVB 15900-529
44
-------
5-l-3 Carbon Monoxide vs Overfire Air
Carbon monoxide (CO) concentration was one of the parameters measured
during the overfire air tests. The test data, presented in Figure 5-5 and
Table 5-4, illustrate the need for increased overfire air flow at high loads
to maintain high combustion efficiency and minimize carbon monoxide emissions.
At high loads the carbon monoxide concentration increased rapidly
as the overfire air flow was reduced. At 88% of design capacity, 12 "H20
static pressure on the overfire air was required to maintain the CO concen-
tration below 400 ppm (Figure 5-5) . At reduced loads of 50-60% of design
capacity, the carbon monoxide concentrations were below 130 ppm at all over-
fire air flow conditions examined, and showed only a slight tendency to de-
crease with increasing OFA flow.
These tests did not isolate the variable overfire air flow from the
variable excess air. Grate air was held constant in each test series while
overfire air was raised or lowered. As a result, excess air varied in direct
proportion to the overfire air flow. Despite this fact, it is quite evident
from the data that at boiler loads in the vicinity of 88% of design capacity
overfire air pressures above 12 "H2O were required to control CO emissions.
5.1.4 Boiler Efficiency vs Overfire Air
Boiler efficiency data are presented in Table 5-5 for three test sets
in which overfire air was the variable. From this data it is clear that in-
creasing overfire air pressure resulted in a decrease in the combustible
fraction of the flyash leaving the boiler. The heat loss due to combustibles
in the flyash was also reduced by up to 0.47% on a heat input basis. Boiler
efficiency, on the other hand, does not show a direct correlation with overfire
air pressure. Variations in other heat loss areas which are believed to be
unrelated to the overfire air mask out the small changes in the flyash com-
bustible heat loss.
The data indicate that maximum overfire air pressure will result in
the lowest heat loss due to combustibles in the flyash.
KVB 15900-529
45
-------
C\]
O o
UJ
O
cc
CO
CE
Q_ °_
S
LU
O
O
21
O
O
GO
cc
cr
o
CENTURY COAL
T
I
5.00 7.50
OVERFIRE RIR
i
10.00
IN.
1
12.50
I
15.00
H20
: HIGH LORD
: MED LmD <> • L0" LORD
FIG. 5-5
CRRBON MONOXIDE
TEST SITE D
VS. OVERFIRE RIR
EACH LINE REPRESENTS A UNIQUE SERIES OF TESTS
PERFORMED IN SEQUENCE
15900-529
46
-------
TABLE 5-4
CARBON MONOXIDE EMISSIONS VS OVERFIRE AIR
TEST SITE D
Test
No.
13A
13B
13D
13C
14
17
18
15
6
8
7
11B
11A
lie
HE
11D
11F
Coal
Perfect
Perfect
Perfect
Perfect
Century
Century
Century
Century
Perfect
Perfect
Perfect
Perfect
Perfect
Perfect
Perfect
Perfect
Perfect
%
Design
°2
Capacity* %
8
8
8
8
8
8
8
8
8
8
8
8
8
89
89
89
89
87
87
79
88
65
66
63
52
52
52
52
52
52
7.
7.
7.
7.
7.
7.
8.
7.
8.
9.
10.
10.
11.
12.
10.
11.
12.
8
4
6
6
5
3
7
8
7
7
2
5
3
5
8
0
0
OFA
UP;
5
7
10
15
7
10
13
15
5
10
15
5
10
15
5
10
15
Pressure ,
per
.0
.0
.3
.0
.0
.0
.0
.0
.0
.0
.5
.0
.0
.0
.0
.5
.0
"H2O
Lower
5
7
10
14
7
10
13
16
5
9
. 15
5
10
14
5
10
14
.0
.0
.0
.0
.0
.0
.0
.0
.3
.5
.5
.0
.0
.0
.0
.0
.0
CO
ppm @ 3% 02
2345
640
378
73
618
885
34
39
87
57
29
126
65
85
129
72
80
* Design capacity of 90,000 Ib stm/hr not met due to draft loss
from retrofit dust collector. Maximum obtainable load averaged
of design capacity.
KVB 15900-529
47
-------
TABLE 5-5
BOILER EFFICIENCY VS OVERFIRE AIR
TEST SITE D
Test % Design OFA Pressure
No. Capacity* "H2O
6
8
7
65
66
63
5
10
15
Combustibles
in Flyash, %
21.9
20.8
16.0
Heat Loss Due to
Comb in Flyash,%
0.22
0.24
0.18
Boiler
Efficiency,
83.21
81.85
82.25
13B
12
89
87
7
10
28.2
22.1
0.35
0.21
83.34
83.26
14
17
18
15
87
87
79
88
7
10
13
15
40.1
30.8
28.1
23.6
0.81
0.47
0.41
0.34
83.97
84.84
84.00
84.55
* Design capacity of 90,000 Ib stm/hr not met due to draft loss
resulting from retrofit dust collector. Maximum obtainable load
averaged 88% of design capacity.
KVB 15900-529
48
-------
5•2 EXCESS OXYGEN AND GRATE HEAT RELEASE
The boiler at Test Site D was tested for emissions and boiler
efficiency over a wide range of loads and excess air conditions. This
section presents the results of these emissions and efficiency tests as
a function of load, expressed as grate heat release, and excess oxygen. Where
other variables, such as coal properties, could influence test results, these
variables were taken into account. Their influence on the performance of the
boiler is also discussed below. The plots presented in this section, which
are best characterized as emissions profiles, will be especially valuable
when comparing the performance of boilers of different designs.
5.2.1 Excess Oxygen Operating Levels
Figure 5-6 depicts the various boiler loads, expressed as grate heat
release, and oxygen levels at which tests were run on Boiler D. Different
symbols are used to distinguish the three coals fired.
Design capacity was not achieved on this unit. The highest boiler
load obtained during testing was 359x10^ Btu/hr-ft grate heat release. This
-3 n
is 90% of the design rating of 401x10° Btu/hr-ft . The load was limited to
90% by the ID fan. The unit was originally designed and installed without a
multiclone dust collector. In its original configuration it was operated
near design capacity. A larger ID fan was added to the unit when the multi-
clone dust collector was installed in order to handle the increased pressure
drop. ' The new ID fan~was too large and could not be effectively controlled.
This necessitated that baffling be installed to reduce the air flow to the
point where the automatic controls could effectively modulate the fan. It was
as a result of this modification that the air flow was restricted to the
point where design capacity could not be met.
The lowest oxygen level obtained was 7.3%, which represents 48% excess
air. This compares to the design level of 31% excess air, or 5% O2. Extra-
polation of the O2 - grate heat release relationship given in Figure 5-6 to
the design capacity of 401xl03 Btu/hr-ft2 yields an oxygen level of no less
than 6.4%. This indicates that the unit was being operated at a higher O2
level than., it was designed for. However, testing did not establish a lower
02 limit on this unit and, therefore, there is no basis for assuming that the
unit could not be operated at design O2.
49 KVB 15900-529
-------
8-
o
UJ
CJ
DZ
„
CD S
>- ^
x
o
O
100.0
200.0
300.0
400.0
500.0
GRflTE HEni RELERSE 1000 BTU/HR-SQFT
S PER 8 COflL -f S CENT. COflL A » VICT- coflL
FIG. 5-6
OXYGEN
TEST SITE D
VS. GRflTE HEflT RELEflSE
CONDITIONS UNDER WHICH EMISSION TESTS WERE RUN
15900-529
50
-------
Figure 5-6 illustrates the conditions under which emission tests
were run at Test Site D. At high load the measured oxygen spread was only
7.3 to 8.4%. Oxygen levels in the flue gas increased linearly as boiler
loading decreased. At the lowest measured load of 141xl03 Btu/hr-ft2, the
oxygen level was 11.5%. Coal properties did not appear to be a factor in
the O2 - grate heat release relationship: load-specific O2 levels were
similar for all three coals, as can be seen in Figure 5-6.
Because particulate loading tests are a principal part of this program,
the oxygen and grate heat release conditions under which the particulate
tests were conducted are given in Figure 5-7. In this plot, the scales are
expanded to provide greater resolution. The data falls inside a band which
is highly dependent on grate heat release and which is less than two percent
C>2 wide.
5.2.2 Particulate Loading vs Oxygen and Grate Heat Release
Figure 5-8 profiles boiler outlet particulate. -loading as a function of
grate heat release. The data points in this plot are keyed to coal type.
The data shows a fairly well defined relationship between particulate loading
-3 o
and boiler load. Below 300x10 Btu/hr-ft grate heat release the particulate
loadings are not influenced by grate heat release and average 0.74io.05
lbs/10^ Btu. Above 300x10^ Btu/hr-ft the particulate loadings average
1.02-0.17 lbs/106 Btu, a 38% increase over particulate loadings below 300xl03
Btu/hr-ft2.
Coal ash did not play a role in the boiler outlet particulate loadings.
