v>EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/3-82-026
October 1982
Air
Gaseous Continuous
Emission Monitoring
Systems -
Performance
Specification
Guidelines for SO2/
NOx, CO2, O2,
andTRS
-------
EPA.-450/3-82-026
Gaseous Continuous Emission
Monitoring Systems -
Performance Specification Guidelines
for SO2, NOx, CO2, O2, and TRS
Emission Standards and Engineering Division
U.S ENVIRONMENTAL PROTECTION AGENCY
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
research 7.-;angie Park, North Carolina 27711
October 1982
-------
This report has been reviewed by the Emission Standards and Engineering Division of the Office of Air Quality
Planning and Standards, EPA, and approved for publication. Mention of trade names or commercial products is not
intended to constitute endorsement or recommendation for use. Copies of this report are available through the
Library Services Office (MD-35), U. S. Environmental Protection Agency, Research Triangle Park, N. C. 27711, or
from National Technical Information Services, 5285 Port Royal Road, Springfield, Virginia 22161.
-------
TABLE OF CONTENTS
Page
1. Introduction 1
2. Definitions 2
3. Performance and Equipment Specifications 4
4. Performance Specifications Test Procedures 7
5. Continuous Emission Monitoring System Calibration 17
6. Equations 23
7. References 25
m
-------
GASEOUS CONTINUOUS EMISSION MONITORING SYSTEMS
PERFORMANCE SPECIFICATION GUIDELINES
S02, NOX, C02, 02, and TRS
1. Introduction
The purpose of these guidelines is to provide vendors, purchasers,
and operators of gaseous continuous emission monitoring systems (CEMS's)
with (a) guidelines for performance and equipment specifications and
(b) suggested test procedures and data reduction procedures for
evaluating the capabilities of gaseous CEMS's.
These guidelines are intended to supplement the requirements in the
CEMS performance specifications (40 CFR Part 60, Appendix B) and
not to be a prerequisite for acceptability of a CEMS for compliance with any
regulation. Some of the suggested test procedures are not applicable
to some specific types of CEMS's, in particular, single-pass, in situ
CEMS's. The manufacturers of CEMS's should provide guidance, applicable
procedures, and specifications for their type of equipment whether or not
that equipment is specifically addressed in these guidelines.
These guidelines are intended for the evaluation of the initial
capabilities of a CEMS. They also provide the user or operator with
guidance that would aid in increasing knowledge of the operation of the
CEMS and provide greater assurance that the CEMS will operate initially
as it is intended.
These guidelines are not designed for the evaluation of
CEMS performance over extended periods of time. The owner or
operator may want to consider longer term evaluation of the CEMS
to ensure satisfactory reliability. Separate quality assurance
-------
procedures are required for this purpose. Such procedures may be
similar to the procedures in the guidelines [e.g., the relative accuracy
(RA) tests], but long-term evaluation procedures must include factors
not part of this document. For example, complete quality assurance
procedures should include CEMS data quality (e.g., frequent RA and
calibration checks), assessment procedures, record keeping requirements,
required maintenance schedules, limits of acceptability, and adjustment
procedures. The user or operator should develop these procedures and
specifications as appropriate for the CEMS and in compliance with the
applicable regulations.
2. Definitions
Terms and expressions used throughout these guidelines are defined
below:
2.1 Continuous Emission Monitoring System. The total equipment
required for the determination of a gas concentration or a gas emission
rate. The system consists of the following major subsystems:
2.1.1 Sample Interface. That portion of a system used for one
or more of the following: sample acquisition, sample transportation,
sample conditioning, or protection of the monitor from the effects of the
stack effluent.
2.1.2 Pollutant Analyzer. That portion of the system that senses
the pollutant gas and generates an output proportional to the gas
concentration.
2.1.3 Diluent Analyzer. That portion of the system that senses
the diluent gas (e.g., 02 or COg) and generates an output proportional
to the gas concentration.
. 2
-------
2.1.4 Data Recorder. That portion of the monitoring system that
provides a permanent record of the analyzer,output. The data recorder
may include automatic data reduction capabilities.
2.2 Types of Monitors.
2.2.1 Extractive Monitor. One that withdraws a gas sample from
the stack and transports the sample to the analyzer.
2.2.2 In-situ Monitor. One that senses the gas concentration in
the stack environment and does not extract a sample for analysis.
2.3 Span Value. The upper limit of a gas concentration measurement
range that is specified for affected source categories in the applicable
subpart of the regulations.
