United States
Environmental Protection
Agency
Environmental Monitoring Systems
Laboratory
Research Triangle Park NC 27711
EPA/600/4-87-001
December 1986
Research and Development
vvEPA
A Summary of the
1985 EPA National
Performance Audit
Program on Source
Measurements
-------
EPA/600/4-87/001
December 1986
A SUMMARY OF THE 1985 EPA NATIONAL PERFORMANCE AUDIT PROGRAM
ON SOURCE MEASUREMENTS
E. W. Streib, T. J. Logan, and M. R. Midgett
Quality Assurance Division
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
Disclaimer
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
-------
FOREWORD
Measurement and monitoring research efforts are designed to anticipate
potential environmental problems, to support regulatory actions by develop-
ing an in-depth understanding of the nature and processes that impact health
and the environment, to provide innovative means of monitoring compliance
with regulations, and to evaluate the effectiveness of health and environ-
mental protection regulations through the monitoring of long-term trends.
The Environmental Monitoring Systems Laboratory, Research Triangle Park,
North Carolina, has responsibility for: assessment of environmental moni-
toring technology and systems; implementation of Agency-wide quality assur-
ance programs for air pollution measurement systems; and supplying technical
support to other groups in the Environmental Protection Agency, including
the Office of Air and Radiation and the Office of Toxic Substances.
The major task of this study was to report the results of the national
quality assurance audit program for stationary source test methods. Audits
were designed to estimate the minimal analytical and computational accuracy
that can be expected with Method 3 (carbon dioxide and oxygen), EPA Method 5
(dry gas meter only), Method 6 (sulfur dioxide), Method 7 (nitrogen oxides),
and Method 19 (coal). Statistical analysis was used to characterize the
data.
John C. Puzak, Deputy Director
Environmental Monitoring Systems Laboratory
Research Triangle Park, North Carolina
iii
-------
ABSTRACT
In 1985, the Quality Assurance Division conducted the National Audits
for Stationary Source Test Methods. The audit materials consisted of: a
disposable gas cylinder for Method 3 (Orsat analyzer), a calibrated orifice
for Method 5 (dry gas meter only), five simulated liquid samples each for
Method 6 (S02J and Method 7 (NOX), and two coal samples for Method 19A.
Participating laboratories sent their data to the Source Branch and in
return received a written report comparing their results to EPA's.
In the Method 3 audit, each gas component had only one concentration.
The mean C02 value for all participants differed by 4.0 percent from the
expected (EPA) value, and the mean for 02 was 0.4 percent from the expected
value.
In the Method 5 audit, the mean value for all participants differed by
10.6 percent from the expected value. For the Method 6 audit, the average
mean differed by 10 percent from the expected value. The average mean in the
Method 7 audit was 4 percent from the expected value.
In the two coal audits, the parameters measured were sulfur, moisture,
ash, and Btu content. On the average for the sulfur analysis, 87 percent of
the participants measured within 10 percent of the expected value; for Btu,
98 percent of the participants measured within 10 percent of the expected value.
This report includes the results of the performance audits done during
the period from January to December 1985.
iv
-------
CONTENTS
Page
Foreword Ill
Abstract 1 v
Figures vl
Tabl es vll
Acknowledgment vl 11
1. Introduction 1
2. Summary 2
3. Method 3 Audit 6
4. Method 5 Dry Gas Meter Audit 8
5. Method 6 Audit 13
6. Method 7 Audit 16
7. Method 19A Audit 19
References 23
Appendixes
A. Frequency Distributions 24
B. Instructions for EPA Audit Materials 30
-------
FIGURES
Number Page
1 Cumulative accuracy for participants in
Method 5 audit, 0685 10
2 Previous results of Method 5 audit 11
3 Results of Method 5 audit, 0685 12
4 S02 expected values compared to participants'
values 15
5 NOX expected values compared to participants'
values 18
VI
-------
TABLES
Number Page
1 Participants' Results from Method 3 Audit
(All Data—No Outliers Removed) 2
2 Method 5 Audit 0685 - Summary Statistics 3
3 Participants' Results from Methods 6 and 7 Audits
(All Data—No Outliers Removed) 3
4 Participants' Results from Method 6 and 7 Audits
(Outliers Removed) 4
5 Participants' Results from Method 19A Coal Audits
(All Data—No Outliers Removed) 5
6 Method 3 Audit Participants 6
7 Source Method 3 Audit—0785 7
8 Method 5 Audit Participants 8
9 Previous Results of Method 5 Audit 9
10 Method 6 Audit Participants 13
11 Summary of Source SOg Audit—0585 14
12 Method 7 Audit Participants 16
13 Summary of Source NOX Audit—0485 17
14 Coal Audit Participants 19
15 Source Coal Audit—0385 21
16 Source Coal Audit—0985 22
17 National Orsat Audit Frequency Distribution of
Absolute Percent Differences of Expected and Reported
Values—Study 0785 25
18 DGM Frequency Distribution of Absolute Percent
Di fference—0685 26
VI1
-------
TABLES
Number Page
19 S0£ Frequency Distribution of Percent Difference—No
Outliers Removed—Audit 0585 27
20 NOX Frequency Distribution of Percent Difference—No
Outliers Removed—Audit 0485 27
21 National Coal Audit Frequency Distribution of Absolute
Percent Differences of Expected and Reported Values-
Study 0385 28
22 National Coal Audit Frequency Distribution of Absolute
Percent Differences of Expected and Reported Values-
Study 0985 29
-------
ACKNOWLEDGMENT
We express our appreciation to the laboratories that participated in
the National Performance Audit Program for Stationary Sources. Thanks also
to the staff of the Standards Laboratory/Performance Evaluation Branch/
Environmental Monitoring Systems Laboratory (EMSL), who did the acceptance
testing of the audit samples, and to the programmers of the Monitoring and
Assessment Division/EMSL for providing the data systems to store and evaluate
the data.
