United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-83-012 May 1983
<&ER& Project Summary
Analysis of Long-Term NO
Emission Data from Pulverized
Coal-Fired Utility Boilers
S. S. Cherry
Long-term NO emission monitoring
data from nine pulverized-coal-fired
utility boilers were analyzed. These data
were in the form of hourly averaged NO,
O2 (or CO2), and load: NO and O2/CO2
were measured with certified
continuous emission analyzers. The
analyses were compared to the
emissions limitations in both the 1971
and 1978 New Source Performance
Standards (NSPS).
The comparison with the 1978 NSPS
showed that seven of the nine boilers
would have produced no periods of
excess NO emissions on a 30-day
rolling average basis. In addition, the
same seven boilers would have
complied with the 1978 NSPS NOX
performance test requirements.
Recommendations are made to more
fully analyze the operation of two of
these boilers to determine if their mode
of operation could be adopted by other
boilers. These boilers use permanently
installed CO monitors to allow
operation at the lowest practical excess
air to achieve either low NO emissions
or better fuel efficiency.
This Project Summary was developed
by EPA's Industrial Environmental
Research Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that is fully docu-
mented in a separate report of the same
title (see Project Report ordering
information at back).
Introduction and Summary
Long-term NO emission data from nine
pulverized-coal-fired utility boilers were
analyzed to determine the degree to
which compliance was achieved with
both the 1971 and 1978 New Source
Performance Standards (NSPS). These
standards are summarized in Table 1.
Pertinent boiler characteristics are sum-
marized in Table 2.
Table 1. Summary of 1971 and 1978 Coal
NO* NSPS
Coal Type
ng/J
lb/101
Btu
1971'
Subbituminous 300 0. 70
Bituminous 300 0. 70
Lignite (Dakotas or Montana) 340 0.80
Other Lignites 260 0.60
1978"
Subbituminous 210 0.50
Bituminous 260 0.60
Lignite f Dakotas or Montana) 340 0.80
Other Lignites _ 260 0.60
'3-hr averaging period trolling or block).
"30-day rolling average.
The NO and 02 (or C02) data were
measured by cert if Fed analyzers.* The
utilities also supplied hourly averaged
unit load, monitor certification reports,
and periodic coal analyses as part of the
data packages.
A KVB computer program was specially
modified for this study to perform the
following computations:
*AII analyzers were measuring NO only, not NOy.
EPA allows the use of NO-only analyzers to satisfy
the requirement to measure "oxides of nitrogen."
-------
Table 2. Boiler Pertinent Information
Utility Station
Oklahoma Gas Muskogee
& Electric
Omaha Public Nebraska City
Power District
Iowa Power & Council Bluffs
Light
Pennsylvania Homer City
Electric
Louisville Mill Creek
f?*jc A Flprtrir
U«o Oi EIGtrlllft
Proprietary
Public Service San Juan
New Mexico
Coal
Unit Type3
4 Sub.
5 Sub.
1 Sub.
3 Sub.
3 Bit.
3 Bit.
A Sub.
B Sub.
1 Sub.
Boiler
Mfgr>>
CE
CE
FW
B&W
B&W
B&W
CE
CE
FW
a Sub. = Subbituminous b\CE = Combustion Engineering
Bit. = Bituminous FW = Foster Wheeler
B&W = Babcock & Wilcox
d Detailed information
Gross NOx Site
MWe Controls* Visit
550 Tang. + OF A Yes
550 Tang. + OFA Yes
575 Interstage air No
720 ORB Yes
680 ORB Yes
425 ORB No"
350 Tang. + OFA Yes
350 Tang. + OFA Yes
3f>0 OFA + ORB Yes
'Tang. = Tilting tangential
OFA = Overfire air
DKB - Dual register burner
1978 NSPS. Note that none of these
boilers are subject to the 1978 NSPS but
were included in the study because they
were equipped with certified continuous
emission monitors (CEM).
