CD
o
EPA PROJECT REPORT NO. 75-LSG-3
TEXAS UTILITY COMPANY, INC.
Big Brown Station
Fairfield, Texas
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Emission Measurement Branch
Research Triangle Park. North Carolina
-------
NITROGEN OXIDES EMISSION MEASUREMENTS
PROM LIGNITE-FIRED POWER PLANT
EMB Projects Report No.
75-LSG-3
Plant Tested
Texas Utility Company, Inc.
Pairfield, Texas
September 30 and October 1, 2, 3 and 4, 1974
October 10, 11 and 12, 1974
Prepared for
Environmental Protection Agency
Office.of Air Quality Planning and Standards
Emission Measurement Branch
Research Triangle Park,
North Carolina 27711
by
W. R. Feairheller
T: L. Peltier
M. T. Thalman
MONSANTO RESEARCH CORPORATION
Dayton Laboratory
1515 Nicholas Road
Dayton, Ohio 45407
Contract No. 68-02-1404, Task No. 3
ii
-------
TABLE OP CONTENTS
I. Introduction 1
II. Process Description and Operation 4
III. Summary and Discussion of Results 9
IV. Sampling Points 25
V. Sampling and Analytical Procedures 30
Appendix
A. Sample Calculations
B. Field Test Data Sheets
C. Analytical Data Sheets
D. Operating Conditions .
E. Sampling Procedures
F. Continuous Monitor Data - Dynascience
G. Sampling Log
H. Project Participants
ill
MONSANTO RESEARCH CORPORATION
-------
LIST OP TABLES
Table Page
1 Summary of Plant Operating Data 7
2 Summary of Averaged Test Results - Unit No. 1 10
3 Summary of Averaged Test Results - Unit No. 2 11
4 Summary of Test Results - September 30, 1974 12
5 Summary of Test Results - October 1, 1974 13
6 Summary of Test Results - October 2, 1974 14
7 Summary of Test Results - October 3, 1974 15
8 Summary of Test Results - October 4, 1974 16
9 Summary of Test Results - October 10, 1974 17
10 Summary of Test Results - October 11, 1974 18
11 Summary of Test Results - October 12, 1974 19
12 Lignite Analysis 23
iv
MONSANTO RESEARCH CORPORATION a
-------
LIST OP FIGURES
Figure Page
1 Air and Gas Flow Schematic for Furnaces 5
2 Sample Port Locations 26
3 Traverse Locations Unit No. 1 27
4 Traverse Locations Unit No. 2 28
MONSANTO RESEARCH CORPORATION
-------
I. INTRODUCTION
Monsanto Research Corporation (MRC) was contracted under the
Environmental Protection Agency's "Field Sampling of Atmos-
pheric Emissions" program t'o provide emission data from the
two lignite fired boilers at the Big Brown Station of the
Texas Utility Company at Fairfield, Texas.
The field test work was directed by Joseph F. Peoples,'Jr.,
Field Testing Section, Emission Measurement Branch. The
sampling was performed by MRC with William R. Feairheller,
Jr. as Team Leader. The sampling schedule was planned by
Arthur D. Little, Inc. Robert Wilson and Keith O'Neal of
A. D. Little, Inc. observed the testing and were responsi-
ble for ensuring that the process variables were changed
according to their schedule. The results of the sampling
program will be used in a study which is being conducted by
A. D. Little, Inc. under contract to the EPA.
This report tabulates the .data collected from Unit No. 1 and
Unit No. 2, two identical 575 MW lignite fired boilers, at
the Big Brown Station. Sampling was conducted on September
30 and October 1, 2, 3 and 4, 197*1 on Unit No. 2. Unit
No. 1 was tested on October 10, 11 and 12, 197^. The primary
goal of this'sampling program was to obtain nitrogen oxides
(NOX) data to develop standards of performance for new lignite
fired steam generators. NOX concentrations have been
MONSANTO RESEARCH CORPORATION
-------
determined by using Method 7, "Determination of Nitrogen
Oxide Emissions from Stationary Sources" Federal Register,
Vol. 36, No. 247, December 23, 1971, and by continuously
monitoring the NOX levels with a Dynasciences Air Pollution
Monitor.
Sampling was performed on two similar exhaust stacks, each
121.92 meters (400 feet) tali. The exhaust gases from each
unit passed through, a Research-Cottrell electrostatic pre-
cipitator before entering the stack. The sampling was per-
formed at the 82.30 meter (270 foot) level where a permanent
platform was located. No modifications were necessary to
prepare the stacks for sampling.
Along with the Method 7 tests (which were run every half
hour) and the continuous monitoring of NOV, the following
.A.
EPA methods we're also followed, as applicable, on each stack:
Method 1 "Sample and Velocity Traverses for Stationary
Sources;" Method 2, "Determination of Stack Gas Velocity and
Volumetric Flow Rate '(Type S Pitot Tube)," performed every
hour; Method 3, "Gas Analysis for Carbon Dioxide,'Excess Air,
and Dry'Molecular Weight," performed every half hour; and
Method 4, "Determination of Moisture in Stack Gases," per-
formed three times a day. All of the above methods are given
in the Federal Register, Vol. 36, No. 247, December 23, 1971.
Lignite samples were also collected from both units. The
samples were brought to MRC laboratories for analysis of
Btu content, sulfur and ash content, moisture, volatility
and ultimate analysis for C, H, N and 0 content.
MONSANTO RESEARCH CORPORATION
-------
This report includes a summary of results, description of
the process, location of sampling points and traverse data,
process operating conditions, and sampling and analytical
procedures. Appendices include all field and analytical
data from this sampling project.
