EPA
American
Boiler Manufacturers
Association
1500 Wilson Boulevard
Arlington VA 22209
United States
Department
of Energy
Division of Power Systems
Energy Technology Branch
Washington DC 20545
J S. Environmental Protection Agency Industrial Environmental Research EPA-600'7-78-1 36a
Office of Research and Development Laboratory July 1978
Research Triangle Park NC 27711
Field Tests of
Industrial Stoker
Coal-fired
for Emissions
Control and
Efficiency
Improvement -
Site A
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.
REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect the
views and policies of the Government, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-78-136a
July 1978
Field Tests of Industrial Stoker
Coal-fired Boilers for Emissions
Control and Efficiency Improvement
Site A
by
J.E. Gabrielson, P.L. Langsjoen, and T.C. Kosvic
KVB, Inc.
600 South County Road 18
Minneapolis, Minnesota 55426
lAG/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 Energy, Minerals, and Industry
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
-------
ACKNOWLEDGMENTS
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) staff members W. B. (Bill) Marx, Executive Director,
W. H. (Bill) Axtman, Assistant Executive Director, and B. C. (Ben) Severs,
Project Manager, and the members of their 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 partici-
pating committee members listed alphabetically are as follows:
F. C. Belsak Island Creek Coal
R. D. Bessette Island Creek Coal
T. Davis Combustion Engineering
J. Dragos Consolidation Coal
T. G. Healey Peabody Coal
W. B. Hoffmann Hoffmann Combustion Engineering Co.
N. H. Johnson Detroit Stoker
K. Luuri Riley Stoker
J. Mullan National Coal Association
E. A. Nelson Zurn Industries
E. Poitrass The McBurney Corporation
P. E. Ralston Babcock and Wilcox
D. C. Reschley Detroit Stoker
R. A. Santos Zurn Industries
W. Sisken U.S. Department of Energy
We would also like to recognize the KVB engineers and technicians
who spend most of their time in the field, often under adverse conditions,
testing the boilers and gathering data for this program. Those involved to
date are George Moilanen, Jim Burlingame, Russ Parker, Jon Cook, John Rech,
and Jim Demont.
. Finally, our gratitude goes to the host boiler facilities who
invited us to test their boilers. At their request, these facilities will
remain anonymous to protect their own interests. Without their cooperation
and assistance this program would not have been possible.
ii
-------
TABLE OF CONTENTS
Section Page
Acknowledgments ii
List of Figures iv
List of Tables v
1.0 INTRODUCTION 1
2.0 EXECUTIVE SUMMARY 3
3.0 DESCRIPTION OF FACILITY TESTED AND COALS FIRED 17
3.1 Boiler A Description 17
3.2 Coals Utilized 26
4.0 TEST EQUIPMENT AND PROCEDURES 28
4.1 Gaseous Emissions Measurements 28
4.2 Gaseous Emission Sampling Techniques 33
4.3 Sulfur Oxides (SOx) Measurement and Procedure ... 35
4.4 Particulates Measurement and Procedures 39
4.5 Particle Size Distribution Measurement and Procedure 39
4.6 Coal Sampling and Analysis Procedures 42
4.7 Ash Collection and Analysis for Combustibles .... 43
4.8 Boiler Efficiency Evaluation 44
4.9 Modified Smoke Spot Number 45
4.10 Corrosion/Deposition Analysis ... 46
4.11 Trace Species Measurement 48
4.12 Flyash Resistivity Measurement 50
5.0 TEST RESULTS AND OBSERVATIONS 51
5.1 Overfire Air 51
5.2 Flyash Reinjection 56
5.3 Excess Air 60
5.4 Boiler Load 70
5.5 Coal Properties 72
6.0 SPECIAL TESTS 79
6.1 Particle Size Distribution 79
6.2 Size Segregation of Combustible Material 85
6.3 Efficiency of Pollution Control Equipment 85
6.4 Modified Smoke Spot Number 87
6.5 Corrosion Probe Study 91
6.6 Flyash Resistivity 94
6.7 Source Assessment Sampling System . . . 94
APPENDIX 96
iii
-------
LIST OF FIGURES
Figure
3-1 BOILER A SECTIONAL SIDE ELEVATION 18
3-2 BOILER A SCHEMATIC 19
3-3 SAMPLE PLANE GEOMETRY 20
3-4 SAMPLE PLANE GEOMETRY 21
4-1 FLOW SCHEMATIC OF MOBILE FLUE GAS MONITORING LABORATORY . 34
4-2 SOx SAMPLE PROBE CONSTRUCTION 36
4-3 SULFUR OXIDES SAMPLING TRAIN 37
4-4 PARTICULATE SAMPLING TRAIN 38
4-5 BRINK CASCADE IMPACTOR SAMPLING TRAIN SCHEMATIC 41
4-6 FIELD SERVICE TYPE SMOKE TESTER 45
4-7 REFLUX BOILER CORROSION PROBE 47
4-8 SOURCE ASSESSMENT SAMPLING SYSTEM FLOW DIAGRAM 49
5-1 PARTICULATE LOADING VS OVERFIRE AIR 53
5-2 PARTICULATE LOADING VS FLYASH REINJECTION 57
5-3 PARTICULATE LOADING VS EXCESS OXYGEN AT MECHANICAL
COLLECTOR OUTLET 63
5-4 CARBON MONOXIDE VS EXCESS OXYGEN AND LOAD 64
5-5 NITRIC OXIDE VS EXCESS OXYGEN AND LOAD 65
5-6 NITRIC OXIDE TRENDS VS EXCESS OXYGEN AND LOAD 67
5-7 DRY GAS LOSS VS EXCESS OXYGEN, LOAD AND COAL 68
5-8 PARTICULATE LOADING AT THE MULTICLONE OUTLET VS GRATE
HEAT RELEASE 71
5-9 AVERAGE AND STANDARD DEVIATION OF 12 KEMMERER COAL SIEVE
ANALYSIS VS ABMA RECOMMENDED LIMITS FOR SPREADER STOKERS 74
5-10 AVERAGE AND STANDARD DEVIATION OF 6 CONSOLIDATION COAL
SIEVE ANALYSIS VS ABMA RECOMMENDED LIMITS FOR SPREADER
STOKERS 75
5-11 COMPARISON OF TWO COAL SAMPLING LOCATIONS FOR SIEVE
TESTING 77
6-1 COMPOSITE OF ALL PARTICLE SIZING TESTS AT SITE A 81
6-2 COMPARISON OF PARTICLE SIZING TECHNIQUES 82
6-3 COMPARISON OF PARTICLE SIZING TECHNIQUES 83
6-4 MODIFIED SMOKE SPOT NUMBER VS PARTICULATE LOADING .... 88
6-5 MODIFIED SMOKE SPOT NUMBER VS COMBUSTIBLE LOADING .... 89
6-6 CORROSION RATE VS TIME 93
iv
-------
LIST OF TABLES
Table
2-1 EMISSION DATA SUMMARY 7
2-2 PARTICULATE EMISSIONS SUMMARY 8
2-3 SUMMARY OF HEAT LOSS ESTIMATES 9
2-4 SUMMARY OF PERCENT COMBUSTIBLES IN REFUSE 10
2-5 COAL SIZING SUMMARY 11
2-6 FUEL ANALYSIS SUMMARY - STANSBURY COAL 12
2-7 FUEL ANALYSIS SUMMARY - KEMMERER COAL 13
2-8 FUEL ANALYSIS SUMMARY - CONSOLIDATION COAL 14
2-9 MINERAL ANALYSIS OF ASH 15
2-10 SUMMARY OF STEAM FLOWS AND HEAT RELEASE RATES 16
3-1 BOILER A DESIGN DATA 23
3-2 BOILER A DESIGN PERFORMANCE SUMMARY 24
3-3 ULTIMATE ANALYSIS OF COAL UPON WHICH PERFORMANCE DATA IS
BASED 25
3-4 AS FIRED PROXIMATE ANALYSIS OF COALS TESTED IN BOILER A . . 26
3-5 TEST NUMBERS CORRESPONDING TO COALS FIRED 27
5-1 EFFECT OF OVERFIRE AIR ON EMISSIONS AND EFFICIENCY .... 52
5-2 EFFECT OF OVERFIRE AIR ON GASEOUS EMISSIONS 55
5-3 EFFECT OF FLYASH REINJECTION ON EMISSIONS & EFFICIENCY . . 59
5-4 EFFECT OF EXCESS O2 ON COMBUSTIBLES IN REFUSE 61
5-5 COAL PROPERTIES CORRECTED TO A CONSTANT 106BTU BASIS ... 72
5-6 SULFUR BALANCE SUMMARY 78
6-1 VARIATION OF PERCENT COMBUSTIBLES WITH PARTICLE SIZE ... 85
6-2 EFFICIENCY OF POLLUTION CONTROL EQUIPMENT 86
6-3 MODIFIED SMOKE SPOT DATA 90
6-4 CORROSION RATE DATA 92
6-5 SASS TESTS RUN AT SITE A 94
-------
1.0 INTRODUCTION
In recent years the vast majority of industrial boiler instal-
lations have been packaged or shop assembled gas and oil fired boiler units
which could be purchased and installed at substantially lower costs than
conventional coal burning boiler-stoker equipment. Because of the decline
in this market area, little or no work has been done in recent years to
improve specification data and information made available to Consulting
Engineers and Purchasers of coal burning boiler-stoker equipment. The
current implementation of more rigid air pollution regulations has made it
difficult for many coal burning installations to comply with required stack
emission limits, and this has become a further negative influence on coal
burning installations.
The American Boiler Manufacturers Association (ABMA), in conjunction
with the Department of Energy (DOE) and the U.S. Environmental Protection
Agency (EPA), have established a field test program to address this problem.
KVB, Inc., a combustion consulting firm, is performing the testing. The
objective of this program is to produce information which will increase manu-
facturers' ability to design and fabricate stoker boilers which are an
economical and environmentally satisfactory alternative to importation and
combustion of expensive oil. In order to do this, it is necessary to define
stoker boiler designs which will provide efficient operation with minimum
gaseous and particulate emissions, and define what those emissions are in
order to facilitate preparation of attainable national emission standards for
industrial size, coal-fired units.
Further objectives are to: provide assistance to stoker boiler
operators in planning for 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 determine
emissions and efficiency as functions of changes in coal analysis and sizing,
degree of flyash reinjection, overfire air admission, ash handling, grate
size, etc., for various boiler, furnace and stoker designs.
KVB 15900-521
-------
This report is the Final Technical report for the first of many
boilers to be tested under the program described above. It contains a des-
cription of the facility tested, the coals fired, the test equipment and
procedures and the results and observations of testing. A companion volume
to this report contains the original "raw" data sheets from the forty
tests conducted on this boiler. As a compilation of all the data obtained
at this test site, it 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
tie together 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.
To protect the interests of the host boiler facilities, each test
site in this program has been given a letter designation. As the first
site tested, this is the Final Technical Report for Test Site A under the
program entitledr "A Testing Program to Update Equipment Specifications and
Design Criteria for Stoker Fired Boilers."
KVB 15900-521
-------
2.0 EXECUTIVE SUMMARY
This Section outlines the major conclusions drawn from the test
program at Site A. Comments are organized into groups according to the
parameter studied.
Overfire Air. Increasing the overfire air effectively increased
fuel-air mixing in the flame zone by inducing turbulence. The result was a
significant reduction in carbon carryover. Increasing the overfire air
pressure from four to ten inches water pressure while maintaining excess
air constant resulted in:
25-^50% reduction in particulate loading at the boiler outlet
due primarily to a reduction in combustible content of the ash.
0.8-2% increase in combustion efficiency as a result of re-
duced carbon carryover.
Reduction in carbon monoxide (CO) concentrations under those
conditions (low excess 02) where significant concentrations
of CO are formed.
* An average 5% or 16 ppm reduction in nitric oxide (NO)
emissions. A reduction which is not statistically significant.
Flyash Reinjection from the multiclone hopper resulted in an
increase in combustion efficiency but also resulted in an increase in particu-
late loading at the boiler outlet. Three sets of tests with and without
reinjection were run with the following results:
Combustion efficiency increased between 1.5 and 2.5% when flyash
was reinjected. This was due to burnout of combustibles in the
flyash.
Particulate loading at the boiler outlet increased between
22 and 39% when flyash was reinjected. This was due to re-
entrainment of the reinjected ash.
There were no significant changes in concentrations of nitric
oxide or carbon monoxide due to reinjection.
Clinkering on the grate due to reinjection was not observed to
be a problem. However, these tests were short in duration and
may not give the full picture.
KVB 15900-521
-------
Excess air reduction improved boiler efficiency and emissions. A
one percent drop in excess oxygen from six to five percent resulted in:
A reduction in particulate loading at the multiclone outlet
of 5, 17 and 21% in three test sets, and a reduction of 8%
in one test set at the boiler outlet.
A reduction in combustible heat losses of 0.15, 0.55 and 0.83%
in three test sets, and an average decrease in dry gas loss of
0.8%.
* An average reduction in nitric oxide emissions of 55 ppm.
Boiler A was able to operate satisfactorily at 4.5 to 5.0% excess oxygen
(25 to 30% excess air) between 60 and 80% of load burning all three coals
tested. Lower loads required increased excess air; higher loads were not
tested.
Boiler Load. In the range of 60 to 80% of maximum continuous
capacity, a ten percent increase in load brought about:
A ten percent increase in particulate emissions at the
boiler outlet.
An average 45 ppm increase in nitric oxide emissions.
Fuel Properties. Three low sulfur western coals were tested. Each
varied in moisture, ash and sulfur content. Their impact on emissions and
efficiency were as follows:
A three percent ash coal from Kemmerer Coal Company produced
1/3 less particulates at the boiler outlet than an eight percent
ash coal from Consolidation Coal Co. Both coals had similar
ash fusion temperatures and sizings.
One percent or less of the fuel sulfur was retained in the
bottom ash. Between one and four percent of the fuel sulfur
was retained in the flyash. The remaining sulfur was emitted
as SCK and SO3.
Coals averaging 60% through a 1/4 inch square mesh screen were
successfully fired on a day-to-day basis at Test Site A. This
is finer than the 20% to 50% range recommended for spreader
stokers by the ABMA.
KVB 15900-521
-------
The 13% moisture Consolidation coal burned 1.9% more
efficiently than the 19% moisture Kemmerer coal due to
the 6% difference in moisture.
Particle Size Distribution was measured at the boiler outlet, the
multiclone outlet and the electrostatic precipitator outlet. Four sizing
techniques were used. The Brink Cascade Impactor was unacceptable as a
method for sizing particles at the boiler outlet of this spreader stoker.
SASS cyclone, BAHCO classifier and Coulter Counter were acceptable methods
but each had limitations. The results are presented in Figure 6-1.
