EPA PROJECT REPORT NO, 75-LS6-
CD
O
fc
BASIN ELECTRIC ROWER COOP
Lei and Olds Station
Stanton, North Dakota
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
-------
REPORT NO.
Y-8'179-6
PAGE
TEST REPORT
of
NITROGEN OXIDE EMISSIONS
at
THE BASIN ELECTRIC POWER COOPERATIVE
LELAND OLDS STATION
STANTON, NORTH DAKOTA
Prepared For
THE ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK
NORTH CAROLINA 27711
UNDER CONTRACT NO. 68-02-1401 TASK 6
REPORT NO. 75-LSG-l
Submitted By
YORK RESEARCH CORPORATION
ONE RESEARCH DRIVE
STAMFORD, CONNECTICUT 06906
REPORT NO. Y-8'479-6
May 19, 1975
STAMFORD, CONNECTICUT
-------
REPORT NO.
Y-8M79-G
PAGE
SECTION
I
II
III
IV
V
TABLE or CONTENTS
TITLE
INTRODUCTION
SUMMARY AND DISCUSSION OF RESULTS
Table I - Summary of Results
Table IB - Average Excess Air
at Sampling Site
Table II - Test Results
Table III - Volumetric Flew
Table IV - Plant Operating Data
Table V - Coal Analysis Results
PROCESS DESCRIPTION AND OPERATION
LOCATION OF SAMPLING POINTS
SAMPLING AND ANALYTICAL PROCEDURE
PAGE
STAMFORD, CONNECTICUT
-------
REPORT NO.
Y-8M79-G
PAGE
NUMBER
1
2
7
8
LIST OF FIGURES
PAGE
Leland Olds Station
Schematic of Port; Location at
Leland Olds Station
Cross Section of Stack
NOX Sampling Train
Chemiluminescent NO-NOX Gas Analyzer
and Conditioning System
Flue Gas Collection by Leveling
Bottle
Preliminary Moisture Determination Train
Pitot Tube - Manometer Assembly and
Thermocouple-Pyrometer
STAMFORD, CONNECTICUT
-------
REPORT NO. Y-8 H79-6 PAGE
I. INTRODUCTION
York Research Corporation, an independent consultant in Environ-
mental Engineering, was retained by the United States Environ-
mental Protection Agency, under Contract No. 68-02-1401, Task 6,
to conduct a series of tests at the Basin Electric Power
Cooperative, Leland. Olds Station, Stanton, North Dakota.
These tests were -performed, on a Babcock and. Wilcox horizontally
opposed, burner designed, boiler burning pulverized, lignite and
equipped, with a cyclone d.ust collector. A York Research team
consisting of a Project Director and four test engineers con-
ducted the series.
The primary purpose of the test program was to determine nitrogen
oxide emission levels. Analysis of the test results will then
assist the Environmental Protection Agency in establishing NOX
stand.ards of performance for new lignite fired boilers.
Sampling of the exhaust gases was conducted from sampling ports
located on the stack to determine nitrogen oxid.e concentrations,
molecular weight, moisture content, velocity, and flow. Con-
currently, samples of the lignite were collected for examination.
Pertinent process data was supplied by A. D. Little. Inc., under
a separate contract, and. was used, to calculate some of the
emission rates presented in this report.
NOX emissions were also recorded, with a Thermo Electron Corpora-
tion "Chemiluminescent NOX Analyzer" to check for variations in
the NOX emissions which would, not.be detected by the EPA Method
7 sampling program. This included, monitoring und.er varied.
excess air situations to confirm that NOX emissions were de-
pend.ent on operating conditions.
CORPORATION (sSi! STAMFORD, CONNECTICUT
-------
REPORT NO. Y-SU79-6
PAGE 2
II. SUMMARY AND DISCUSSION OF RESULTS
YORK RESEARCH CORPORATION JS1 STAMFORD, CONNECTICUT
-------
REPORT NO. Y-S'179-6 PAGE
II. DISCUSSION OF RESULTS
In determination of the NOX emissions the following results
were obtained:
The NOx emissions (gm per 106 joules) were reduced by an average
of 31.8% by a reduction in excess air (at the test sampling point)
from 28.2% to 17.8% or a reduction in excess air at the furnace
outlet from 25.25% to IM.0%. The reduction in excess air re-
sulted in a 10.7% reduction in gas flow. Although these re-
ductions in NOX were obtained by changing boiler operating condi-
tions, no long runs were made under these conditions. Therefore,
York Research can riot ascertain if the boiler can operate under
these conditions permanently and. maintain the same level of
reduction.