-------
SECTION 1
INTRODUCTION
The Environmental Protection Agency's (EPA's) Environmental Monitoring
Systems Laboratory (EMSL) at Research Triangle Park, North Carolina, estab-
lished an audit program in 1977 to evaluate the performance of companies
that conduct compliance testing using EPA reference methods. The audits
check the participants' analytical accuracy in applying the analytical
phase of EPA Reference Methods 3. 6, 7, and 19A and the calibration accuracy
of the Method 5 control console.(1) Accuracy is defined as the percent
difference between a participant's analytical results and the EPA expected
value. By participating in this free and voluntary program, testing companies
can compare their performance to other laboratories conducting similar
measurements.
Source Test Methods 3, 5, 6, and 7 were each audited once and Method 19A
twice in 1985. Each participating laboratory received an audit package
consisting of the audit sample, a data card, instructions, and an envelope
for returning the data to EPA. A label for returning the audit device was
included with the Method 5 audit package. Participants had 8 weeks to use
the audit material and return their data to EPA. At the end of this period,
all data received were statistically analyzed to determine the accuracy
with respect to the EPA expected value and those obtained by the participants
(see Appendix A).
The Quality Assurance Division of EMSL also maintains a limited
repository of samples for the EPA Methods 3, 6, 7, and 19A that are available
to source testing laboratories for such purposes as training new personnel
or conducting quality control checks. Because the expected values for
these samples are included with the analysis instructions, there is no
requirement for the data to be returned to EPA. We recommend that source
testing laboratories use this sample repository to help improve their overall
analytical performance.
This report summarizes the results obtained in the 1985 source audits.
-------
SECTION 2
SUMMARY
In 1985, EPA's EMSL at Research Triangle Park, North Carolina, conducted
National Quality Assurance Audits for Stationary Source Test Methods 3
(Orsat analyzer), 5 (dry gas meter only), 6 ($02), 7 (NOX), and 19A (coal).
Industrial laboratories, contractors, foreign laboratories, as well as
local, State, and Federal agencies participated.
The results of the 1985 audit of Method 3 are summarized in Table 1.
Participants analyzed the gas sample twice for percentages of carbon dioxide
(C02), oxygen (03), and carbon monoxide (CO). The mean values of C02 and
62 differed by 4 percent and 0.4 percent from the expected values, respectively.
In contrast, the mean values for CO differed as much as 63 percent from the
expected values.
TABLE 1. PARTICIPANTS1 RESULTS FROM METHOD 3 AUDIT
(ALL DATA—NO OUTLIERS REMOVED)
Type of Audit
sample date
Small 0785
cylinder
(gas)
Parameter
% C02
% 02
% CO
No. of
analyses
46
45
47
46
37
36
Repli-
cate
1
2
1
2
1
2
EPA
true
value
7.00
7.00
12.00
12.00
0.60
0.60
Participant results
Mean
6.76
6.68
11.98
12.07
0.96
1.00
Median
6.60
6.60
12.00
12.00
0.50
0.50
Std.dev.
0.97
0.88
1.38
1.33
2.92
2.97
One audit of Method 5 was conducted in 1985. The overall results (no
outliers removed) are summarized in Table 2. The mean for all participants
was 10.6 percent from the expected value and the standard deviation was
102.1 percent. After the removal of 7 percent of the data statistical outliers,
the mean was 2.3 percent from the expected values and the standard deviation
was 1.7 percent. The participants' performance based on the standard deviation
of all data was poor; however, when the 7 percent of the outliers was removed,
the performance was consistent with previous years.
-------
TABLE 2. METHOD 5 AUDIT 0685 - SUMMARY STATISTICS
Mean
Median
Std. dev
All data
Outliers removed
633
588
10.6
2.3
2.2
2.0
102.1
1.7
Table 3 represents the data (no outliers removed) from the Methods 6
and 7 audits. In the Method 6 audit, the procedure requires the partici-
pants to determine the sulfate content in five aqueous solutions using the
titration procedure. For four out of five concentrations, the mean of the
participants was 10 percent higher than the expected value. In contrast, the
median differed by 2 percent. In the Method 6 audit, 60 to 85 percent of
the participants achieved an accuracy within 5 percent of the expected value.
The Method 7 audit procedure requires that the participants determine
the nitrate content in five aqueous solutions. For each concentration, the
mean of the participants was less than 4 percent from the expected value,
and the median differed by less than 2 percent. In this audit, 70 percent
of the participants achieved an accuracy within 10 percent of the expected
value for all five concentrations.
TABLE 3. PARTICIPANTS' RESULTS FROM METHODS 6 AND 7 AUDITS
(ALL DATA—NO OUTLIERS REMOVED)
Type of Audit
sample date
Aqueous 0585
sulfate
Aqueous 0485
nitrate
No. of
analyses
89
91
90
91
89
72
73
73
72
73
EPA true
value
(mg/DSCM)
198.3
221.2
701.6
1250.7
1326.9
99.5
119.5
298.6
338.4
537.5
Participant results
Mean
243.8
245.5
723.7
1332.6
1426.1
97.9
118.3
293.8
344.8
518.4
Median
201.5
225.0
698.3
1240.1
1323.1
101.1
119.0
298.0
339.4
534.0
Std.
dev.
263.8
104.5
247.9
547.5
595.3
24.4
23.0
49.5
73.8
112.9
-------
Table 4 represents the data from Methods 6 and 7 after the removal of
outliers. Two percent of the Method 6 data and 3 percent of the Method 7
data were classified as outliers. It is important to note that even with
the statistical outliers removed, the performance of this year's Method 6
data (based on the standard deviation) is not as good as in previous years.