Combustibles in Ash and the resultant heat losses were found to
be as follows on the average:
Bottom ash: 0.8% Combustibles 0.03% Heat Loss
Multiclone Catch: 66% Combustibles 2.00% Heat Loss
Flyash Passing Multiclone: 30% Combustibles 0.50% Heat Loss
The combustible content of the ash is related to particle size with large
particles containing a larger percent of the combustibles.
Efficiency of Pollution Control Equipment. The plants physical
layout was such that, for sampling purposes, the mechanical collector could
not be isolated from the air heater's settling hopper for determining its
collection efficiency. The two combined removed 94% of the boiler outlet
particulates. The ESP removed 97-98% of the remaining participates. The
sulfur scrubber removed 60% of the SOx entering it. Its design efficiency
was 90%.
Modified Smoke Spot Number did not correlate with either
particulate loading or combustible loading at the boiler outlet.
Corrosion Probe Data was inconclusive. Corrosion coupons were
not exposed long enough to produce repeatable or meaningful corrosion/erosion
rates.
KVB 1,5900-521
-------
Flyash Resistivity. No data was obtained due to application
problems with the Wahlco resistivity device.
Source Assessment Sampling System. SASS test results will be
reported on under separate cover at the conclusion of the testing portion
of this program.
The following summary tables present reduced data from the
testing performed at Site A:
Table Title
2-1 Emission Data Summary
2-2 Particulate Emission Summary
2-3 Summary of Heat Loss Estimates
2-4 . Summary of Percent Combustibles in Refuse
2-5 Coal Sizing Summary
2-6 Fuel Analysis Summary - Stansbury Coal
2-7 Fuel Analysis Summary - Kemmerer Coal
2-8 Fuel Analysis Summary - Consolidation Coal
2-9 Mineral Analysis of Ash
2-10 Summary of Steam Flows and Heat Release Rates
KVB 15900-521
-------
TABLE 2-1
EMISSION DATA SUMMARY
TEST SITE A
Test
No.
1
2A
:B
2C
3D
2E
:F
2J
2H
3
4
5
6
7A
7B
7C
?D
71
8A
SB
9
10
11A
HE
lie
X1D
HE
11F
1C
13A
13B
14A
14B
15
16
17
ISA
16B
19
20
21
22
23
24
25A
2fB
26A
2CB
27
28
29
30
31;.
318
31C
3in
31E
3ir
3;
3-1A
3-J5
15,\
:''"
' '-
Load
Date
8/09/77
8/10/77
8/11/77
8/17/77
8/18/77
8/18/77
S/19/77
O/ it f 1 1
8/20/77
8/23/77
8/24/77
8/24/77
8/24/77
8/25/77
8/26/77
8/27/77
8/30/77
8/31/77
9/11/77
9/12/77
9/14/77
9/23/77
9/23/77
9/24/77
9/24/77
10/05/77
l'VOC/77
10/OH/77
10/10/77
10/10/77
10/13/77
10/14/77
10/15/77
10/29/77
10/30/77
1 1 r'r 1 ^77
11A 1/77
11'C:/T7
i i/V/^
i: ;.'"
%
64
59
61
60
72
70
68
65
52
61
37
41
45
77
74
69
73
59
52
46
44
42
82
83
59
59
74
70
74
81
84
B5
80
53
-.6
33
76
76
SO
It
f..1)
'"
NOTES t Caseous
Coal Typo of Test
s
s
s
s
s
s
s
s
s
s
s
s
s
s
S)
s
s
K
K
K
K
K
K
K
K
K
K
K
K.
K
C
C
r
C
7
K
data
Gaseous drtta
i^aije
ous
data
Part ic-jlat*.
Coal
SOx mech out
Vary excess air
Part mech out
Part mech out
-------
TABLE 2-2
PARTICULATE EMISSIONS SUMMARY
TEST SITE A
Test
No
3
4
5
6
10
14
15
16A
16B
17A
17B
21
22
23
24
26A
26B
28
29
34A
34B
35A
35B
36
37
38
39
%
Load
61
60
72
70
37
74
73
59
59
52
52
82
83
59
59
70
74
84
85
76
76
76
76
79
87
59
60
%
02
4.1
6.3
4.1
5.8
9.9
5.4
4.5
6.1
6.1
6.3
6.3
4.5
4.1
4.4
4.8
5.1
4.6
3.9
3.8
5.9
5.9
6.0
6.0
4.7
5.1
4.7
4.9
Sample Location
Mech Outlet
Mech Outlet
Mech Outlet
Mech Outlet
Mech Outlet
Mech Outlet
Mech Outlet
Mech Outlet
ESP Outlet
ESP Outlet
Stack
Blr Outlet
Blr Outlet
Blr Outlet
Blr Outlet
Blr Outlet
Mech Outlet
Blr Outlet
Blr Outlet
Blr Outlet
ESP Outlet
Blr Outlet
ESP Outlet
Blr Outlet
Blr Outlet
Blr Outlet
Blr Outlet
EMISSIONS
Ib/lO&BTU
0.582
0.646
0.660
0.948
0.537
0.663
0.572
0.512
0.0166
0.0194
0.0128
12.12
16.63
11.67
8.43
9.77
0.600
20.48
15.41
11.89
0.0576
20.48
0.0296
18.51
22.54
6.29
13.26
gr/SCF
0.329
0.317
0.373
0.481
0.199
0.345
0.315
0.259
0.0076
0.0090
0.0057
6.82
9.58
6.60
4.65
5.31
0.335
11.93
9.03
6.10
0.0283
10.44
0.0146
10.23
12.15
3.48
7.24
Ib/hr
153
164
194
271
86
196
174
134
4.04
4.25
2.82
4275
6719
3098
2288
3034
201
8026
5839
3740
16.0
6705
9.0
5819
7181
1568
3223
Velocity
ft/sec
33.25
37.34
38.54
41.95
29.85
42.05
41.16
37.52
67.13
62.12
33.48
34.21
36.96
23.68
24.92
29.95
45.78
37.32
35.95
32.15
78.97
34.65
89.32
30.21
32.11
22.88
22.36
Plow
SCF/sec
905
1006
1014
1095
837
1104
1071
1004
1032
915
958
1220
1365
913
956
1111
1168
1308
1258
1193
1095
1249
1220
1106
1149
878
866
KVB 15900-521
-------
TABLE 2-3
SUMMARY OF HEAT LOSS ESTIMATES
TEST SITE A
g
u
>H
g
pq
en
H
Cfl
in
^
4J
CO
0)
H
3
4
5
6
10
15
co
(0
0
fi
6.94
9.00
7.54
8.75
10.73
7.72
rH
§
c
H
CD
H
4J
U)
H
1.64
1.63
1.63
1.63
1.61
1.63
1 CN
6 SC
0
U 4J
rj Q
fCQ
u-S
0.03
0.03
0.03
0.03
0.04
0.03
H Q)
P Cfl
2 3
3 14-1
jQ QJ
8c
H
H
ID CO
P 4)
1.41
1.75
2.17
3.58
1.05
1.90
6
0
M
[l4
O
H
^J Ij
m CD
r4 r-4
O -H
(5 O
« pq
0.56
0.57
0.47
0.48
0.50
0.46
B
o
Ij ll
fa 0
4-1
rt -H
4J On
a -H
-rH O
O fl>
0.70
0.70
0.70
0.70
0.70
0.70
a
i
co
a)
Q)
I
1.50
1.50
1.50
1.50
1.50
1.50
CO
cu
CO
CO
s
H
10
£
17.15
19.53
18.42
21.03
20.40
18.28
ft
u
a>
r)
U
H
<4-l
H
82.85
80.47
81.58
78.97
79.60
81.72
2
u
a:
3
i
1
21
22
23
24
26
28
29
7.10
7.76
6.78
7.00
8.49
8.52
8.81
2.38
2.40
2.36
2.36
2.39
2.41
2.31
4.23
4.24
4.18
4.18
4.22
4.27
4.29
2.30
0.00
0.00
1.46
1.34
3.83
2.26
0.57
0.72
0.35
0.36
0.34
0.95
0.56
0.01
0.01
0.01
0.01
0.04
0.02
0.00
2.88
0.73
0.36
1.83
1.72
4.80
2.82
0.41
0.40
0.58
0.58
0.46
0.40
0.40
0.70
0.70
0.70
0.70
0.70
0.70
0.70
1.50
1.50
1.50
1.50
1.50
1.50
1.50
19.20
17.73
16.46
18.15
19.48
22.60
20.83
80.80
82.27
83.54
81.85
80.52
77.40
79.17
g
H
EH
§3
88
§
u
34
35
36
37
38
39
7.57
8.02
6.95
7.41
6.07
6.13
1.46
1.47
1.46
1.46
1.43
1.43
4.13
4.16
4.13
4.14
4.05
4.05
1.74
3.18
2.66
0.00
1.00
1.53
0.43
0.79
0.66
0.82
0.25
0.38
0.05
0.09
0.03
0.04
0.01
0.15
2.22
4.06
3.35
0.86
1.26
2.06
0.44
0.44
0.44
0.39
0.58
0.57
0.70
0.70
0.70
0.70
0.70
0.70
1.50
1.50
1.50
1.50
1.50
1.50
18.02
20.35
18.53
16.46
15.59
16.44
81.98
79.65
81.47
83.54
84.41
83.56
KVB 15900-521
-------
TABLE 2-4
SUMMARY OF PERCENT COMBUSTIBLES IN REFUSE
TEST SITE A
o
o
c2
D
w
EH
CO
Test
No.
3
4
5
6
10
14
15
Boiler Multiclone Multiclone
Outlet Outlet Catch
26.3
29.2
37.6
39 . 1
20.1
^
37.6
Bottom
Ash
0.74
£
u
tf
1
1
16A
17
21
22
23
24
26A
26B
28
29
21.4
--
72.1
58.4
43.7
60.8
52.3
30.8
65.8
54.6
76.70
58.81
74.17
80.59
58.73
0.12
0.28
0.26
0.53
1.87
0.17
0.00
.§
H
EH
S ,
Q i-j
W rij
J O
o u
10
z
8
34A
35A
36
37
38
39
__
^»-
60.1
65.6
61.3
47.2
66.6
70.0
68.45
__
54.71
48.06
0.82
1.66
v.
0.23
2.80
AVERAGE
58
.4
30.
3
66
.00
0
.79
10
KVB 15900-521
-------
TABLE 2-5
COAL SIZING SUMMARY
TEST SITE A
% Passing:
TEST NO.
21
22
23
24
26A
26B
27
28
29
30
32
34
35
36
37
38
39
40
1"
96
97
99
98
98
99
99
96
97
95
99
98
98
95
97
98
97
97
1/2"
78
87
83
88
78
89
88
74
76
77
91
82
82
75
88
90
87
86
1/4"
45
70
49
67
47
66
64
44
49
52
72
53
52
50
70
74
72
68
#8
23
40
23
38
23
35
33
22
26
28
38
27
32
27
42
41
44
43
#16
15
26
13
22
14
22
20
14
16
17
23
17
19
17
27
28
28
28
11 KVB 15900-521
-------
TABLE 2-6
FUEL ANALYSIS SUMMARY - TEST SITE A - STANSBURY COAL
TEST NO.
PROXIMATE (as rec.)
% Moisture
% Ash
% Volatile
% Fixed Carbon
BTU/lb
% Sulfur
ULTIMATE (as rec.)
% Moisture
% Carbon
% Hydrogen
% Nitrogen
% Chlorine
% Sulfur
% Ash
% Oxygen (diff.)
ASH FUSION (reducinq)
Initial Deformation
Soft. (H=W)
Soft. (H=1/2W)
Fluid
158
14.21 15.33 13.10
7.00 5.21 5.79
33.96 34.29 35.18
44.83 45.27 45.93
10365 10588 10838
1.05 1.02 0.97
14.21
60.64
4.24
1.01
0.00
1.05
7.00
11.85
9 10 12 14 15 18 19
14.57 13.70 13.36 14.61 13.56 15.05 15.64
5.95 6.70 5.70 4.49 6.13 5.90 8.04
34.79 34.33 35.15 35.03 35.69 34.72 33.97
44.69 45.27 45.79 45.87 44.62 44.33 42.35
10637 10505 10548 10780 10567 10538 10107
0.91 0.92 0.88 0.90 0.86 0.88 0.92
13.70
60.30
4.08
0.96
0.02
0.92
6.70
13.32
2000
2190
2240
2640
AVG.
14.31
6.09
34.71
44.90
10547
0.93
2000
2190
2240
2640
KVB 15900-521
-------
TABLE 2-7
FUEL ANALYSIS SUMMARY - TEST SITE A - KEMMERER COAL
TEST NO. 16* 17
PROXIMATE (as rec.)
% Moisture 17.85 17.80
% Ash 6.97 3.94
% Volatile 33.20 35.72
% Fixed Carbon 42.16 42.54
BTU/lb 9969 10350
% Sulfur 0.62 0.63
ULTIMATE (as rec.)
% Moisture
% Carbon
% Hydrogen
% Nitrogen
% Chlorine
% Sulfur
% Ash
% Oxygen (diff)
ASH FUSION (reducing)
Initial Deformation
Soft. (H=W)
Soft. (H=3/2W)
Fluid
21 22 23
17.83 18.94 20.43
2.65 2.63 2.86
35.99 34.45 34.61
43.53 43.98 42.10
10514 10338 10242
0.49 0.54 0.56
20.43
58.97
4.02
0.72
0.00
0.56
2.86
12.44
24 26A
19.56 18.09
2.92 3.27
34.67 35.31
42.85 43.33
10323 10487
0.58 0.75
2260
2360
2460
2565
26B 27 28
20.42 17.17 18.29
3.14 3.54 3.02
33.92 35.04 34.78
42.52 44.25 43.91
10158 10393 10475
0.77 0.85 0.65
17.17
60.44
4.12
1.01
0.01
0.85
3.54
12.86
29 30
19.79 20.34
2.80 2.81
34.75 34.63
42.66 42.22
10372 10202
0.59 0.53
19.79
60.30
4.09
0.83
0.02
0.59
2.80
11.58
2180
2240
2325
2410
40
17.83
4.58
34.05
43.54
10206
0.63
17.83
58.99
4.08
0.84
0.00
0.63
4.58
13.05
AVG
18.64
3.46
34.70
43.05
10310
0.63
2220
2300
2393
2488
U)
* Not included in averages
KVB 15900-521
-------
TABLE 2-8
FUEL ANALYSIS SUMMARY - TEST SITE A - CONSOLIDATION COAL
TEST NO.
PROXIMATE (as rec.)
% Moisture
% Ash
% Volatile
% Fixed Carbon
BTU/lb
% Sulfur
ULTIMATE (as rec.)