.As can be seen in Figure 5-8, particulate loadings at the boiler outlet were
basically the same for all three coals . The Table on page 54 shows the amount of
ash in the coal, and the amount of coal ash which was carried out of the
boiler in the flyash. The data show that where the ash content of the coal
was low, the percent carryover in the flyash was high, thus negating the
potential benefits of firing a low ash coal to reduce particulates. This
conclusion, however, only applies to the coals fired in these tests and to
the boiler at Site D. The results should not be generalized to other coals
or boilers.
KVB 15900-529
51
-------
8-J
LLJ
CJ
DC
LiJ
0- 8-
„
CD S-
>- co
x
o ,
4-
~T~f T
150.0
T
4-
200.0 250.0 300.0 350.0
GRflTE HERT RELEflSE 1000 BTU/HR-SQ FT
; PER 8 COflL -f S CENT. COflL A • VICT-
FIG. 5-7
OXYGEN
TEST SITE D
VS. GRflTE HEflT RELERSE
CONDITIONS UNDER WHICH PARTICULATE TESTS WERE RUN
15900-529
52
-------
CD
R 8-
OQ
8-
OJ
cc
o
cc
o
CQ
-f
4-
150.0
r
200.0
250.0
300.0
350.0
GRflTE HEflT RELERSE 1000 BTU/HR-SQFT
S PER 8 COflL + S CENT. COflL A ' VICT- coflL
FIG. 5-8
BOILER OUT PflRT.
TEST SITE D
VS. GRRTE HERT RELERSE
15900-529
'53
-------
ASH CARRYOVER VS COAL TYPE
TEST SITE D
Average Ash Content Average Ash Content Average Ash
Coal of Coal, lbs/106 Btu of Flyash, lbs/106 Btu Carryover, %
Perfect 8
Century
Victoria
6.73
4.31
5.19
0.627
0.719
0.560
9.3
16.7
10.8
Particulate measurements were made at the outlet of the multiclone dust
collector simultaneously with the measurements made at the boiler outlet.
Figure 5-9 plots the multiclone outlet particulate loadings as functions of
grate heat release. Again, the data points are keyed to coal type.
There is more scatter in the data from this sample location than is
the case with the test data from the boiler outlet. The scatter can be
accounted for by variations in the collection efficiency of the dust collector.
The dust collector was not operating at design efficiency. This problem will
be discussed further in section 5.5.
At the multiclone outlet, the particulate loading averaged 0.42io.lO
lbs/106 Btu below 300x103 Btu/hr-ft2 grate heat release, and 0.53JI0.14 lbs/106
Btu above this value. Thus, the average particulate "loading was 26% greater
at high load than at low load when measured after the multiclone.
It is evident from an examination of -Figure 5-9 that coal type may
be a factor in the particulate loadings after the multiclone. The apparent
stratification by coal type may or may not be real. The scatter in the data
makes it difficult to draw any conclusions about the effect of coal type.
5.2.3 Nitric Oxide vs Oxygen and Grate Heat Release
Nitric oxide (NO) concentration was measured during each test in
units of parts per million (ppm) . It is presented here in units of lbs/106 Btu
to be more easily compared with existing and proposed emission standards.
The parts per million values may be found in Table 2-1 of this report.
Figure 5-10 plots nitric oxide as a function of grate heat release,
with coal type designated for each data point. The range in nitric oxide
KVB 15900-529
54
-------
DQ
^
O
§-
CO
CD
4-
CC
CE
D_
o
LU
s:
o
•f
4-
. 4*
4-
150.0
200.0
r
250.0
300.0
350.0
GRRTE HEflT RELEflSE 1000 BTU/HR-SQFT
O ; PER 8 COflL + 5 CENT. COfiL A • VICT- coflL
FIG. 5-9
MULTICLONE OUT PRRT. VS. GRflTE HEflT RELEflSE
TEST SITE D
15900-529
55
-------
§1-
DO
UJ 8'
O 1
i—i
X
o
o
H-4
F §
150.0
200.0
250.0
300,0
350.0
GRflTE HEflT RELERSE 1000 BTU/HR-SQFT
: PER 8 COflL + S CENT. COflL A " VICT. CORL
FIG. 5-10
NITRIC OXIDE
TEST SITE D
VS. GRRTE HERT RELERSE
15900-529
56
-------
loading at high load was 0.232 to 0.312 lbs/106 Btu. NO concentrations
tended to increase slowly as load decreased. The maximum nitric oxide
level of 0.464 lbs/106 Btu was created artificially by forcing the air up
to 12.5% 02 at low load. Under normal firing conditions on Boiler D, the
upper limit on nitric oxide was about 0.35 lbs/106 Btu. Neither load nor
coal type had a measurable influence on nitric oxide levels at Site D.
Oxygen level, on the other hand, had a very strong influence on nitric oxide
emissions. Figure 5-11 plots the nitric oxide data as a function of oxygen
with the data points separated into these distinct grate heat release ranges.
This plot clearly shows the strong relationship between NO and O2 found at
Test Site D.
Figures 5-12, 5-13 and 5-14 show the same data on expanded scales for
each of the three grate heat release ranges. Through each data set-a line
that best "fits" the data has been drawn. The slopes of the lines were
determined by means of linear regression analysis. The results of these
tests can be interpreted as a flatening or leveling of the NO-02 relationship
as O2 declines. The data indicate that grate heat release does not have a
significant effect on nitric oxide emissions on this boiler. Nitric oxide
trend lines are being determined for each of the boilers tested under this
program. Based on the data obtained at Test Site D, the nitric oxide trend
line appears to be a function only of oxygen as shown in Figure 5-15.
KVB 15900-529
57
-------
DQ
O H
CQ
LU 8"
O -™
i—i
X
O
CJ
I—I
n~ 8-
O
O
O
A
I
4.00
OXYGEN
6.00
i \
8.00 10.00
PERCENT
12.00
O : 100-199GHR (J
FIG. 5-11
NITRIC OXIDE
TEST SITE D
° 200-299GHR
S 3QO-399GHR
VS. OXYGEN
15900-529
58
-------
s
CD
-ZL
O
CQ
UJ g
Q ™
i—i
><
O
LJ
i—i
8.00
OXYGEN
9.00
10.00 11.00
PERCENT
12.00
O I 100-199GHR
FIG. 5-12
NITRIC OXIDE
TEST SITE D
VS. OXYGEN
LINEAR REGRESSION APPLIED BY METHOD OF LEAST SQUARES
15900-529
59
-------
CD
CD
LJ E
Q *
i—i
X
o
LJ
i—i
F §H
I
8.00
OXYGEN
9,00
T
T
10.00 11.00
PERCENT
\
12.00
I 200-299GHR
FIG. 5-13
NITRIC OXIDE
TEST SITE 0
VS. OXYGEN
LINEAR REGRESSION APPLIED BY METHOD OF LEAST SQUARES
15900-529
60
-------
CD
O 8-
CD
I O
—' o-
o
O
i—i
X
O
CJ
I—I
QZ
A
A
A
8.00
OXYGEN
9.00
10.00 11.00
PERCENT
r~
12.00
I 300-399GHR
FIG. 5-14
NITRIC OXIDE
TEST SITE D
VS. OXYGEN
LINE INDICATES AVERAGE VALUE
15900-529
61
-------
EH
H
0.50
g 0.40
«
10
o
0.30
H
O
o
0.20
0.10
100-199 GHR
200-299 GHR
300-399 GHR
I
I
I
9 10 11
OXYGEN, PERCENT
12
FIGURE 5-15.
TREND IN NITRIC OXIDE EMISSIONS AS A FUNCTION
OF OXYGEN AT SITE D
KVB 15900-529
62
-------
5-2.4 Carbon Monoxide vs Oxygen and Grate Heat Release
Carbon monoxide (CO) was measured during each test and the data
are presented here in units of parts-per-million (ppm) by volume on a dry
basis, corrected to 3% G>2 - Carbon monoxide is a by product of incomplete
combustion, but if it is kept below 400 ppm it is considered insignificant for the
purposes of this report. As a reference, 400 ppm CO is equivalent to
0.04% CO and represents a 0.20% heat loss in a coal fired boiler operating
at 8% O2.
Figure 5-16 presents the carbon monoxide data gathered under a
variety of firing conditions and plotted as a function of grate heat release.
Each data point is keyed to the coal type being fired. This plot shows that
carbon monoxide was only found in significant concentrations at loadings
above 300xl03 Btu/hr-ft2 grate heat release. Below 300xl03 Btu/hr-ft2 the
concentrations were all below 130 ppm.
Figures 5-17 and 5-18 present the carbon monoxide concentrations as
a function of oxygen for Perfect 8 coal and Century coal, respectively. Both
plots show that carbon monoxide concentrations were significant only when
firing at below 8% 02- It must be kept in mind, however, that oxygen operating
levels and grate heat release are closely related at this site so that
it is not clear from this data whether the low 02 or high grate heat release
is responsible for the sudden increase in carbon monoxide at low 02 levels.
Finally, the data show that coal properties did not significantly
affect the carbon monoxide emission levels of Boiler D.
5.2.5 Combustibles vs Oxygen and Grate Heat Release
In this report the term "combustibles" refers only to the solid com-
bustibles in the various ashes leaving the boiler. Combustibles are described
here in terms of their percent by weight in the flyash at the boiler outlet
and in the bottom ash collected from the ash pit.