TABLE 4. PARTICIPANTS' RESULTS FROM METHOD 6 AND 7 AUDITS
(OUTLIERS REMOVED)
Type of
Sample
Audit
Date
No. of
Analyses
EPA true value
(mg/DSCM)
Participant
Mean
Median
results
Std. dev
Aqueous
sulfate
0585
88
88
89
89
87
198
221
701
1250
1326
,3
.2
,6
.7
.9
217.6
227.6
699
1252
1338
201.3
224.6
697.8
1240.0
1322.2
91.9
25.2
95.9
103.6
117.0
Aqueous 0485
nitrate
69
71
71
70
71
99.5
119.5
298.6
338.4
537.5
101.5
120.8
299.4
342.9
530.9
101.9
119.1
298.5
339.4
534.0
17.5
17.7
36.3
49.5
84.8
Table 5 summarizes the results of the two coal audits that were con-
ducted in 1985. Participants analyzed each coal sample in duplicate for
percentages of sulfur, moisture, and ash, and for gross calorific value
(Btu/lb). The means of the ash, moisture, and sulfur content were within
6 percent of the expected value except for a 20 percent difference on the
lowest sulfur concentration. An accuracy of 1 percent was achieved on the
Btu content. The Btu content as measured by the participants is lower than
the EPA value because coal oxidation is not prevented in the audit sample.
The Btu content is a good example of the accuracy to which a measurement can
be made when that measurement receives good laboratory quality control.
-------
TABLE 5. PARTICIPANTS' RESULTS FROM METHOD 19A COAL AUDITS
(ALL DATA—NO OUTLIERS REMOVED)
Type of Audit
sample date
Coal 0385
0985
0385
0985
0385
0985
0385
0985
0385
0985
0385
0985
0385
0985
0385
0985
No. of
Parameter analyses
% S 111
108
110
106
% H20 111
106
112
108
% Ash 111
107
110
105
Btu/lb 108 11
102 12
109 13
104 12
EPA
value
1.62
0.72
4.97
2.35
3.86
1.61
3.93
17.45
11.36
4.33
20.32
17.92
,414.0
,300.0
,054.0
,374.0
Participant results
Mean
1.60
0.86
4.75
2.29
3.60
1.75
3.97
16.73
11.37
4.25
19.92
17.59
11,379.7
12,252.9
13,011.1
12,179.3
Median
1.59
0.70
4.76
2.29
3.70
1.53
3.95
17.41
11.33
4.37
20.07
17.83
11,383.5
12,273.0
13,025.0
12,319.0
Std.dev.
0.14
1.56
0.35
0.23
0.52
1.86
1.65
2.61
0.45
0.44
1.52
2.34
256.61
183.77
261.53
573.78
-------
SECTION 3
METHOD 3 AUDIT
The Method 3 audit checks participants' ability to analyze a gas sample
using an Orsat analyzer. The audit package consists of a disposable cylinder
that contains a 4-liter (L) sample of COg, Og, and CO. The analyst expels
the gas into the Orsat analyzer using the positive pressure of the cylinder.
The gas sample is quantitatively analyzed for percentage of C02, 02, and CO.
In the 1985 audit, 44 percent of the 103 laboratories receiving the audit
package returned data. Table 6 shows the total number of laboratories
requesting participation and the number that returned data for the Method 3
audit.
TABLE 6. METHOD 3 AUDIT PARTICIPANTS
No. receiving No. returning
Category samples data
Contractors
Industry
Foreign
Federal
State
Local
57
28
1
2
12
3
16
15
1
2
8
3
Total 103 45
Table 7 summarizes the Method 3 audit results. Each laboratory was
asked to analyze the sample in duplicate. Five and ten percent accuracy
were chosen for the reporting criteria for each of the parameters. Each
parameter had only one concentration.
In the 1985 audit, 35 percent of the reporting laboratories achieved
an accuracy within 5 percent for the C02, which was a decrease from the
last audit.'4) Ninety percent of the laboratories achieved an accuracy within
-------
5 percent of the expected value for the 02 analysis. For the CO analysis,
only 33 percent of the laboratories achieved an accuracy within 5 percent,
and 21 percent of the laboratories did not report a value for CO.
TABLE 7. SOURCE METHOD 3 AUDIT--0785
Expected
value
7.00
12.00
0.60
No. of
analyses*
(1) 46
(2) 45
(1) 47
(2) 46
(1) 37
(2) 36
Laboratories
accurate within 5%
(%)
35
30
89
91
32
36
Laboratories
accurate within 10%
(%)
85
89
92
94
32
36
* Numbers in parentheses indicate first and second analyses.
-------
SECTION 4
METHOD 5 DRY GAS METER AUDIT
In the Method 5 audit procedure, participants use a calibrated orifice
to check the calibration of the dry gas meter in their EPA Method 5 control
console (meter box). They insert the orifice in the Method 5 meter box,
allow the box to warm up, and then make three 15-min volume measurements.
The participants convert each of the three volumes to cubic meters at
standard conditions using the formula specified in Eq.# 5.1 of Method 5
(Appendix A, 40 CFR 60) and record them on the data card. Then they return
the orifice and the data card to EPA, where the data undergoes statistical
analysis.
In the 1985 audit, 59 percent of the 160 laboratories that received the
audit package returned data. Table 8 shows the categories of participants,
the number of participants who requested participation in the Method 5 audit,
and the number who actually returned data.