% Moisture
% Carbon
% Hydrogen
% Nitrogen
% Chlorine
% Sulfur
% Ash
% Oxygen (diff.)
ASH FUSION (reducing)
Initial Deformation
Soft. (H=W)
Soft. (H=1/2W)
Fluid
34 35 36 37 38 39
12.72 13.06 12.72 11.29 12.87 14.28
8.79 7.81 6.14 8.23 8.71 7.68
33.82 34,37 34.60 34.34 33.96 33.79
44.67 44.76 46.54 46.14 44.46 44.25
10534 10683 10920 10768 10479 10386
0,73 0.35 0.31 0.28 0.32 0.40
13.06
62.19
4.11
0.82
0.01
0.35
7.81
11.65
2170
2265
2355
2420
AVG.
12.82
7.89
34.15
45.14
10628
0.40
2170
2265
2355
2420
-------
TABLE 2-9
MINERAL ANALYSIS OF ASH
TEST SITE A
Coal
Test No,
MINERAL ANALYSIS OF ASH
Silica, SiO2
Alumina, A1203
Titania, Ti02
Ferric Oxide, Fe2O3
Lime, CaO
Magnesia, MgO
Potassium Oxide, K20
Sodium Oxide, Na2O
Sulfur Trioxide, 803
Phos. Pentoxide, P2°5
Undetermined
Silica Value
Base: Acid ratio
T250 Temperature
Stansbury
10
59.32
12.90
0.55
11.10
5.90
2.16
1.06
0.32
5.71
0.19
0.77
75.57
0.28
2590°F
Kemmerer
40
51.88
18.49
0.75
5.02
8.12
2.76
0.90
0.32
10.77
0.18
0.81
76.54
0.24
2665°F
Consolidation
35
46.02
18.65
0.70
6.96
14.80
1.62
0.65
1.02
8.23
0.78
0.57
66.31
0.38
2435°F
15
KVB 15900-521
-------
TABLE 2-10
SUMMARY OF STEAM FLOWS AND HEAT RELEASE RATES
TEST SITE A
Test
No.
1
2
3
4
5
6
7
a
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26A
26B
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Capacity
%
64
59
61
60
72
70
68
65
61
37
41
45
77
72
73
59
52
46
44
42
82
83
59
59
74
70
74
81
84
85
80
53
83
79
76
76
80
87
59
60
66
Steam Flow
10 3 Ib/hr
191
177
182
180
217
210
203
196
182
110
124
136
232
217
220
178
156
139
132
127
246
250
176
176
222
211
221
243
251
255
239
159
248
237
289
227
240
259
176
180
198
Heat Input
106BTU/hr
228
211
217
208
255
223
240
236
210
127
144
159
272
261
262
210
181
161
155
155
295
300
210
210
267
255
268
294
304
310
290
192
301
289
277
271
289
316
212
219
240
Front Foot
Heat Release
104BTU/ft/hr
839
776
801
768
939
822
886
871
774
470
532
587
1001
962
965
776
668
595
570
570
1088
1106
774
774
984
938
983
1086
1122
1143
1068
708
1111
1066
1022
1000
1066
1166
782
808
886
Grate Heat
Release
103BTU/ft2/hr
441
408
421
404
494
433
466
458
407
247
280
309
527
506
508
408
352
313
300
300
572
582
408
407
518
494
517
571
591
601
562
373
585
561
538
526
561
614
412
425
466
Furnace Heat
Release
102BTU/ft3/hr
136
126
130
125
153
134
144
142
126
76
87
95
163
156
157
126
109
97
93
93
177
180
126
126
160
153
160
177
182
186
174
115
181
173
166
163
173
190
127
131
144
16 KVB 15900-521
-------
3.0 DESCRIPTION OF FACILITY TESTED AND COALS FIRED
This Section discusses the general physical layout and operational
characteristics of Boiler A. The coals utilized in this test series are also
discussed.
3.1 Boiler A Description
Figure 3-1 shows a sectional side elevation of the boiler unit
used during the tests. The boiler is a Foster Wheeler unit. It is designed
to produce 300,000 pounds per hour of steam at 320 psi. The steam is not
superheated and thus leaves the boiler at its saturation temperature of
427°F. The boiler was built and first operated in 1976.
The stoker is a Detroit Stoker unit. It has seven feeders and a
split traveling grate with front ash discharge. Overfire air is introduced
through two rows of jets on the back wall and two rows of jets on the front
wall. Each row of jets is dampered manually. The overfire air system is
completely separate from the forced draft system and is not heated by the
air heater.
The arrangement of the boiler's economizer, air heater and collection
equipment is shown in Figure 3-2. This figure shows a block diagram of how
the individual components are arranged. The gas flows from the boiler through
a tubular type air preheater (with a settling chamber), into a multiclone
dust collector, into an electrostatic precipitator, then through the economizer
followed by an induced draft fan, and finally through a scrubber for sulfur
oxide removal. From there it exits up the stack.
Flyash is reinjected continuously from the boiler hopper and the
air heater hopper. Flyash from the mechanical collector hopper can be either
reinjected or dumped into a surge hopper and discarded. Flyash reinjection
air comes from the overfire air system.
Sample ports were installed at the locations shown on the boiler
schematic. Figure 3-2. These sample locations were not ideal relative to
EPA testing recommendations or for performing boiler heat balances. Because
17 KVB 15900-521
-------
I
i
m
ft
;!
i
(i
P
i|«!
' 1
i '.
it
I
Ois
^ DKuM-i
rr
r
U^5
1 wrs
FIGURE 3-1. BOILER A SECTIONAL SIDE ELEVATION
18
KVB 15900-521
-------
BOILER OUTLET
SAMPLE PLANE
MECHANICAL
COLLECTOR OUTLET
SAMPLE PLANE
STACK
SAMPLE
PLANE
ESP
SAAA/
STACK
ESP OUTLET
SAMPLE PLANE
FIGURE 3-2. BOILER A SCHEMATIC
(Not Drawn to Scale)
KVB 15900-521
-------
BOILER OUTLET SAMPLING PLANE
7
II II
0*
f
* +
6
II
A
f
II II
i :
+ +
4
6
+
^ ., "1C, \ T> >
3'7'
Cross Sectional Area = 95.26 ft2
MECHANICAL COLLECTOR OUTLET SAMPLING PLANE
3 2
II II | 1 II 1 |
1
II II
A ^^ A ^\ ^ Id ^
» ty » » ^S> ^
2 '2"
1
Cross Sectional Area = 57.04
KEY: + Particulate Sampling Point
O Gaseous Sampling Point
SCALE: 1 inch = 5 feet
NOTE: Numbers refer to probe numbers on original gaseous
data sheets for Tests 21-40. For Tests 1-20 all
seven probes were in mechanical outlet duct
numbered 1-7 right to left.
FIGURE 3-3.
SAMPLE PLANE GEOMETRY
TEST SITE A
20
15900-521
-------
PRECIPITATOR OUTLET SAMPLING PLANE
_J
_^ . + + * +
-,<'*
I
4
>
I
8.5
f
Cross Sectional Area = 26.48 ft2
STACK SAMPLING PLANE
KEY: + Particulate Sampling Point
O Gaseous Sampling Point
SCALE: 1 inch =2.5 feet
Radius = 3'4"
Cross Sectional Area = 34.91 ft2
FIGURE 3-4.
SAMPLE PLANE GEOMETRY
TEST SITE A
21
15900-521
-------
of physical limitations on the installation of sample ports, heat loss
calculations required assumptions to be made concerning the collection
efficiency of the dust collector which could not be isolated by a separate
sampling station from the air heater. Sample plane geometry is shown in
Figures 3-3 and 3-4.
The facility has two identical boilers. When both are on line,
the one being tested can be put on manual operation and process demand
fluctuations can be met by modulation of the second boiler. Unfortunately,
the other boiler was down during much of the test program so that getting and
maintaining the desired loads was a problem. The test boiler had to meet
process demands. This problem was aggravated by the fact that furnace
excess air was always on manual control because of automatic control problems
with the FD fan dampers.
The boiler design data is summarized in Table 3-1 following. The
design performance and pressure drops are given in Table 3-2. This data
was estimated utilizing a design coal whose properties are given in Table
3-3.
22 KVB 15900-521
-------
TABLE 3-1
BOILER A DESIGN DATA
Foster Wheeler Boiler
Design Steaming Capacity 300,000 Ib/hr
Design Pressure 350 psi
Final Steam Temperature 427°F
Boiler Heating Surface 20,186 ft2
Waterwall Heating Surface 2,982 ft2
Economizer Heating Surface 13,276 ft2
Furnace Volume 16,712 ft3
Year Built 1976
Type VOSP
Detroit Stoker
Number of feeders 7
Grate Type Split, Continuous Front Discharge
Grate Length (shaft to shaft) 20'8"
Grate Width 27'1-1/2"
Effective Grate Area 515.4 ft2
Recommended Coal Sizing 3/4"xO with no more than 25% thru 1/4"
Grate Thermocouple Location 6 in air seal beams, 23-1/8" forward
of rear grate shaft
Overfire Air
Upper Rear: 25 jets @ 12" spacing, 6'0" above grate, 9° below
horizontal
Lower Rear: 27 jets @ 8-16" spacing, 2'2" above grate except end
two which are I'O" above grate, horizontal
Upper Front: 25 jets @ 12" spacing, 6'6" above grate, 19° below
horizontal
Lower Front: Underfeeder air located about 10" above grate
Discussion: The OFA system is completely separate from the FD
system. The overfire air is nonheated ambient air
from inside the plant. Maximum obtainable OFA pressures
are 12-15 in. H2O.
Flyash Reinjection
Boiler Hopper: 6 injectors @ 4'0" and 4'4" spacings
Air Heater Hopper: 7 injectors @ 4'0" and 4'4" spacings
Mechanical Hopper: 12 injectors @ 2'0" and 2'4" spacings
Discussion: Mechanical hopper ash can be stored in a surge hopper
and discarded rather than reinjected at the boiler
operators option.
23 KVB 15900-521
-------
TABLE 3-2
BOILER A DESIGN PERFORMANCE SUMMARY
Max. Load Base Load
Steam Flow, 103lb/hr 300 250
Steam Temperature Boiler Outlet, °F 427 427
Pressure Boiler Drum, psig 320 320
Temp. Feedwater Entering Unit, °F 240 240
Temp. Feedwater Leaving Economizer, °F 313 291
Temp. Air Entering Unit, °F 80 80
Temp. Air Leaving Air Heater, °F 317 346
Temp. Gas Leaving Furnace, °F 1,960
Temp. Gas Leaving Boiler, °F 765 687
Temp. Gas Leaving Air Heater, °F 600 498
Temp. Gas Leaving Economizer, °F 350 310
Excess Air Leaving Furnace, % 25 25
Wet Gas Entering Air Heater, 103lbAr 364 302
Wet Gas Leaving Air Heater, 103lb/hr 364 302
Air Entering Air Heater, 103lb/hr 326 271
Air Leaving Air Heater, 103lb/hr 326 271
Draft in Furnace, in. H20 -.2 -.2
Gas Side Loss Thru Boiler, in. H2O 0.56 0.39
Gas Side Loss Thru Mechanical D.C., in. H20 2.50 1.72
Gas Side Loss Thru Air Heater, in. H20 2.90 2.00
Gas Side Loss Thru ESP, in. H20 0.64 0.44
Gas Side Loss Thru Economizer, in. H20 2.07 1.42
Gas Side Loss Thru Flues, in. H20 0.64 0.44
Air Side Loss Thru Air Heater, in. H20 2.74 1.89
Air Side Loss Thru Ducts, in. H20 0.76 0.53
Air Side Loss Thru Grate, in, H20 3.17 2.19
Air Side Loss Thru Steam Coil, in. H20 1.00 0.69
Air Side Loss Thru Air Measuring Device, in. H20 1.27 0.88
Fuel Flow, 103/lb/hr 31,726 26,000
Liberation, BTU/hr/ft3 total vol. 21,452 17,584
Dry Gas Loss, % 6.47 5.51
Hydrogen & Moisture in Fuel Loss, % 5.50 5.41
Moisture in Air Loss, % 0.11 0.09
Unburned Combustible Loss, % 1.70 1.41
Radiation Loss From Boiler, % 0.34 0.41
Radiation Loss From Precipitator, % 0.70 0.70
Manufacturers Margin, % 1.50 1.50
Unit Efficiency, % 83.68 84.97
24 KVB 15900-521
-------
TABU: 3-3
ULTIMATE ANALYSIS OF COAL UPON WHICH PERFORMANCE DATA IS BASED
Ash 2.67% (8% max.)
S 0.81% (1.5% max.)
H2 4.71%
C 64.83%
N2 1.08%
02 14.90%
H20 11.00%
BTU/lb as Fired 11,300
25 KVB 15900-521
-------
3.2
Coals Utilized
During the test program on Boiler A three coals were utilized.
These were a Wyoming coal from Stansbury Coal Company, a Wyoming coal from
Kemmerer Coal Company, and a Colorado coal from Consolidation Coal Company.
Each of these coals has different thermal and ash properties. Table 3-4
below summarizes the average as fired proximate analysis of the coals
tested in Boiler A. A complete fuel analysis summary for each coal is
given in Tables 2-6 thru 2-9 in the Executive Summary, Section 2.0.
TABLE 3-4
AS FIRED PROXIMATE ANALYSIS OF COALS TESTED IN BOILER A
Coal Company:
% Moisture
% Ash
% Volatile
% Fixed Carbon
BTU/lb
% Sulfur
Initial Deformation
of Ash (Reducing) 2000°F
Stansbury Kemmerer Consolidation
14.31
6.09
34.71
44.90
10547
0.93
18.64
3.46
34.70
43.05
10310
0.63
12.82
7.89
34.15
45.14
10628
0.40
2220°F
2170°F
The Kemmerer coal differed from the others by its higher moisture
(19°i) and lower ash (.3.5%). By comparison, the Consolidation coal averaged
13% moisture and 8% ash. The Stansbury coal differed primarily in its higher
sulfur content. All three coals had low ash fusion temperatures.
The coal handling system at Test Site A operates as follows.
Bottom dump hopper cars discharge the coal onto a system of three belt con-
veyors which deliver the coal to a standpipe. The standpipe is a ported
cylinder which builds a conical coal pile and minimizes wind-related losses
during stacking operations.
26
KVB 15900-521
-------
A front end loader transfers the coal from the active pile or
from dead storage to a reclaim hopper and an apron feeder. The coal is
conveyed from the apron feeder to a tripper which fills the bunker above the
boiler.
The dates of introduction for each of the three coals are given
in Table 3-5 below. Also shown are the test numbers corresponding to each
of these coals.