Figure 5-19 shows the combustibles in the boiler .outlet flyash as a
function of grate heat release. The data points are keyed to coal type. In
these tests combustibles ranged between 14.4% and 40.1% with the Century
KVB 15900-529
63
-------
d
8
UJ
CJ
DC
UJ
00 I
cr
z:
o_
LU
Q
I—I o
x a'
o f
o
(T"
cr
o
i
100.0
200.0
I
300.0
1^
400.0
n
500.0
GRRTE HEflT RELERSE 1000 BTU/HR-SQFT
O S PER 8 COflL -+- S CENT. COflL
FIG. 5-16
CRRBON MONOXIDE
TEST SITE D
°, VICT. COflL
VS. GRRTE HERT RELERSE
15900-529
64
-------
C\J
O o
t- I"
2: "
LU
CJ
CC
^ I
Q_
Q_
Q
•—' o
X d
o ?
o
CD
CC
GC
CJ
O
8.00
OXYGEN
9.00
I I
10.00 11.00
PERCENT
\
12.00
S PER 8 COflL
FIGD 5-17
CRRBON MONOXIDE
TEST SITE D
VS. OXYGEN
15300-529
65
-------
CM
O
UJ
CJ
QC
UJ
Q_
I
Q_
0-
LU
Q
I—I o
X Q-
O ?
O
CD
CJ
I I I I I
8.00 9.00 10.00 11.00 12.00
OXYGEN PERCENT
CENT. COflL
FIG. 5-18
CflRBON MONOXIDE VS. OXYGEN
TEST SITE D
15900-529
66
-------
o
s"
o
s"
CJ
DC
LJ
0_ <=.
(O
CD
51
O o
O o"
o
oc
A +
o
CQ
T
T
100.0 200.0 300.0 400.0 500.0
GRRTE HERT RELERSE 1000 BTU/HR-SQFT
I PER 8 COflL + S CENT. COflL A ' VICT- coflL
FIG. 5-19
BOILER OUT COMB.
TEST SITE D
VS. GRRTE HERT RELERSE
15900-529
67
-------
coal (average 29.7%), only slightly higher than the Perfect 8 coal (average
23.5%). The single Victoria coal data point (26.5%) falls between the averages
for the other two coals. The average heat losses due to combustibles in the
flyash at the boiler outlet were 0.24% for Perfect 8 coal, 0.47% for Century
coal and 0.29% for the single Victoria coal test. These losses are relatively
small and suggest that flyash reinjection, were it installed on this unit,
would result in fairly insignificant energy savings.
Figure 5-20 describes combustibles in the bottom ash as a function of
grate heat release. Again the data points are keyed to coal type. The
absolute range of values was between 8.4% and 35.9%. Perfect 8 coal and
Century coal showed similar results, averaging 18.4% and 16.4%, respectively.
Victoria coal had the two highest measurements, averaging 34.8%.
The heat losses due to the bottom ash combustibles were higher than
those due to the flyash combustibles. The average heat losses were 1.63%,
0.90%, and 3.51%, respectively, for Perfect 8, Century and Victoria coals.
Grate heat release did not have a major influence on the levels of bottom ash
combustibles.
5.2.6 Boiler Efficiency vs Oxygen and Grate Heat Release
Boiler efficiency was determined for each test that included a boiler
outlet particulate loading measurement. The efficiency determinations were
made by the ASTM heat loss method.
Figure 5-21 shows the calculated boiler efficiencies as a function of
grate heat release. Data points are keyed to coal type. The efficiencies
appear to be independent of grate heat release, but dependent upon coal type.
Table 5-6 shows the average heat loss differences between the coal
types tested. This data is compared with the results of a 1964 performance
acceptance test run at 370xl03 Btu/hr-ft2. The Table shows that boiler efficiency
was highest with Century coal, but this may be explained in part by the fact
that the tests with Century coal were run at high load and low 02- The low
efficiency of the single Victoria coal test was a result of its high combustibles
heat loss. According to the boiler's predicted performance data, the
guaranteed minimum efficiency is 84.6%.
KVB 15900-529
68
-------
o
8
o
g"
LJ
CJ
QC
LJ
Q_
§
O o
CJ o-
co
cc
o
CG
I I I I I
100.0 200.0 300.0 400.0 500.0
GRRTE HERT RELER5E 1000 BTU/HR-SQFT
; PER s CORL + : CENT. CORL A ° VICT. CORL
FIG. 5-20
BOTTOM RSH COMB.
TEST SITE D
VS. GRRTE HERT RELERSE
15900-529
-------
DC
LiJ
n
u—
>-
8-
CC
o
CD
A
i i i i i
100.0 200.0 300.0 400.0 500.0
GRRTE HERT RELERSE 1000 BTU/HR-SQFT
: PER 8 coflL
: CENT. CORL
: VICT. CORL
5-21
BOILER EFFICIENCY
TEST SITE D
VS. GRRTE HERT RELERSE
15900-529
70
-------
TABLE 5-6
AVERAGE HEAT LOSSES BY COAL TYPE
Coal
Perfect 8
Century
Victoria
Acceptance
Test, 1964
Dry
Gas
8.53
7.82
8.58
7.76
Moisture
in Fuel
0.29
0.23
0.28
0.39
H20 From
H2 in Fuel
4.05
3.99
3.97
4.13
Total
Combustibles
1.96
1.49
3.85
0.21*
Radiation &
Unmeasured
2.20
2.09
2.27
0.80**
Total
Losses
17.04
15.62
18.95
13.29
BOILER
EFFICIENCY
PERCENT
82.96
84.38
81.05
86.71***
Ash sample taken from "ash hopper" only and apparently does not include flyash
This is radiation loss only and does not include unmeasured losses of 1.5%
*** Allowing for unmeasured losses of 1.5% and flyash combustible losses of 0.3%
the acceptance test efficiency would be 84.91%.
5.3 COAL PROPERTIES
Coals from three different mines were tested in Boiler D. Repre-
sentative coal samples were taken from the unit's two coal feeders during
each test that included either a particulate measurement or SASS sample
catch. These samples were analyzed for proximate analysis, Ultimate analysis
and size consistency. A composite sample consisting of increments from
each test with each coal was also analyzed for ash fusion temperature, Hard-
grove grindability index, free swelling index, and minerals in the ash. This
section will bring into focus all of the coal related relationships which
have been discussed in the previous two sections and will discuss the coal
size consistency and sulfur balance data.
5.3.1 Chemical Composition of the Coals
The chemical analyses of each coal sample are grouped by coal and
presented in Tables 5-7, 5-8, and 5-9. Coal ash analyses are given in Table
5-10. These Tables also show the average analysis for each coal and the
standard deviation which is a measure of the variation in composition from
sample to sample. From comparing these Tables, it is evident that all three
coals were similar in makeup. However, some differences are notable.
KVB 15900-529
71
-------
TABLE 5-7
FUEL ANALYSIS - PERFECT 8 COAL
TEST SITE D
TEST NO.
PROXIMATE (As Rec'd)
% Moisture
% Ash
% Volatile
% Fixed Carbon
BTU/lb
% Sulfur
ULTIMATE (As Rec'd)
% Moisture
% Carbon
% Hydrogen
% Nitrogen
% Chlorine
% Sulfur
% Ash
% Oxygen (Diff.)
ASH FUSION (Reducing)
Initial Deformation
Soft (H=W)
Soft (H=1/2W)
Fluid
2
2.69
8.44
39.47
49.40
13124
2.35
2.69
72.89
5.08
1.27
0.13
2.35
8.44
7.15
3
9
38
48
3
.37
.26
.98
.39
12835
2
3
71
5
1
0
2
9
7
.54
.37
.25
.02
.24
.12
.54
.26
.20
4
3.23
8.08
39.45
49.24
13081
2.79
3.23
72.71
5.06
1.27
0.12
2.79
8.08
6.74
5
3.
8.
39.
48.
10
47
53
90
13034
2.
3.
72.
5.
1.
0.
2.
8.
7.
44
10
38
10
24
11
44
47
16
6
3.
8.
39.
48.
06
83
82
29
13034
2.
3.
72.
4.
1.
0.
2.
8.
6.
85
06
32
97
09
11
85
83
77
7
3.35
9.00
39.15
18.50
12978
2.57
3.35
71.80
5.01
1.12
0.12
2.57
9.00
7.03
8
3.04
8.65
39.01
49.30
13084
2.59
3.04
72.44
4.96
1.24
0.13
2.59
8.65
6.95
9
3.40
8.24
39.16
49.20
13068
2.23
3.40
72.05
5.04
1.32
0.12
2.23
8.24
7.60
10
3.45
8.13
39.17
49.25
13019
2.50
3.45
72.38
5.02
1.33
0.10
2.50
8.13
7.09
12
3
9
37
49
.50
.14
.90
.46
12935
2
3
71
4
1
0
2
9
7
.82
.50
.14
.90
.31
.12
.82
.14
.07
13
3
9
39
47
.31
.57
.22
.90
12839
2
3
71
4
1
0
2
9
6
.94
.31
.40
.91
.15
.12
.94
.57
.60
Hardgrove Grindability Index
Free Swelling Index
COMP
2.77
8.97
38.25
50.01
12945
2.73
2.77
72.04
4.89
1.42
0.09
2.73
8.97
7.09
2050
2320
2380
2520
47
4
AVG
3
8
39
48
.23
.71
.17
.89
13003
2
3
72
5
1
0
2
8
7
.60
.23
.07
.01
.23
.12
.60
.71
.03
STD
DEV
0.24
0.49
0.49
0.53
97
0.22
0.24
0.59
0.07
0.08
0.01
0.22
0.49
0.27
KVB 15900-529
-------
TABLE 5-8
FUEL ANALYSIS - CENTURY COAL - TEST SITE D
TEST NO.