TABLE 8. METHOD 5 AUDIT PARTICIPANTS
No. receiving No. returning
Category samples data
Contractors
Industry
Foreign
Federal
State
Local
81
51
4
2
18
4
37
35
2
2
14
4
Total 160 94
Figure 1, a cumulative histogram, shows the accuracy obtained by
participants in the Method 5 audit, expressed as the percent difference
from the expected (EPA) value at various levels of accuracy. The Code of
Federal Regulations^) requires that the dry gas meter be calibrated within
an accuracy of 2 percent. Figure 2 shows that 47 percent of the reporting
laboratories attained this accuracy. These results are not as good as
those from 1984. Table 9 is a summary of previous audits that, like
Figure 2, shows a decline from 1984.
-------
TABLE 9. PREVIOUS RESULTS OF METHOD 5 AUDIT
Audit
0382
0982
0383
0983
0684
0685
Number of
analyses
827
769
763
614
631
633
Mean Median
(% from EPA
7.6
12.5
5.7
4.1
3.1
10.6
value)
2.5
2.5
2.2
1.9
1.9
2.2
Std. dev.
39.5
81.4
32.7
21.3
4.9
102.1
The histogram in Figure 3 depicts the individual results from all
participants of the 1985 audit with the mean and median values. The majority
of the laboratories reported values lower than the EPA value.
The standard deviation of the triplicate analysis (repeatability) by
each laboratory indicated that 71 percent of the standard deviations for
each set were within 0.3 percent. Four percent of the 1985 data were identi-
fied as outliers using Chauvenet's Criterion.(5) Before the outliers were
removed, the mean value (absolute) differed by 10.6 percent from the expected
value. After deletion of outliers, this value was reduced to 2.3 percent.
-------
LABORATORIES ACHIEVING SPECIFIED ACCURACY, percent
N>
O
U
O
(II
O
O)
o
Nj
o
00
o
(O
o
o
o
(Q
C
-^
CD
3
II
O)
w
w
o
c
3
c_
D)
31
m
Z
O
m
O
CD
0)
O
O
C
3
o
•a
to
o
•6'
Q>
3
W
3
CO
m
CO
c
3)
m
m
Z
CO
Tl
m
•o
a>
o
a.
01
0)
c
a.
rn
•o
CD
00
O
en
oo
en
(O
-------
55
50
CO
ill
I"
oc
o
CO
40
35
DEGREE OF ACCURACY = 2%
0382/827 0982/769 0383/763 0983/614 0684/631 0685/633
AUDIT / NUMBER OF SAMPLES
Figure 2. Previous results of Method 5 audit.
11
-------
20
15
V
UJ
c 10
oc
o
CD
NUMBER OF VOLUMES = 633
MEDIAN = -1.0
OUTLIERS
REMOVED
Mean = 0.7
OUTLIERS
REMOVED
<-12 -10 -8-6-4-20246
DIFFERENCE FROM EPA VALUE, percent
8 10 >10
Figure 3. Results of method 5 audit, 0685.
-------
SECTION 5
METHOD 6 AUDIT
The Method 6 audit checks participants' ability to quantitatively analyze
Method 6 samples for sulfur content. The audit set consists of five aqueous
dilutions of 10 N sulfuric acid in 25 milliliters (ml) sealed glass ampoules.
The analyst withdraws 5 ml from each ampoule, adds 30 ml of 3 percent hydrogen
peroxide, and dilutes this sample to 100 ml with distilled water. A 20-mL
aliquot is then withdrawn from the diluted sample, 80 ml of 100 percent
isopropanol and thorin indicator are added, and the sample is titrated with
barium perchlorate to a pink endpoint. In calculating the results, the
participants assume an original sample volume of 100 mL and a sample volume
of 0.021 dry standard cubic meter of stack gas.
Table 10 shows the categories of the participants and compares the
total number of participants requesting participation with the number return-
ing data. In the 1985 audit, 66 percent of the 137 laboratories that received
the audit package returned data.
TABLE 10. METHOD 6 AUDIT PARTICIPANTS
Category
No. receiving
samples
No. returning
data
Contractors
Industry
Foreign
Federal
State
Local
74
40
3
1
13
6
42
30
2
1
11
5
Total
137
91
Table 11 shows the percentage of laboratories that achieved 2 percent
and 5 percent accuracy for each of the five different concentrations in the
Method 6 audit. At least 62 and 67 percent of the reporting laboratories
achieved an accuracy within 5 percent for the two lower concentrations, but
83 percent of the laboratories showed an improvement on the higher concentra-
tions. Five percent is used as the criterion because it was established as
the criterion for source SOg compliance audit samples.
13
-------
TABLE 11. SUMMARY OF SOURCE S02 AUDITS—0585
Concentration ±2% ±5%
(mg/DSCM)
198.3
221.2
701.6
1250.7
1326.9
28.1
34.1
58.9
39.6
64.0
61.8
67.0
83.3
87.9
86.5
91
Figure 4 shows the mean values for each S0£ concentration and the
confidence intervals (CI). Each CI is calculated using the 95 percent
confidence level. The CI for the lowest concentration is ± 22.5 percent;
whereas the CI for the highest concentration is ± 8.7 percent. Overall,
the CI's for the SOg were higher than expected for this method because of
the high standard deviations.
14
-------
PARTICIPANTS'VALUES. mg/DSCM
198.3 221.2 701.6 1250.7 1326.9
EXPECTED VALUES, mg/DSCM
Figure 4. SC>2 expected values compared to participants' values.