TABLE 3-5
TEST NUMBERS CORRESPONDING TO COALS FIRED
Test No. Coal Source Date Introduced
1-15 Stansbury August 3, 1977
16r-17 Kemmerer I August 28, 1977
18-19 Stansbury September 9, 1977
20-33 Kemmerer II September 12, 1977
34-39 Consolidation October 25, 1977
40 Kemmerer II November 3, 1977
The Kemmerer coal burned during tests 16 and 17 reportedly came
from a different seam than the coal burned during tests 20-33. During both
tests 16 and 17 a high ash,low BTU coal was blended into the Kemmerer coal.
This blending was especially evident in test 16's coal analysis as shown
in Table 2-7,
27 KVB 15900-521
-------
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 was
collected.
4.1 Gaseous Emissions Measurements (NO, NO», CO, CO2, O^/ HC).
A description is given below of the analytical instrumentation and
related 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 over five years of testing, and are
operational and fully checked out.
A. Analytical Instruments and Related Equipment. The analytical
system consists of five instruments and associated equipment for simultaneously
measuring the composition of the flue gas. The analyzers, recorders, valves,
controls, and manifolding are mounted to 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 (C02), oxygen (©2)/ and gaseous hydrocarbons (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:
0 Nitric Oxide/total oxides of nitrogen (NO/NOx)
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
0 Carbon Monoxide
Beckman Model 315B NDIR Analyzer
Range: 0-500 and 0-2000 ppm CO
Accuracy: ±1% of full scale
28 KVB 15900-521
-------
0 Carbon Dioxide
Beckman Model 864 NDIR Analyzer
Range: 0-5% and 0-20% CO2
Accuracy: ±1% of full scale
0 Oxygen
Teledyne Model 326A Fuel Cell Analyzer
Range: 6-5, 10 and 25% 02 full scale
Accuracy: ±1% of full scale
0 Hydrocarbons
Beckman Model 402 Flame lonization Analyzer
Range: 5 ppm full scale to 10% full scale
Accuracy: ±1% of full scale
The oxides of nitrogen monitoring instrument used is a Thermo
Electron chemiluminescent nitric oxide analyzer. The operational basis of
the instrument is the chemiluminescent reaction of NO and 03 to form NO2.
Light emission results when electronically excited !K>2 molecules revert to
their ground state. This 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 through an air dryer
and a 10 micron 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 which insures that no water collects 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, N©2 molecules in the sample gas are
reduced to NO molecules, and the analyzer now reads NOx. NO2 is obtained by
the difference in readings obtained with and without the converter in operation.
29 KVB 15900-521
-------
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
Sensitivity 0.5 ppm
Linearity ±1% of full scale
Vacuum detector operation
Rar.ge: 2.5, 10, 25, 100, 250, 1000, 2500, 10,000
ppm full scale
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 infrared 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 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 115 ± 15 V rms
Response: 90% of full scale in 0.5 or 2.5 sec.
Precision: ±1% of full scale
Output: 4-20 ma
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
30
KVB 15900-521
-------
infrared 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 CC>2
analyzer are 0-5% and 0-20%.
Specifications: Span Stability ±1% of full scale in 24 hours
Zero Stability il% of full scale in 24 hours
Ambient Temperature Range 32°F to 120°F
Line Voltage 115 t 15 V rms
Response: 90% of full scale in 0.5 or 2.5 sec.
Precision: ll% of full scale
Output: 4-20 ma
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: il% of full scale
Response: 90% in less than 40 sec.
Sensitivity: 1% of low range
Linearity: tl% of full scale
Ambient Temperature Range: 32-125°F
Fuel cell life expectancy: 40,000%-hrs.
Power Requirement: 115 VAC, 50-60 Hz, 100 watts
Output: 4-20 ma
Hydrocarbons are measured using a Beckman Model 402 hydrocarbon
analyzer which utilizes the flame ionization method of detection. The sample
is drawn through a heated line to prevent the loss of higher molecular weight
31 KVB 15900-521
-------
hydrocarbons to the analyzer. 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
+ 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 an electronic measuring circuit. This ionization current is pro-
portional 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
Specifications: Full scale sensitivity, adjustable from 5 ppm
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
Reproducibility: ±1% of full scale for successive
identical samples
Analysis Temperature: Ambient
Ambient Temperature: 32°F to 110°F
Output: 4r*20 ma
Air Requirements: 350 to 400 cc/min of clean,
hydrocarbons-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
32
KVB 15900-521
-------
Recording Instruments. The outputs of the four analyzers are
presented 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:
Specifications: Strip Chart display
Chart Size: 9-3/4 inch
Accuracy; _0.2 5%
Linearity: <0.1%
Line Voltage: 120V i 10% at 60 Hz
Span Step Response: 1 sec.
B. Gas Sampling and Conditioning System. The gas sampling and
conditioning system consists of the probes, sample line/ valves, pumps,
filters and other components necessary to deliver a representative, conditioned
sample gas to the analytical instrumentation. The following section describes
the system and the components which make up the system. The entire gas
sampling and conditioning system shown schematically in Figure 3-1 is con-
tained in the emission test vehicle.
4.2 Gaseous Emission Sampling Techniques, (NOx, CO, CC>2, 02, HC)
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 micron Mott Metal-
lurgical Corp. sintered stainless steel filter is attached to each probe for
removal of particulate material.
Gas samples to be analyzed for ©2, 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 temperature
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
outside. Gas samples are drawn sequentially from all probes for each test.
The average emission values are reported in this report.
33 KVB 15900-521
-------
.......... SSiv.r: sfssssaam t imli
FIGURE 4-1.
FLOW SCHEMATIC OF MOBILE FLUE GAS MONITORING LABORATORY
KVB 15900-521
-------
4.3 Sulfur Oxides (SOx) Measurement and Procedure
Measurement of SO2 and 803 concentrations are 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.
Some of the sulfur trioxide is removed by the first absorber, while the re-
mainder, which passes through as a sulfuric acid mist, 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 trans-
fer 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 tri-
oxide 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.
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.
35 KVB 15900-521
-------
Flue Wall
Asbestos Plug
Ball Joint
Insulation
Support Tube
Vycor
Sample Probe
Heating I *
Tape Pryoraeter
and
Thermocouple
FIGURE 4-2. SOx SAMPLE PROBE CONSTRUCTION
15900-521
36
-------
Trap
Dial Thermometer
Pressure Gauge
Volume Indica
Vapor Trap Diaphragm
Pump
Dry Test Meter
FIGURE 4-3. SULFUR OXIDES SAMPLING TRAIN
15900-521
37
-------
00
Heated Probe
Sampling Nozzle
Reverse Type
Pitot Tube and
Gas Temp T/C
Control
Unit
T/C (Impinger
Out Temp)
T/C (Impinger
in Temp)
Check Valve
Velocity
Pressure
Gage
Impingers ^Ice Bath
Fine Control Valve
Sample
Vacuum
Gage
Orifice
Gage (AP)
Dry Test Meter
Coarse
Control
Valve
Air-Tight
Pump
Umbilical
Cord
FIGURE 4-4. PARTICULATE SAMPLING TRAIN
15900-521
-------
4.4 Particulates 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 5 microns and a 100 mm glass fiber
filter for retention of particles down to 0.3 microns. Condensible parti-
culates 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 to.lmg), 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.
4,5 Particle Size Distribution Measurement and Procedure
The measurement of particle size distribution of the flyash is
performed using a Brink Model "B" Cascade impactor. The Brink impactor is
a five stage, low sample rate, cascade impactor suitable for measurements in
high mass loading situations. A schematic of the Brink sampling train is
shown in Figure 4-5.
Samples are pulled isokinetically from a single sample point. The
flow rate through the impactor is held constant during sampling to preserve
the impaction cut points.
Gelman type A-E binderless glass fiber filter paper is used as
the collection substrate. The main purpose of the glass mats is to reduce
39 KVB 15900-521
-------
re-entrainment due to particle bounce. The 5/8 inch diameter mats are cut
from larger stock with a cork bore and inserted in the collection plates.
The collection plates with mats installed are desiccated 24 hours before
tare weighing. After sampling, all particles adhering to the impactor walls
are brushed down onto the collection plate immediately below. The plates
are again desiccated 24 hours before weighing.
The cyclone catch is brushed onto a tare weighed paper, desiccated
and weighed. The final filter, cut from the same fiber glass stock as the
collection plate substrates, is treated the same as the collection plates.
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 mano-
meter, thermocouple and indicator are used.
Second, a nozzle size is selected which will maintain 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 cali-
bration curve. This pressure drop is corrected for temperature, pressure
and molecular weight of the gas to be sampled.
The impactor is placed in the duct for 20-30 minutes prior to
sampling to allow it to be heated to stack temperature. During this warm up
period, the sample nozzle is turned away from the direction of gas flow so
that no particulates will be collected. Once hot, the stages are re-tightened
with pipe wrenches to prevent leakage. The impactor's nozzle is then turned
into the gas stream for collecting the particulate sample.
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 re-entrainment, a rule of thumb is that no stage should be loaded
above 10 mg.
The volume of dry gas sampled is measured with a dry gas meter.
This allows calculation of actual isokinetics. The dry gas volume is also
40 KVB 15900-521
-------
FIGURE 4-5
PRESSURE TAP
FOR 0-20"
MAGNAHELIX
BRINK CASCADE IMPACTOR
SAMPLING TRAIN SCHEMATIC
CYCLONE
STAGE 1
STAGE 2
STAGE 3
STAGE 4
STAGE 5
FINAL FILTER
EXHAUST
i
ELECTRICALLY HEATED PROBE
DRY GAS
METER
FLOW CONTROL
VALVE
DRYING
COLUMN
41
15900-521
-------
used to convert test results to concentration units. Stack moisture used
for calculating isokinetics is measured with the EPA Method 5 sample train
during concurrent particulate sampling.
Data reduction involves a time-consuming iterative process and
is best accomplished with the aid of a computer. For this purpose KVB
developed a 223 step program for the Texas Instruments SR-52 card program-
mable calculator. With this program, Brink data reduction can be easily
done in the field.
In addition to the Brink Cascade Impactor, particle sizing is
accomplished by several other methods. The SASS train utilizes three sized
cyclones and a final filter under controlled temperature and flow rates to
achieve gravimetric separation at ten, three and one microns.
Selected flyash samples are sent to independent laboratories for
sizing using the BAHCO centrifugal classifier (PTC 28) or the Coulter Counter.
4.6 Coal Sampling and Analysis Procedures
Coal samples are taken from the base of the non-segregating (conical)
hopper immediately above the feeders. The reason for selecting this sampling
location is discussed in Section 5.5.
Samples are collected by lifting the feeder inspection doors and
allowing 10-20 pounds of coal to flow into a rectangular bucket. The first
sample is discarded to purge the area near the inspection door. The second
sample taken immediately after the first, is quartered with a sample splitter.
One quarter is saved. ...........
This process is repeated in a random pattern for each of the seven
feeders. Samples are taken at fifteen to twenty minute intervals over the
course of testing the boiler. At the completion of each test the cumulative
sample is passed thru a sample splitter several times until the last two
splits are approximately six pounds each. One sample is sealed in a plastic
bag for chemical analysis. The other sample is processed for sieve analysis.
Sieve analysis is accomplished with a Gilson Porta-Screen Model
PS-3. This device holds five trays with 14"xl4" screen areas, and a dust pan.
42 KVB 15900-521
-------
Approximately six pounds of air-dried coal is placed in the top tray. After
shaking for one minute, the sample in the top tray is removed and weighed. The
second tray is shaken for two minutes, the third for four, the fourth for
eight, the fifth for sixteen. In this way, wearing down of the particles is
minimized while allowing sufficient time for size segregation. Screen sizes
used are 1", 1/2", 1/4", #8 and #16 mesh.
Coal analysis is performed by 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,
selected samples receive ultimate analysis ash fusion and mineral analysis
of the ash.
4,7 Ash Collection and Analysis for Combustibles
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 desiccated and tare weighed ceramic crucible. The crucible with sample
is heated in an oven at 110°C 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 750°C 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 110°C weight.
Bottom ash samples are collected from the bottom ash hopper within
two hours after completion of each test. The ash hopper is cleared just
prior to the test to insure that the hopper contains only ash generated during
the test. Four five-pound samples are collected representing a cross-section
of the ash hopper. These samples are mixed, quartered, and sent to Commercial
Testing and Engineering Company, South Holland, Illinois, for combustible
determination.
43 KVB 15900-521
-------
Multiclone ash samples are taken from four ports near the bottom of
the hopper into which the multiclone ash is dumped prior to being discarded.
This sample, approximately two liters, is sent to Commercial Testing and
Engineering Company for combustible determination.
4.8 Boiler Efficiency Evaluation
Boiler efficiency is calculated using the ASME Test Form for
Abbreviated 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,
combustible 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 form where combustible losses are lumped
into one category, combustible losses are calculated and reported separately
for combustibles in the bottom ash, combustibles in the mechanically
collected ash which is not reinjected, and combustibles in the flyash leaving
the mechanical collector.
KVB has developed a program for the Texas Instrument's SR-52 card
progrartmable calculator to compute the above heat losses. Use of this program
helps minimize human error in the calculations.
44 KVB 15900-521
-------
4.9
Modified Smoke Spot Number
Modified Bacharach smoke spot numbers are determined using a
Bacharach field service type smoke tester. ASTM procedures for this measure-
ment apply only to oil fired units. Therefore, KVB has 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 and two strokes
were taken at the boiler 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 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 36,900 ±1650 cubic centimeters of gas at 16°C and 1 atmosphere pressure
through an enclosed filter paper for each 6.5 square centimeters effective
surface area of the filter paper.
Filter Paper
Handle
FIGURE 4-6. FIELD SERVICE TYPE SMOKE TESTER
45
KVB 15900-521
-------
The standard smoke scale consists of a series of ten spots numbered
consecutively from 0 to 9, and ranging in equal photometric steps from white
through neutral shades of gray to black. The standard spots are imprinted
on white paper having an absolute surface reflectance of between 82.5 and
87.5%, determined photometrically. The smoke scale spot number is defined
as the reduction (due to smoke) in the amount of light reflected by a spot
divided by 1O.
The smoke density is reported as the Smoke Spot Number of the spot
on the standard smoke scale that most closely corresponds to the color of
the soiled spot on the sample filter paper. Differences between two standard
Smoke Spot Numbers are interpolated to the nearest half number.
4.10 Corrosion/Deposition Analysis
The method used to determine corrosion rates on boiler water
tubes is to insert metal coupons representing a segment of water tube
into the boiler furnace. The coupons are attached to the end of a probe
which controls coupon temperature. These are inserted into the furnace
through existing viewport openings and exposed to the corrosive action of
the flue gas for about one month. The coupons are then removed and examined
for corrosion effects.
The corrosion probes used in this program were designed and
developed by KVB, Inc. The basic concept for cooling the coupons has
evolved from early air cooled models through heat pipe models to the current
reflux boiler concept shown in Figure 4-7.