MONSANTO RESEARCH CORPORATION
-------
II. PROCESS DESCRIPTION AND OPERATION
Two similar units were tested during this test program. Each
unit was a CE dual furnace steam generator with a 575 MW rated
load. Each of the twin furnaces in each unit was fired by
four vertical columns of eight burners. One column is placed
in each corner in such a way that tangential flame motion is
induced in the furnace.
Coal is carried from eight bowl mills by heated primary air
to the 64 burners through a network of ducting as illustrated
in Figure 1. The primary air makes up about 35% of total air
at full load.. Since each bowl mill feeds one level through-
out the eight corners, it is possible to cut off the coal
feed to any level by shutting down the mill. It is conven-
tional procedure at Big Brown to fire with the uppermost level
off. Since secondary air continues to flow through the idle
burner level, a form of overfire air occurs under normal
operations. Secondary air is preheated to 760°F .to assist
in lignite volatilization..
The excess air was normally set to give between 2.8 and 3-6$
02 at the furnace exit. Values in the range 2.9 to ^.^% 02
could be reached for limited duration.
Three basic parameters characterize furnace operation: the
chemical energy feed rate, the overall air flow and the air
distribution to active burners. Although gross load (MW)
MONSANTO RESEARCH CORPORATION
-------
o
z
z
H
O
73
m
in
m
o
i
o
o
10
D
O
Z
VJ1
FORCED DRAFT FAN
HOT WATER AIR HEATER
AIR HEATER (
BURMtRttlNDBOX
I'JOUCEO OHAFT FAN
PIUCIPITATOR
STACK
POIMARV AIR FAN
BQ-.1L Mill
MILL FEEDER
ICMroH AIM FAN
SCA
SEAL AIH BOOSTER FAN
l~| SECONDARY AIR
II PRIMARY AIR - COCO
jI PRIMARY AtR - HOT
r~| IGNITOR & SCANNER AIR
FLUE CAS
BM.
ua.
Figure 1. Air and Gas Flow Schematic for Furnaces
-------
and excess 02 are "output" variables, they were used as con-
venient and reliable indices for "input" chemical energy and
overall air flow. This interchangeability is justified be-
cause combustion is essentially complete and boiler efficiency
is nearly constant.
The air flow to active burners (as percent of stoichiometric)
was controlled and estimated as follows: burner air flow was
varied by withholding fuel (but not air) from the top burner
level. Three conditions could be set up: no overfire with
all burners operating, moderate overfire with secondary air
to the top idled burner level, or maximum overfire with
primary and secondary air to the top idled burner level.
These air flows were recorded and estimated from the plant's
instruments.
Additional variables known to affect NOY were also recorded:
A.
windbox temperature and pressure, ambient humidity, and fuel
nitrogen content. Sufficient steam cycle measurements were
recorded to construct an energy balance and verify normal
operation of each boiler. A copy of the process data col-
lected 'can be found in Appendix D. Table 1 gives a summary
of plant operating conditions.
Operating conditions during any of the identified test phases
were subject to changes because of the nature of plant opera-
tion. For example-, it was not unusual for excess oxygen to
fluctuate between 2.7 and 3-3 percent within one-half hour
when set at 3-0 percent. The reason for this drift is that
electrical output and steam flow typically are maintained
constant within about ±0.5 percent by continually adjusting
excess air or burner tilt to compensate for transient slag
buildup, coal heating value, or air flow variations. This
drift contributes to the scatter in successive NOY measurements
.A.
MONSANTO RESEARCH CORPORATION
-------
o
z
z
-{
o
m
en
m
O
I
O
o
O
ID
z
SUMMARY OF PLANT OPERATING DATA (Big Brown Units
GROSS
DATE INTERVAL . CONDITION LOAD(MW)
Big Brown 9/30/74 1200-1545
Unit 92 1600-1630
1700-1800
10/1/74 . 0900-1045
1100-1410
1430-1600
. 10/2/74 0800-0930
0945-1300
1315-1500
10/3/74 0700-0945
1000-1100
1120-1315
1330-1400
10/4/74 0900-0930
0945-1230
V .
. 1245-1530
Big Brown 10/10/74 1300-1800
Unit VI
10/11/74 0945-1040
-. - . 1050-1345
1400-1600
' 10/12/74 0900-1410
/ 1420-1600
Baseline (Moderate overfire)
Low load
Baseline
Baseline
Low excess air
Baseline
High excess air
No overfire (all burners on)
Baseline
High excess air
Baseline
Max overfire (level-H
primary air) .
Baseline
High excess air
Max overfire and low
excess air
Baseline . . .
Special test (level B off)
. Max overfire
No overfire (all burners on)
Baseline
Baseline
Max overfire and high air
601
483
. 597
598
.599
593
586
595
601
602
601
600
595
594
592
577
596
595
596
598
597
1 and 2) <
COAL FEED RATE BURNER-AIR*
(Ibs/hr, as rec'd) (% Stoich}
839,000
666,000
823,000
801 ,000
807,500
825,000
801,000
800,500
810,000
827,500
825,000
840,000
828,000
840,000
845,500
831,500
826,500
860,000
853,500
836,500
866,000
869,000
' 105
105
104
103
105
111
116
104
109
106
101
104
107
114
104
105
101
117
104
108
104
*Estimated based on excess oxygen measurements before the air preheater and
the number (and location) of inactive burners.