Each utility submitted a data package
consisting of: hourly averages of NO, O2
(or C02)*; CEM certification reports; and
periodic coal analyses. These packages
were reviewed for completeness and
legibility. Discrepancies were resolved
with the utility by telephone.
The hourly averages were entered on
computer input sheets for analysis by a
specially modified computer program.
Plots were prepared of NO and O2 (or CO2)
as a function of unit load prior to the site
visits in order to identify any unusual
results.
Seven of the nine boilers analyzed
were visited during this assignment.
Details on an eighth boiler were obtained
by letter. The purpose of these visits was
to hold detailed discussions with the
appropriate station personnel. Topics
discussed included:
1. Daily means and standard deviations
of NO, 02 (or C02), and load.
2. Means and standard deviations of NO
and 02/C02 as functions of unit load.
3. Load probability.
4. NO probability.
5. Number of periods of excess NO
emissions with respect to the 1971
NSPS on both a 3-hour rolling
average and 3-hour block average
basis.
6. Number of periods of excess NO
emissions with respect to the 1978
NSPS on a 30-day rolling average
basis.
Sites were visited (see Table 2) to
discuss details of monitor and boiler
operation. As a result of these visits,
revised computer analyses were
performed for certain boilers when it was
determined that some of the monitor data
was of questionable validity.
The results of this study are
summarized in Table 3. Note that seven of
the nine boilers would have complied
with the 1978 NSPS performance test
and that the same boilers would not have
produced any periods of excess
emissions on a 30-day rolling average
basis. (The "Maximum" column is the
number of 30-day rolling averages
obtained from the data, each of which is a
potential period of excess emissions.)
Discussion
Long-term NO emissions data from
nine pulverized-coal-fired utility boilers
were analyzed and compared with the
Table 3. Analysis Summary
1978
1. Boiler Design-
a. Boiler description, number of
burners and burner arrangement,
and heat release rate.
b. Operation of dual register burners
and/or overfire air system.
•The 1971 NSPS allows the use of either 02 or C02 as
a dilution correction. CC>2 was used on only one
boiler.
Boiler
Muskogee 4
Revised
Muskogee 5
Revised
Nebraska City 1
Council Bluffs 3
Revised
Homer City 3
Mill Creek 3
Unit A
Unit B
San Juan 1
NSPS Per-
Coal for ma nee
Type" ng/J
Sub.
Sub.
Sub.
Sub.
Bit.
Bit.
Sub.
Sub.
Sub.
208
2O7
179
b
228°
141
138
274°
210
189
158
175
Number
of Valid
Days
94
78
82
21
31
96
63
61
54
75
78
57
1971 NSPS NO,
Exceedances
Rolling Block
17
17
16
5
30
0
0
161
0
75
2
0
6
6
6
2
12
0
0
60
0
31
2
0
1978 NSPS NO,
Exceedances
Actual Maximum
0
0
0
2
0
0
32
0
0
0
0
65
49
53
2
67
34
32
25
46
49
28
* Sub. = Subbituminous Bit. = Bituminous
* Insufficient data to calculate 30-day rolling average.
c Would not have complied with 1978 NSPS performance test.
-------
2.
3.
Boiler Operation-
a. 02 versus load.
b. Heat rate.
c. Steam temperature problems.
d. Slagging/fouling characteristics.
e. Unit availability/load limitations.
f. Ash carbon characteristics.
g. Tube wastage.
h. NO excess emissions.
CEM System--
a. Make and model.
b. Sampling location.
c. Data reduction procedure.
d. Recertification.
e. Calibration procedure, frequency,
out-of-calibration flag, and
corrective action.
f. Quality assurance program.
SOOer
4.
Coal Supply—
a. Source.
b. Nominal characteristics
variability.
and
If the site visit identified some of the
data as being of questionable validity, a
revised computer analysis was
performed, deleting these data.
The analysis is illustrated by the results
computed for Mill Creek Unit 3. Figure 1
shows the boiler characteristics (NO and
02) developed as a function of load. These
characteristics were computed by sorting
the data into 10 load "cells" and
averaging the associated values of NO
and 02. Note the relatively flat NO
emission factor across the load range.