In the reflux boiler design concept, condensed fluid is returned
to the evaporator section via gravitational force instead of capillary action
through a wetted wick as in the heat pipe design. This is accomplished by
tipping the probe about 15 degrees off horizontal with the evaporator end
down.
The evaporator section of the probe is insulated with wet felt.
The function of the insulation is to reduce the heat load into the probe
in those areas away from the coupon location. Thus, the coupons are
the only evaporator segment exposed to the boiler heating environment.
46 KVB 15900-521
-------
PRESSURE
RELIEF
VALVE
FIGURE 4-7. REFLUX BOILER CORROSION PROBE
15900^521
-------
The condenser section of the corrosion probe has shrouded external
fins and forced air cooling. The cooling air is supplied by a Rotran model
TN 3AZ fan driven by an 85 watt, 155 VAC, 50/60 Hz, single phase motor.
Thus, electrical power is the only requirement for operation of the unit.
New coupons are washed in acetone and air dried prior to weighing.
After exposure in the boiler, the coupons are carefully cleaned with a wire
brush. They are then cleaned in an ultrasonic bath of 1.0 normal inhibited
hydrochloric acid. After rinsing with acetone, any remaining carbon deposits
are removed with a brass brush. The coupons are next cleaned in an ultrasonic
bath of benzene or trimethylene chloride for fifteen more minutes. They
are then cleaned one more time in an ultrasonic bath of 1.0 normal inhibited
hydrochloric acid and rinsed with acetone prior to drying and weighing.
4.11 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 4-8). The "catch" from the SASS train
can be analyzed for Poly Chlorinated Biphenyls (PCB's), Poly Organic Matter
(POM) and other trace species.
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) > loy B) 3y to loy c) ly to 3y
Together with a filter, a fourth cut (
-------
vO
Convection
oven
Filter
Gas cooler
trace element
collector
Stack velocity (AP)
magnehelic gauges
Gas
T6**1" Coarse adjustment
'y" Fine valve
\\ adjustment
« valve O" \ (\
Implnger
T.C.
Vacuum
gage
Orifice AH, V \
wgnehellc gauqe ^ -"
Dry test meter
FIGURE 4-8. SOURCE ASSESSMENT SAMPLING SYSTEM FLOW DIAGRAM
KVB 15900-521
-------
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 a 10 cfm high volume vacuum
pump, while required pressure, temperature, power and flow conditions are
obtained from a main controller.
4.12 Flyash Resistivity Measurement
The Wahlco Resistivity Probe is an in situ field device for use
in investigating problems with electrostatic precipitators. The means of
collection is mechanical so no dust characteristics are destroyed during the
sampling process. The ^robe can be used in temperatures up to 450°F. An
integral cleaning system allows repeated tests in a single location without
removing the probe from the duct. All instrumentation and probe hardware
are contained in a single carrying case suitable as a shipping container.
The Resistivity Probe consists of a small cyclone inserted in the
duct which collects a dust sample in a cylindrical stainless steel cup. A
high voltage discharge pin mounted axially and electrically insulated from
the cup serves as the energizing electrode. The steel cup serves as the
receiver. The high voltage supply is held at 1,000 volts and resistivity of
the collected sample is then determined as a function of the current.
An integral cleaning system permits emptying of the cup without
removing the probe from the duct. Air is blown through a separate tube to
a purge coil and then into the bottom of the cup, thus discharging the dust
back into the flue gas stream. The purpose of the purge coil is to pre-heat
incoming air to prevent condensation in the cup.
50 KVB 15900-521
-------
5.0 TEST RESULTS AND OBSERVATIONS
This Section presents the results of the tests performed on
Boiler A. Observations are made regarding the influence on gaseous and
particulate emissions and efficiency as the control parameters were
varied. A total of forty tests were conducted in a defined test matrix
to develop this data.
5.1 Overfire Air
Overfire air was varied to determine its effect on both emissions
and boiler efficiency. It was found to effectively increase boiler efficiency
and reduce particulate carryover by increasing combustible burnout in the
flyash. High overfire air was also found to reduce carbon monoxide (an
indicator of incomplete combustion) under those conditions (low excess air)
where carbon monoxide is found; and, it was found to slightly decrease
nitric oxide emissions from Boiler A.
The effectiveness of overfire air is based on how well it promotes
mixing of the product gases in the flame zone. If local fuel rich pockets
exist in the flame, combustion will be incomplete. Particulate emissions
and combustible losses will increase. This increase will be evidenced by
increasing CO levels and smoke. Turbulence induced by overfire air jets is
effective if it penetrates deep into the flame zone. Some of the questions
which this test program hopes to answer are:
A. Where in the flame zone is turbulence most effective?
(Elevation off grate, angle, number of rows, spacing.)
B. What jet velocities are required for effective penetration?
(This relates to fan sizing and header pressures required.)
C. Do extensive overfire air systems justify their cost?
Many of these questions will not be answered until data from several different
overfire air systems are compared. Test results from Boiler A are presented
in Tables 5-1 and 5-2, and in Figure 5-1. A discussion of this data follows.
51 KVB 15900-521
-------
TABLE 5-1
TEST NO.
Overfire Air Pressure, "H20
Firing Conditions
Coal Supplier
Load, % of Capacity
Grate Heat Release, 103BTU/ft2/hr
Coal Sizing, % passing 1/4"
Excess Air, %
Boiler Outlet Emissions
Pajrticulate Loading, Ib/lO^TU
Combustible Loading, Ib/lO^TU
Inorganic Ash Loading, Ib/lO^BTU
Combustibles in Flyash, %
02, % (dry)
CO, ppm (dry) @ 3% O2
NO, ppm (dry) @ 3% O2
Heat Losses,%
Combustibles in Refuse
Dry Gas Loss
Boiler Efficiency
AIR ON
1 28
4
Kern
84
591
44
22
20.5
13.5
7.0
65.8
3.9
1076
408
4.80
8.52
77.40
EMISSIONS
Set 1
29_ 1
9
Kern
85
601
49
21
15.4
8.4
7.0
54.6
3.8
480
398
2.82
8.81
79.17
AND EFFICIENCY
1 35
5.5
Con
76
526
52
38
20.5
14.3
6.2
70.0
6.0
313
394
4.06
8.02
79.65
Set 2
3£ '
10.5
Con
76
538
53
38
11.9
7.9
4.0
66.6
5.9
300
353
2.22
7.57
81.98
1 39
3
Con
60
425
74
29
13.3
6.3
7.0
47.2
4.9
670
335
2.06
6.13
83.56
Set 3
38 I
10
Con
59
412
72
28
6.3
3.9
2.4
61.3
4.7
61
313
1.26
6.07
84.41
52
KVB 15900-521
-------
o
£
*
o
25
20
15
10
J L
.0 O
35 28
400 450 500 550
GRATE HEAT RELEASE 103BTU/FT2/HR
600
FIGURE 5-1. PARTICULATE LOADING VS OVERFIRE AIR
53
KVB 15900-521
-------
Particulates were run during three sets of overfire air tests.
This data is presented in Table 5-1. Load, excess air, coal type and sizing
remained relatively constant for each set of tests. The two tests in each
set were run as close to each other in time as was practical. The two tests
in set 1 and set 3 were run on the same day. The two tests in set 2 were
separated by 24 hours. From this data, the following general conclusions
are drawn:
A. Particulate loading at the boiler outlet is reduced by
25-50% when overfire air pressure is doubled.
B. Much or all of this reduction is due to a more complete burn-
out of combustibles in the flyash. Combustible loading
was reduced by an average 40%.
C. Combustion efficiency is improved by 0.8-2% when overfire
air pressure is doubled. This may be a conservative estimate.
Gaseous emissions were also examined as possible functions of over-
fire air. This data is presented in Tables 5-1 and 5-2. In this case, more
credibility should be given to tests 25 and 31 (Table 5-2) where changes in
overfire air pressure were made at 10-20 minute intervals rather than over a
period of hours or days.
In examining this data, two conclusions can be reached. They are:
A. Carbon monoxide concentration is reduced by the use of more
overfire air. (This is a direct indication that the overfire
air is doing its job of helping to complete combustion.)
B, Nitric oxide concentration is slightly reduced by increased
overfire air on this unit. (After correcting for the effect
of excess 02 with the average trends shown in Figure 4-6, the
average NO reduction was found"to be 16 ppm'or about 5%. '
Since the random variations in NO are large, this may not be
a statistically significant change.)
The reduction in CO is encouraging as an indication of improved
fuel burnout but really on its own has little impact on overall unit efficiency-
For one thing, 1000 ppm of CO represents approximately a 0.33% heat loss.
Under normal operating conditions with or without overfire air, CO emissions
remain at or below this figure. Also, it has not been demonstrated that
increased use of overfire air allows operation of the boiler at a lower total
54 KVB 15900-521
-------
TABLE 5-2
EFFECT OF OVERFIRE AIR ON GASEOUS EMISSIONS
TEST NO. 25, 74% LOAD, KEMMERER COAL
Vary all O.F. Air Headers Together
O.F. AIR, "H2O 3 9
02, % (dry) 4.4 5.0
CO, PPM (dry) @ 3% 02 802 450
NO, PPM (dry) @ 3% O2 423 439
TEST NO. 31, 54% LOAD, KEMMERER COAL
Vary Front Lower O.F. Air, All Others 5"
O.F. AIR, "H20 2 5 10
02, % (dry) 2.6 2.6 2.6
CO, PPM (dry) @ 3% 02 293 181 117
NO, PPM (dry) @ 3% O2 196 196 201
TEST NO. 31, 54% LOAD, KEMMERER COAL
Vary Front Upper O.F. Air, All Others 5"
O.F. AIR, "H20 2 5 1Q
02, % (dry) 3.5 3.6 3.45
CO, PPM (dry) @ 3% O2 231 233 231
NO, PPM (dry) @ 3% O2 244 243 239
55 KVB 15900-521
-------
air (which would reduce the dry gas losses and improve unit efficiency). With
the low-fusion coals tested on this unit, clinkering often sets the lower
limit on undergrate air rather than the onset of smoke or CO.
A reduction in nitric oxide concentration by increased use of
overfire air was expected. The reductions found on this unit averaged 16 ppm
or about 5%. When firing coal, random variations in NO concentration of
this magnitude often occur. Thus, 16 ppm may not be statistically significant.
One last comment should be made concerning the use of the upper
rows of overfire air during low loads. The upper front row of overfire air
jets was ineffective in reducing carbon monoxide during test number 31
(.Table 5-2) at 54% of load capacity. It was observed during this test
that the upper rows of jets were above the flame zone.
5.2 Flyash Reinjection
Flyash reinjection from the mechanical collector hopper on Boiler
A was found to increase combustion efficiency by an estimated 1.5% or more.
However, it also increased the particulate concentration at the boiler outlet
by an average 33%, Increased tendency of grate clinkering with reinjection
was not a problem during these tests although operators reported that this
has been a problem in the past. In fact, it is because of the risk of
clinkers that the mechanical collector ash is not routinely reinjected at
Test Site A.
Three sets of tests were attempted on this unit to assess the
effects of flyash reinjection on emissions and combustion efficiency. Tests
21 and 22 were run at 80% load on Kemmerer coal. Tests 23 and 24 were run
the next day at 60% load on the same coal. Both sets were run with and with-
out reinjection from the mechanical collector and showed similar results.
Tests 36 and 37, which were run on Consolidation coal at 80% load, failed to
produce satisfactory results. The unit tripped out briefly during Test
No. 36 due to a low drum water level. Although testing was stopped until
the load was restored, the load continued to fluctuate drastically during the
remainder of the test.
56 KVB 15900-521
-------
CQ
«o^
o
25
20
With Reinjection
Consolidation Coal
Without Reinjection
Consolidation Coal
37
15
10
With Reinjection
Kemmerer Coal
23
24
O
22
21
Without Reinjection
Kemmerer Coal
I I
I
400 450 500 550
GRATE HEAT RELEASE, 103BTU/FT2/HR
600
FIGURE 5-2. PARTICULATE LOADING VS FLYASH REINJECTION
57
KVB 15900-521
-------
These test results are presented in Table 5-3 and Figure 5-2.
In Table 5-3 the emissions are given. Here it is shown that the percent
combustibles in the boiler outlet ash decreased significantly while the total
particulate loading increased. The net result was that the combustible
loading at the boiler outlet, expressed in terms of mass per unit heat
input, remained basically the same while the inorganic ash loading doubled
when reinjecting. The logical explanation is that a considerable fraction
of the reinjected ash is re«-entrained, i.e., carried out of the boiler
without settling on the grate.
When looking at the mass flow rates of the ash, it was found for
Test 23, for example, that 587 Ib/hr ash was entering the boiler with the fuel
and 1,744 Ib/hr ash was exiting the boiler outlet. This again shows that a
substantial portion of the reinjected ash was circulating through the boiler
continuously without being deposited on the grate.
Heat loss calculations were made to assess the effect of flyash
reinjection from the mechanical collector on combustion efficiency. Several
assumptions were necessary in order to compute the mass flow rate of ash
collected by the multiclone, This measurement could not be made directly
because of the physical impossibility of sampling particulates between
the air heater twhich reinjects continuously) and the multiclone inlet. The
assumptions made were these:
A. 93% of the boiler outlet particulates are collected by the air
heater hopper and the mechanical collector combined. This
number was established by comparing boiler outlet dust
loadings with mechanical collector outlet dust loadings under
similar test conditions (i.e., Test 26A and 26B) .-
B. 70% of particulates entering the mechanical collector are
collected. This is the design efficiency of the collector.
Because less than 10% of the particles entering the collector
are smaller than 10 microns, the collector efficiency may
be greater than 70%. Therefore, the calculated combustible
heat losses may be conservative.
Combustible contents of the flyash and bottom ash were measured
directly. The heating value of the combustible material was calculated to
be 14,250 BTU/lb. Appendix A-l shows how this value was established. Flyash
exiting the mechanical collector was assumed to have 45% less combustibles
58 KVB 15900-521
-------
TABLE 5-3
EFFECT OF FLYASH REINJECTION ON EMISSIONS AND EFFICIENCY
Set 1
Set 2
Set 3
TEST NO.