-------
taken at one-half hour intervals. Therefore, the averaged
NOX data corresponds to an average condition representative
of the range over which the boiler conditions drifted.
MONSANTO RESEARCH CORPORATION
-------
III. SUMMARY AND DISCUSSION OP RESULTS
The N.OX, C02, 02 and moisture \content, and stack velocity
and temperature were obtained at the stacks of the Big Brown
steam plant during both normal operation and under various
conditions of modified operation, A summary of the averaged
results for Unit No. 1 with the corresponding run conditions
is given in Table 2. The averaged results from Unit No.. 2
are given in Table 3. The results of the sampling program
on a day by day basis are given in Tables 4-11. The Method
7 NOX data are given in ppm of N02 on a dry basis. ' The
Dynasciences data recorded on a continuous basis was read
from the chart every 10 minutes and the value obtained at a
time corresponding to each EPA-7 'result was recorded on the
Table. .
Emissions from Boiler No. 2 were measured during the first
week of sampling '(Sept. 30 to Oct. 4). The boiler was
operated under normal conditions on the first day to pro-
vide baseline emission rate data. The initial samples,
analyzed for C02 and 02 content, indicated only air. The
monitor values for NOX were about 100 ppm rather than about
350 ppm as expected. The probe system, consisting of two
stainless probes and one glass probe in a common sheath, was
sealed around the probes and also between the sheath and
port. The monitor and integrated gas samples were initially
withdrawn at the rate required for'each instrument. After
it was found that the gas samples were contaminated with
MONSANTO RESEARCH CORPORATION
-------
Table 2
s
o
z
>
z
o
X)
m
m
Summary of Averaged Test Results -.-Big
No. of
Gas Plow Samples
Date Test Phase
10/10 Special Test
(1300-1800)
10/11 Max Overfire
(0915-1010)
O No Overfire
I "
O
O
O
XI
TJ
O
3)
H
(1050-1315)
Baseline
(1100-1600)
10/12 Baseline
(0900-1110)
5 1 Max Overfire
Z and High Air
e
(1120-1600)
DSCPM
106
1.11
1.21
1.26
1.35
1.17
1.19
Averaged
EPA-7
10
1
6
3
11
3
Dyna,
11
1
6
3
11
1
. Brown -Station Steam Generating
ppm
EPA-7
301
295
337
319
290
268
Dyna.
389 '
310
368
313
333
317
Ib/hr
EPA-7
3101
2605
3018
3375
3159
2883
Dyna.
- 1007
2738
3331
3030
3520
3390
Plant -
kg/hr
EPA-7
1107
1182
1383
1531
1133
1308
Dyna.
1818
1212
1512
1375
1597
1538 '
Unit No.
1
lb/106 Btu
EPA-7
.58
.18
.57
.65
.57
.52
Dyna.
.75
.51
.63
.58
.61
.61
gra/106
EPA-7
1.01
.86
1.03
1.17
1.03
-91
cal.
Dyna.
1-35
.92
1.13
1.01
1.15
1.10
-------
Table 3
Summary of Averaged Test Results - Big Brown Station Steam Generating Plant - CJnlt No. 2
.
Gas Plow
DSCPM
Date Test Phase 106
9/30 Baseline
(1200-1545)
Low Load
(1600-1630)
Baseline
(1700-1800)
o
^ : 10/01 Baseline
> ' (0900-1045)
Z
-1
O
XI
m
in
m
Low Excess Air
(1100-1410)
Baseline
(1430-1600)
X) 10/02 High Excess Air
o (0800-0930)
I
O I1
O
XI
T)
O
H
5
No Overfire
(0945-1300)
Baseline
(1315-1500)
10/03 High Excess Air
(0700-0945)
z
Baseline
, (1000-1100)
Max Overfire
. , (1120-1315)
! Baseline
(1330-1100) .
10/04 High Excess Air
(0900-0930)
Max Overfire &
Low Excess Air
(0945-1230)
Baseline
(1245-1530)
1.32
1.32
1.26
1.28
1.28
J..38
1.33
1.32
1.56
1.39
1.12
1.15
1.50
1.36
1.40
No. of
Samples
Averaged
EPA-7 Dyna.
8
2
2
6
3
1
4
2
3
1
2
_
_
3
2
-
1
4
4
1
8
5
_
2
2
3
1
4
1
ppm
EPA-7 Dyna.
' 296
334
314
325
370
268
335
307
305
267
252
_
_
319
273
-
411
378
376
225
326
316
403
200
248
375
293
55
Ib/hr
EPA-7 Dyna.
2803
2154
2816
2957
3380
2650
3189
2965
3441
2659
2582
_
_
3330
2914
-
3651
3448
3430
2224
3107
3062
_
4008
2006
2580
4029
1510
587
kg/hr
EPA-7 Dyna.
1271
1431
' 1277
1341
1533
1202
1447
1345
1561
1206
1171
_
_
3196
1322
-
1656
1564
1556
1009
1409
1389
-
^
1818
910
1170
1828
1450
266
lb/106 Btu
EPA-7 Dyna.
.54
.76
.58
.61
.68
51
.61
.57
.67
.52
.90
_
_
.68
.60
-
.75
.71
.69
-43
.59
.58
_
.79
.39
.51
.82
.62
.12
gm/106 cal.
EPA-7 Dyna.