* 00 N> 0
O O O C
0 0 O C
II I I I I I I I I I II
LLJ
II I I I I I I I I I I
-------
SECTION 6
METHOD 7 AUDIT
The Method 7 audit checks participants' ability to quantitatively
analyze Method 7 samples for nitrate content. The NOX audit set consists
of five aqueous dilutions of a potassium nitrate solution in 2b-mL glass
ampoules that are autoclaved after sealing so that bacteria that
might attack the nitrate are destroyed. The analyst withdraws 5 ml of
solution from an ampoule, adds this with 25 ml of Method 7 absorbing
solution to a flask, adjusts the pH with sodium hydroxide, and dilutes to
50 ml with distilled water. A 25-mL aliquot is withdrawn from the diluted
sample, placed in an evaporating dish, and analyzed as described in
Section 4.3 of Method 7.U) After this treatment is completed, the
absorbance is measured at 410 nanometers (nm) with a calibrated spectro-
photometer. In calculating the concentrations present, the participant
assumes that 2000 ml of stack gas was sampled.
Table 12 shows the total number of laboratories requesting participa-
tion and the number that returned data for the 1985 Method 7 audit. Sixty-
two percent of the 117 laboratories receiving the audit package returned
data.
TABLE 12. METHOD 7 AUDIT PARTICIPANTS
No. receiving No. returning
Category samples data
Contractors 67 36
Industry 36 25
Foreign 2 2
Federal 1 0
State 6 6
Local 5 4
Total 117 73
16
-------
The percentage of laboratories that achieved 5 and 10 percent accuracy
for each of the five concentrations is shown in Table 13. Ten percent is
used as the criterion because it was established as the criterion for the
source NOX compliance audit samples. Sixty percent of the reporting
laboratories achieved an accuracy within 10 percent on the lowest concentra-
tion, and 70 to 80 percent achieved an accuracy within 10 percent on the
highest concentrations.
TABLE 13. SUMMARY OF SOURCE NOX AUDIT—0485
Concentration
(mg/DSCM)
99.5
119.5
298.6
338.4
537.5
± 5%
(*)
41.7
49.3
54.8
51.4
53.4
± 10%
(%}
58.3
69.9
80.8
72.2
71.2
73
Figure 5 shows the means for each NOX concentration and the CI's. The
CI's are calculated the same tfs for SOg. The CI for the lowest concentration
is ± 5.8 percent, whereas the CI for the highest is 5.0 percent. Overall,
the CI's for the NOX were very low because of low standard deviations.
17
-------
600
5 500
O
(/>
O
N.
O)
Ul
D
(A
O
DC
400
300
200
100
- E
99.5
537.5
119.5 298.6 338.4
EXPECTED VALUES, mg/DSCM
Figure 5. NOX expected values compared to participants' values
18
-------
SECTION 7
METHOD 19A COAL AUDIT
Standards of performance for newer electric utility steam generators
(Subpart Da of 40 CFR 60) allow coal sampling and analysis to serve as an
acceptable method for determining scrubber inlet flue gas sulfur concentra-
tion. The coal audit checks participants' ability to analyze coal samples
for sulfur, ash, moisture, and Btu content.
The coal audit samples consisted of two samples each with 50 grams (g)
of 60-mesh coal but with different parameter levels. The following American
Society for Testing and Materials (ASTM) procedures^6) were recommended, but
not required, for participants1 use in analyzing the coal samples:
• ASTM D-3177 (Standard Test Method for Total Sulfur in the
Analysis of Coal and Coke)
• ASTM D-3174 (Standard Test Method for Ash in the Analysis
Sample of Coal and Coke)
• ASTM D-3173 (Test for Moisture in the Analysis Sample of
Coal)
• ASTM D-2015 (Standard Test Method for Gross Calorific Value
of Solid Fuel by the Adiabatic Bomb Method).
Participants measured the parameters and reported their results for
moisture (%) on an as-received basis, and their results for sulfur (%),
ash (%), and gross calorific value (Btu/lb) on a dry basis.
In both audits, 80 percent of the laboratories that received the audit
package returned data. Seventy-three of the same laboratories participated
in both audits and returned data. Table 14 shows the total number of
laboratories requesting participation and the number that returned data for
coal audits 0385 and 0985.
TABLE 14. COAL AUDIT PARTICIPANTS
No. requesting samples No. returning data
Category 0385 0985 0385 0985
Contractors
Industry
Federal
State
Local
Total
57
42
1
11
4
115
52
37
1
12
5
107
42
37
0
10
3
92
42
33
0
10
5
90
19
-------
Tables 15 and 16 summarize the coal audit results. The number of
analyses is greater than the number of participants because some companies
had more than one laboratory participating. In this case, each laboratory
received its own set of samples and was asked to analyze the samples in
duplicate. Accuracies of 5 and 10 percent were chosen as the reporting
criteria for each of the four parameters (sulfur, moisture, ash, and gross
calorific content).
In the 0385, 90 percent of the laboratories were able to analyze both
sulfur samples within 10 percent of the expected value. In the 0985 audit,
76 percent of the laboratories were able to analyze the sulfur content of
the lower concentration sample within 10 percent of the expected value;
however, 90 percent were able to analyze the high level sample within 10
percent of the expected value. Seventy percent were within the 10 percent
criterion for both moisture concentrations in the 0385 audit, whereas 89
percent in the 0985 audit achieved 10 percent for the higher moisture
concentration, but only 55 percent of the laboratories achieved 10 percent
on the lower concentration. For the ash analysis and Btu content, 87 to
100 percent of the reporting laboratories were able to achieve an accuracy
within 10 percent for both sample concentrations.
Comparing the 0385 audit to the 0985 audit, there was improvement in
the latter audit except for sulfur and moisture parameters. In the 0985,
the sulfur and moisture levels were considerably lower, therefore causing a
greater chance for error.
20
-------
TABLE 15. SOURCE COAL AUDIT—0385
Expected
value
1.
4.
3.
3.
11.
20.