PROXIMATE (As Rec'd)
% Moisture
% Ash
% Volatile
% Fixed Carbon
BTU/lb
% Sulfur
ULTIMATE (As Rec'd)
% Moisture
% Carbon
% Hydrogen
% Nitrogen
% Chlorine
% Sulfur
% Ash
% Oxygen (Diff)
ASH FUSION (Reducing)
Initial Deformation
Soft (H=W)
Soft (H=1/2W)
Fluid
Hardgrove Grindability
Free Swelling Index
14
3.09
6.86
38.75
51.30
13342
2.04
3.09
74.52
5.02
1.26
0.10
2.04
6.86
7.11
Index
2
4
39
53
15
.69
.92
.14
.25
13829
1
2
77
5
1
0
1
4
7
.15
.69
.18
.24
.23
.13
.15
.92
.46
2
5
37
53
16
.62
.89
.96
.53
13622
1
2
75
5
1
0
1
5
8
.11
.62
.62
.14
.32
.13
.11
.89
.17
17
2.59
5.66
37.96
53.79
13691
0.86
2.59
76.63
5.15
1.34
0.15
0.86
5.66
7.62
2
5
38
53
18
.39
.99
.39
.23
13662
0
2
76
5
1
0
0
5
7
.98
.39
.46
.19
.50
.13
.98
.99
.36
COMP
2.94
5.46
38.79
52.81
13709
1.00
2.94
76.65
5.18
1.32
0.04
1.00
5.46
7.41
2680
2700
2700
2700
40
3
AVG
2.68
5.86
38.44
53.02
13629
1.23
2.68
76.08
5.15
1.33
0.13
1.23
5.86
7.54
STD
DEV
0.26
0.70
0.51
0.99
178
0.47
0.26
1.04
0.08
0.10
0.02
0.47
0.70
0.40
KVB 15900-529
-------
TABLE 5-9
FUEL ANALYSIS - VICTORIA COAL
TEST SITE D
TEST NO.
PROXIMATE (As Rec'd)
% Moisture
% Ash
% Volatile
% Fixed Carbon
Btu/lb
% Sulfur
ULTIMATE (As Rec'd)
% Moisture
% Carbon
% Hydrogen
% Nitrogen
% Chlorine
% Sulfur
% Ash
% Oxygen (Diff)
ASH FUSION (Reducing)
Initial Deformation
Soft (H=W)
Soft (H=1/2W)
Fluid
Hardgrove Grindability Index
Free Swelling Index
22
3.05
6.47
37.43
53.05
13453
1.10
3.05
75.08
4.98
1.27
0.06
1.10
6.47
7.99
2480
2700
2700
2700
43
5-1/2
23
3.16
6.96
37.06
52.82
13357
1.11
3.16
74.84
5.02
1.32
0.07
1.11
6.96
7.52
2700
2700
2700
2700
45
4
AVG
3.11
6.72
37.25
52.94
13405
1.11
3.11
74.96
5.00
1.30
0.07
1,11
6.72
7.76
KVB 15900-529
-------
TABLE 5-10
MINERAL ANALYSIS OF COAL ASH
TEST SITE D
Coal
Test No.
MINERAL ANALYSIS OF ASH
Silica, SiO2
Alumina, A12O3
Titania, Ti02
Ferric Oxide, Fe2O3
Lime , CaO
Magnesia, MgO
Potassium Oxide, K2O
Sodium Oxide, Na2O
Sulfur Trioxide, 803
Phos Pentoxide, P2°5
Undetermined
Silica Value
Base: Acid Ratio
T25Q Temperature
SULFUR FORMS (As Rec'd)
% Pyritic Sulfur
% Sulfate Sulfur
% Organic Sulfur
Perfect 8
Comp
45.24
23.83
1.18
24.72
1.04
0.62
1.82
0.42
0.62
0.13
0.25
63.17
0.41
2400°F
1.46
0.12
1.15
Century
Comp
47.07
32.74
1.42
9.24
2.23
1.10
1.99
0.63
1.85
0.32
0.92
78.92
0.19
2730°F
0.19
0.02
0.79
1 22
49.83
29.97
1.55
10.15
1.88
0.86
1.61
0.54
1.38
0.36
1.48
79.45
0.18
2765°F
0.27
0.07
0.76
Victoria
23
50.60
31.28
1.55
9.39
1.91
0.72
1.71
0.42
0.96
0.70
0.35
80.80
0.17
2785°F
0.29
0.06
0.76
|
Avg . 1
50.22
30.63
1.55
9.77
1.90
0.79
1.66
0.48
1.17
0.53
0.92
80.13
0.18
2775°F
0.28
0.07
0.76
KVB 15900-529
-------
The more significant coal properties are presented in Table 5-11 on a
heating value basis in order to allow for a more meaningful comparison. From
this table it can be seen that moisture was relatively constant at 2.5, 2.0
and 2.3 lbs/106 Btu for the three coals. Coal moisture, therefore, was not
considered a variable in these tests. Ash content of the coal shows some
variation (6.7, 4.3, 5.0 lbs/106 Btu), but not enough to be considered really
significant. Sulfur content does vary significantly — between two percent for
Perfect 8 coal and less than one percent for the other two coals.
TABLE 5-11
COAL PROPERTIES CORRECTED TO A CONSTANT 106 BTU BASIS
Moisture, lbs/106 Btu
Ash, lbs/106 Btu
Sulfur, lbs/106 Btu
Perfect 8
Coal
2.5
6.7
2.0
Century
Coal
2.0
4.3
0.90
Victoria
Coal
2.3
5.0
0.83
The influence of coal properties on emissions and boiler efficiency is
summarized below, with references to the relevant figures. Each of these
relationships has been addressed elsewhere in the report but is reviewed here
for convenience.
RELATIONSHIP BETWEEN COALS FIRED AND EMISSIONS
Figure
Parameter
1. Excess 02
2. Particulates (Boiler Outlet)
3. Particulates (Multiclone Outlet)
4. Nitric Oxide
5 . Carbon Monoxide
6. Combustibles (Boiler Outlet Flyash)
7. Combustibles (Bottom Ash)
8. Boiler Efficiency
9. Particle Size Distribution of
Flyash
10. SOx Emissions Table
No.
5-6
5-8
5-9
5-10
5-16
5-19
5-20
5-21
5-25
5-13
Coal
Perfect 8
Century
Victoria
Perfect 8
Century
Victoria
Perfect 8
Century
Victoria
Perfect 8
Century
Victoria
Relationship
to Coal
None
None
Highest Average Particulates
Lowest Average Particulates
Insufficient Data
None
None
None
Lowest Average Combustibles
Lowest Average Combustibles
Highest Average Combustibles
Average Efficiency
Highest Efficiency
Lowest Efficiency
Highest % Passing 20 Micrometers
Lowest % Passing 20 Micrometers
Insufficient Data
Proportional to Sulfur Content
of Coal
KVB 15900-529
76
-------
5.3.2 Coal Size Consistency
The individual coal samples and the composite coal samples were
screened at the site using 1", 1/2", 1/4", #8 and #16 square mesh screens.
The results of these screenings are presented in Table 5-12. The average
coal size consistency and standard deviation for each of the three coals were
determined and are plotted against the ABMA recommended limits for mass fed
stokers in Figures 5-22, 5-23, and 5-24.
This parameter was not varied for test purposes and its natural
fluctuations from test to test were small. All three coals were coarse, having
low fines and falling at the lower limit of the ABMA curve. A generally
accepted definition of "coal fines" is the percent by weight passing a 1/4"
square mesh screen. By this definition, Perfect 8 coal had the lowest fines
at 9.3% but also had the greatest spread in values at -4.5%. Century coal
had 15.212.0% fines, and Victoria coal had 15.8+0.2% fines.
5.3.3 Sulfur Balance
Sulfur oxides — SC>2 and 503 — were measured in the flue gas during
one test on Century coal and one test on Victoria coal. Coal and ash samples
taken at the same time thdfee tests were run were analyzed for sulfur content
in order to determine sulfur balance. The results of this sulfur balance
are presented in Table 5-13.