-97 -
1.37
1.05 1.36
1.09 1-27
1.22 1.24
0.92 .77
1.10 . 1.06
1.02 1.04
1.'21
.94 1.42
1.61 .69
92
1.48
1.22 1.12
1.07 -21
-------
Table
Summary of Test Results - September 30, 1974
2
O
z
en
>
Z
H
O
3
m
t/>
m
n
i
o
o
3J
T)
O
H
O
IV)
P.S. -
Time
1200
123C
1300
1330
1400
1430
1500
1530
1611
1613
29-86 - Coal Heating Value - 6240 Btu/lb
Moisture Dry MW DSCFM NmVsec.
106
12.5 30.3 1.31 620
1.32 625
1.34 631
1.32 624
1.32
. Coal' Feed Rate . .
10 3 Ib/hr ppm
839 254.
262
299
279
309
338
314
310
666 . 331
336
Ib/hr
2384
2459
2827
2638
2966
3245
2969
2932
3130
3177
Method
kg/hr
1081
1115
1282
1197
1345
1472
1347
1330
1420
1441
7 NOX
lb/106 Btu
.46
.47
.54
.50
.57
.62
.57
.56
.75
.76
gm/106 cal.
.82
.85
.97
.91
1.02
1.12
1.02
1.01
1.36
1.38
-------
Table 5
Summary of Test Results - October 1, 1974
c
5
?.S. - 29.18 -
Time Moisture
O
Z , 9115 16.0
(/)
> ; 1010
H 1015
o
R 1120
> 1115 H.9
0 120°
1 oo i 12I)5
g 1330
30
T)
0
H
O
Z i
1315
1130
1115 16.0
. ' 1500
1530
1515
1600
Coal Heating Value - 6010 Btu/lb
DSCFM Coal Feed Rate
Dry MW 106 NmVsec.
30.1 1.27 602
1.21 585
30.2 1.26 595
1.27 599
1.30 611
30.1 1.27 598
1.29 608
10 3 Ib/hr ppm
801
289
338
807.5 315
335
317
328
316
280
825
357
113
339
Ib/hr
2629
3003
3111
3021
2861
2981
3118
2608
3218
3758
3133
. Method 7 NOX
kg/hr lb/106 Btu
1192
1362
1192
1372
1298
1351
1128
1183
1173
1705
1121
.51
.62
.61
.62
.59
.61
.65
.53
.65
.75
.63
Dynascience Continuous
gm/106 cal. ppm Ib/hr
.98
1.12
1.15
1.12
1.06
1.10
1.16
.96
1.17
1.36
1.13
111 3651
370 3310
390 3518
385 3503
365 3399
365 3321
372 3385
385 3503
. 380 3512
kg/hr lb/10b Btu
1656
1515
1609
1589
1512
1506
1535
1589
1593
.75
.68
.73
.72
.70
.67
.68
.70
.70
Monitor
gm/100 cal.
1.36
1.23
1.31
1.29
1.25
1.20
1.22
1.27
1.27
-------
Table 6
Summary of Test Results - October 2, 1971
O
z
z
H
O
m
in
m
>
u
n
i
o
O
3)
z
p.s. - 29.60
Time Moisture
915 13-1
1015
1030 11.5
1050
1115
1130
1215
1230
1300
1310 15.6
1400
1120
' 1110
1515
- Coal Heating Value - 6180 Btu/lb
DSCFM Coal Feed Rate -- -
Dry MW 106 NmVsec.
30.1 1.38
30.5 1-32
1.33
1.31
30.1 1.31
1.36
1.36
650
625
629
631
631
612
612
10 3 Ib/hr ppm
801 268
807.5 331
307
351
316
810
301
309
Ib/hr
2650
3158
2903
3373
3322
2918
3011
- Method 7 NOX
kg/hr lb/106 Btu
1202
1132
1317
1530
1507
1321
1366
.51
.60
.55
.61
.63
.56
.57
gm/106 cal . ppm
.92
1.09
1.00
1.16
1.11
1.00
1.03
225
350
380
370
320
310
315
315
. 250
280
320
320
' 325
335
Dynascience Continuous
Ib/hr
2221
3310
3593
3199
3019
2951
3021
3021
2100
2688
3072
3118
3166
3261
kg/hr lb/106 Btu
1009
1501
1630
1587
1383
1310
1372
1372
1089
1219
1393
llll
1136
1181
.13
.63
.69
.67
.58
.56
.58
.58
.16
.51
.59
.60
.60
.62
Monitor
gm/106 cal.
77
1.11
1.21
1.20
1.05 '
1.02
1.01
1.01
.83
92
1.05
1.08
1.09
1.12
-------
Table 7
o
z
3)
m
en
m
>
3)
O
I
O
O
3)
TJ
O
3)
O
Z
VJI
\
Summary of Test Results - October 3, 1971
P.S. - 29.60 - Coal Heating Value - 6190 Btu/lb
Coal Peed Rate
Method 7 'HDX
Dynascience Continuous Monitor
\ uaufw uoai reea Kate mcunuu i IMUX ^_^ uynascience uontinuous Monitor
I Time Moisture Dry MW 10s NmVsec. 103 Ib/hr ppm Ib/hr kg/hr lb/10b ₯tu gm/10b cal. ppm Ib/hr kg/hr lb/10b Btu gm/10b cal
; 725 1.9 30.1 1.57 711 827.5
', 800
! 820 1.59 752
! 900
1 910 1.53 723
' 1000 825
, 1020 18.2 30.3 1.39 655
j 1100
1125 16.2 30.1 1.13 673 810
:' 1200
1 1215 1.40 661
' 1300
!