2.4 Calibration Gases. A known concentration of a gas in an
appropriate diluent gas.
2.5 Calibration Gas Cells or Filters. A device that, when
inserted between the transmitter and .detector of the analyzer, produces
the desired output level on the data recorder.
2.6 Zero Drift. The difference between the CEMS response to a
zero-level calibration gas and the reference value after a stated period
of operation during which no unscheduled maintenance, repair, or
adjustment took place.
2.7 Calibration Drift. The difference between the CEMS response
to a span-level calibration gas and its reference value after a stated
period of operation during which no unscheduled maintenance, repair, or
adjustment took place.
-------
2.8 Response Time. The amount of time required for the CEMS to
display on the data recorder 95 percent of a step change in pollutant
concentration.
3. Performance and Equipment Specifications
Tables 1, 2, and 3 list guideline specifications for S02> NOX> C02>
02, and TRS (total reduced sulfur) CEMS performance and equipment
parameters. Section 4 provides explanation and description of the
parameters and the test procedures for determining these values.
The CEMS output range is defined in terms of the span value that is
given in the applicable subpart of the regulations. This span value is
generally 1.5 to 2.5 times the expected emission measurement so that the
majority of the output data are produced in the mid-range area.
If no span value is provided, use the same criteria of 1.5 to 2.5
times the emission level in determining the CEMS range. For example, a
coal-fired boiler fires 2.5 percent sulfur coal and employs a flue gas
desulfurization scrubber for S0« removal. The S02 concentration at the
inlet to the scrubber would average about 1500 ppm, the oxygen (02)
concentration about 7.0 percent, and the carbon dioxide (C02) concentration
about 12.0 percent. The appropriate span values for this location would
be 3000 ppm for S02, 15.0 percent for 02 (air at 20.9 percent 02 is
often used, but depending on the analyzer, this is not always the
best value), and 20 to 25 percent for CO,,.
The scrubber designed for 90 percent S02 removal would produce
S02 concentrations of about 150 ppm in the exhaust. The oxygen and
carbon dioxide concentrations would remain constant through the
scrubber provided no in-leakage of air occurs. The appropriate span
-------
TABLE 1. S02 AND NOX CEMS GUIDELINE
PERFORMANCE SPECIFICATIONS
Parameter
Specifications
1. Response time
2. Zero drift, 2-hour
3. Zero drift, 24-hour
4. Calibration drift, 2-hour
5. Calibration drift, 24-hour
6. Calibration error
7. Relative accuracy3
£15 minutes
£2.5 percent of span value
<.2.5 percent of span value
£2.5 percent of span value
£2.5 percent of span value
£5 percent of span value
£20 percent or 10 percent of
emission standard, which-
ever is less
Expressed as the sum of the absolute mean of the difference plus
the 2.5 percent error confidence coefficient of a series of tests
divided by a reference value.
TABLE 2. 02 AND C02 CEMS GUIDELINE
PERFORMANCE SPECIFICATIONS
Parameter
Specifications
1.
2.
3.
4.
5.
6.
Response time
Zero drift, 2-hour
Zero drift, 24-hour
Calibration drift, 2-hour
Calibration drift, 24-hour
Calibration error
7. Accuracy
£15 minutes
£0.5 percent 0« or C02
£0.5 percent 02 or C02
£0.5 percent 02 or C02
£0.5 percent 02 or C02
£0.5 percent 02 or C02
£l.O percent 02 or C02
Expressed as the sum of the absolute mean of the differences plus
the 2.5 percent error confidence coefficient of a series of tests.
-------
TABLE 3. TRS CEMS GUIDELINE
PERFORMANCE SPECIFICATIONS
Parameter
Specifications
1. Response time
2. Zero drift, 2-hour
3. Zero drift, 24-hour
4. Calibration drift, 2-hour
5. Calibration drift, 24-hour
6. Calibration error
7. Relative accuracy9
<_15 minutes
<2 percent of span value
<2 percent of span value
<5 percent of span value
-------
value for SO^ should allow for the measurement of some excursions from
90 percent control, so a factor of 2.5 is used. This results in a span
value of about 375 ppm for SCL.
Use these span values in determining the calibration gas values
and the pollutant analyzer response ranges.