The sulfur balance was not good. It shows more sulfur being emitted
from the boiler than is being input from the coal. It is not possible at
this point to determine which measurement is in error, the sulfur input (fuel
sulfur and coal flow) or the sulfur output (SOx in flue gas, sulfur in ash,
and gas flow). However, it is safe to say that the most accurate determination
of sulfur retention is to measure the sulfur in the ash directly rather than
compare the sulfur input with the sulfur output. The measured sulfur retentions
were as follows:
Century coal - 1.9% sulfur retained in the ash
Victoria coal - 2.1% sulfur retained in the ash
KVB 15900-529
77
-------
TABLE 5-12
AS FIRED COAL SIZE CONSISTENCY
TEST SITE D
<
u
oo
PERFECT
Test
Number
2
3
4
5
6
7
8
9
10
12
13
Composite
Average
Std. Dev.
PERCENT PASSING STATED
1"
99.5
98.4
99.0
99.8
99.8
99.2
99.8
98.3
99.0
99.8
99.8
99.3
99.3
0.6
1/2"
59.5
65.8
45.4
36.3
46.0
38.9
49.0
38.6
43.1
54.2
57.5
47.0
48.6
9.6
1/4"
15.2
20.1
9.0
5.5
8.6
4.7
7.5
7.0
6.4
9.0
9.4
8.5
9.3
4.5
SCREEN
#8
6.4
7.6
4.1
2.8
3.5
2.4
3.5
3.9
3.4
3.5
3.8
4.3
4.1
1.5
SIZE
#16
4.0
4.3
3.2
2.3
2.7
2.0
2.6
3.0
2.7
2.4
2.6
3.2
2.9
0.7
O
u
>H
CENTUF
14
15
16
17
18
Composite
Average
Std. Dev.
98.5
94.6
96.1
96.0
95.0
94.9
96.0
1.5
50.2
48.2
48.2
49.7
52.3
44.2
49.7
1.7
13.3
13.7
14.3
16.4
18.1
12.2
15.2
2.0
6.3
5.7
5.2
7.0
7.6
4.8
6.4
1.0
4.5
4.1
2.9
5.1
5.4
3.3
4.4
1.0
VICTORIA
COAL
22
23
Average
Std. Dev.
88.2
89.4
88.8
0.8
39.5
39.6
39.6
0.1
15.9
15.6
15.8
0.2
7.7
8.1
7.9
0.3
4.1
5.7
4.9
1.1
KVB 15900-529
78
-------
w
H
s
95
80
50
30
20
10
W
ABMA RECOMMENDED
SIZING
/-[PERFECT 8 COAL!
16# 8# 1/4" 1/2"
SIEVE SIZE DESIGNATION
FIGURE 5-22.
SIZE CONSISTENCY OF "AS FIRED" PERFECT 8 COAL
VS ABMA RECOMMENDED SIZING FOR OVERFEED STOKERS
KVB 15900-529
79
-------
ABMA RECOMMENDED
SIZING
16 # 8# 1/4" 1/2" 1"
SIEVE SIZE DESIGNATION
2"
FIGURE 5-23.
SIZE CONSISTENCY OF "AS FIRED" CENTURY COAL
VS ABMA RECOMMENDED SIZING FOR OVERFEED STOKERS
KVB 15900-529
80
-------
w
H
EH
a
8
«
w
95
80
50
30
20
10
ABMA RECOMMENDED
SIZING
I VICTORIA COAL]
16# 8# 1/4" 1/2" 1"
SIEVE SIZE DESIGNATION
2"
FIGURE 5-24.
SIZE CONSISTENCY OF "AS FIRED" VICTORIA COAL
VS ABMA RECOMMENDED SIZING FOR OVERFEED STOKERS
KVB 15900-529
81
-------
TABLE 5-13
SULFUR BALANCE
TEST SITE D
CO
to
Test
No.
IS
22
SULFUR
Fuel
Sulfur
%
0.98
1.10
IN FUEL
As SO2
lbs/106Btu
1.435
1.635
SULFUR IN BOTTOM ASH
Ash
Sulfur
%
0.16
0.16
As S02
lbs/106Btu
0.018
0.023
Retention
%
1.25
1.41
SULFUR IN BLR HPR ASH
Ash
Sulfur
%
0.40
0.29
As SO2
lbs/106Btu
0.005
0.003
Retention
%
0.34
0.21
Ash
Sulfur
%
0.62
0.47
SULFUR IN FLYASH
As SO2 Re tention
lbs/106Btu %
0.005 0.35
0.007 0.43
SULFUR IN FLUE GAS
SOx
ppm(dry)
1271
1293
As SO 2
lbs/106Btu
1.799
1.815
Fuel Sulfur
Emitted, %*
125
111
* The imbalance between the sulfur in the fuel and the sulfur emitted
can be attributed to measurement error.
KVB 15900-529
-------
5.4 PARTICLE SIZE DISTRIBUTION OF FLYASH
The purpose of the particle size -distribution tests carried out under
this program is to accumulate a data bank of particle size distribution data
from all types of stoker boilers firing a variety of coals under a variety of
firing conditions. This data will be valuable to manufacturers of dust col-
lection equipment and to consulting engineers faced with the tasks of specifying
such equipment.
Five particle size distribution tests were run at the boiler outlet
employing the methodologies outlined in Table 5-14; three by Bahco classifier,
one by Brink cascade impactor, and one by SASS cyclones. Each methodology has
its own drawback, so the results of each test should be examined in light of
the methodology used. A discussion of each method is included in Section 4.5.
Test results are presented in Table 5-15 and Figures 5-25, 5-26 and 5-27.
At this test site it is especially important to point out that the
size distribution data from the Bahco classifier tests require a correction for
the lost portion of the sample. A significant fraction of the sampled flyash
was lost because it was not collected by the cyclone. This lost fraction was
collected on a filter following the cyclone and can be used to apply a correction
to the data. For the three tests involved, the classified sample represents the
following percent of the total sample:
Test 5 — 62.2%
Test 17 -- 81.7%
Test 23 — 70.8%
TABLE 5-14
DESCRIPTION OF PARTICLE SIZE DISTRIBUTION
TESTS AT THE BOILER OUTLET - TEST SITE D
Test
No.
5
15
17
22
23
Coal
Perfect 8
Ce ntury
Century
Victoria
Victoria
% Design
Capacity*
86
88
87
59
61
02
%
7.9
7.8
7.3
10.2
10.4
OFA
"H20
10.0
15.5
10.0
12.0
12.5
Particle Size Distri-
bution Methodology Used
Bahco - Sieve
Brink Impactor
Bahco - Sieve
SASS Cyclones
Bahco - Sieve
*Design capacity is 90,000 Ib stm/hr. Maximum obtainable load
averaged 88% of design capacity due to draft loss from retrofit
dust collector.
KVB 15900-529
83
-------
TABLE 5-15
RESULTS OF PARTICLE SIZE DISTRIBUTION TESTS
AT THE BOILER OUTLET - TEST SITE D
Size Distribution
Size Concentration
Test Description
Test 5 High Load - Bahco
Test 15 High Load - Brink
Test 17 High Load - Bahco
Test 22 Low Load - SASS
Test 23 Low Load - Bahco
% Below
3ym
3.5
13.0
2.3
35.0
1.3
% Below
18
—
12
48
5.4
lb/10bBtu
Below 3ym
0.029
0.013
0.025
0.270
0.010
lb/10bBtu
Below 10 Um
0.15
—
0.13
0.37
0.04
KVB. 15900-529
84
-------
oo
Ul
99.9 -
BAHCO CLASSIFIER!
o.i
10 30 100 300
-EQUIVALENT PARTICLE DIAMETER, MICROMETERS
1000
FIGURE 5-25.
PARTICLE SIZE DISTRIBUTION OF THE BOILER OUTLET FLYASH
FROM BAHCO CLASSIFIER AND SIEVE ANALYSIS
KVB 15900-529
-------
H
N
H
CO
IS
95
80
50
tf
CO
EH
3
01
w
CM
20
0.1
0.3 1 3
EQUIVALENT PARTICLE DIAMETER, MICROMETERS
FIGURE 5-26.
PARTICLE SIZE DISTRIBUTION AT THE BOILER OUTLET
FROM BRINK CASCADE IMPACTOR - TEST SITE D
KVB 15900-529
86
-------
99.9
H
01
£
IS
10
98
90
5 70
EH
a
W
30
10
1 3 10
EQUIVALENT PARTICLE DIAMETER, MICROMETERS
FIGURE 5-27.
PARTICLE SIZE DISTRIBUTION AT THE BOILER OUTLET
FROM SASS GRAVIMETRICS - TEST SITE D
KVB 15900-529
87
-------
5.5 EFFICIENCY OF MULTICLONE DUST COLLECTOR
The multiclone dust collector efficiency was determined in sixteen
tests under various boiler operating conditions. In each case the multiclone
collection efficiency was well below its design efficiency of 90%. The test
data is presented in Table 5-16 and plotted as a function of grate heat
release in Figure 5-28.
The collection efficiency was greatest while burning Century coal,
when it averaged 56.2% ±7.5%. While burning Perfect 8 coal, the collection
efficiency averaged 39.9 ill.9%. The collection efficiency for the single
Victoria coal test was 21.9%.