1310 11. 1 30.5 1.15 681 828
306 3112 1561 .67 1.21
300 3371 1530 .66 1.19
308 3508 1591 .68 1.23
370
135
267
295 3022 1371 .58 1.05
209 2111 971 .11 .71
200
200
220
250
275
3681
1332
2659
2006
2006
2285
2597
2857
1671
1965
1206
910
910
1036
1178
1296
.72
.85
-52
.39
.39
.15
.51
.56
1-30
1.53
. .91
.69
.69
.80
-.91
1.00
-------
Table 8
2
O
z
z
-I
O
3)
m
c/>
m
O
I
O
O
33
TJ
O
O
Z
Summary of Test Results - October 1, 197*1
' P.S. - 29-53 - Coal Heating Value - 5820 Btu/lb
Time Moisture Dry MW 106 Nmvsec. 10 3
Ib/hr ppra
Ib/hr
kg/hr lb/10° Btu gm/10" cal. ppm
915 11-7 30.3 1-50 707 810
1015 12.1 30.1 1.17 695' 815-5
1030 358
; 1115 1.15 681
1130
1215 1.11 679
1230
1310 12.1 30.3 1.19 703
1320
288
311
275
271
3769
2991
3230
2935
2892
1710
1357
1165
1331
1312
.77
.61
.66
.60
.59
1.38
1.09
1.18
1.07
1.06
375
500
310
275
85
55
Ib/hr
1029
5261
3219
2836
877
587
kg/hr lb/10" Btu gm/10" cal.
1828
2388
1160
1286
398
266
.82
1.07
.65
.58
.18
.12
1.18
1.93
1.18
1.01
.32
.21
-------
Table 9
o
z
z
-I
o
71
m
O
I
O
o
JJ
-a
o
z
Summary of Test Results - October 10, 1971
P.S. - 29.37 - Coal Heating Value - 6150 Btu/lb
DSCFM Coal 1
Time Moisture Dry MW 106 NmVsec. 103
1300
1330 12.8
iloo
1130
1500
1510
1600
1630
1700
. 1710
1750
*eed Rate
Ib/hr ppm
Method 7 NOX
Ib/hr
kg/hr lb/10b Btu gm/10b cal. ppm
862.5
30.1 1.15
1.12
1.13
1.15
1.11
681
668
671
681
679
312
318
197
180
278
337
357
311
305
378
3553
3303
2001
1831
2818
3152
3708
3262
3116
3900
1612
1198
909
831
1292
1566
1682
1180
1127
1769
.67
.62
.38
.31
.53
.65
.70
.61
-59
.73
1.20
1.12
.68
.62
.96
1.17
1.25
1.10
1.06
1.32
380
375
375
365
390
380
100
105
395
105
105
Dynascience Continuous Monitor
Ib/hr
3917
3895
3895
3713
3967
3893
1098
1207
1103
1178
1178
kg/hr lb/106 Btu gm/106 cal.
1790
1767
1767
1681
1799
1766
1859
1908
1861
1895
1895
.71
.73
.73
.70
.71
.73
.77
.79
-77
.78
-78
1.33
1.32
1.32
1.25
1.31
1.31
i.38
1.12
1.39
1.11
. 1.11
-------
Table 10
Summary of Test Results - October 11, 1971
2
O
z
in
>
Z
H
O
m
co
m
>
73
O
I
O
O
73
TJ
O
H
O
Z
o
oo
P.S. - 29. DM
Time Moisture
900
930 11.6
1000
1030
1100
1130 12.0
, 1200
1230
'1300
1330
Hoo
1130 11.2
1500'
- uoai Heating value -
DSCFM
Dry MW 10s Nra3/sec.
30.1 1.23
1.21)
30.1 1.26
1.27
1.26
30.5 1.35
580
587-
596
601
591
635
DOU HCU/1D
Coal Peed Rate
10 3 Ib/hr ppm
860 271
288
301
317
853-5 319
368
361
327
295
321
836.5 120
335
292
Method 7 NO*
Ib/hr
2111
2538
2652
2816
3150
3322
3259
2975
2681
2898
1062
3210
2821
Dynsasclence
kg/hr lb/10° Btu gm/10" cal. ppm
1095
1151
1203
1277
1429
1507
1178
1319
1217
1315
1812
1170
1281
.15
.17
.19
.52
.59
.62
.61
.56
.50
.51
.78
.62
.51 '
.81
-85
.89
.91
1.06
1.12
1.10
1.00
91
.98
1.10
1.12
.97
295
285
300
360
370
375
370
365
370
360
310
310
320
Ib/hr
2599
2511
2611
3198
3310
3385
3310
3321
3366
3250
2998
2998
3095
Continuous Monitor
kg/hr lb/10b Btu
1179
1139
1199
1151
1515
1535
1515
1506
1527
1171
1360
1360
Hoi
.18
.17
.19
.59
-63
.63
.63
.62
.63
.61
.57
-57
.59
gm/10D cal.
.87
.81
.89
1.07
1.13
1.11
1.13
1.12
1.11
1.10
1.03
1.03
1.07
-------
Table 11
Summary of Test Results - October 12, 1971
P.S. - 29.64 - Coal Heating Value - 6390 Btu/lb
o
z
in
Z
H
O
3)
m
DSCFM
Time Moisture
900 11.7
930
1000
1030
w ; 1100
m
> 1130
0 ,_, , I200
1 VD 1 1230
8 '; "O0
TJ ; 1330
o j iloo
> i
- j 1130 12.5
Z J 1500
: 1530 '
1600
Dry MW 10fa NmVsec.
30.1 1
1
1
1
1
1
30.1
1
1
-17
.18
.18
.11
.51
.16
.50
.17
693
697
697 .