These calibration gases should be introduced to the CEMS so as to
involve as much of the CEMS system as possible. For example, an
extractive CEMS should be equipped to introduce calibration gases at
the connection between the probe and the sample line near the stack or
duct port location. It is not necessary to use this connection for
daily drift checks, but the calibrations should be and response time
checks must be made through the entire CEMS.
4. Performance Specifications Test Procedures v
4.1 Location of Monitor. The proper location of the sample
interface of the monitoring system is important for several reasons:
(1) ease of access will facilitate proper maintenance, (2) CEMS
measurement must be representative of the total emissions, and (3)
proper location of the sample point(s) or path within the duct or stack
is vital in meeting the RA requirements.
The CEMS should be located where it would most likely meet the
RA requirements. It is suggested the measurement location be at least
two equivalent diameters downstream from the nearest control device,
the point of pollutant generation, or other point at which a change in
the pollutant concentration or emission rate occurs and at least 0.5
-------
equivalent diameter upstream from the effluent exhaust or control
device. Individual subparts of the regulations may contain additional
requirements. For example, for steam generating facilities, the location
must be downstream of the air preheater.
Stack locations for GEMS's are generally farther downstream of
control devices than most duct locations. Gas mixing is usually more
complete meaning less stratification of pollutant concentrations and
an easier task for determining RA. Disadvantages of stack locations
include remoteness of the sample probe or the entire CEMS, as in the
case of in-si.tu monitors. This remoteness often leads to poor
maintenance and frequent breakdowns. For extractive GEMS's stack
locations often mean long, heated sample lines from the probe
to the analyzer. Long sample lines require more energy to
maintain the heat and are subject to electrical failure more frequently
than short lines.
Sample point(s) or paths within the stack or duct may be
determined a number of ways. Performance Specifications 2 and 5
suggest that the location of the sample point(s) or the majority of the
sample path for the CEMS be within the concentric inner
50 percent of the stack or duct cross section. Generally, any
sample point(s) or path that include the centroid of the stack or
duct and meets the other measurement location criteria above will
provide samples representative of the average emission concentrations.
Citation 7.7 provides further information in selecting measurement
points for coal-fired sources.
-------
The CEMS locations in duct work are usually the most accessible
for maintenance and repair purposes, but often provide very difficult
situations in terms of representative sampling. Because duct work
is minimized for efficiency of space and cost, few duct locations
provide a well-mixed condition desirable for CEMS installation.
Frequently, by-passed gas is mixed with cleaned or scrubbed gas for
reheating purposes, and the result is significant stratification of
emissions within the duct work. Stratification can also occur due
to poor scrubber operation, especially plugging of gas passageways
and liquid-slurry spray nozzles. This latter type of stratification
is most difficult to measure and is often temporally variable.
If stratification is suspected, the following procedure to locate
a point or path of average emissions is suggested. For rectangular
ducts, locate at least nine sample points in the cross section such
that sample points are the centroids of similarly-shaped, equal
area divisions of the cross section. Measure the pollutant concentration
and, if applicable, the diluent concentration at each point using
appropriate reference methods or appropriate instrument methods that
give relative responses to pollutant concentrations. Then calculate
the mean value for all sample points, and select a point, points, or path
that provides a value equivalent to the mean. The sample location should
be within the inner 50 percent area of the cross section.
For circular ducts, conduct a 12-point traverse (i.e., six points
on each of the two perpendicular diameters) locating the sample points as
described in 40 CFR 60, Appendix A, Method 1. Perform the measurements
-------
and calculations as described above, and determine a point, points, or
path that provides a value equivalent to the mean. The sample location
should be within the inner 50 percent area of the cross section.
The performance specifications specify the location of the traverse
points for reference method tests for RA. It is not necessary that the
CEMS measurement location and the reference method traverse location
be the same. For example, the CEMS may be located in the exhaust duct
of a control device for accessibility purposes, but ports for reference
method tests are not available or are inadequate in terms of the
requirements in the performance specifications. In this case, the RA
tests would be conducted farther downstream of the control device,
probably in the stack, where ports are available and meet the requirements
of the performance specifications. The results of the CEMS would be
compared with the results of the reference method tests to determine RA.
4.2 Response Time. Determination of response time is important
only when a short averaging time is specified in the applicable regulation
or when time sharing makes quick response a necessity. The response time
procedure provides a check of the effect of the sample interface on the
calibration of the CEMS's. For this reason, it is necessary to check
the entire sample transport line (if applicable), sample conditioning
equipment (if applicable), the gas analyzer, and the data recorder. The
procedure is as follows:
Introduce zero gas (or zero cell or filter into the CEMS). For
extractive systems, introduce the specified calibration gas at the
sample probe. If a gas analyzer is used to monitor more than one
10
-------
source, perform the response time test for each sample interface.