Apparently, the decrease in 02 with load
is sufficient to nearly offset the higher
heat release rate. Figure 2 is the load
100
0.01 0.1 0.5 1 2 5 10 20 40 60 80
Percent Less Than
90 95 98 99 99.8 99.99
Figure 2. Mill Creek Unit 3 - load probability.
probability distribution, representable
only in terms of its median value of 270
gross MWe (63.5 percent of rating).
Figure 3 indicates that the hourly
averaged NO is normally distributed with
a mean of 215 ng/J and a standard
deviation of 27 ng/J.
Muskogee Units 4 and 5 are nominally
identical, yet their NO characteristics are
significantly different (Figure 4). Note that
the maximum NO emission factor occurs
at less than full load, and that Unit 5
indicates a significant NO decrease with
loads above 200 gross MWe. The
difference in 02 characteristics (Figure 5)
300
200
700
^Performance /VO,
200
300
400
12
10
500
Generator Output, gross MWe
Figure 1. Mill Creek Unit 3 - boiler characteristics.
was deemed to be insufficient to
completely resolve the difference in NO.
The NO emissions from Council Bluffs
3 and San Juan 1 were low, compared to
the other boilers analyzed. Visits to these
two units indicated that they were being
operated with the minimum air flow
practical, and that both units were
equipped with CO monitors as operating
adjuncts. Reasons for this mode of
operation were:
Council Bluffs 3-
• Induced draft fan limitation at
high load
• Fuel economy
San Juan 1--
• More stringent State of New
Mexico NOx limitation
In addition, neither unit was experiencing
any unusual conditions due to low air
operation; e.g., slagging, fouling,
unburned carbon.
Conclusions and
Recommendations
The major conclusion of this study was
that seven of the nine boilers analyzed
would have complied with the 1978
NSPS performance test requirements
and would have produced no periods of
excess NO emissions on a 30-day rolling
average basis. The main exception
-------
10
0.01 0.1 0.5 1 2 5 10 20 40 60 80 90 95 98 99 99.8 99.99
Percent /.ess Than
Figure 3. Mill Creek Unit 3 - NO probability.
300
200
O'
700
700
200 300 4OO
Generator Output, gross MWe
500
600
*
I-
> Unit 4
' 5~
\\
i
0 10O 200 300 400 500 60C
Generator Output, gross MWe
Figure 5. Muskogee Units 4 and 5 -
comparison of Oz characteristic.
(Homer City 3) is operated in a base-
loaded manner; therefore, it is not possible
to "average in" potentially lower NO
emissions at reduced load.
It was also established that Council
Bluffs 3 and San Juan 1 are low NO
emission sources across their respective
load ranges. This is attributed to their
being operated with the lowest practical
excess air with the assistance of a CO
monitor.
It is recommended that the operation of
these two units be documented in detail
and then discussed with the respective
boiler manufacturers to determine if this
mode of operation could be adopted by
other boilers.
It is also noted that base-loaded units
behave differently with respect to 30-day
NOx rolling averages since they do not
have the operational flexibility to
"average in" periods of potentially lower
NO emissions at lower loads to meet the
1978 NOX standard.
Figure 4. Muskogee Units 4 and 5 - comparison of NO characteristic.
4
-------
S. S. Cherry is with KVB, Inc., Irvine, CA 92714.
Robert E. Hall is the EPA Project Officer (see below).
The complete report consists of two volumes, entitled "Analysis of Long- Term NO
Emission Data from Pulverized Coal-Fired Utility Boilers,"
"Volume I. Technical Analysis." (Order No. PB 83-175 000; Cost: $19.00,
subject to change)
"Volume II. Appendices," (Order No. PB 83-175 018; Cost: $32.50, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
. S. GOVERNMENT PRINTING OFFICE: 1983/659-095/1937
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
------- |