680
711
689
706
691
Coal Feed Rate
10 3 Ib/hr ppm
866 290
291
291
308
255
321
309
313
215
329
333
866 316
191
298
Method 7 NOX
Ib/hr
3051
3096
3117
3266
2701
3103
3188
3229
2650
3559
3183
3396
2052
3202
kg/hr lb/10" Btu gm/10b cal. ppm
1385
ilol
llll
1181
1227
1511
1116
1165
1202
1611
1580
1510
931
1152
.55
.56
.56
.59
.19
.62
.58
.58
.18
.61
.63
.61
.37
.58
-99
1.01
1.01
1.06
.88
1.11
l.Ol
1.05
.86
1.16
1.13
1.10
.67
1.01
330
315
328
330
331
323
335
325
315
350
350
330
320
305
313
Dynsacience Continuous
Ib/hr
3175
3317
3178
3199
3511
3125
3156
3353
3732
3786
3661
3516
3139
3278
3296
kg/hr lb/10" Btu
1576
1505
1578
1587
1606
1551
1568
1521
1693
1717
1661
1608
1560
1187
1195
.63
.60
.63
.63
.61
.62
.62
.61
.67-
.68
.66
.61
.62
.59
.60
Monitor
gm/10b cal.
1.13
1.08
1.13
1.11
1.15
1.11
1.12
1.09
1.21
1.23
1.19
1.15
1.12
1.07
1.07
-------
ambient air, the sampling interface was modified by connecting
the probes to metal bellows pumps operated at maximum capacity
and samples were then withdrawn from "T"s in these lines at
flow rates comparable to those of the monitor (28 1/hr) and
the integrated gas sampler (28 1/min.). After this correction,
reasonable values were obtained from the monitor and the gas
sampler.
Numerous problems were encountered with the monitor. It was
observed that the indicated value increased rapidly with in-
creasing ambient temperature and that low ambient (10°C and
below), temperature prevented stabilization on the calibration
gas. This effect was observed on both Oct. 3 and Oct. 4.
According to the literature supplied with the monitor, opera-
tion is possible from +5 to ii9°C. However, if the ambient
temperature varies, the calibration will change. For the
second week the monitor was heated to about 40°C to maintain
a constant sensor temperature and thus eliminate major drifts.
Operation during the second week was reliable and valid data
were .obtained.'
The results obtained from the Method 7 NOX determinations
are quite varied and indicate that the scatter of data during
an operating condition is greater than the variation in NOX
due to changes in the operating conditions. The procedures
used were as specified in.the Federal Register. Identical
techniques were employed for sampling and the analysis of all
samples with one exception. During the analysis of one group
of samples, representing for the most part baseline conditions
from Oct. 1 to Oct. 4, it was found that the hood fan ceased
operation and condensate from the hood dripped into the
samples, giving invalid high NOX results. The data for these
samples are shown in parentheses on the data sheets in the
appendices but have been omitted from the summary tables in
this section as they do not represent actual stack emissions.
20
MONSANTO RESEARCH CORPORATION e
-------
Line voltage variations were a major problem during'the
testing of Unit No. 2. At the times when the sampling equip-
ment was in operation, the line voltage was observed to drop
as low as 92 volts. In order to lessen the problem of voltage
drop, the various pieces of sampling equipment were not run
simultaneously. The digital thermometer was only used when
the other equipment was not being used. The calibration of
the thermometer was checked at least twice a day to insure
that'the voltage drop was not affecting the temperature read-
out . The Dynascience monitor and the recorder seemed to be
noticeably affected. The continuous readout from the recorder
was noticeably lower when other equipment or heaters were
running. When values were read from the chart paper, they
were adjusted to account for sudden drops due to a lower
voltage. An additional power line was used for the Dynascience
monitor and recorder on October 3 and 4. A battery powered
recorder was also used to help alleviate the problem of
changing voltage.
The test data from September 30 indicate the following NOX
baseline data: an average of 303 ppm, a moisture level of
12.9$, and gas emissions contents of 6.3% 02 and 12.9$ C02.
On successive days the baseline data vary greatly. The
studies on Boiler No. 1 indicate less variation in the NOX
baseline data, especially as measured with the monitor, but
even on this unit the variation is significant. As a result,
it is necessary to observe the baseline data obtained before
and after an operating parameter was changed in order to
evaluate the change in NOX concentration resulting from that
change.
21
MONSANTO RESEARCH CORPORATION
-------
The problems encountered during the first week of sampling
were corrected before the program began on Boiler No. 1.
As a result, the data obtained from the second week are
more reliable, and the effects of process variables can be
observed.
During the sampling program coal samples were gathered for
proximate and ultimate analysis and BTU content to be com-
pleted at the laboratory. The proximate analysis included
percent moisture, volatile, ash and fixed carbon. The
ultimate analysis included percent carbon, hydrogen, nitrogen,
sulfur and ash. These samples were gathered from the con-
veyors feeding the silos above the pulverizers at one half-
hour intervals. The samples were composited on site to
provide a single sample daily and a portion of this daily
sample was returned to Dayton in a sealed plastic bottle for
analyses. All analyses were performed on a sample ground to
pass a 60 mesh sieve. The results are shown in Table 12.
All the results are reported on the as determined basis except
the BTU value. The BTU samples were oven dried and analyzed
and the results were calculated to the as determined basis
using the percent moisture figure.