For in-situ systems, introduce the specified calibration gas at the
sample interface (if applicable), or introduce a calibration gas cell
or filter at an appropriate location in the gas analyzer. The
appropriate location is one that includes all the components active
in the measurement. This includes sample and reference measurement
cells, electronics, lenses, and reflective surfaces. In some cases,
substitute reflective devices are used to facilitate this use of
calibration gases or cells.
When the CEMS output has stabilized, switch to monitor the
stack effluent, and wait until a "stable value" is reached. A "stable
value" is a CEMS response that shows a change of less than 1 percent
of span value for 30 seconds. Record the upscale response time.
Introduce a high-level concentration gas or cell to the CEMS and
repeat the above steps recording the downscale response time. Repeat
the entire procedure three times. Record the results of each test
on a data sheet similar to that shown in Figure 1. Determine the
mean values of the upscale and downscale response times. The slower
or longer time is the CEMS response time.
4.3 Zero and Calibration Drift Measurements. The purpose of
the drift checks is to determine the ability of the CEMS to maintain
its calibration over a specified period of time. The performance
specifications require a 24-hour drift determination and set the limit
of acceptability for seven consecutive checks.
11
-------
Date
High-level =
*
Test Run
1
2
3
Average
Upscale,
min
A =
Downscale,
min
B =
System Response Time (slower of A and B) = min.
Figure 1. Response time.
-------
Upon installation, the CEMS usually requires a break-in period
during which the operating parameters can be adjusted to meet the
conditions of the particular test site. The CEMS drift determinations
during this break-in period often show excessive drift in the CEMS
because of the frequent adjustments and the conditioning of the equipment.
It is recommended that a CEMS conditioning period be undertaken prior to
the performance specification test (PST) to allow for the initial
adjustments and conditioning of the CEMS to be completed and provide greater
assurance that the performance specifications will be met. This conditioning
period should be 1 week or longer of CEMS operation.
Drift checks and calibrations (Section 5) with calibration gases or
cells should be conducted so that all components active in the measurement
of the gas concentration are included. For CEMS's using calibration gases,
all sample conditioning equipment, including external filters and moisture
traps, should be included in the measurement check. If calibration cells
or filters are used, the cell or filter should be inserted in a
measurement path that is the same as that used for stack gas measurement
and that includes all the components of the measurement path, e.g.,
lenses, optical filters, mirrors, beam splitters, etc. The measurement
signal should be a result of the same electronics involved in measuring
the emission concentration. That should include special compensation
components, such as for pressure and temperature.
In addition to the 24-hour drift check required by the performance
specifications, 2-hour zero and span drift checks are suggested. The
zero drift and calibration drift procedures are as follows:
13
-------
4.3.1 Two-hour Drift. Introduce consecutively zero gas (or
gas cell or filter) and high-level calibration gas-(or gas cell or
filter) at 2-hour intervals for a total of seven 2-hour periods.
Determine and record the amount the CEMS output differs from the
calibration reference value at the end of each 2-hour period (Example
data sheet is shown in Figure 2). Conduct this test before making
any zero or calibration adjustments to the CEMS during the test period.
The 2-hour periods over which the measurements are conducted need not
be consecutive, but must not overlap. Calculate the differences between
the CEMS responses and the certified gas or cell value. The difference
must not exceed the applicable levels listed in Tables 1, 2, and 3.
4.3.2 Twenty-four Hour Drift. At 24-hour intervals, introduce
consecutively zero gas (or gas cell or filter) and high-level calibration
gas (or gas cell or filter); make no adjustments to the CEMS during the
test period. Record the amount that the CEMS output differs from the
calibration reference value at the end of each 24-hour period (an example
data sheet is shown in Figure 2).
No CEMS response for a calibration value should differ from the
calibration reference value by more than 2.5 percent of span value for
S02 and NOX analyzers or 0.5 percent 02 or C02 for diluent analyzers.
For TRS monitors, at least six of the seven consecutive tests for a
calibration value should not differ from the calibration reference
value by more than 5 percent of span value.
4.4 Relative Accuracy Test. The relative accuracy test (RAT)
is the primary check of the operation of the CEMS. The response of the
14
-------
Test
no.