Reinjection from Multiclone
Firing Conditions
Coal Supplier
Load, % of Capacity
Grate Heat Release,
Coal Sizing, % passing 1/4"
Excess Air, %
Boiler Outlet Emissions
Particulate Loading, Ib/lO^TU
Combustible Loading, Ib/lO^TU
Inorganic Ash Loading, lb/10^BTU
Combustibles in Flyash, %
02, % (dry)
CO, ppm (dry) @ 3% O2
NO, ppm (.dry). @ 3% O2
Heat Losses, %
Combustibles in Collected Flyash
Combustibles in Emitted Flyash
Combustibles in Bottom Ash
Total Combustibles in Refuse
Dry Gas Loss
Boiler Efficiency
1 21
NO
Kern
82
572
45
26
12.1
8.7
3.4
72.1
4.5
998
392
2.30
0,57
0.01
2.88
7.10
30.80
22 1
YES
Kern
83
582
70
23
16.6
9.7
6.9
58.4
4.1
1600
362
0.00
0.72
0.01
0.73
7.76
82.27
1 24
NO
Kern
59
407
67
29
8.4
5.1
3.3
60.8
4.8
150
370
1.46
0.36
0.01
1.83
7.00
81.85
23 1
YES
Kern
59
408
49
26
11.7
5.1
6.6
43.7
4.4
104
388
0.00
0.35
0.01
0.36
6.78
83.54
1 36
NO
Con
80
561
50
27
18.5
11.1
7.4
60.1
4.7
2000
344
2.66
0.66
0.03
3.35
6.95
81.47
37 |
YES
Con
87
614
70
31
22.5
14.8
7.7
65.6
5.1
276
347
0.00
0.82
0.04
0.86
7.41
83.54
59
KVB 15900-521
-------
than at the boiler outlet in those instances where it was not measured.
This value was established by comparing values from the two sample
locations as tabulated in Table 2-4.
However the assumptions are drawn, flyash reinjection from the
mechanical collector did improve combustion efficiency by as much as two
percent. It also increased particulate loading at the boiler outlet by
33%. Its effect on particulate loading after the mechanical collector was
not measured.
No trends were observed for nitric oxide or carbon monoxide
emissions when flyash was reinjected from the dust collector (see Table 5-3).
5.3 Excess Air
At loads above 60% (400x10^BTU/ft2/hr grate heat release) Boiler A
was able to operate continuously without undesirable operating anomalies
at 4.5 to 5.0% excess C>2. This is in the range of 25 to 30% excess air
which is very good for a spreader stoker. This Section will discuss the
influence of excess air on emissions and efficiency at Test Site A. Some
general observations on the optimization of excess air for this boiler will
also be discussed.
Total particulates were measured as a function of excess C>2 at
the mechanical collector outlet and are shown graphically in Figure 5-3.
It stands to reason that as combustion air velocity through the grate is in-
creased, furnace velocities will increase and more ash from the grate and sus-
pension burning will be carried out of the boiler. This was, in fact, observed
at the mechanical collector outlet.
It was also observed that decreasing the excess 02 reduced the
heat loss due to combustibles in the refuse. As seen in Table 5-4, decreasing
the excess oxygen an average 1.7% decreased the combustible content of the
flyash and the particulate concentration, resulting in an average 0.8%
increase in combustion efficiency.
60 KVB 15900-521
-------
TABLE 5-4
EFFECT OF EXCESS 02 ON COMBUSTIBLES IN REFUSE
Set 1 Set 2 Set 3
TEST NO. I 43 I I 65 I I 3536 I
Excess O2, % (.dry) 6.3 4.1 5.8 4.1 6.0 4.7
Firing Conditions
Coal Supplier Stan Stan Stan Stan Con Con
Load, % of Capacity 60 61 70 72 76 80
Grate Heat Release, 103BTU/ft2/hr 404 421 433 494 526 561
Excess Air, % 41 23 37 23 38 27
Measured Combustibles/ %
Collected Flyash 70.0 68.5
Emitted Flyash 29.2 26.3 39.1 37.6
Bottom Ash 1.66
Heat Losses, %
Combustibles in Collected Flyash 1.39* 1.12* 2.87* 1.73* 3.18 2.66
Combustibles in Emitted Flyash 0.33 0.26 0.68 0.41 0.79* 0.66*
Combustibles in Bottom Ash 0.03* 0.03* 0.03* 0.03* 0.09 0.03*
Total Combustibles in Refuse 1.75 1.41 3.58 2.17 4.06 3.35
* Indicates heat loss was estimated from relationships
developed in Section 5.2
61 KVB 15900-521
-------
Carbon monoxide represents a heat loss but is more useful as an
indicator of combustion problems. As a heat loss, it is generally very
small. It has been calculated that 1000 ppm of CO represents 0.33% efficiency
loss. Under normal and satisfactory operating conditions, carbon monoxide
concentrations are well below this level.
Figure 5-4 presents all of the carbon monoxide data from Boiler A
tests as a function of excess O2. Percent steam loads are indicated for
each test. The three test series specifically run to determine the relation-
ship of CO to 02 are connected by solid lines. The following observations
are made:
A. Carbon monoxide begins to rise rapidly below 5-6% excess
02 (about 30% excess air) at high loads.
B. The lower the load, the lower the excess O2 before
significant concentrations of CO are formed.
C. From an efficiency standpoint, CO would have to rise 2500 ppm
per 1% 02 decrease to offset the reduction in dry gas loss
(discussed later). Thus, the breakeven point in terms of
boiler efficiency is 1.5-2% O2 at 60% load, 2.5-3% 02 at 70%
load, and 4.5-5% O2 at 80% load. These points do not represent
realistic operating conditions for reasons of clinkering,
slagging and safety.
Figure 5-5 presents all the nitric oxide data from Boiler A as
a function of excess O2. Data points are also differentiated by load. The
three tests designed specifically to find the relationship between NO and 02
are connected by solid lines. Proposed nitric oxide trend lines are shown in
Figure 4-6. These are based on the data presented in Figure 4-5. The
following observations are madet
A. Nitric oxide concentration is primarily a function of excess
O2 on this unit. Nitric oxide concentration increases an
average of 55 ppm for each one percent increase in O2.
B. Nitric oxide is secondarily a function of boiler load. It
increases an estimated 45 ppm for each ten percent increase
in load although the exact amount is different under dif-
ferent conditions of load, O2, etc.
C. Fuel properties, especially fuel nitrogen, may play an impor-
tant role in nitric oxide formation but data on this variable
is insufficient to make any correlation at Site A.
62 KVB 15900-521
-------
1.2
u>
D
^
O.B
D
U
H
0.6
0.4
FIGURE 5-3.
345678
EXCESS OXYGEN, PERCENT
PARTICUIATE LOADING VS EXCESS OXYGEN AT MECHANICAL COLLECTOR OUTLET
KVB 15900-521
-------
ffi
1,000
CM
O
n
g
p«^
8
B
H
g
S
2
U
800
600
400
200
A 80% Load
70% Load
D 60% Load
O 50% Load
V 40% Load
5 6
EXCESS OXYGEN, %
FIGURE 5-4.
CARBON MONOXIDE VS EXCESS OXYGEN AND LOAD
TEST SITE A
KVB 15900-521
-------
600
in
<*>.
500
400
g
g
H
u
H
M
2
300
V
V V
A 80% Load
0 70% Load
Q 60% Load
O 50% Load
V 40% Load
200
456
EXCESS OXYGEN, %
8
FIGURE 5-5.
NITRIC OXIDE VS EXCESS OXYGEN AND LOAD
TEST SITE A
KVB 15900-521
-------
D. It is theorized that on a day-to-day basis, coal moisture or
sizing may significantly affect NQx emissions, even on the
same coal.
Excess air influences boiler operating efficiency because the
dry gas loss increases with an increase in excess air. The additional air
absorbs heat which could otherwise be absorbed by the boiler, and carries
the heat out the stack. In the same way, heat losses increase with in-
creased moisture in the fuel.
Figure 5-7 shows the dry gas losses for Boiler A as a function of
excess (>2. Boiler load and coal type are indicated. The following obser-
vations are made:
A. The dry gas loss increases by about 0.8% for each one percent
increase in excess O2.
B. The dry gas loss increases about 0.33% for each ten percent
increase in load,
C. The dry gas loss is greater for the Stansbury coal than for
the Consolidation coal. This is because of the lower moisture
in the Consolidation coal. Stansbury averaged 19% moisture
compared to 13% for Consolidation.
Excess Air Cjptiinizatiqn. It is known that dry gas losses (heat
out the stack), can be reduced by lowering the excess air, thus increasing
unit efficiency. It is also known that there are practical lower limits to
the excess air established by the onset of clinkering, smoking and/or high
carbon monoxide emissions.
The key is to operate with the lowest practical excess air without
getting into trouble. However, several problems force operators to operate
consistently above optimum air levels. One problem is that the lower air
limit is a. variable. Some coals clinker up at higher air settings than others.
Air settings which produced a clean stack one day may not the next. Another
problem is that automatic controls cannot be relied upon to hold the air at
low levels. Allowance must be made for air density changes due to temperature
and pressure variations and for small mechanical deviations in the controls.
KVB 15900-521
DD
-------
CM
-J
O
*
PI
g
O
O
Q
I
Pi
X
O
H
600
80%
BOILER LOAD
70% 60% 50%
40%
500
400
300
200
FIGURE 5-6,
5678
EXCESS OXYGEN, %
NITRIC OXIDE TRENDS VS EXCESS OXYGEN AND LOAD
TEST SITE A - ALL COALS
KVB 15900-521
-------
40% Load
Stansbury Coal
a
10
10
en
oo
<*>
en
8
70% Load
Stansbury Coal
O
O
60% Load
Stansbury Coal
80% Load
Consolidation Coal
Op
38 \,
39
60% Load
Consolidation Coal
FIGURE 5-7.
3456
EXCESS OXYGEN, %
DRY GAS LOSS VS EXCESS OXYGEN, LOAD AND COAL
TEST SITE A
KVB 15900-521
-------
Because of these two problems, it is very helpful for operators
to have a reliable means of measuring the excess air on a regular basis.
From the observations at Boiler A, the following suggestions are made:
A. The 02 measurement should be made at the boiler outlet, not
at the stack or after tubular air heaters where dilution will
affect the readings,
B. Ideally, the sample should be pulled from several points
across the duct. If a single point system is used, it should
be centered as much as possible on the duct. Often, because
of poor feeder adjustments/ clinkers, coal size maldistribution,
or other reasons, one side of the boiler will be burning
well while the other side is unsatisfactory. This can lead to
erroneous readings if a single sample point system is used.
C. A reliable continuous monitor should be used and checked
periodically with an Orsat or other O2 measuring device.
Reliability in continuous monitors has been their biggest
problem. Modern equipment is gradually reducing these
problems. With proper maintenance and calibration checks
these monitors are still better than intermittent sampling
with an Orsat. When problems start, boiler performance often
deteriorates rapidly. A continuous monitor would warn the
operator in time to take corrective action.
D. A carbon monoxide monitor at the boiler outlet or an opacity
monitor on the stack are indispensable companions to the 02
monitor if optimum combustion conditions are to be maintained.
They let the operator know the minute a problem develops.
69 KVB 15900-521
-------
5.4
Boiler Load
At Test Site A, peak boiler loads were not available. One reason
was that process demand was seldom high enough to obtain peak loads. Also,
the boiler would often run out of FD fan capacity before peak loads could be
obtained. The latter may have been a control linkage problem. A study of
emissions at .and near peak loading would have been informative in establishing
maximum heat release rates for grates and furnace volumes, and in
establishing emissions at design load for comparison with other units.
The effect of boiler loading on particulate emissions in the 60-80%
load range was studied at both the boiler outlet and mechanical collector
outlet. From Figures ?*-! and 5-2 it can be seen how boiler load (expressed as
grate heat release) affected emissions at the boiler outlet. At the mechani-
cal collector outlet the data was very poorly defined, as seen in Figure 5-8.
It is hard to discern what effect load has on particulate emissions but it
appears to be much less sensitive to load than at the boiler outlet. This
is to be expected. Dust loading increases with grate heat release but so
does collector efficiency. The two effects tend to cancel each other out.
The following exercise was undertaken to establish the extent of this relation-
ship. The test pairs selected are considered representative of the particu-
late loading trend with changing grate heat release.
Test 23
Test 22
% Increase
BOILER OUTLET
Particulate Loading
lb/10.6BTU.
11.7*
16.6
42
Grate Heat Release
lQ3BTU/ft2/hr
408
582
43
MECHANICAL COLLECTOR OUTLET
Particulate Loading
lb/106BTU
Test 10 0.54
Test 5 0.66
% Increase 22
Grate Heat Release
103BTU/ft2/hr
247
494
100
70
KVB 15900-521
-------
1.0
0.9
E-"
vo"
o
0.8
H
9
S 0.7
I
8
H
I
0.6
0.5
0.4
O
o
.00
5 14
o
10
o
16A
O
26B
15
I I
1
I
200 300 400 500
GRATE HEAT RELEASE, 103BTU/ft2/hr
600
FIGURE 5-8.
PARTICULATE LOADING AT THE MULTICLONE OUTLET
VS GRATE HEAT RELEASE
71
KVB 15900-521
-------
From this exercise the following conclusions can be drawn. At
the boiler outlet, a 10% increase in load brings a 10% increase in particu-
late loading. At the mechanical collector outlet, a 10% increase in load
brings a 2.2% increase in particulate loading. Admittedly, this is only
an estimate because the data trend is not well defined at the mechanical
collector outlet.
The effect of boiler load on carbon monoxide emissions was pre-
sented in Section 5.3. To paraphrase, as excess air is reduced a point is
reached at which carbon monoxide concentrations begin to increase very
rapidly. This "carbon monoxide limit" is reached at higher excess oxygen
conditions as the grate heat release increases (Figure 5-4). Thus, if carbon
monoxide were the only limit to low air operation, the unit could operate at
increasingly lower air as the load was reduced. This is not the case in
actual boiler operation because clinkering on the grate often occurs before
the CO limit is reached and necessitates higher air settings at the lower
loads.
The effect of boiler load on nitric oxide emissions is shown in
Figures 5-5 and 5-6 of the previous section. Nitric oxide concentration
increases with load. This is a direct result of higher flame temperatures
at higher loads.
5.5 Coal Properties
Three coals were studied at Test Site A. They differed in ash
content, moisture, and sulfur. Complete coal analysis can be found in
Tables 2-6, 2-7, 2-8, and 2-9. The analyses are shown on a constant
heating value' basis in Table 5-5 below so that direct comparisons can be made.
TABLE 5-5
COAL PROPERTIES CORRECTED TO A CONSTANT 106BTU BASIS
Stansbury Kemmerer Consolidation
Moisture Ib/lO^BTU 13,6 18.1 12.1
Ash lb/106BTU 5.8 3.4 7.4
Sulfur lb/106BTU 0.88 0.61 0.38
72 KVB 15900-521
-------
The largest effect observed for fuel moisture in this test series
was its detrimental effect on boiler efficiency. Figure 5-7 demonstrates
that high moisture Kemmerer coal averaged about a one percent greater dry
gas loss than the lower moisture Consolidation coal. Table 2-3 shows that
the heat loss due to moisture in the fuel was an average 0.9% greater for
the Kemmerer coal than for the Consolidation coal. These losses were ex-
pected and are well understood. The effect of fuel moisture on nitric
oxide emissions and on particulate emissions will not be speculated on in
this report. The reason is that this variable was not isolated from other
fuel variables.