The TECO NOX analyzer averaged M-0-50 ppm higher than the EPA-7
test methods at this plant. The comparison includes only the
latter part of the test program, however, as initial analyzer
opeiMtion was poor. Inasmuch as TECO's purpose was solely to
indicate significant trends in the NOX concentration, its re-
sults should not be considered equivalent to the EPA-7 data.
YORK RESEARCH CORPORATION (ram) STAMFORD, CONNECTICUT
-------
SUMMARY OF
TABLE I
AVERAGED TEST RESULTS - LELAND OLDS
CONCENTRATION
Date &
Phase
10/1
Baseline
10/2
Baseline
10/2
Low Air
10/3
Baseline
10/3
Lew Air
10/3
Baseline
AVERAGE
Baseline
AVERAGE
Low Air
NOTE 1:
NOTE 2:
NOTE 3:
Flow
(SCFMD)
533,982
520,126
465,868
523,640
477 ,192
53"4, 387
528,034
471,530
Number of
values.
These emis
JJ..I. and IV
Tboso emis
report.
(ppm, dry)
EPA-7 TECO
568
507 ' -
403
556
427 47 9
631 672
566 I
415 I
samples used, to
sion rates are
, respectively.
siori rates were
(ppm i
02,
EPA-7
635
554
408
622
444
705
629
426
EMISSION RATES2
1 3% (Gm/10D
dry) (Lb/Hr) (Lb/MMBTU) joules) Lb/ Gm/10&
TECO EPA-7 TECO EPA-7 TECO EPA-7 TECO MMBTU3 ioules3
21591
(13)
1900
(4)
1334
(9)
2083
(5)
494' 1479
(5)
7 51 237 4
(9)
I 2129
I 1407
determine averages
based on volumetric
determined, using "
I = Insufficient
.963 - .414 - .883 .380
.924 - .397 - - .718 .309
.660 - .284 - .530 .228
.972 - .418 - .870 .374
1659 .690 .773 .297 .332 .620 .267
(5)
2529 1.10 1.17 .473 .203 .985 .424
(9)
I .990 I .765 I .864 .671
I .675 I .525 I .575 .447
are shown in parenthesis below the "Lb/Hr "
flow rates and process d.ata as shown in Table
F Factors" as described in Section V of this
Data
-------
TABLE IB
AVERAGE EXCESS AIR AT SAMPLING
Phase
Baseline
Baseline
Low Air
Baseline
Low Air
Baseline
Avg. Baseline
Avg. Low Air
*Excess air (figures
% F,A = 10° *
Date
10/1
10/2
10/2
10/3
10/3
10/3
computed
% 02
co2
-14.4
15.3
16.2
14.98
15.8
14.8
14.88
. 16.0
02
4.9
4.5
3.1
4.7
3.67
4.9
4;75
3.39
with the following
SITE - LELAND OLDS
ORSAT DATA
CO
..22
..32
.55
-
.2
. .18
'.24
.38
equation:
N2
80.52
79.88
80.27
80.25
80.33
80.13
80.2
80.3
EA*
29.1
25.9
15.4
28.5
20.2
29.4
28 . 2 '
17.8
'
0.264 x % N2 - % 02
-------
TABLE TI
TEST RESULTS
CONCENTRATION -1-
Date &
Phase Time
10/1
Baseline 1030
1100
1130
1200
1230
1300
1330
1400
1430
1500
1530
1600
1630
10/2 0830
0900
0930
100.0
1030
Low Air 1400
1430
1500
1530
1600
1630
1700
1730
1800
* NO READINGS DUE
(ppm, dry)
EPA -7 TECO
529 *
544
516
518
541
585
585
606
620
594
625
557
576
529
506
475
516
Void
425
396
444
383
377
378
391
411
426
(ppm @> 3%
02, dry)
EPA-7 TECO
587.6 *
608.4
580.3
5S3.-4
597.6
658.5
650.3
678.2
698.1
652.6
703.9
604.9
648.7
576.9
552.8'
512.6
574.1
Void.