Date
Time
begin
end
Zero input
reading
Arithmetic mean (Eq. 1)
Drift (Eq. 4)
difference
High-level input
reading
difference
Figure 2. Zero and calibration drift (2-hour and 24-hour)
15
-------
CEMS to the stack gas emission levels is compared with reference method
measurements that are representative of the total emissions from the source.
To pass the RAT, the CEMS response must agree with the reference measurements
to within 20 percent using the sum of the mean of the differences plus the
confidence coefficient divided by the mean of the reference method
values. Because of some variability inherent in the results of
reference method tests and of CEMS responses, the confidence
coefficient usually accounts for about half of the calculated RA
value. Therefore, the limit of direct comparison between reference
method results and CEMS responses is about +10 percent.
The respective performance specifications provide detailed information
on the number of test points, the number of test runs, the sampling
and analytical methods, and the calculation procedures required for
the RAT. The most important factor in performing and passing
the RAT is quality assurance — both for the CEMS and the reference
method tests. It is strongly suggested that audits be included
in the testing program and that experienced, qualified personnel conduct
the RAT's.
There are a few suggestions for conducting the RAT that may
facilitate the testing and reporting. First, note that the location
of the ports for the RAT need not be the same as for the CEMS, but
must be such a location that is representative of the total process
emissions. An exhaust stack location is most likely to meet this
criterion. Should no stack location be available or should control
equipment inlet testing be necessary, schemes for sampling multiple
16
-------
ducts might be required. In such a case, there must be coordination
between test crews for simultaneous testing.
Another suggestion that could facilitate the satisfactory completion
of the RAT regards the allowance in the performance specifications that
more than nine test runs may be collected with the option to reject up
to three sets of test results, so long as at least nine test results are
used to determine the RA. It is suggested that more than nine test
runs (up to 12) be conducted for each CEMS RAT, so that this option
is available should one or more of the test runs be contaminated or
otherwise determined to be unacceptable following the test period. All
measurement data must be reported in the PST report.
Orsat data collected in the field can be checked to determine
whether the ratios of diluent concentration corresponds to the
stoichiometric combustion products ratio for the fuel being used.
Citation 7.8 details this procedure and acceptable limits for these
comparisons.
5. Continuous Emission Monitoring System Calibration
In preparation of the CEMS for undergoing the PST's, proper
calibration of the analyzers is one of the most important functions.
Manufacturers have procedures specific for their instruments and usually
supply a manual with a calibration curve or chart prepared for the
analyzer by the manufacturer in the home laboratory. Many purchase
agreements include a provision for an on-site calibration to be performed
by the manufacturer upon installation and startup.
The two purposes of performing calibrations on a CEMS are: (1) to
resolve whether the CEMS can accurately determine the concentration of
17
-------
a known source, and (2) to determine whether the CEMS response over
the range of measurements is predictable, e.g., linear. The wide range
of instrument types and calibration procedures precludes including
specific procedures for performing these calibrations in these guidelines.
In general, a known concentration source (e.g., compressed gas, gas cell,
or optical filter) is introduced to the analyzer, and the analyzer response
is compared with the known value.
If a CEMS response curve is supposed to be linear, this may be
checked by introducing a zero or low-level gas followed by a high-level
gas (see Section 5.1) and preparing a linear calibration curve between
the two responses. Then, a mid-level gas is introduced to the analyzer,
and the analyzer response is compared to the calibration curve. The
calibration curve should predict the known mid-level concentration to
within +5 percent of span value (or 0.5 percent Og or C02 for diluent
analyzers).
If a CEMS is designed so that the response curve is nonlinear, an
alternative calibration error test must be used. The CEMS manufacturer
may recommend a procedure for determining the response curve. At a
minimum, three calibration points (low-level, mid-level, and high-level
values), and a zero value should be used in establishing the calibration
curve. Introduce the calibration gases and the zero gas one at a time,
and record the responses. Prepare a calibration curve from the responses
according to the manufacturer's recommendations. Introduce a fourth
calibration gas with a known value between the mid-level and the
high-level values, and record the response. The CEMS response for this
gas as determined from the calibration curve should agree with the
certified gas value to within +5 percent of span value.
18
-------
The manufacturer's operating manual should provide at a minimum
a three-point calibration check or a suitable alternative for
determining the CEMS response to known concentrations. The recommended
calibration levels and concentration verification procedures are as
follows:
5.1 Calibration Levels. A minimum of three calibration
concentration levels are recommended to perform the checks on the
CEMS. These concentrations are defined below.