The high ash Consolidation coal emitted more particulates than the
low ash Kemmerer coal. Figure 5-2 illustrates this observation. This should
not be interpreted as an absolute relationship. Coking properties may play a
significant role for other coals. Coal fines may also be a factor.
Sulfur oxides in the flue gas are directly related to the sulfur
content of the coal. Generally, 80% or more of the fuel sulfur is converted
to sulfur oxides and emitted while the remainder is retained in the ash.
In Table 5-6 a sulfur balance was performed on the six SOx tests for which
fuel sulfur and bottom ash sulfur data were available. It is clear that an
insignificant fraction (1% or less) of the sulfur is retained in the bottom
ash. One to four percent of the fuel sulfur is retained in the flyash at the
boiler outlet. The remainder is converted to sulfur oxides and carried out
with the flue gas.
Coal sizing was not one of the variables at Test Site A. The
average and standard deviation of coal sieve test results for the Kemmerer
and Consolidation coals are presented in Figures 5-9 and 5-10. They are both
very similar. They are plotted along with the current ABMA recommended
limits for stoker fired boilers.
Neither of these coals is a stoker coal. They both fall outside
the recommended limits for spreader stokers on the high fines side. Yet,
they burned quite well with a minimum of problems. The coal did pile up
below the feeders occasionally. Past experience has shown that a high surface
moisture will aggravate this piling problem, but the coals at Site A were not
overly wet. At Test Site A, frequent checks by the operators prevented
serious problems from developing.
73 KVB 15900-521
-------
Fuel to be delivered across stoker hopper without size segregation
AVERAGE SIEVE ANALYSIS OF
KEMMERER COAL WITH
STANDARD DEVIATION LIMITS
ABMA RECOMMENDED LIMITS
FOR SPREADER STOKERS
CO
-------
Fu«l to b o delivered across stoker hopper without size segregation
8
:
1
I
I
AVERAGE SIEVE ANALYSIS
OF CONSOLIDATION COAL WITH
STANDARD DEVIATION LIMITS
ABMA RECOMMENDED LIMITS
FOR SPREADER STOKERS
US Std sieve designation
Square mesh screen, inches
FIGURE 5-10
AVERAGE AND STANDARD DEVIATION OF 6 CONSOLIDATION COAL SIEVE
ANALYSIS VS ABMA RECOMMENDED LIMITS FOR SPREADER STOKERS
KVB 15900-521
-------
The coal sampling technique should be discussed here to establish
its credibility. Normally, coal samples would be taken off the coal scales
apron feeder. Boiler A did not have coal scales. Also, a practical method
of taking routine samples off the conveyor belt could not be found.
Therefore, samples were taken from the observation ports above the feeders
out of necessity, This coal sampling procedure was previously discussed in
Section 4,6, The following test was run to determine the representativeness
of these samples.
On September 14th there was a minor fire in the coal bunker. The
bunkers were allowed to burn down overnight. As they were reloaded on the
15th, coal was sampled from the belt. The beltwas stopped five times and
each time a sample was obtained by taking all the material from an eighteen-
inch section of the belt. The composite sample which weighed over 100 pounds
was placed into a conical pile and divided by quartering and adding opposite
quarters. When this procedure had produced a small enough sample, it was
screened. In a couple of hours when the freshly bunkered coal started
feeding, samples were taken from the seven feeders. The arrival of the
fresh coal was distinguished by the surface moisture which was present as a
result of a rain on September 14th. The composite feed sample was treated
in the same manner as the belt sample. The results are shown in Figure 5-11.
The feeder sample had 60% fines (passing 1/4") while the belt sample had 65%
fines. The samples appear to be very similar. Subsequent samples were taken
from the feeders, since the bunkers are not burned down every day.
76 KVB 15900-521
-------
99.99
si
-J
tt
w
I
H
::
u
a
-
:
99.9
99.99
PERCENT WEIGHT ABOVE STATED SIZE
FIGURE 5-11. COMPARISON OF TOO COAL SAMPLING LOCATIONS FOR SIEVE TESTING
KVB 15900-521
-------
TABLE 5-6
SULFUR BALANCE SUMMARY
TEST SITE A
-j
00
Test Load
No. *
SULFUR IN FUEL
O2 Fuel Sulfur
% %
As S02 Ash
Ib/lO^BTU
SULFUR IN BOTTOM ASH SULFUR IN
Sulfur As SO2
% lb/106BTU
Retention Ash Sulfur As
% % lb/1
STANSBURY COAL
8 65
9 61
12 45
18 46
4.8 0.97
4.9 0.91
6.1 0.90
7.4 0.88
1.790 0.05 0.005
1.711 0.06 0.007
1.670 0.06 0.005
1.670 0.16 0.018
0.3
0.4
0.3
1.1
FLYASH SULFUR EMISSIONS
SO2 Retention SOx
06BTU % ppm/dry
754
796
795
896
SOx
lb/106BTU
1.452
1.533
1.531
1.726
Emitted, %
81
90
92
103
CONSOLIDATION COAL
34 76 5.9 0.73 1.386 0.01 0.002
35 76 6.0 0.35 0.655 0.02 0.003
0.1 0.37 .016 1.2 514 0.975 70
0.5 0.32 .025 3.8 417 0.791 121
KVB 15900-521
-------
6.0 SPECIAL TESTS
This Section presents the results of certain tests which because
of their unique nature are kept separate from the main body of test results
in Section 5.0.
6.1 Particle Size Distribution
Figure 6-1 shows the particle size distribution at the boiler
outlet, the multiclone outlet and the electrostatic precipitator outlet of
Boiler A. Four test methods were used to compose this graph. Each method
had its advantages and disadvantages. None were ideal. It is felt important
enough for future particle size testing to devote the remainder of this
section to a discussion of these four test methods and their peculiarities.
There are several developed technologies for measuring particle
sizing. However, very little particle sizing has been done at the boiler
outlet of spreader stokers. Therefore, the technology in this area is not
developed. At the start of this program, KVB consulted several sources
including Southern Research Institute and the Process Measurements Branch
of EPA's Industrial Environmental Research Laboratories - RTF. Both sources
recommended the Brink Cascade Impactor equipped with a precutter cyclone
as the best available technology to use. Therefore, a Brink Impactor was
obtained by KVB for these tests.
Unfortunately, the Brink Cascade Impactor was found to be unsat-
isfactory for measuring particle size distributions at the outlet of Boiler A.
The particles being measured were often as large as the sample nozzle (1.5-2.0 mm)
making it nearly impossible to obtain a representative sample. Also, the
impaction range of the Brink is 0.3 to 3.0 microns which included only the
lower six percent or less of the total catch.
Three alternatives to the Brink Cascade Impactor were investigated
at Test Site A. They were the SASS cyclones, the BAHCO centrifugal classifier,
and the Coulter Counter. All four methods are discussed below.
Coulter Counter. With this method a sample is collected and sent
to a laboratory for analysis. Its range of size classification is greatest
79 KVB 15900-521
-------
of all methods tested. Ten to 250 microns are measured with the counter.
Data from 250 to 1,680 microns is obtained by means of a sieve analysis.
The lower size detection limit of this device is set by the largest particles
being measured. Although it can measure particles in the sub-micron range,
the large size of the particles collected from Boiler A dictated using a
560 micron aperature on the counter. This limited the lower end of its
detection range to ten microns.
Figures 6-2 and 6-3 present the Coulter Counter test data and
compare it with the BAHCO Classifier test data for two ash samples. The
loss of particle detection below ten microns is evident in these two figures.
Thus, the Coulter Counter is not a viable sizing device when used by itself if
there is interest in particle size distribution below ten microns.
BAHCO Classifier. This method, described in PTC 28, has the
advantage of being recognized in the power industry as an established
particle sizing procedure. However, like the Coulter Counter, BAHCO classifi-
cation is a laboratory technique. Thus, it shares some of the same limitations
as the Coulter Counter and all other laboratory techniques. One limitation is
the difficulty in collecting particles below about five microns. These
usually end up on a final filter or are discarded entirely, thus changing the
size distribution of the sample catch. Also, because the sample catch must
be transported and stored, the particle sizes may be reduced (by breaking)
or increased (by agglomeration).
The range of size classification for the BAHCO is approximately
1,5 to 25 microns. To compare the BAHCO Classifier with the Coulter Counter,
two flyash samples were split with half of each sample being size classified
by each of these two methods. The results are presented in Figures 6-2 and
6-3. By combining the two methods, the Coulter Counter's loss of detection
below ten microns can be corrected as it was in the composite plot shown in
Figure 6-t-l.
SASS Cyclones. The SASS train contains three cyclones upstream of
the filter which are sized for ten, three and one micron cut points. This
device has an advantage over the previous two devices in that all data re-
duction can be done in the field. It has the disadvantage of being a single
80 KVB 15900-521
-------
oo
w
N
H
CO
Q
W
EH
in
EH
§
W
99.9
99
98
95
90
80
70
50
30
20
10
5
2
1
0.1
TEST
NO. METHOD LOCATION
0.3
3 10
EQUIVALENT PARTICLE
30 100
DIAMETER - MICRONS
300
1000
FIGURE 6-1. COMPOSITE OF ALL PARTICLE SIZING TESTS AT SITE A
KVB 15900-521
-------
:-
-
L;
v
C
EH
C
_
a
.:
.
.9
.8
99
98
ys
yu
8U
70
60
50
40
30
20
10
5
1
0.5
0.1
.OE
.01
^^
>"
,
^
"
^
1 i i 1 i 1 i « i -
lit;? \ ft 8
^*
i
/
/
/
./
"
BAHCO TEST JUH
KEMMERER COAL
3
X
SAMPLE FROM METHO
2YCLONE
/
1
1
'
,x
,x
^x
ix
D 5
,/
J
15 20 30 40 60 80 100
» _
»-^
.x*'
'7
/
/'
*
/
COULTER TEST 30B
KEMMERER COAL
SAMPLE PROM METHOD 5
CYLCONE
^M
/
m~^~ ^mm
i
A
|
»* "
200 300 500 1000
EQUIVALENT PARTICLE DIAMETER - MICRONS
TTIGUBE 6-2 COMPARISON OF PARTICLE SIZING TECHNIQUES
-------
K
_
N
. .
-
I
v.
J. J
).8
99
9R
95
90
80
70
60
50
40
30
20
:
:
;
0.5
0.1
.05
01
>-*-^
.^
-
- -
^
1
^p_ i
V
SASS TEST 32
KEMMERER COAL
83% LOAD, 5.2% O2
i
x^
.
/
^
^ {
1
***
^--
.
^I
^
^
^
^'
^J»
/
/
7
/
/
BAJ
SAI
ICO TEST 32
1PLE FROM
MULTICLONE ASH
/
^
/
/
.,
^
^
x
^
p
'
.-
P^-
r -
COULTER COUNTER TEST 32
SAMPLE FROM MULTICLONE ASH
v~»-
l
,. '
^^l
'
s
1
,-'
1.5 2
FIGURE 6-3
8 10 15 20 30 40 60 80 100
EQUIVALENT PARTICLE DIAMETER - MICRONS
COMPARISON OF PARTICLE SIZING TECHNIQUES
200 300
500
1000
KVB 15900-521
-------
point sampler, a problem shared by Cascade impactors. A large particle size
stratification is expected at any stoker fired boiler outlet. This is
because of turns in the gas stream, and because the particulate matter contains
relatively large particles which do not turn with the gas. Therefore, it
is always desirable to sample at several points over the cross section of the
duct to get a representative sample. The gravimetric data from five SASS
catches are plotted beside the Coulter and BAHCO curves in Figure 6-1. Also,
one of the SASS tests is compared with the BAHCO and Coulter methods in
Figure 6r-3. Its higher fines can be attributed to the sample location.
Br ink Impactor. Impactors have an advantage over competing
techniques in that they are compact and can be inserted directly into the
s
duct, avoiding the problem of sample loss in a probe. Size classification
is made in the duct so that all conditions are realistic. Impaction also
allows all data reduction to be done in the field. The main disadvantage of
the Brink impactor is its restricted range of classification (0.3 to three
microns} and its demonstrated inability to collect the largest particles in
the flue gas stream at the boiler outlet of spreader stokers.
Although impaction sampling was not feasible at the boiler outlet,
two tests were run at the mechanical collector outlet. This data is pre-
sented in Figure 6-1. Test 14 was run at 69% load with 35% excess air (5.6% '
O2). Test 19 was run at 44% load with 46% excess air (6.8% O2). Both tests
were with Stansbury coal. Both tests showed a similar size distribution.
84 KVB 15900-521
-------
6.2 Size Segregation of Combustible Material
A multiclone ash sample from Test 24 was cut into three size
fractions. Each fraction was analyzed for percent combustibles. The test
results are presented in Table 6-1 below.
TABLE 6-1
VARIATION OF PERCENT COMBUSTIBLES WITH PARTICLE SIZE
Screen Size Weight, % % Ash (Dry) % Combustibles (Dry)
+ 20 mesh 4.3
20 x 100 mesh 34.4
100 mesh x 0 61.3
Weighted Average 72.72 27.28
In this test, the smaller particles had the smaller combustible
fraction. This agrees with the combustible data shown in Table 2-4 of the
appendix where flyash at the boiler outlet averaged 58.4% combustibles
compared to only 30.3% at the multiclone outlet after the largest particles
had been removed.
6.3 Efficiency of Pollution Control Equipment
Several test series were run to measure the efficiency of the
pollution control equipment at Test Site A. The results are. presented in
Table 6-2. A brief discussion follows.
The mechanical collector could not be isolated from the air heater
hopper for testing. However, the efficiency of the two combined was found to
be 93.9% at 70% load on Kemmerer coal. The mechanical collector by itself
has a design efficiency of 70%.
The electrostatic precipitator (ESP) was found to be 96.8%
efficient in removing particulates at 60% load on a coal believed to be a
blend of Stansbury and Kemmerer coals. Its design efficiency is 97.83%.
85 KVB 15900-521
-------
TABLE 6-2
EFFICIENCY OF POLLUTION CONTROL EQUIPMENT
Test
No.
26A
26B
16A
16B
17A
17B
34A
34B
35A
35B
8A
8B
ISA
18B
Location
Boiler Outlet
Mech. Coll. Outlet
Mech Coll. Outlet
ESP Outlet
ESP Outlet
Stack
Boiler Outlet
ESP Outlet
Boiler Outlet
ESP Outlet
Mech Coll Outlet
Stack
Mech Coll Outlet
Stack
Load
%
70
74
59
59
52
52
76
76
76
76
65
52
46
46
Excess O2
%
5.1
4.6
6.1
6.1
6.3
6.3
5.9
5.9
6.0
6.0
4.8
6.8
7.4
7.1
Part.