The reason for the low dust collection efficiency may be related
to the particle size distribution of the flyash. When corrected for the
"lost sample" described in Section 5.4, the Bahco tests yield the following
results:
Test 5 — Perfect 8 coal — 58% below 20um
Test 17 — Century coal — 39% below 20um
Test 23 — Victoria coal — 38% below 20ym
This data indicates that a large fraction of the particulates were small
enough to pass through the collector without being captured.
5.6 MODIFIED SMOKE SPOT NUMBER
Smoke spot readings were taken with a Bacharach Smoke Spot Tester
at the multiclone outlet during nine of the particulate tests (Table 5-17).
One, two and three pump smoke spot readings were taken for each of these tests.
The purpose of this exercise was to determine if a correlation existed between
the smoke spot readings and either the particulate loading or opacity of the
boiler's emissions. As seen in Figures 5-29 and 5-30, no correlation was
found. Previous tests on spreader stokers had also shown no correlation.
KVB 15900-529
88
-------
TABLE 5-16
EFFICIENCY OF MULTICLONE DUST COLLECTOR
TEST SITE D
Test
No.
4
5
6
7
8
9
10
12
13B
14
15
16
17
18
23
Coal Type
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Perfect 8
Century
Century
Century
Century
Century
Victoria
%
Design
Capac-
ity*
64
86
65
63
66
45
63
87
89
87
88
64
87
79
61
02
%
9.3
7.9
8.7
10.2
9.7
11.5
9.9
8.4
7.4
7.5
7.8
8.7
7.3
8.1
10.4
Particulate Loading
lbs/106 Btu
Coll. Inlet
0.664
0.832
0.694
0.793
0.788
0.732
0.707
1.097
0.837
1.336
1.000
0.798
1.072
0.969
0.762
Coll. Outlet
0.557
0.574
0.372
0.488
0.328
0.456
0.354
0.667
0.490
0.754
0.424
0.319
0.393
0.421
0.595
Collector
Efficiency/ %
16.1
31.0
46.4
38.5
58.4
37.7
49.9
39.2
41.5
43.6
57.6
60.0
63.3
56.6
21.9
*Maximum obtainable load averaged 88% of design capacity due to
draft loss from retrofit dust collector.
KVB. 15900-529
89
-------
o
o
8
LU
CJ
CC
LU
Q_
U_
U_
LU
LU
21
O
_l
o
f f
150.0
—r
200.0
—r
250.0
300.0
350.0
GRflTE HERT RELERSE 1000 BTU/HR-SQFT
O : PER 8 CORL -}- ". CENT. CORL
FIG. 5-28
MULTICLONE EFF.
TEST SITE D
i VICT. CORL
VS. GRRTE HERT RELERSE
15900-529
90
-------
TABLE 5-17
MODIFIED SMOKE SPOT DATA
TEST SITE D
Test
No.
2
3
4
5
9
12
13
15
16
AVERAGE SMOKE SPOT READING
1 Pump
3.5
4.0
0.5
4.8
1.0
4.3
3.0
3.0
1.0
2 Pumps
4.5
4.8
2.5
5.8
1.5
6.0
5.0
3.3
1.8
3 Pumps
5.0
5.5
2.8
6.5
2.0
7.0
5.5
4.0
2.0
Multiclone Out
Particulate
lb/106 Btu
__
0.355
0.557
0.574
0.456
0.667
0.490
0.424
0.319
Stack
Opacity
%
___
—
30
23
7
19
29
19
12
KVB 15900-529
91
-------
DC
LiJ
CD
8-
O
O
o
Q_
CO
O CM
21
CO
O O
10.00
OPflCITY
20.00
I I
30.00 40.00
PERCENT
50.00
3 PUMPS
FIG. 5-29
SMOKE SPOT NUI
TEST SITE D
1BER
VS. OPflCITY
15900-529
-------
g-l
8-
00
CD
O o-
21 to
CC
LjJ
QQ
51
i s.H o
O
o_
CO
O ' °
CO
.200 .400 .600 .800 1.000
MULTICLONE OUT PflRT. LB/MILLION BTLI
O I 3 PUMPS
FIG. 5-30
SMOKE SPOT NUMBER VS. MULTICLONE OUT PflRT
TEST SITE D
15900-529
93
-------
5.7 SOURCE ASSESSMENT SAMPLING SYSTEM
One SASS test was run at Test Site D. This test, No. 22, was
originally intended to be run on both Perfect 8 coal and .Century coal.
However, both of these coals became unavailable when it came time to test
and Victoria coal was tested instead. A single particulate test, No. 23,
was run on this coal at the same firing conditions as the SASS test to
supplement the data taken during that test.
All SASS test results will be reported under separate cover at the
conclusion of this test program. The SASS sample catches will be analyzed
by combined gas chromatography/mass spectroscopy for total polynuclear con-
tent. In addition, seven specific polynuclear aromatic hydrocarbons (PAH)
will be sought. These are given in Table 5-18.
TABLE 5-18
POLYNUCLEAR AROMATIC HYDROCARBONS
ANALYZED IN THE SITE D SASS SAMPLE
Element Name
7,12 Dimethylbenz (a) anthracene
Dibenz (a,h) anthracene
Benzo (c) phenanthrene
3-methyl cholanthrene
Benzo (a) pyrene
Dibenzo (a,h) pyrene
Dibenzo (a,i) pyrene
Dibenzo (c,g) carbazole
Molecular
Weight
256
278
228
268
252
302
302
267
Molecular
Formula
C20H16
C22H14
C18H12
C21H16
C20H12
C24Hi4
C24H14
C20H13N
5.8 DATA TABLES
Tables 5-19 through 5-23 summarize the test data obtained at Test
Site D. These tables, in conjunction with Table 2-1 in the Executive
Summary, are included for reference purposes.
94
KVB 15900-529
-------
TABLE 5-19
PARTICULAR EMISSIONS
TEST SITE D
EH
EH
D
O
A
H
O
ffl
Test
No.
4
5
6
7
8
9
10
12
13B
14
15
16
17
18
23
Coal*
P
P
P
P
P
P
P
P
P
C
C
C
C
C
V
% Des
Cap**
64
86
65
63
66
45
63
87
89
87
88
64
87
79
61
02
9.3
7.9
8.7
10.2
9.7
11.5
9.9
8.4
7.4
7.5
7.8
8.7
7.3
8.1
10.4
EMISSIONS
lb/10°Btu
0.664
0.832
0.694
0.793
0.788
0.732
0.707
1.097
0.837
1.336
1.000
0.798
1.072
0.969
0.762
gr/SCF
0.264
0.371
0.291
0.293
0.305
0.239
0.267
0.476
0.387
0.615
0.452
0.339
0.503
0.427
0.276
Ib/hr
35
72
44
50
44
26
44
90
73
114
91
53
96
78
45
Velocity
ft/sec
6.57
10.14
7.29
9.32
8.92
5.32
8.16
9.55
8.86
9.39
9.81
8.12
9.39
10.94
8.80
EH
H
;D
O
rt
8
u
^
0
o
u
H
<;
6
i
3
4
5
6
7
8
9
10
12
13B
14
15
16
17
18
23
P
P
P
P
P
P
P
P
P
P
C
C
C
C
C
V
69
64
86
65
63
66
45
63
87
89
87
88
64
87
79
61
10.1
9.3
7.9
8.7
10.2
9.7
11.5
9.9
8.4
7.4
7.5
7.8
8.7
7.3
8.1
10.4
0.355
0.557
0.574
0.372
0.488
0.328
0.456
0.354
0.667
0.490
0.754
0.424
0.319
0.393
0.421
0.595
0.126
0.208
0.246
0.142
0.168
0.118
0.138
0.127
0.268
0.212
0.329
0.179
0.125
0.176
0.176
0.205
24
29 •
50
24
31
18
16
22
55
43
64
39
21
35
34
35
18.59
13.36
16.07
13.74
15.50
17.09
10.82
14.03
17.37
17.77
14.97
16.25
14.62
17.14
18.09
13.67
* P - Perfect 8 Coal
C - Cent lory Coal
V - Victoria Coal
** Maximum obtainable load averaged 88% of design capacity
due to draft loss from retrofit dust collector.
KVB 15900-529
95
-------
TABLE 5-20
HEAT LOSSES AND EFFICIENCIES
TEST SITE D
o
a
EH
CO
B
4
5
6
7
8
9
10
12
13
14
15
16
17
18
23
to
s
1
K.J
s
8.35
8.76
8.20
8.80
9.20
9.54
8.60
8.10
7.25
7.85
8.11
7.78
7.63
7.76
8.58
g ^
CO Cq
H
o a
0.29
0.28
0.28
0.30
0.27
0.30
0.31
0.32
0.30
0.27
0.23
0.22
0.22
0.20
0.28
1 CN
H
& O
*H
o to
(ND
EC ffl
4.03
4.18
4.03
4.05
4.03
4.02
4.05
4.02
4.01
4.00
4.03
3.95
3.97
3.98
3.97
CO
a <
M o
H ^
||
D 8
S O
8g
0.13
0.18
0.10
0.14
0.12
0.00
0.03
0.05
0.09
0.11
0.09
0.11
0.02
0.27
0.05
co
H
m tc
PQ Pq
§
o a
U H
0.22
0.34
0.22
0.18
0.24
0.15
0.24*
0.21
0.35
0.81
0.34
0.30
0.47
0.41
0.29
a
H
S
CQ CO
H O
EH
CO g
D S
no EH
§ EH
O O
o m
1.14
1.36
1.74
2.04
2.07
1.01
0.68
2.00
2.63
0.95
0.61
0.85
0.81
1.28
3.51
§
m
£1
B
S EL]
SCO
D
£-1
o a
EH H
1.49
1.88
2.06
2.36
2.43
1.16
0.95
2.26
3.07
1.87
1.04
1.26
1.30
1.96
3.85
S
rt
§
H
EH (*!