5.1.1 High-level. A level that is equivalent to 80 to 90
percent of the span value.
5.1.2 Mid-Level. A level that is equivalent to 45 to 55 percent
of the span value.
5.1.3 Zero Level. A level that is equivalent to less than 0.25
percent of the span value.
5.2 Gas Cylinder Analysis. There are several, equally acceptable
alternatives for checking or obtaining known concentrations of calibration
cylinder gases. The principle behind each procedure is to correlate
the gas concentration in the cylinder with a known reference value. The
reference value can be based on the measurement by a recognized materials
and measurements authority, such as the National Bureau of Standards, or
can be measured more directly with established reference procedures,
usually wet chemical methods. The alternative procedures are described
below:
5.2.1 Manufacturer-Certified Cylinder Gases. These are calibration
gases prepared according to the protocol defined in Citation 7.9.
19
-------
5.2.2 Tagged Cylinder Gases. Tagged cylinders are those supplied
by the vendor with a label indicating the vendor's measurement of the
gas concentration, but may not have been performed in the manner specified
in the protocol. The concentration of these cylinder gases may be
checked by performing reference method tests on the cylinder gases.
Procedures for sampling and analyzing these calibration cylinder gases
are described in Citations 7.1 and 7.10. Triplicate analyses are
required for each calibration gas with each analysis in a set within
10 percent (or 15 ppm for S0« and NO , whichever is greater) of the
£ /\
average. If this criterion is not met, the measurements should be
discarded and the analyses repeated. If the average of the triplicate
reference method tests is within 5 percent of the vendor's tag value,
use either the measurement average or the tag value as the correct
concentration. If the triplicate average is not within 5 percent of
the tag value, use a measured average for the cylinder concentration
value. (An example data sheet is shown in Figure 3.)
5.3 Calibration Gas Cell or Filter. Cells or filters used to
calibrate GEMS's are usually short-path cells of high concentration gases
designed to simulate concentrations of a long-path, stack measurement.
20
-------
Date
Cylinder Tag Value(s)
Reference Method Used
Analyzer
_(S02, NOX, TRS)
Sample run
1
2
3
4
5
6
Average
Maximum % Deviation
Mid- level
ppm
High-level
ppm
Not necessary if the protocol in Citation 7.a is used to prepare
the cylinder.
Must be <_ +_ 10 percent of the average of all runs.
Figure 3. Analysis of calibration gases .
21
-------
Determining the pollutant concentration in the gas cell is difficult
and would serve little purpose in the calibration of the.cell. More
directly, what must be demonstrated is that the cell produces a
known analyzer response and that the response is within the range
desired.
This can be accomplished by comparing the analyzer response
produced by the calibration cell with a known stack concentration.
The known stack concentration can be produced in a laboratory source
simulator using certified calibration gases as described in Section
5.1 or by direct reference method measurements. The effects of
temperature, and to some extent, pressure, can be quantified as well,
with this type of calibration check.
Another approach is to compare the CEMS response to calibration
cells after the CEMS is installed on the stack and its accuracy has
been verified with the reference method test measurement of the stack
gases. This approach automatically accounts for stack gas
interferences and temperature conditions, but does not provide certified
calibration values for pretest calibrations.
22
-------
6. Equations
6.1 Arithmetic Mean. Calculate the mean of the differences of
a data set as follows:
_ 1
n
d=-^ E d.
n ._, i
(Eq. 1)
Where:
d = Difference between the measured value and a reference
value.
n = Number of data points.
z d.. = Algebraic sum of the individual differences, d^.
6.2 Standard Deviation. Calculate the standard deviation,
S., as follows:
-rV2
n-1
(Eq. 2)
6.3 Confidence Coefficient. Calculate the 2.5 percent error
confidence coefficient (one-sided), CC, as follows:
CC = t
.975
(Eq. 3)
Where:
t g75 = t-values (see Table 4).
23
-------
Table 4; t-Values
na
2.
3
4
5
6
t.975
12.706
4.303
3.182
2.778
2.571
na
7
8
9
10
11
t.975
2.447
2.365
2.306
2.262
2.228
na
12
13
14
15
16
t.975
2.201
2.179
2.160
2.145
2.131
a The values in this table are already corrected for n-1 degrees of
freedom. Use n equal to the number of individual values.