11,
13,
62
97
86
93
36
32
414
054
Laboratories
No. of accurate
analyses* within 5%
(%)
(1)
(2)
(1)
(2)
(1)
(2)
(1)
(2)
(1)
(2)
(1)
(2)
(1)
(2)
(1)
(2)
111
107
110
106
111
104
112
105
111
105
110
104
108
103
109
104
Sulfur
76
78
53
60
Moisture
43
39
40
45
Asji
96
97
93
96
Gross Calorific
98
99
98
98
.6
.5
.6
.4
.2
.4
.2
.7
.4
.1
.6
.2
.1
.0
.2
.1
Laboratories
accurate
within 10%
(%)
93.
95.
88.
85.
70.
68.
72.
74.
97.
97.
99.
99.
99.
99.
99.
99.
7
3
2
8
3
3
3
3
3
1
1
0
1
0
1
0
* Numbers in parentheses indicate first and second analyses.
21
-------
TABLE 16. SOURCE COAL AUDIT—0985
Expected
value
0.72
2.35
1.61
17.45
4.33
17.92
12,300
12,374
No. of
analyses
(1) 108
(2) 102
(1) 106
(2) 100
(1) 106
(2) 100
(1) 108
(2) 102
(1) 107
(2) 101
(1) 105
(2) 99
(1) 102
(2) 97
(1) 104
(2) 99
Laboratories
accurate
within 5%
Sulfur
46.3
50.0
62.3
65.0
Moisture
31.1
35.0
66.7
70.6
Ash
68.2
69.3
94.3
92.9
Gross Calorific
99.0
100
89.4
92.9
Laboratories
accurate
within 10%
77.8
75.5
87.7
93.0
51.9
57.0
88.0
90.2
86.9
86.1
95.2
96.0
100
100
93.3
92.9
Numbers in parentheses indicate first and second analyses,
22
-------
REFERENCES
1. U.S. Environmental Protection Agency. Standards of Performance for
New Stationary Sources—Appendix A. Title 40, Part 60, Code of
Federal Regulations.
2. Streib, E.W., and M.R. Midgett. A Summary of the 1982 EPA National -
Performance Audit Program on Source Measurements. EPA-600/4-83-049,
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina, April 1984.
3. Streib, E.W., T.J. Logan, and M.R. Midgett. A Summary of the 1983
EPA National Performance Audit Program on Source Measurements. EPA-
600/4-85-004, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, January 1985. 38 pp.
4. Streib, E.W., T.J. Logan, and M.R. Midgett. A Summary of the 1984
EPA National Performance Audit Program on Source Measurements. EPA-
600/4-86-005, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, February 1986. 34 pp.
5. Chauvenet, W. Manual of Spherical and Practical Astronomy: Volume II-
Theory and Use of Astronomical Instruments (Method of Least Squares).
J.P. Lippincott and Co., Philadelphia, Pennsylvania, 1863.
6. American Society for Testing and Materials. Annual Book of ASTM
Standards—1984. Volume 5.05. 01-050584-13, American Society for
Testing and Materials, Philadelphia, Pennsylvania.
23
-------
APPENDIX A
FREQUENCY DISTRIBUTIONS
24
-------
en
TABLE 17. NATIONAL ORSAT AUDIT FREQUENCY DISTRIBUTION
OF ABSOLUTE PERCENT DIFFERENCES OF EXPECTED AND REPORTED VALUES—STUDY 0785
C02
Sample
no. No.
5000 91
Min. 10% 20% 30% 40% 50% 60% 70% 80% 90% Max. Mean Std.de v.
0.00 2.86 2.86 4.29 5.71 5.71 7.14 7.14 10.00 11.43 71.43 8.51 10.85
02
Sample
no. No.
5000 93
Min. 10% 20% 30% 40% 50% 60% 70% 80% 90% Max. Mean Std.dev.
0.00 0.00 0.00 0.83 0.83 1.67 1.67 1.67 3.33 4.17 51.67 4.24 10.44
CO
Sample
no. No.
Min. 10% 20% 30% 40% 50% 60% 70% 80% 90% Max. Mean Std.dev.
5000 73 0.00 0.00 0.00 0.00 16.67 16.67 16.67 33.33 50.00 66.67 2950.00 102.06 480.43
-------
TABLE 18. DGM FREQUENCY DISTRIBUTION OF ABSOLUTE PERCENT DIFFERENCE—0685
(S3
CT>
Sample
no.
633
630
617
609
600
591
588
Min.
0.0
0.0
0.0
0.0
0.0
0.0
0.0
10%
0.6
0.6
0.6
0.6
0.6
0.6
0.5
20%
1.0
1.0
1.0
1.0
1.0
1.0
0.9
30%
1.3
1.3
1.3
1.2
1.2
1.2
1.2
40%
1.8
1.7
1.7
1.7
1.7
1.7
1.7
50%
2.2
2.2
2.2
2.1
2.1
2.0
2.0
60%
2.6
2.6
2.5
2.5
2.5
2.4
2.4
70%
3.2
3.2
3.1
3.1
3.0
2.9
2.9
80%
4.1
4.1
4.0
3.9
3.7
3.6
3.6
90%
6.3
6.1
5.7
5.2
5.0
4.8
4.7
Max.
1,562.9
75.2
20.1
11.4
9.6
8.6
7.7
Mean
10.6
3.6
2.8
2.6
2.5
2.4
2.3
Std.dev.
102.1
6.5
2.6
2.1
1.9
1.7
1.7
Skewness
14.34
-0.00
-0.00
-0.00
-0.00
-0.01
-0.01
Median
2.2
2.2
2.2
2.2
2.1
2.0
2.0
-------
TABLE 19. S02 FREQUENCY DISTRIBUTION OF PERCENT DIFFERENCE—NO OUTLIERS REMOVED—AUDIT 0585
Sample no.