429.4
419.1
449 . 4
389.4
380.9
377.9
386.3
411.2
428.6
(Lb/Hr)
EPA-7 TECO
1898 *
1954
1982
1993
208-1
2185
2197
2277
2406
2306
2407
2146
2239
1998
1914
1767
1919
-
1402
1306
1410
1215
1209
1213
1351
1423
1478
EMISSION RATES2
(Lb/MMBTU)
EPA-7 TECO
. . 8 47 *
.841
.884
.889
.928
.975
.980
1.01
.939
1.03
1.07
.957
.999 .
.971 '
.930
.859
.933
.690
.643
.694
.598
.596
.603
.671.
.707
.735
(Cm/100
joules)
EPA-7 TECO
.364 *
.362
.380
.382
.399
.419
.421
. .434
.404
.4M3
.460
.412
.430
.418
.400
.369
.401
-
.297
.276
.298
.257
.256
.259
.289
.304
.316
Lb/
MMBTU3
.818
.848
.808
.812
;840
.895
,905
.945'
.972
.908'
.980
.842
.903
.748
.717
.664
.744
.557
.543
.582
.505
.494
..490
.500
.533
.566 -
Gm/10G
joule s^
.352
.365
.347
.349
'.361
.385
.389
.406
.418
.390
.421
.362
.388
.322
.308
.286
.320
.240
.233
.250
.217
.212
.211
.215
.229
. 2 43
TO PLUGGED CAPILLARY
-------
TABLE II ('CONTINUED')
TEST
RESULTS
CONCENTRATION l
Date &
Phase
10/3
Baseline
Low Air
Baseline
(ppin y 3%
(ppm, dry) 02, dry)
Time
0830
0900
0930
1000
1030
1230
1300
1330
1400
1430
1500
1530
1600
1630
1700
1730
1800
1830
1900
*NO READINGS DUE
NOTE : 1
2
-
3
- Maxima
- These
Tables
- These
EPA-7
531
505
535
617
590
453
420
425
457
381
620
623
636
623
592
626
670
657
634
TO PLUGGED
and. Minima
TECO EPA-7
* 587 . 5
558.3
602.7
685.8
540 676'. 5
510 468.2
420 427 . 5c
440 449.8
505 483;7
520 390.0
630 698.1
. 660 701.9
660 729.4
700 683.6
680 662.1
680 691.2
680 745.4
680 730.9
680 704.8
CAPILLARY
TECO
*
620
528
427
466
535
390
709
743
757
769
761
751
756
756
756
values underlined.
emission rates are based.
Ill and. IV
, respectively
EMISSION RATES2
(Lb/IIr) (Lb/MMHTU)
EPA-7
2004
1904
2021
2338
2159
1657
1441
1456
1548
1294
2104
2371
2418
. 2377
2260
2394
2532
2483
2429
for each
on volumetric
9
emission rates were determined
using "
TECO E.PA-7 TECO
* .935
.888
.944
1.09
1976 1.00
1865 .773
1441 .672
1507 .679
1710 .722
1766 .609
2137 .972
2511 1.10
2509 1.11
2670 1.09
2595 1.04
2600 1.10
2569 1.17
2569 1.15
2605 1.12
phase
flow rates
F-Factors"
A
.922
.870
.672
.703
.798
.824
.987
1.16
1.15
1.23
1.20
1.20
1.18
1.18
1.20
(Cm/100
joules)
EPA-7
-.402
.382 .
.406
.469
.430
'..332
.289
.292
.310
.262
.418
.473
.477
.469
.447
.473
.503
.494
.482
Lb/ Cm/10 6
TECO MMBTU3 joule s3
*
.396
.374
.289
.302
.343
.354
.424
.499
.494 1
.529
.516
.516
.507 1
.507 1
.516
and. process, data shown
as described, in
Section
.826
.779
.843
.958
.946
,655
.593
.629
.676
.545
.976
.981
.02
.954
.926
.967
.04
.02
.984
.in
V
.355
.335
.362
.412
.407
.282
.255
.270
.291
.234
.420
.422
.439
.410
.398
.416
.447
.439
.423
of this report. ' '
-------
REPORT NO.