In reviewing the sampling program, there are three recommenda-
tions we feel are important to future programs of this type:
1. . The monitor should be heated and maintained .at a con-
stant temperature above ambient temperature to prevent
calibration drift.
22
MONSANTO RESEARCH CORPORATION
-------
Table 12
Lignite Analysis
90 n v )i
- jU / 4
% Moisture 30.53
<*
^
o
z
(ft
>
z
H
O
3)
m
en
m
j>
33
O
I IV)
OJ
o
o
33
T3
O
33
j>
H
5
z
o
i $ Volatile
. % Ash
5? Fixed Carbon
BTU/lb
(Oven-dried)
BTU/lb
(As Determined)*
% Carbon
% .Hydrogen
% Nitrogen
% Oxygen
% Sulfur
11.49
13-16
44.82
8980
6240
52.66
4.07
1.01
28.23
0.87
10-01-74
33-45
11.15
11.41
43.99
9060
6040
46.16
3-82
2.60
35-24
0.77
10-02-74
33.
10.
10.
46.
09
07
35
49
9680
6480
55.
4.
0.
13
16
84
28.74
0.78
10-03-74
32.47
4.82 - .
12.64
50.17
9160
6190
49.80
4.22
0.89
31.70
0.75
10-04-74
33-
12.
14.
39.
86
18
01
95
8810
5820
53-
4.
0.
27.
08
14
95
01
0.81
10-10-74
32.75
8.67
10.93
47.65
9600
6450
49-53
4.18
1.05
33.57
0.74
10-11-74
31.77
20.96
10.81
36.46
9160
6250
51.76
4.19
1.14
31.17
0.93
10-12-74
30.67
5-43
12.74
51.16
9230
6390
50.90
4.13
1.27
30.37
"..0.59
*A11 results are as determined basis and this BTU value
oven-dried results.
was calculated from the
-------
2. All electronic instrumentation including the continuous
NO monitor and recorders should have a power circuit
A.
isolated from that for instrument heating and other
.fluctuatingsources. When fluctuating and/or reduced
line voltage can be expected, a portable power supply is
suggested.
Consideration should be given to the design of an adequate
sample probe collection system to interface the analyzer
to the stack to eliminate ambient air being drawn into
the system and contaminating the sample. Drawing a large
volume of sample into the probe and then removing a
portion of this sample for analysis is recommended when
high stack velocity situations are encountered. This
method reduces the response time of the system and also
assures that adequate stack gas will be supplied to the
instrument provided there are no leaks in the sampling
system.
MONSANTO RESEARCH CORPORATION
-------
IV. SAMPLING POINTS
Sampling was conducted on the outlet stacks of both units.
Unit No. 2 was tested Sept. 30-Oct. 4 and Unit #1 was tested
Oct. 10-12. The combustion gases from each unit passed
through an electrostatic precipitator (ESP) before being
exhausted out the stack. Method 1, 2, 3, 4, and 7 tests
were performed on each stack. (NOX concentrations were
continuously monitored with a Dynasciences Air Pollution
Monitor) during the time the EPA Method 7 testings was
being performed.
Sampling was done through 4 ports, 90° apart from each other.
The ports were located 73.2 meters (240 feet) above the
inlet to the stack (approximately 9.9 stack diameters) and
39-6 meter (130 feet) upstream of the outlet (approximately
5.4 stack diameters). EPA Method 1 criterion was satisfied
by using a total of 12 traverse points.
Both stacks have inside diameters of 7-39 meters (291 inches),
However, Unit No. 1 has an outside diameter of 10.23 meters
(403 inches) and Unit No. 2 has an outside diameter of 10.11
meters (398 inches) at the sampling location. Figure 2 shows
the location of the ports on both stacks, while Figures 3
and 4 show the location of the 12 traverse points'. The
ports consisted of a large square door on each outer stack
and a 0.15 meter (6 inches) cyclindrical port in the stack
of Unit No. 2. A 0.1 meter (4 inches) cyclindrical port was
25
a MONSANTO RESEARCH CORPORATION
-------
INLET
Figure 2. Sample Port Locations
26
MONSANTO RESEARCH CORPORATION
-------
IN LET
INLET
29l" 403"
I
Figure 3- Traverse Locations Unit No. 1
27
MONSANTO RESEARCH CORPORATION o
-------
INLET
INLET
Figure 4. Traverse Locations Unit No. 2
28
MONSANTO RESEARCH CORPORATION a
-------
located in the stack of Unit No. 1. A pipe the same dimen-
sion as the port extended from each inner wall to within
.31 meter (1 foot) of the outer wall.
Lignite samples were collected at the inlet of the storage
silos from the conveyor which feeds the silos.
29
MONSANTO RESEARCH CORPORATION
-------
V. SAMPLING AND ANALYTICAL PROCEDURES
SAMPLING PROCEDURES
The exhaust stacks of both Unit No. 1 and Unit No. '2 were
generally sampled in accordance with the Federal Register
methods. Some exceptions to these methods were used and
they are listed below:
1. The Type S pitot tubes used-to determine the stack
gas velocity were calibrated at the same velocity as
the stack being measured. Four pitot. tubes were used
to determine the velocity, one per port. The pitot
tube coefficients were averaged for the four tubes and
the average was used in the calculation of the stack
velocity.
2. Two heated glass probes and a stainless steel probe
were used to extract the exhaust gases from the stack.
One glass probe with a glass wool filter was used for
continuous monitor. Another probe without the filter
was.used for the Method 7 NO samples and the Method 3
A.
gas analysis. The stainless steel probe was used for
Method 4.