Q H
< O
K ffl
0.74
0.55
0.72
0.74
0.72
1.05
0.74
0.54
0.53
0.55
0.54
0.73
0.54
0.60
0.77
Q
§
D
3
"
a
D
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
CO
PL)
co
S
3
EH
16.40
17.15
16.79
17.75
18.15
17.57
16.15
16.74
'16.66
16.03
15.45
15.44
15.16
16.00
18.95
w
H
U
H '
FLI
EM
H
83.60
82.85
83.21
82.25
81.85
82.43
83.85
83.26
83.34
83.97
84.55
84.56
84.84
84.00
81.05
* Average used where data was missing
KVB 15900-529
96
-------
TABLE 5-21
PERCENT COMBUSTIBLES IN REFUSE
TEST SITE D
1
CO
E-i
0
w
ft
Test
No.
2
3
4
5
6
7
8
9
10
11
12
13
Average
Boiler
Hopper
._
—
12.3
17.5
8.9
11.8
11.1
0.0
3.8
3.9
6.6
8.4
Boiler
Outlet
__
35.2
23.7
28.8
21.9
16.0
20.8
14.4
22.1
28.2
23.5
Mechanical
Collector
Hopper
16.9
11.9
11.1
13.6
10.7
8.9
17.5
6.1
8.9
8.9
11.5
Mechanical
Collector
Outlet
—
—
28.4
—
::
—
—
28.4
Bottom
Ash
14.9
33.1
19.4
15.1
17.5
19.6
20.7
11.8
8.4
18.7
22.6
18.3
o
u
•B
EH
y
14
15
16
17
18
Average
16.6
18.7
17.9
3.3
40.9
19.5
40.1
23.6
25.9
30.8
28.1
29.7
11.4
16.3
9.7
—
14.9
13.1
__
—
—
27.8
27-7
27.8
15.3
14.5
15.0
15.9
21.3
16.4
EH O
U O
H
22
23
"Average
20.9
4.8
12.9
26.5
26.5
10.8
10.8
33.7
35.9
34.8
KVB 15900-529
97
-------
TABLE 5-22
AS FIRED COAL SIZE CONSISTENCY
TEST SITE D
1
03
EH
O
W
PH
Test
Nuiriber
2
3
4
5
6
7
8
9
10
12
13
Composite*
Average
1"
99.5
98.4
99.0
99.8
99.8
99.2
99.8
98.3
99.0
99.8
99.8
99.3
99.3
PERCENT
1/2"
59.5
65.8
45.4
36.3
46.0
38.9
49.0
38.6
43.1
54.2
57.5
47.0
48.6
PASSING SCREEN
1/4"
15.2
20.1
9.0
5.5
8.6
4.7
7.5
7.0
6.4
9.0
9.4
8.5
9.3
SIZE
#8
6.4
7.6
4.1
2.8
3.5
2.4
3.5
3.9
3.4
3.5
3.8
4.3
4.1
#16
4.0
4.3
3.2
2.3
2.7
2.0
2.6
3.0
2.7
.. 2.4
2.6
3.2.
2.9
J
o
o
CENTURA
14
15
16
17
18
Composite*
Average
98.5
94.6
96.1
96.0
95.0
94.9
96.0
50.2
48.2
41.2
49.7
52.3
44.2
49.7
13.3
13.7
14.3
16.4
18.1
12.2
15.2
6.3
5.7
5.2
7.0
7.6
4.8
6.4
4.5
4.1
2.9
5.1
5.4
3.3
4.4
EH O
O U
H
22
23
Average
88.2
89.4
88.8
39.5
39.6
39.6
15.9
15.6
15.8
7.7
8.1
7.9
4.1
5.7
4.9
* The composite sample includes a coal sample from each test
on a given coal. It is not included in the coal's average
size consistency.
KVB 15900-529
98
-------
TABLE 5-23
STEAM FLOWS AND HEAT RELEASE RATES
TEST SITE D
Test
No.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
22
23
% Design
Capacity*
72.9
69.3
63.8
85.5
65.2
63.3
65.6
44.5
63.2
52.1
87.4
89.3
86.8
88.1
64.1
87.5
79.0
59.1
61.0
Steam Flow
103lb/hr*
65.6
62.4
57.4
77.0
58.6
57.0
59.0
40.1
56.9
46.9
78.6
80.3
78.0
79.3
57.7
78.7
71.1
53.2
54.9
Heat Input
106Btu/hr
69.2
67.5
52.6
86.4
63.2
63.1
56.0
35.0
62.5
49.8
82.5
87.2
85.1
90.8
66.8
89.8
80.6
60.8
59.3
Front Foot
Heat Release
105Btu/hr-ft
4.58
4.35
3.40
5.58
4.08
4.07
3.62
2.30
4.03
3.25
5.32
5.62
5.41
5.86
4.31
5.80
5.20
3.92
3.83
Grate
Furnace
Heat Release Heat Release
103Btu/hr-ft2 103Btu/hr-ft3
280
266
208
341
250
249
221
141
247
199
326
344
331
359
264
355
318
240
234
20.2
19.2
14.9
24.5
18.0
17.9
15.9
10.1
17.8
14.3
23.4
24.8
23.8
25.8
19.0
25.5
22.9
17.3
16.9
* Maximum obtainable load averaged 79,000 Ib stm/hr or 88% of design
capacity. Load restriction was due to draft loss from retrofit
dust collector.
KVB 15900-529
99
-------
APPENDICES
Page
APPENDIX A English and Metric Units to SI Units 102
APPENDIX B SI Units to English and Metric Units 103
APPENDIX C SI Prefixes 104
APPENDIX D Emissions Units Conversion Factors 105
APPENDIX E EPA Method 7 vs Thermo Electron Instrument . . 106
APPENDIX F Flyash Combustible Content vs Particle Size . 107
APPENDIX G Unit Conversion from ppm to lb/105Btu .... 108
KVB 15900-529
101
-------
APPENDIX A
CONVERSION FACTORS
ENGLISH AND METRIC UNITS TO SI UNITS
To Convert From
in
ft
ft2
ft3
Ib
Ib/hr
lb/106BTU
g/Mcal
BTU
BTU/lb
BTU/hr
J/sec
J/hr
BTU/ft/hr
BTU/ft/hr
BTU/ft2/hr
BTU/ft2/hr
BTU/ft3/hr
BTU/ft3/hr
psia
"H20
Rankine
Fahrenheit
Celsius
Rankine
FOR TYPICAL COAL FUEL
ppm @ 3% O2 (SO2)
ppm @ 3% 02 (S03)
ppm @ 3% O2
ppm @ 3% O2
ppm @ 3% O2
(NO)*
(N02)
(CO)
ppm @ 3% O2 (CH4)
To
cm
m
Kg
Mg/s
ng/J
ng/J
J
JAg
W
w
W
W/m
J/hr/m
W/m2
J/hr/m2
W/m3
J/hr/m3
Pa
Pa
Celsius
Celsius
Kelvin
Kelvin
ng/J
ng/J
ng/J
ng/J
ng/J
ng/J
(lb/106Btu)
(lb/!06Btu)
(lb/106Btu)
(lb/106Btu)
(lb/106Btu)
(lb/106Btu)
Multiply By
2.540
6.452
0.3048
0.09290
0.02832
0.4536
0.1260
430
239
1054
2324
0.2929
1.000
3600
0.9609
3459
3.152
11349
10.34
37234
6895
249.1
C
C
K
K
5/9R-2 73
5/9(F-32)
C+273
5/9 R
0.851
1.063
0.399
0.611
0.372
(1.98xlO~3)
(2.47xlO~3)
(9.28xlO~4)
(1.42xlO~3)
(8.65xlO~4)
0.213 (4.95xlO~4)
*Federal environmental regulations express NOx in terms of NO2;
thus NO units should be converted using the NO2 conversion factor.