Y-8479-6
PAGE
8
TABLE III
LELAND OLDS - NORTH DAKOTA
Date &
Phase
10/1
Baseline
Plant
Gross Duct
Load Temp.
Time (MW) (° F)
Volumetric Flow
ORSAT DATA1
GAS FLOW RATES
%
CO
02 CO (%v/v)
10/2
Baseline
10/2
Low Air
10/3
Baseline
10/3
Low 'Air
10/3
Baseline
1030
"1130
1230
1330
1430
1530
1630
0830
0930
1030
1400
1500
1600
1700
1800
0830
0930
1030
1230
1330
1130
1500
1600
1700
1800
1845
205
203
201
199
198
198
200
373
378
382
380
377
383
384
380
385
380
387
392
392
391
388
383
388
384
396
395
394
398
393
400
399
398
13.6
14.5
14.6
14.6
14.3
14.6
14.6
15.7
15.3
15.3
15.8
16.2
16.0
16.4
16.2
15.0
14.6
14.5.
15.8
15.8
1-6.0
14.8
14.5
14.9
15.0
15.0
4.9
5.0
4.7
4.8
4.9
5.0
5.0
.4
.2
.4
.2
.2
.2
.2
4.5 .6
4.3 .3
4.4 .1
3.2
3.2
3.2
2.8
3.1
4.7
5.0
5.3
.6
.0
.4
.6
.3
.1
.0
.0
3.3 .2
4.0 .3
3.4 .1
5.0
5.3
4.9
4.8
4.8
.2
.0
.3
.6
.1
15.0
13.2
13.3
20.2
16.0
13.7
15.4
14.4
Actual
(ACFMlv1)
981,215
1,055,935
1,061,956
1,034,586
1,043,344
1,043,071
1,055,114
1,034,039
1,023,638
995,994
994,078
962,329
974,098
997,637
996,542
1,046,355
1,053,198
1,017,617
965,614
972,456
972,456
1,082,210
1,081,115
1,091,789
1,076,736
1,091,242
Standard
Dry
(SCFMD)
503.699
539.201
539^509
526,670
544.128
539,734
544,934
529.925
521^118
509,335
462.869
444^978
450.096
48 5',107
456,289
528.447
529.147
513^325
480.609
475^205
475,762
533.218
535^417
536,495
529,523
537.283
NOTE; 1 - A grab sample Orsat was taken every half hour. The values re-
sulting from the samples taken at the times shown above were used in com-
puting the flow rates for that particular hour. The only exceptions are
that the Orsat values for 1030 and 1130 on 10/1 are the averages of the.
values for 1030^1100 and 1100-1130, respectively.
NOTE; 2 - Two or three flue gas moisture samples were taken each test
day. The first moisture value was used in the flow computa-
tions until a new sample was taken, and this new value was
then used until the next sample was taken, etc..
YORK RESEARCH CORPORATION
STAMFORD, CONNECTICUT
-------
TABLE IV
SUMMARY OF PLANT OPERATING DATA
LELAND OLDS BOILER NO. 1
BTU/Lb Gross Heat % %
Gross Coal Feed. Rate Coal Input Excess Excess
Date Time (MW) (Ibs/hr, as reed) (as reed) MMBTU/Hr. 02- Air **
10/1/71 10:30-16:30 205 322,000 6960 22M-1 4.5 27
10/2/71 08:30-10:30 . 203 . 318,000 6165 2056 4.4 26
14:00-16:00 201 314,000 6465 2030 2.9 ' 16
16:00-18:00 199 311,, 000 6465 2011 2.5 13
10/3/74 08:30-10:30 198 310,000 6909 2142 4.3 25
12:30-14:30 198 310,000 6909 2142 2.5 13
15:00-19:00 200 313,000 6909 2163 3.9 23
''Measured, before Air Heater. Excess 02 is measured with L&N paramagnetic analyzer.