Loading
lb/106BTU
9.77
0.600
0.512
0.0166
0.0194
0.0128
11.89
0.0576
20.48
0.0296
SOx
Ib/lO^TU
1.452
0.586
1.726
0.612
Equipment, Efficiency
and Coal
Air Heater & Mech, Coll.
93.9% efficient, Kern Coal
Electrostatic Precipitator
96.8% efficient, Coal blend
Sulfur Scrubber
34% part removal Coal blend
A.M., Mech Coll & ESP
99.5% eff. Consolidation
A.H. Mech Coll & ESP
99.86% eff, Consolidation
Sulfur Scrubber, 59.6% eff
Stansbury Coal
Sulfur Scrubber, 64.5% eff
Stansbury Coal
86
KVB 15900-521
-------
If the air heater and mechanical collector together remove 93.9%
of the particulates and the precipitator removes an additional 96.8%, then
the combination should remove 99.8% of the particulates exiting the boiler.
During Tests 34 and 35 particulate loading was measured simulta-
neously across the air heater, mechanical collector and the precipitator.
During these tests, three of the four precipitator fields were not operating.
The measured efficiencies were 99.5 and 99.86% respectively at 76% load on
the Consolidation coal.
The sulfur scrubber was found to be 34% efficient in removing
particulates at 50% load on a blend of Stansbury and Kemmerer coals.
In two tests the sulfur scrubber was found to remove 59.7 and
64.5% of the sulfur in the flue gas. This is well below the 90% design
efficiency of the unit. Test loads were 60% and 46% respectively. In both
cases Stansbury coal (about one percent sulfur) was burned.
6.4 Modified Smoke Spot Number
Smoke spot readings were taken with a Bacharach Smoke Spot tester
at the boiler outlet. The pump was stroked once or twice each time as
opposed to the specified ten times required on an oil fired unit by ASTM
D2156-65. The smoke spot results are tabulated in Table 6-3 below. They
are plotted against particulate loading in Figure 6-4, and against combustible
loading in Figure 6-5.
87 KVB 15900-521
-------
CO
CO
10
8
35
w
Q
W
H
CM
H
24
O
26
6
10
21
23
O
O
22
O
29
O
2 Strokes
1 Stroke
12 14 16 18
PARTICULATE LOADING, lb/106BTU
FIGURE 6-4. MODIFIED SMOKE SPOT NUMBER VS PARTICULATE LOADING
28
O
20
22
KVB 15900-521
-------
10
26
£ 2 Strokes
O 1 Stroke
22
03
23 24
21
O
O
29
O
28
O
8 10 12
COMBUSTIBLE LOADING, lb/106BTU
14
16
FIGURE 6-5. MODIFIED SMOKE SPOT NUMBER VS COMBUSTIBLE LOADING
KVB 15900-521
-------
TABLE 6-3
MODIFIED SMOKE SPOT DATA
Test Average Part. Loading Combustible Loading
No. No. Pumps Reading lb/106BTU lb/106BTU
21 1 3.5 12.1 8.7
22 1 6.0 16.6 9.7
23 1 2.3 11.7 5.1
24 1 2.5 8.4 5.1
26 1 8.3 9.8 5.1
28 1 3.3 20.5 13.5
29 1 1.0 15.4 8.4
21 2 6.5 12.1 8.7
22 2 7.5 16.6 9.7
23 2 4.0 11.7 5.1
24 2 4.0 8.4 5.1
26 2 9.0 9.8 5.1
28 2 6.3 20.5 13.5
29 2 3.0 15.4 8.4
The purpose of this exercise was to develop a quick and easy method
of estimating either particulate loading or combustible loading from stoker
fired boilers. It is observed in Figures 6-4 and 6-5 that no correlation
can be made.
Based on this data, the modified smoke spot technique is not a
viable method for estimating particulate or combustible loadings at the
boiler outlet of spreader stokers. A primary reason is its inability to collect
on filter paper the large particles which contain the majority of the partic-
ulate and combustible mass.
90 KVB 15900-521
-------
6.5 Corrosion Probe Study
A corrosion probe was installed in the convective section of
Boiler A. This device is intended to give comparative information between
different coals on the same boiler and between similar boilers burning the
same coal. It does not give absolute boiler corrosion/erosion rates.
Table 6-4 summarizes the corrosion rate data and indicates which
coals were burned during the residence of each coupon in the boiler. Kemmerer
coal appears to be more corrosive than Stansbury coal after one month of
exposure. Consolidation coal which was very low in sulfur could not be
tested because it was only burned for a one-week period. Figure 6-6 shows
the corrosion rate in mils/year as a function of cumulative time in the furnace.
Again, Kemmerer coal appears to be more corrosive than Stansbury. However,
scatter in the data prevents drawing quantitative conclusions.
The mechanism most often attributed to corrosion is the formation
of ferric sulfide on the boiler tube surfaces. Thus, iron and sulfur are
important fuel properties to examine. Sodium in the ash may also contribute
to corrosion because it causes deposits to stick to the tubes. A correlation
of coal properties with corrosion rate will be attempted when more data is
available.
The corrosion rate appears to start out high and then decrease
with time. Even after two months, it does not appear to have leveled off.
The overall average corrosion rate was 3.2 mils/year, yet the rate for the
two coupons which were exposed for the longest period of time averaged only
1.0 mils/year. The effect of this phenomenon on our ability to draw meaningful
conclusions about long range corrosion potential is not known at this time.
91 KVB 15900-521
-------
TABLE 6-4
CORROSION RATE DATA
Coupon Time in Boiler Weight Loss
Number years mg.
Corrosion Rate
miIs/year
Coals Burned
19
22
27
29
0.038
0.038
0.090
0.090
34.6
1.1
98.9
199.4
4.1
0.1
5.0
10.1
100% Stansbury
100% Kemmerer
21
23
20
25
24
26
0.099
0.099
0.156
0.156
0.195
0.195
63.5
71.9
33.5
108.0
73.6
12.9
2.9
3.3
1.0
3.1
1.7
0.3
50% Stansbury
50% Kemmerer
95% Kemmerer
5% Stansbury
75% Kemmerer
25% Stansbury
92
KVB 15900-521
-------
10
0 Kemmerer Coal
O Stansbury Coal
(f) 1/2 Kemmerer 1/2 Stansbury
£ 3/4 Kemmerer 1/4 Stansbury
U)
H
W
I
g
H
O
U
o I
0.05
I
I
I
0.10 0-15
CUMULATIVE TIME IN BOILER, YEARS
0.20
FIGURE 6-6. CORROSION RATE VS TIME
KVB 15900-521
-------
6.6
Flyash Resistivity
An unsuccessful attempt was made to measure flyash resistivity
with the Wahlco resistivity probe. Three factors were responsible for the
problems with this device. First, the dust loading at the precipitator
inlet (mechanical collector outlet) was too light to collect an adequate
sample. Second, the sample cup was made of Teflon which has an upper
temperature limit of 450°F. The gas temperatures at our sample point
sometimes approached 500°F. The cup was damaged during testing because
of the high temperature. Third, the device became coated on the outside
with soot which caused the two electrodes to short out and give erroneously
low readings.
No data was obtained from this device at Test Site A.
6.7
Source Assessment Sampling System
Seven Source Assessment Sampling System (SASS) tests were run at
Test Site A. Three tests were run to satisfy the requirement that two SASS
tests be run on the first coal tested and one for each coal thereafter.
Three additional tests were run as a result of a joint test venture with the
Aerotherm Division of Acurex Corporation under an EPA contract. A seventh
test was rejected because of a leak in the sampling system which was detected
at the conclusion of that test. The conditions under which the seven SASS
tests were run are shown in Table 6-5 below.
TABLE 6-5
SASS TESTS RUN AT SITE A
Test
No.
27
32
34A
34B
35A
35B
40
Sample
Location
Boiler Out
Boiler Out
Boiler Out
ESP Out
Boiler Out
ESP Out
Boiler Out
Coal
Origin
Kemmerer
Kemmerer
Consolidation
Consolidation
Consolidation
Consolidation
Kemmerer
Load
%
81
83
76
76
76
76
66
Excess 02
%
4.2
5.2
5.9
5.9
6.0
6.0
6.9
O.F.A.
"H20
6.0
11.5
10.5
10.5
5.5
5.5
5.5
Contractor
For Analysis
AD Little, Inc
Rejected
Aerotherm
Aerotherm
Aerotherm
Aerotherm
AD Little, Inc
94
KVB 15900-521
-------
POM, the total POM as percent of participates will be reported. The identity
and quantity of the following POM will be determined.
7, 12 - Dimethylbenz (a) anthracene
Dibenz (a,h) anthracene
Benzo (c) phenanthrene
3-Methylocholanthrene
Benzo (a) pyrene
Dibenzo (a,h) pyrene
Dibenzo (a,i) pyrene
Dibenzo (c,g) carbazole
95 KVB 15900-521
-------
APPENDIX A-l
HEATING VALUE OF FLYASH COMBUSTIBLES
Most flyash samples collected were analyzed for percent
combustibles only. In order to assign a heating value to the combustibles
it was necessary to determine their relative fractions of volatiles and
fixed carbon.
To do this, four flyash samples from the boiler outlet were
analyzed for percent volatiles and percent fixed carbon using ASTM methods
D3172-75. The results are tabulated below:
Heating Value
Sample # % Vol. % Ash % F.C. % Comb. of Comb.
A
B
C
D
1.9
0.7
1.0
0.9
61.2
65.5
47.1
60.0
36.9
33.8
51.9
39.1
38.8
34.5
52.9
40.0
14358
14201
14196
14217
Average 14243
The heating value of fixed carbon (FC) is taken as 14093 BTU/lb. We assume
the volatiles to be similar to a #2 fuel oil having a heating value of
19500 BTU/lb. Thus, the average heating value for combustibles in these
samples is 14243 BTU/lb with a standard deviation of *77 BTU/lb.
Based on these four tests the value 14250 BTU/lb has been
assigned to all combustibles measurements for heat loss calculations.
96 KVB 15900-521
-------
APPENDIX A-2
ENGLISH AND METRIC UNITS TO SI UNITS
To Convert From
in
ft
Lb
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
To
cm
cm2
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
Multiply By
2.540
6.452
0.3048
0.09290
0.02832
0.4536
0.1260
430
239
1054
0.002324
0.2929
1.000
3600
0.9609
3459
3.152
11349
10.34
37234
6895
249.1
C = 5/9R-273
C = 5/9(F-32)
K = C+273
K = 5/9R
COAL FUEL ONLY
ppm @ 3% O2 (SO2)
ppm @ 3% O2 (SO3)
ppm @ 3% 02 (NO)
ppm @ 3% O2 (NO2)
ppm @ 3% O2 (CO)
ppm @ 3% 02 (CH4)
ng/J
ng/J
ng/J
ng/J
ng/J
ng/J
0.851
1.063
0.399
0.611
0.372
0.213
97
KVB 15900-521
-------
APPENDIX A-3
SI UNITS TO ENGLISH AND METRIC UNITS
To Convert From
cm
cm2
m
Kg
Mg/s
ng/J
ng/J
J
JAg
J/hr/m
JAr/m2
W
H
W/m2
W/ra3
Pa
Pa
Kelvin
Celsius
Fahrenheit
Kelvin
TO
in
in2
ft
ft2
ft3
Ib
lb/106BTU
g/Mcal
BTU
BTU/Jb
BTU/ftAr
BTU/ft2/hr
BTU/ft3Ar
BTU/hr
BTU/ftAr
BTU/ft2Ar
BTD/ft3Ar
psia
nH2o
FeLhrenheit
Fahrenheit
Rankine
Rankine
Multiply By
0.3937
0.1550
3.281
10.764
35.315
2.205
7.937
0.00233
0.00418
0.000948
4.303
0.000289
0.0000881
0.0000269
3.414
0.000278
1.041
0.317
0.0967
0.000145
0.004014
F
F
R
R
1.8K-460
1.8C+32
F+460
1.8K
COAL FUEL ONLY
ng/J
ng/J
ng/J
ng/J
ng/J
ng/J
ppm j
ppm «
ppm «
ppm (
ppm 1
ppm *
* 3% 02 (S02)
1 3% 02 (S03)
? 3% 02 (NO)
J 3% 02 (N02)
> 3% 02 (CO)
J 3% 02 (CH4)
1.18
0.941
2.51
1.64
2.69
4.69
98
KVB 15900-521
-------
APPENDIX A-4
SI PREFIXES
Multiplication Factor Prefix SI Symbol
10 12 tera T
giga G
mega M
103 kilo k
102 hecto* h
101 deka* da
10"1 deci* d
10~2 centi* c
10-3 milli m
10~*> micro y
10~" nano n
10~12 pico p
10~15 femto f
10~18 atto a
*Not recommended but occasionally used
99 KVB 15900-521
-------
o
o
APPENDIX A-5
EMISSIONS UNITS CONVERSION FACTORS
FOR TYPICAL COAL FUEL (HV - 13,320 BTU/LB)
graini/scr
(Cry 12% CO.)
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 15900-521
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/7-78-136a
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 A
5. REPORT DATE
July 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
J.E.Gabriels on, P.L.Langsjoen, and T.C.Kosvic
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
KVB, Inc.
600 South County Road 18
Minneapolis, Minnesota 55426
10. PROGRAM ELEMENT NO.
EHE624
11. CONTRACT/GRANT NO.
EPA-IAG-D7-E681 and
DOE-EF-77-C-01-2609
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; 6-12/77
14. SPONSORING AGENCY CODE
EPA/600/13 and DOE
15. SUPPLEMENTARY NOTES IERL-RTP project officer is R.E.Hall. (*)Cosponsors are
(W.T.Harvey, Jr.) and the American Boiler Manufacturers Assoc.
DOE
. ABSTRACT
report gives results of field measurements made on a 300,000 Ib/hr
spreader stoker boiler. The effect of various parameters on boiler emissions and
efficiency was studied. Parameters studied included overfire air, flyash reinjection,
excess air, boiler load, and fuel properties. Measurements included gaseous emis-
sions, particulate emissions, particle size distribution of the flyash, and combus-
tible content of the ash. Gaseous emissions measured were excess O2, CO2, CO,
NO, SO2, and SOS. Sample locations included the boiler outlet, multiclone outlet,
electrostatic precipitator outlet, and wet sulfur scrubber outlet. In addition to test
results and observations, the report describes the facility tested, coals fired, test
equipment, and procedures.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
cos AT I Field/Group
Air Pollution
Boilers
Combustion
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
Flvash Reinjection
13B
13A
21B
21D
14B
11G
07B
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
106
20. SECURITY CLASS (Tliispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
101
------- |