KVB 15900-529
102
-------
APPENDIX B
CONVERSION FACTORS
SI UNITS TO ENGLISH AND METRIC UNITS
To Convert From
cm
m
m3
Kg
Mg/s
ng/J
ng/J
J
JAg
J/hr/m
J/hr/m2
J/hr/m3
W
W
W/m
W/m2
W/m3
Pa
Pa
Kelvin
Celsius
Fahrenheit
Kelvin
FOR TYPICAL COAL FUEL
ng/J
ng/J
ng/J
ng/J
ng/J
ng/J
To
Multiply By
in
in2
ft
ft2
ft3
Ib
Ib/hr
lb/106BTU
g/Mcal
BTU
BTU/lb
BTU/ft/hr
BTU/ft2/hr
BTU/ft3/hr
BTU/hr
J/hr
BTU/ft/hr
BTU/ft2/hr
BTU/ft3/hr
psia
"H9O
0.3937
0.1550
3.281
10.764
35.315
2.205
7.937
0.00233
0.00418
0.000948
0.000430
0.000289
0.0000881
0.0000269
3.414
0.000278
1.041
0.317
0.0967
0.000145
0.004014
Fahrenheit
Fahrenheit
Rankine
Rankine
ppm @ 3% 02 (SO2)
ppm @ 3% O2 (SO3)
ppm @ 3% O2 (NO)
ppm @ 3% O2 (NO2)
ppm @ 3% 02 (CO)
ppm @ 3% O2 (CH4)
F = 1.8K-460
F = 1.8C+32
R = F+460
R = 1.8K
1.18
0.941
2.51
1.64
2.69
4.69
KVB 15900-529
103
-------
APPENDIX C
SI PREFIXES
Multiplication
Factor
10
10
10
18
15
12
10
103
10
10-1
10
ID'2
10-3
10"6
10-9
10-12
10-15
lo-18
Prefix
exa
peta
tera
giga
mega
kilo
hecto*
deka*
deci*
centi*
milli
micro
nano
pico
femto
atto
SI Symbol
E
P
T
G
M
k
h
da
d
c
m
V
n
P
f
a
*Not recommended but occasionally used
KVB 15900-529
104
-------
EMISSION UNITS CONVERSION FACTORS
FOR TYPICAL COAL FUEL (HV = 13,320 BTU/LB)
Multiply
TO "\. BY
Obtain
% Weight
In Fuel
Weight in Fuel
S N
lbs/106Btu
SO2 NO2
0.666
0.405
grams/106Cal
SO2 N02
0.370
0.225
PPM
(Dry @ 3% 02)
SOx NOx
13.2xlO~4
5.76xlO~4
Grains/SCF.
(Dry @ 12% CO2)
SO2
1.48
.903
SO-
lbs/106Btu
1.50
NO-
SO,
2.70
grams/106Cal
SOx
PPM
(Dry @ 3% 02)
NOx
758
SO-
.676
Grains/SCF
(Dry@12% CO2)
NO-
(.556)
19.8xlO~4
(2.23)
2.47
(.556)
14.2xlO~4
(2.23)
(1.8)
4.44
505
1736
(.448)
1.11
35.6xlO~4
1.
(4.01)
(1.8)
25.6x10
-4
(4.01)
281
1127
704
!.448)
(.249)
391
8.87x10
-4
(.249)
6.39xlO~4
1566
a
H
X
a
NOTE: 1. Values in parenthesis can be used for all flue gas constituents such as oxides of carbon,
oxides of nitrogen, oxides of sulfur, hydrocarbons, particulates, etc.
2. Standard reference temperature of 530°R was used.
KVB 15900-529
-------
APPENDIX E
EPA METHOD 7 VS. THERMO ELECTRON INSTRUMENT
A special set of tests were run to determine the accuracy of our
nitric oxide instrument. Eight NOx sample flasks were prepared and sent
out by Truesdail Laboratories of Los Angeles-, California. They were filled
with flue gas by KVB and returned for analyses by the phenol disulfide (PDS)
method, EPA method 7. At the same time the samples were taken, the nitric
oxide concentration was measured with the Thermo Electron instrument.
The results are shown in the table below. The averages obtained
by the instrument were from -2% to +2% off from the averages obtained by the
PDS method. The PDS flasks were not completely filled and a high residual
vacuum remained which made the oxygen analyses suspect. The difficult oxygen
analyses may have caused two flasks to have poor results when the NO concen-
tration was corrected to 3% ©-
Analyses by
Truesdail
Test
No.
19
20
21
Flask
No.
1344
1187
977
Average
1349
972
1435
Average
1457
1465
Average
NOx
ppm
(dry)
177*
147
149
148
136
140*
143
140
159
147
153
NOx , ppm
dry at
3% 07
327*
227
247
237
254
216*
241
248
306
278
292
Analyses by
KVB Instrument
NO
ppm
(dryl
153
155
155
154
147.5
147.5
145
147
155
156
155
NO , ppm
dry at
3%. O2
228
233
233
231
254
254
252
253
289
300
295
Delta
ppm
-6
+5
+ 3
* Not included in average
KVB 15900-529
106
-------
APPENDIX F
FLYASH COMBUSTIBLE CONTENT VS PARTICLE 'SIZE
Four boiler hopper ash samples from Site D were selected at random,
screened into five size classifications, and baked for combustible deter-
mination. The results are presented below.
% Mass by Size
Size
0-75
75-180
180-425
425-1000
+1000
Test 4
ym
ym
ym
ym
ym
9.
21.
43.
20.
4.
6
5
5
8
5
Test 6
11
22
43
18
4
.4
.6
.6
.3
.1
Test 7
11.
21.
47.
17.
2.
3
3
5
4
5
Test 9
16
30
41
9
1
.9
.8
.5
.3
.5
% Combustibles by Size
Test 4
12.2
11.5
6.2
36.7
71.2
Test 6
6.9
6.7
13.8
21.6
74.4
Test 7
7.0
6.6
4.7
24.5
77.3
Test 9
5.4
6.3
8.5
33.1
74.5
100 300
SIEVE SIZE, MICROMETERS
1000
3000
KVB 15900-529
107
-------
APPENDIX G
UNITS CONVERSION FROM PARTS PER MILLION (PPM) TO
POUNDS PER MILLION BTU INPUT (LB/106BTU)
- SCF
Ib/lO^Btu = (ppm) (fuel factor, b ) (02 correction, n.d. ) (density of
emission, -r-r) (10~6)
i-
Fuel factor, ^0+. " = 106[1.53C + 3.61H2 + -14N2 + .575 - -46O2] ^
(Btu/lb)
where C, H2, N2, S, O2 & Btu/lb are from ultimate fuel analysis;
(a typical fuel factor for coal is 9820 SCF/106Btu ±1000)
O2 correction, n.d. = 20.9 ^ (20.9 - %O2)
where %O2 is oxygen level on which ppm value is based;
for ppm @ 3% 02, O2 correction = 20.9 T 17.9 = 1.168
Density of ' emission = SO2 - 0.1696 Ib/SCF*
NO - 0.0778 Ib/SCF
CO - 0.0724 Ib/SCF
CH4 - 0.0415 Ib/SCF
to convert lbs/106Btu to ng/J multiply by 430
* Standard conditions are 70°F, 29.92 "Hg barometric pressure
KVB 15900-529
108
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-237a
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Field Tests of Industrial Stoker Coal-fired Boilers for
Emissions Control and Efficiency Improvement—Site D
5. REPORT DATE
November 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
J.E. Gabrielson, P.L. Langsjoen, and T.C. Kosvic
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
KVB, Inc.
6176 Olson Memorial Highway
Minneapolis, Minnesota 55422
10. PROGRAM ELEMENT NO.
EHE624
11. CONTRACT/GRANT NO.
IAG-D7-E681 (EPA)
EF-77-C-01-2609
and
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
Final; 7/78 - 9/78
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES IERL_RTP project officer is R.E. Hall. (*) Cosponsors are DoE (W.T.
Harvey, Jr.) and the American Boiler Manufacturers Association. EPA-600/7-78-136a, -79-
Q41a. and -79-130a are similar Site A. B. and C reports.
16. ABSTRACT ^& report g±v&s results of field measurements made on a 90,000 Ib/hr
vibrating-grate-stoker boiler. The effect of various parameters on boiler emissions
and efficiency was studied. Parameters included overfire air, excess air, boiler
load, and fuel properties. Measurements included gaseous emissions, particulate
emissions, particle size distribution of the flyash, and combustible'content of the
ash. Gaseous emissions measured were 02, C02, CO, NO, S02, and S03 in the flue gas.
Sample locations included the boiler outlet and the multiclone dust collecter outlet.
In addition to test results and observations, the report describes the facility
tested, coals fired, test equipment, and procedures. Increasing the overfire air
flow above the minimum required to control opacity resulted in a slight increase in
nitric oxide emissions, a slight increase in boiler efficiency, and mixed results
with particulate loading. At maximum boiler load, the boiler outlet emissions
averaged 1.02 Ib/million Btu for particulate loading and 0.275 Ib/million Btu for
nitric oxide.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
COS AT I Field/Group
Air Pollution
Boilers
iombustion
Coal
Field Tests
Dust
Stokers
Improvement
Efficiency
Flue Gases
Fly Ash
Particle Size
Nitrogen Oxides
Sulfur Oxides
Air Pollution Control
Stationary Sources
Combustion Modification
Spreader Stokers
Particulate
Overfire Air
Flyash Reinjection
13B
13A
21B
21D
14B
11G
07B
3. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
115
20. SECURITY CLASS (This page}
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
109
------- |