** Per cent excess air was determined, from a nomograph supplied by A. D. Little. (See attached
Test
Phase
Baseline
Baseline
Loiv Air
Low Air
Baseline
Low Air
Baseline
copy)
-------
:^>SM^Sf^
TEST DATE I
REPORT NO. Y-8M-79-6
PAGE- -13
-------
TABLE V
COAL ANALYSIS RESULTS
Date
Sample No.
BTU/Lb. (Dry)
BTU/Lb.
(as received)
Proximate as
received
Volatile C
Fixed C
Ash
Moisture
Proximate Dry
Volatile C
Fixed C
Ash.
Ultimate (Dry)
Ash
S
N
C
II
0
F Factors
(DSCF/101* BTU)
10/1
1
9,414
6,836
35.32
31.76
5.53
27.39
48.64
43.74
7.62
7 . 62
0.51
0.90
58.53
4.70
27.74
100.2
10/1
2
9,219
7,084
34.88
33.95
8.01
23.16
45.40
55.83 .
10.43
10.43
0.79
0.93
50.50
4.82
26.53
100.4
(DSCM/1014- joules) .00268 .00270
10/2
3
9,824
6,461
25.68
33.84
6.25
34.23
39.04
i)8 . 55
9.51
9.51
0.03
0.90
5G.30
"4 . 07
28.59
89.9
.00241
10/2
4
9,194
6,469
29.76
33.88
7.02
29.34
42.12
52.06
9. --94
9.94
0.63
0.93
56.13
'1.28
28.09
96.8
.00260
10/3
5
8,989
7,123
38.34
36.23
4.67
20.76
48.39
45.72
5.89
5.89
0.43
0.86
58.32
4.95
29.55
104.6
.00279
10/3 10/1' 10/2
678
9,899
6,695
27.40 27.28 31.18
33.16 29.72 33.11
7.07 7.85 9.13
32.37 35.15 26.58
40.52 42.07 42.47
49.02 45.82 45.09
10.46 12.11 12.'44
10.46
0.57
1.04
58.87
4.55
24.51
96.8
.00260
10/3
9
30.78
32.27
10.07
26.88
42.10
44.13
13.77
-------
REPORT NO. Y-8M79-6 PAGE. i2
SECTION III - PROCESS 'DESCRIPTION AND OPERATION
The Leland Olds Unit #1 is a 216-MW steam-electric plant. The
boiler, designed by Babcock and Wilcox, burns pulverized lignite
which is fired through horizontally-opposed burners, as shown in
Figure 1. The lignite is pulverized in one of ten pulverizers,
each pulverizer feeding two burners. The burners are arranged
in three rows of four burners each on the front wall, and two
rows of four burners e-ach on the rear wall. The plant was first
put into operation in 196G.
During the test program, the gross electrical load and excess 62
before the air preheaters were record.ed from the company instru-
ments. No overfire was attempted, and total air was assumed.
equally distributed over all burners; thus, total excess air was
assumed Lo equal burner excess air. Steam flow and coal feed
rates were also recorded during the testing. Some oil was used
on October 2, 197 M-, and October 3, 197 M-, but not during the
emissions testing program.
Table IV summarizes the boilei-1 conditions which xvere tested for
Leland Olds #1. Essentially, only baseline and low air data
were obtained, although on the first day a certain burner "J"
was off and this constituted a "special test". Copies of the
company's log sheets are included in Appendix ft.
Operating conditions during any of the identified test phases
were subject to changes because of the nature of plant operation.
An example of the reason for this drift is that the 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 NOX measurements taken at one-half hour intervals.
Therefore, the averaged. NOX data corresponds to an average con-
dition representative of the range over which the boiler con-
ditions drifted.
CORPORATION KE^B) STAMFORD, CONNECTICUT
-------
REPORT NO.
Y-SM79-G
PAGE. 13
IV. LOCATION OF SAMPLING POINTS
The sampling sites for the emission tests were located at the
234 foot level of the exhaust stack. The inside diameter of
the stack at this point was 18.157'. This location is 204-T
downstream from the nearest disturbance which is the breeching
entry to the stack at the 30 foot level. The stack outlet is
11G feet above the sampling ports. The sampling ports are
located at 11 stack diameters downstream and 6.2 stack diameters
upstream from the nearest disturbance as described by Method #1
December 23, 1971, Federal Register.
STAMFORD, CONNECTICUT
-------
REPORT NO.