3- The Method *J moisture samples were collected in two
midget impingers, each containing 10 ml of distilled
water. Sampling was conducted at a continuous rate
of approximately 4 ft3 per hour until the dry test
meter registered 0.06 cubic meter (2 cubic feet) of
dry gas sampled.
30
« MONSANTO RESEARCH CORPORATION
-------
The guidelines proposed in the Federal Register
September 11, 1974, Vol. 39, No. 177, entitled
"Performance Specification 2 - Performance Specifi-
cation and Specification Test Procedures for Monitors
of S02 and NOX from Stationary Sources," were followed
when possible. A Dynasciences Air Pollution Monitor
with a NOX sensor was used. The instrument is de-
signed to give results within ± 5 ppm of the actual
concentration.
The continuous monitoring sampling train consisted
of a heated probe, filter, condensing coil and flask,
and pump. After the exhaust gases passed through the
pump they were pushed through an S02 scrubbing solu-
tion and a filter before they reached the NOX sensor.
Approximately 1 cubic foot per hour of stack gas was
sampled.
The system was calibrated in the morning and again
checked .at the end of the day to insure there was
no instrument drift. The 3^7 ppm NO in nitrogen
calibration gas was certified by the manufacturer
and hence no Method 7 check was run. The calibration
gas was placed into the system before the pump which
pushed the gas through scrubbing solution and filter
into- the NOX monitor.
The Dynascience unit- was connected to a continuous
strip chart recorder and the data was continuously
recorded. A listing of the reading for every 10
minutes is given in Appendix P. The value at the
time corresponding to the Method 7 grab flask was
used in the various tables.
The negative pressure in the stack caused difficulties
in obtaining a representative sample the first day of
sampling. Fresh air was being drawn into the stack
31
MONSANTO RESEARCH CORPORATION o
-------
and causing a dilution to occur. This problem was
eliminated by sealing the port area around the probe,
drawing off a larger volume of the stack gas, and
directing a portion of this gas to the instrument or
sampling system.
Cooler weather on October 3 and 4 caused morning
problems with NOX sensor. The accuracy of the sensor
is very dependent on the temperature of its electro-
lyte. On October 10, 11, and 12 a heating tape was
wrapped around the monitor to keep the sensor warm.
No early morning problems were encountered with
calibration or sampling after the heating tape was
used.
Voltage drop at the platform during the first three
days of sampling may have resulted in incorrect
results. The voltage problem was corrected for the
completion of the test by providing a separate power
line for the electronic instruments involving the
NOX monitor and temperature measurement devices.
Zero gas was used initially for the zero setting of
the Dynasciences monitor. However, zeroing the
monitor with the zero gas resulted in inconsistent
readings. When ambient air was used, no problems
were encountered. Therefore, ambient air was used
as a zero gas for the last six days of the sampling
program.
Lignite samples were taken every half hour. These
samples were composited and at the end of the
sampling day were quartered down into a single
sample for that day. There was no sampling location
at the inlet to the boilers, so the samples were
taken before the lignite was deposited in storage
silos. It was estimated that there was a 3-hour
32
MONSANTO RESEARCH CORPORATION
-------
time lag from the time the lignite went into the
silos until it was fired into the boilers. Sampling
of the lignite began as soon as MRC personnel reached
the job site each morning and ended 3 hours before
the NOX sampling was completed.
ANALYTICAL PROCEDURES
Analytical procedures used generally followed the methods
outlined in the Federal Register. Some exceptions are
discussed below.
The NOX grab samples were analyzed by the phenoldisulfonic
acid method in the December 23, 1971 Federal .Register, Vol.
36, Number 247- The following are exceptions to the method.
The samples were evaporated in 250 ml beakers, not eva-
porating dishes as described in the proposed Method 7
changes. The samples collected on October 10, 11, and 12,
1974 were diluted to 100 ml and a 50 ml aliquot was used
to perform the analysis. These samples were evaporated on
the steam bath with cover glasses on the beakers to pre-
vent contamination of the samples. The absorbance of each
sample was measured at 410 nm using the blank solution as
zero. The observances of the calibration standards were
also measured at. this wavelength.
The percent moisture, volatile, ash, and heat of combustion
were determined on the daily composites of lignite coal.
The samples were ground to pass a 60 mesh sieve and all
analyses were performed on the ground sample. After
grinding the composite sample, ASTM D271-70 was followed
for moisture, ash, volatile, BTU, carbon,hydrogen, nitrogen,
and sulfur determinations.
33
MONSANTO RESEARCH CORPORATION
-------
Specifically, the volatiles were measured by the Meker
burner method; the BTU values were obtained with the Parr
Bomb Calorimeter; the nitrogen determined by the Kjeldahl-
Gunning method; and the carbon and hydrogen were analyzed
with a Hewlett Packard Model 185 Carbon and Hydrogen
Analyzer. All results are reported on an "as determined"
basis, except the BTU values which were obtained from the
oven dried sample.
For the calorific value, the sample was oven dried at
100-110°C so that reproducible results were obtained.
The ground sample had a high moisture content which caused
the BTU value to vary when duplicate ground samples were
run. After the samples were dried the BTU values were
in agreement described by the Parr Bomb method.
The Orsat analysis of Method 3 was performed in accordance
with the methods outlined in the Federal Register with two
exceptions. 'Only two consecutive samples were analyzed if
the results varied no more than 0.5 percent by volume of
each component being analyzed. The carbon monoxide analysis
was only performed once per sample if none was detected in
the first sample.
MONSANTO RESEARCH CORPORATION o
------- |