6.4 Relative Accuracy. The RA is a measure of the difference between
the mean value or expected value and the desired value. Calculate the
RA as follows:
RA = I?IR* ICCI x 100 (Eq. 4)
Where:
RV = Reference value, as defined in Sections 4.3.1, 4.3.2, and 4.4.
24
-------
7. References
7.1 Curtis, Foston. A Method for Analyzing NO Cylinder Gases -
/\
Specific Ion Electrode Procedure. Source Evaluation Society Newsletter.
4(1). February 1979.
7.2 Department of Commerce. Experimental Statistics.
Handbook 91. 1963. p. 3-31, paragraphs 3-3.1.4.
7.3 Jaye, F.C. Monitoring Instrumentation for the Measurement
of Sulfur Dioxide in Stationary Source Emissions. U.S. Environmental
Protection Agency. Office of Research and Monitoring. Washington, D.C.
Publication No. EPA-650/2-73-163. February 1973.
7.4 Nader, J.S., F. Jaye, and W. Conner. Performance Specifications
for Stationary-Source Monitoring Systems for Gases and Visible Emissions.
U.S. Environmental Protection Agency. Research Triangle Park, N.C.
27711. Publication No. EPA-650/2-74-013. January 1974.
7.5 Parts, L.P., P.L. Sherman, and A.D. Snyder. Instrumentations
for the Determination of Nitrogen Oxides Content of Stationary Source
Emissions. Volume I. U.S. Environmental Protection Agency. Research
Triangle Park, N.C. 27711. Publication No. APTD-0847. October 1971.
7.6 Parts, L.P., P.L. Sherman, A.D. Snyder. Instrumentations for
the Determination of Nitrogen Oxides Content of Stationary Source
Emissions. Volume II. U.S. Environmental Protection Agency. Research
Triangle Park, N.C..27711. Publication No. APTD-0942. January 1972.
7.7 Shigehara, R.T. Sampling Location of Gaseous Pollutant
Monitoring in Coal-fired Power Plants. Source Evalution Society
Newsletter. 1(2). May 1979.
25
-------
7.8 Shigehara, R.T., R.M. Neulicht, and W.S. Smith. Validating
Orsat Analysis Data from Fossil-Fuel-Fired Units. Stack Sampling
Technical Information - A Collection of Monographs and Papers.
Volume I. Publication No. EPA-450/2-78-042a. October 1978. p. 44-55.
7.9 Von Lehmden, D. Traceability for Establishing True
Concentrations of Gases Used for Calibration and Audits of Continuous
Source Emission Monitors (Protocol No. 1). June 15, 1978. Available
from U.S. Environmental Protection Agency, Environmental Monitoring
Systems Laboratory, Research Triangle Park, N.C. 27711.
7.10 Westlin, P.R. and J.W. Brown. Methods for Collecting and
Analyzing Gas Cylinder Samples. Source Evaluation Society Newsletter.
_3(3). September 1978.
7.11 Blosser, Russell 0. Observation of Field Performance of
TRS Monitors on a Kraft Recovery Furnace. Technical Bulletin No. 91
by the National Council of the Paper Industry for Air and Stream
Improvement, Inc. New York, N.Y. January 1978.
26
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2. •
EPA-450/3-82-026
4. TITLE AND SUBTITLE
Gaseous Continuous Emission Monitoring Systems
Performance Specification Guidelines - SO-, NO ,CO_,0
and TRS X
7. AUTHOR(S)
Emission Standards and Engineering Division
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Emission Measurement Branch (MD-19)
Emission Standards and Engineering Division
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
12. SPONSORING AGENCY NAME AND ADDRESS
DAA for Air Quality Planning and Standards (MD-10)
Office of Air, Noise, and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
December 1982
, 6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
*
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
1
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This document serves as a guideline for vendors, purchasers, and operators of
gaseous continuous emission monitoring systems and offers suggestions for equipment
performance and specifications, test procedures, and data reduction procedures for
evaluating the capabilities of these systems.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS b.lOENTIFI
18. DISTRIBUTION STATEMENT 19. SECURI
_ , „ , . . , Uncl
release uiu.imxcea 20. SECURI
Uncl
ERS/OPEN ENDED TERMS C. COSATI Field/Group
13B
TV CLASS (This Report) 21. NO. OF PAGES
assilled 29
TY CLASS .-This page) ::. =>RICE
assified
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION i s OBSOLETE
------- |