Y-8U79-6
PAGE 14
FIGURES
YORK RESEARCH CORPORATION
STAMFORD, CONNECTICUT
-------
!i«^
'
^s;^v* ,;:|o |'''''-<5 --VS
r-iU:^^--rf;VlT^-:nlLt1 fil" ll Hi!
CO
CD
cn
ELECTRIC POWER CO-0='£rWTIVE
uiSVA^CK. VOTIH OAKOT4
LELAND OLD STATION
F/OURE I
-------
REPORT NO. Y-8M79-6
PAGE
16
22
9
/IG
-17' 6"-
ELEV.350
18'8"-
ELEV.234
21'6"-
593/;6
ELEV: 234
6" CAPPED PORTS
ELEV: 350'
SCHEMATIC OF PORT LOCATION AT LEL&ND OLDS STATION
FIGURE 2
STAMFORD, CONNECTICUT
-------
Page 17
H20 a Orsat
Velocity Traverse
Velocity Traverse
NOK Teco
Velocity Traverse
Velocity Traverse
LELAND-OLDS
Stack Dia Id's"
NO, Method 7
CROSS SECTION
FIGURE 3
-------
0 PURGG
senate
.STBCK
PUDP
-------
EXHAUST
ICE
BATH
NO 'MODE
.?.«<>*«!.
» o .'<*>> . :
DRY AIR
OR
OXYGEN
PHOTO-
MULTIPLIER
jr
NOX -NO
CONVERTER
^
^
STACK
FILTER
KNOCK
OUT
NOK MODE
CHEMILUMINESCENT N0-N0a GAS
ANALYZERS CONDITIONING SYSTEM
STRIP CHART
RECORDER
(OPTIONAL)
POWER
SUPPLY
FIGURE 5
-------
REPORT NO. Y-8479-6
PAGE20 OF PAGES
STACK
FLUE GAS COLLECTION/ BY
LEVELING BOTTLE
GAS FLOW
SAMPLE
GAS
WATER
FLOW
ORSAT 5AMRLE ANALYSIS
VENT
WATER
FLOW
GAS FLOW
SAMPLE
3AS
ORSAT
FIGURE #
YORK RESEARCH CORPORATION
STAMFORD, CONNECTICUT
-------
m
T3
o
?o
H
Z
O
oo
ID
I
ST9CK
O
m
N)
O
FJGURG
O
m
C/)
-------
REPORT NO. Y-8 47 9-6
PAGE 22
PIPE COUPLING^ TUBING ADAPTER
f
' Pilot tube-manometer assembly
5
YORK RESEARCH CORPORATION
STAMFORD, CONNECTICUT
-------
REPORT NO. Y-8479-6 PAGE 23
V. SAMPLING AND ANALYTICAL PROCEDURES
The sampling at each plant consisted of the following tests:
Nitrogen Oxides - EPA Method 7
Nitrogen Oxides - NOX Analyzer
Orsat - EPA Method 3
Moisture Determination - EPA Method M-
Velocity and Flow - EPA Method 2
Nitrogen Oxides
The nitrogen oxide emissions were determined by two methods.
The first, EPA Method 7, made use of grab flasks. The second
involved, utilization of a Thermo Electron Corporation "Chemi-
luminescent NOX Analyzer".
It should be noted that EPA Method 7 procedures followed in our
nitrogen oxide sampling were those specified in a revised Method
7 dr-aft given York Research's Project Director before testing
began. This draft is included in the Appendix 7. Furthermore,
the sampling schedule consisted of a grab sample taken every
half-hour during the three days of testing, as is recorded, in
the Field Data sheets.
As a check for trends in the NOX concentration that might have
gone unnoticed with this half-hour sampling routine, the con-
tinuous NOX analyzer was utilized. The principle of operation
for this monitor is the chemiluminescent reaction of NO and 0--;
a synopsis appears below and the procedure is in Appendix 9.
To measure NO concentrations, the gas sample to be analyzed is
blended with 03 in.a flow reactor. Light emission results when
electrically excited. N02 molecules revert to their ground state.