2
4
7
8
9
No.
90
89
89
91
91
Min.
0.00
0.00
0.01
0.00
0.00
10%
0.40
0.35
0.35
0.38
0.54
20%
0.77
1.01
0.60
0.73
0.99
30%
1.08
2.12
0.88
0.99
1.81
40%
1.51
2.67
1.12
1.22
2.31
50%
1.62
3.68
1.40
1.66
3.35
60%
2.15
4.69
1.80
1.97
3.48
70%
2.94
5.65
2.55
2.44
5.06
80%
4.22
7.31
3.93
3.17
6.69
90%
11.05
15.33
6.17
5.83
14.69
Max.
309.06
1186.89
304.61
296.58
301.67
Mean
8.18
26.47
9.93
9.38
13.72
Std.dev.
34.50
132.35
44.38
43.25
46.52
ro
TABLE 20. NOX FREQUENCY DISTRIBUTION OF PERCENT DIFFERENCE—NO OUTLIERS REMOVED—AUDIT 0485
Sample no.
1
3
5
6
8
No.
73
72
73
72
73
Min.
0.03
0.10
0.00
0.03
0.00
10%
0.37
1.51
0.65
0.62
0.50
20%
1.51
2.41
1.77
1.63
1.26
30%
2.31
2.61
2.75
2.63
2.09
40%
3.45
4.52
3.48
3.63
4.44
50%
4.39
6.63
4.52
4.43
4.60
60%
5.66
10.05
6.27
6.62
7.95
70%
7.23
14.87
9.08
9.04
9.62
80%
9.21
26.43
15.40
15.48
19.33
90%
20.73
33.97
32.76
27.45
32.13
Max.
79.60
95.18
98.81
117.55
79.33
Mean
9.23
15.01
11.69
11.61
11.58
Std.dev.
13.81
19.43
17.77
18.50
15.31
-------
TABLE 21. NATIONAL COAL AUDIT FREQUENCY DISTRIBUTION
OF ABSOLUTE PERCENT DIFFERENCES OF EXPECTED AND REPORTED VALUES—STUDY 0385
ro
oo
Sample no.
1000
2000
Sample no.
1000
2000
Sample no.
1000
2000
Sample no.
1000
2000
No.
218
216
No.
217
215
No.
216
214
No.
213
211
Min.
0.00
0.00
Min.
0.00
0.00
Min.
0.00
0.00
Min.
0.00
0.00
10%
0.62
0.60
10%
0.76
1.04
10%
0.09
0.25
10%
0.08
0.09
20%
1.23
1.61
20%
1.27
2.07
20%
0.26
0.44
20%
0.18
0.16
30%
1.85
2.21
30%
3.05
3.37
30%
0.35
0.79
30%
0.25
0.25
40%
1.85
3.22
40%
4.58
4.66
40%
0.44
0.94
40%
0.36
0.33
Sulfur
50%
3.09
4.23
Moi
50%
6.62
6.22
50%
0.53 0
1.28 1
Gross
50%
0.47
0.46
60%
3.70
5.23
sture
60%
8.14
8.03
Ash
60%
.79
.37
Calori
60%
0.59
0.63
70%
4.32
6.84
70%
9.41
10.10
70%
1.06
1.87
fie
70%
0.78
0.77
80%
5.56
7.85
80%
11.45
15.28
80%
1.41
2.26
80%
1.03
1.05
90%
8.02
11.47
90%
22.39
22.54
90%
2.64
3.54
90%
1.55
1.63
Max.
66.67
42.25
Max.
417.05
89.12
Max.
24.12
77.36
Max.
17.59
19.98
Mean
4.34
5.76
Mean
13.01
10.05
Mean
1.47
2.32
Mean
0.92
0.92
Std.dev.
6.98
6.22
Std.dev.
40.70
12.30
Std.dev.
3.51
7.46
Std.dev.
1.86
2.07
-------
TABLE 22. NATIONAL COAL AUDIT FREQUENCY DISTRIBUTION
OF ABSOLUTE PERCENT DIFFERENCES OF EXPECTED AND REPORTED VALUES—STUDY 0985
ro
vo
Sulfur
Sample no.
3000
4000
No. Min.
210 0.00
206 0.00
10% 20%
1.39 1.39
0.43 0.85
30%
2.78
1.70
40%
4.17
2.98
50%
5.56
3.83
60%
6.94
4.26
70%
8.33
5.96
80%
11.11
8.09
90%
18.06
9.79
Max
2256.
55.
Mean
94 31.37
74 5.51
Std.dev.
219.04
7.60
Moisture
Sample no.
3000
4000
No. Min.
210 0.11
206 0.00
10% 20%
0.63 1.43
1.24 2.48
30%
2.12
4.35
40%
2.58
6.21
50%
3.15
8.70
60%
4.01
11.18
70%
5.27
14.91
80%
6.76
19.88
90%
10.09
32.30
Max. Mean
99
1009
.08 6.64
.94 25.77
Std.dev.
14.28
106.36
Ash
Sample no.
3000
4000
No. Min.
208 0.00
204 0.00
10% 20%
0.46 1.15
0.22 0.45
30%
1.62
0.61
40%
2.31
0.78
50%
3.00
1.17
60%
3.70
1.45
70%
5.08
2.01
80%
6.93
2.51
90%
14.09
4.13
Max.
83.83
87.67
Mean Std.
5.35 9.
3.82 12.
dev.
10
53
Gross Calorific
Sample no.
3000
4000
No. Min.