The resulting chemiluminescence is monitored through an optical
filter by a high sensitivity photomultiplier positioned at one
end of the reactor. The filter-photomultiplier combination re-
sponds to light in a narrow wavelength band, unique to the above
reaction. The flow parameters can be adjusted, in such a way
that the output from the photomultiplier is linearly proportional
to the NO concentration.
To measure NOX concentrations as was done for this test, the
sample gas flow was first diverted, through an N02 to NO converter.
By transforming any N02 in the NOX concentration to NO, an
effluent of NO was created which was linearly proportional to the
NOX concentration entering the converter. This flow could then
be analyzed by the monitor to give a reading for ppm concentration
of NOX not just NO.
CORPORATION (sale) STAMFORD, CONNECTICUT
-------
REPORT NO. Y-8'179-6 PAGE24
A conditioning system, depicted in Figure 5, was setup before
the TECO unit for gas preparation. On extraction from the
duct, the gas was sent through a filter that removed, particulate
above 20 u. It then flowed down the probe to a cooling coil
submerged in an ice bath. At the exit of the coil a water knock-
out separated, and. stored, the condensate. The resultant dry gas
traversed a Thomas pump and a 1 u filter, and, subsequently, was
fed into the TECO unit.
The monitoring data from the analyzer was xnecorded. on a Rustrak
Recorder. Data reduction was accomplished by obtaining the
arithmetic mean for each period of time recorded and tabulating
this as the concentration, as is shown in Table II. Maxima and.
minima for each phase are also noted in Table II.
Daily calibration was performed on the TECO unit using the manu-
facturer's guidelines. Generally, this took place in the morning,
though a calibration was also conducted after each repair. It
consisted of injecting a gas, analyzed as 292 ppm of NO? on 9/27
at the York Research lab, into the analyzer after the instrument
had. been zeroed.. The analyzer cal ad.just setting was then
changed so that the instrument output read 290. Full calibration
directions are in the instruction manual for the Model 10B Rack
Chemiluminescent NO-NOX Gas Analyzer. November, 1973, Thermo
Electron Corporation.
Orsat
During the testing an Orsat grab sample was taken every half hour,
at approximately the same time as the Method. 7 NOX sample. The
field data sheets have a single number record.ed. for each gas com-
ponent of these Orsat analyses.
The leveling bottle technique was used by York Research to extract
the sample from the stack and. draw it into the analyzer. Figure 6
provides an illustration. Though this system differs from EPA
Method. 3 procedures, where a squeeze bulb is utilized, it is the
only dissimilarity between the sampling techniques. The analysis
was performed, with an Orsat unit.
As noted, above, a single sample was collected, analyzed, and
record.ed every half hour. The Orsat values entered at the times
indicated, in Table III were used, to calculate the volumetric flow
rate for that hour. The concentration values, adjusted, to three
percent 02 and shown in Tables I and. II, were corrected, using
the Orsat 0? value of the sample taken at approximately the same
time as the Method 7 NOX sample. This deviates slightly from the
requirements of EPA Method 3, which stipulates that grab sampling
and. analysis be repeated, until three consecutive samples vary by
no more than 0.5 percent, by volume, for any gas component sampled.
if t *»'" x>
CORPORATION h3SOE STAMFORD, CONNECTICUT
-------
REPORT NO. Y-8M79-6 PAGE 25
However, this deviation does not significantly affect results
since there was little variation in the Orsat analyses during
any specific test phase.
Moisture
The percent moisture in the flue gas was determined, with EPA
Method M- (See Figure 7). Sampling was conducted, at a single
point located one foot from the stack wall for 1 1/2 to 3 hours
at a constant flow rate of 0.05 cfm_ It was repeated during
each of the NOX test phases.
Velocity and Flow
Velocity and. flow were established with EPA Method 2 (See Figure
8). Twenty-four traverse points, meeting minimum EPA Method. 1
requirements, were tested.. The measurements were performed with
four different pitot tubes, and in reducing the data, the individ-
ual Cp values were averaged (to .85 rounded, from .851) for utili-
zation in calculations and. measurements. A record of each pitot
tubes Cp value and. port placement has been made in the field, data
sheets.
Velocity traverse measurements were taken every hour during the
NOX sampling, starting with the first NOX test in the morning and
continuing throughout the test day.
Table III is a compilation of computed velocity and flow rates.
Moisture values for each calculation were determined, with EPA
Method M-. utilizing the moisture data included in the field data
sheets. Orsat values were applied to the relevant flow computa-
tions as described, in the previous subsection. And the stack
temperatures for every calculation were averaged from the 2M points
of the appropriate velocity traverse. The average temperatures
are recorded at the bottom of the velocity traverse sheets.
Coal Sampling
Nine of the ten feed.ers were sampled. The cap on one of the ports
could, not be removed, and hence that feed.er could, not be sampled..
During each excess air change and baseline test phase, approxi-
mately 5-10 Ibs. of coal were collected from each feeder in
regular intervals over a two hour period.. Each sample was riffled
twice after collection from a feed.er and. the pretained. portion was
ad.d.ed to a composite pile. At the end. of each test day the com-
posite pile was riffled to obtain a sample for that day.
The quartering procedure, as outlined in ASTM Method. D271-68,
was used, as the basis for this collection procedure.
CORPORATION p-nfe) STAMFORD, CONNECTICUT
-------
REPORT NO. Y-8479-6 PAGE. 26
In addition, a single grab sample was obtained daily from one
of the feeders and. immediately sealed to prevent moisture loss.
This was felt to be more accurate than using the moisture
analysis of the daily sample because the probability and amount
of moisture loss was reduced.
As mentioned above, ASTM Method D271-68 was applied for coal
sampling and analysis. Moisture corrections were incorporated
into the coal analysis tabulation according to this method. The
HHV (high heating value, BTU/Lb) results depicted in Table V
(Section II) were established by ASTM Method D2015-66, while
the elemental analysis values were obtained, through the auto-
mated Pregl Method (ASTM) with a Perkin-Elmer Model 240
elemental analyzer.
F-Factor
In Tables 1 and II of Section II, F-Factor emission rates have
been included. They were calculated, according to the equation;
E = CF 2MO
f
\
20.9 - % 02
displayed in the Federal Register, Wednesday, September 11, 1974,
Vol. 39, No. 177, Part II. Also outlined in this register (60,, 46)
were procedures for calculating the F-Factor of the coal. The
equation involved, was:
F = 100 [(364) (% H) + (153) (% C) + (57) (% S) + 14 ^ N) -(46) (%02))
HHV
where the HHV (High heating values) and. H, C, S, N and Op values
were determined by the ASTM methods noted above. Example calcu-
lations with the F-Factor are shown in Appendix 5.
As illustrated by the results in Tables I and. II, there is a
difference between the emission rates calculated, with F-Factors and.
those determined through measured values. The probable explanation
lies in the advantage of the F-Factor itself.
Emission rates calculated, by F-Factor do not require measurements
for heat input and. flow. Both are difficult to gauge and. errors
in their assessment can easily be incorporated, in the subsequent
measured emission rate determinations. It is felt that this was
the reason for the difference noted, in the report.
However, it is possible that the lignite samples analyzed were
not representative of the lignite being used as fuel since Type
II, Condition C sampling procedures (as described in ASTM D223M-72)
YORK RESEARCH CORPORATION fecfe] STAMFORD, CONNECTICUT
-------
REPORT NO.
Y-81-17 9-6
PAGE
27
were followed. Similarly, there is also the possibility that the
Orsat readings were inaccurate. However, the close correlation
between the Orsat readings attained, at North Dakota and the nomo-
graph supplied by A. D. Little seems to preclude this latter
possibility.
STAMFORD, CONNECTICUT
-------
REPORT NO. Y-8 47 9-0
PAGE
28
Project participants and assignments
Donald Fraser - (Project Director) - N'0X (analyzer)
Louis Millspaugh - (Test Engineer) - NQX EPA Method 7
Ross Kittrell - (Test Engineer) - Orsat'and Moisture Tests
Kurt Mesedahl - (Test Engineer) - Velocity Traverse
Jonathan Gardner - (Test Engineer) - Velocity Traverse and.
Coal Sampling
Prepared By:
Roy j
Dir2;ctor\Env(lrcnrjiental Sciences
Reviewed By
Anthony Licata
Vice President
YORK RESEARCH CORPORATION
STAMFORD, CONNECTICUT
------- |