203 0.00
199 0.00
10% 20%
0.28 0.45
0.07 0.15
30%
0.61
0.24
40%
0.79
0.37
50%
0.99
0.50
60%
1.12
0.64
70%
1.48
0.89
80%
1.92
1.08
90%
3.51
1.63
Max.
18.52
9.34
Mean Std
2.30 4.
0.80 1.
.dev.
19
04
-------
APPENDIX B
INSTRUCTIONS FOR EPA AUDIT MATERIALS
30
-------
Page 1 of 2
INSTRUCTIONS FOR USING EPA METHOD 3 AUDIT MATERIALS
Equipment Supplied with Audit Kit
(1) Small gas cylinder containing four liters of gas
(2) Nozzle for cylinder (taped on cylinder)
(3) Tygon tubing
Equipment to be Supplied by Participant
(1) Orsat analyzer
(2) Clamp
(3) Extra Tygon or surgical tubing
Procedure
(1) Leak-check apparatus by clamping off tubing.
SEE DIAGRAM.
(2) Fill up burette with gas by using positive pressure. Vent this sample
through the manifold to the atmosphere.
(3) Repeat Step 2.
(4) Fill up burette past fill mark with the gas and carefully vent out
excess to the atmosphere, until the fill mark is reached.
(5) Analyze for C09, 09 and CO as described in Sections 4.2.5, 4.2.6
and 4.2.7 of EPA Method 3.
(6) Record the results on the data card enclosed with the sample.
(7) Repeat Steps 4 through 6.
CAUTION: If the tubing is punctured excess times, leakage can occur.
Replace if necessary.
Send the data card to the address below. (The cylinder gas can should
not be returned.)
Ms. Ellen Streib
Quality Assurance Division (MD-77A)
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
31
-------
Page 2 of 2
Note: Site number will always be 001 except when other Orsat apparatus
or participants are using the same gas sample. The extra
apparatus or participants should be labeled 002, 003, etc.
APPARATUS SET-UP
inn
o
o
NJ
o
32
-------
Page 1 of 2
INSTRUCTIONS FOR USE OF ENVIRONMENTAL PROTECTION AGENCY
METHOD 5 DRY GAS METER PERFORMANCE TEST DEVICE
NOTE: All procedures referred to are from revised Method 5 published in the
Federal Register, Volume 12, Number 160, Part II, Thursday, August 18,
1977, pp. 41776-41782 and references contained therein. This revised
method should be adhered to in all details in the use of this quality
assurance performance device.
EQUIPMENT: The participant in this study should possess the following
equipment, including the performance test device supplied by EPA.
Quantity Item
1 Method 5/Source Sampling Meter Box
1 Stopwatch, preferably calibrated in decimal minutes
1 Thermometer, ambient range
1 Barometer. If unavailable, call nearest National Weather Service and
request the ABSOLUTE barometric pressure. (Corrected for temperature
and acceleration due to gravity, but not corrected for altitude.)
1 Performance Test Device. A calibrated flow orifice housed in a quick-
connect coupling and identified with an engraved three-digit serial
number.
WARNING: THE DEVICE MUST NOT BE DISASSEMBLED UNDER ANY CIRCUMSTANCES,
Use these devices at room temperature.
PROCEDURE:
Calibration of Vacuum Gauge — The vacuum pressure gauge on the
meter box must be calibrated in the range of use (11-22" Hg) against
a standard (Hg Manometer) to ensure accurate results.
Remove the performance test device from its case and insert it into
the gas inlet quick-connect coupling on the source sampling meter box.
Turn the power to the meter box on and start the pump.
Adjust the coarse flow rate control valve and the fine flow rate
control valve to give a reading of 19" Hg (vacuum reading).
CAUTION: The vacuum reading must be accurate and stable for the
test period.
Allow the orifice and source sampling meter box to warm up for 45
minutes with flow controls adjusted as described in Step 3 before
starting quality assurance runs.
33
-------
Page 2 of 2
PROCEDURE: (continued)
Make triplicate quality assurance runs. For each run, record
initial and final dry gas meter volumes, dry gas meter inlet and
outlet temperatures, internal orifice pressure drop (AH), ambient
temperature, and barometric pressure. Run duration should be
slightly greater than 15 minutes. The following procedure is
recommended. Fifteen minutes after a run is started, the partici-
pant watches the dry gas meter needle closely. As the needle reaches
the zero (12 o'clock) position, the pump and stopwatch are stopped
simultaneously. The dry gas meter volume and time are recorded.
This complete run procedure is performed three times to provide the
required triplicate quality assurance runs.
Calculate the corrected dry gas volume for each run using equation 5.1
of the above-referenced Method 5. For each replicate, record the
corrected dry gas volume in dry standard cubic meters, the sampling
time in decimal minutes, the barometric pressure in mm Hg, and the
ambient temperature in degrees Celcius on the enclosed data card. Be
sure to record the performance test device serial number on the data
card in the column headed "Orifice Number."
NOTE 1: If you calculate dry gas volume in English Units, use the
conversion factor of 0.02832 m3ft3 to obtain the volume in
metric units.
NOTE 2: If your stopwatch is not in decimal minutes, be sure to
convert (e.g. 15 minutes 20 seconds is reported as 15.33
minutes).
After recording the requested data on the enclosed data form, return
the data form and the performance test device to:
Quality Assurance Division (MD-77A)
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
ATTN: Ellen W. Streib
A postpaid return envelope and label are enclosed for this purpose.
34
-------
INSTRUCTIONS FOR USE OF ENVIRONMENTAL PROTECTION AGENCY STATIONARY
SOURCE QUALITY ASSURANCE S02 REFERENCE SAMPLES
NOTE: All Method 6 procedures referred to are from the amended method pub-
lished in the Federal Register Vol. j*2, No. 160, Part II, Thursday,
August 18, 1977, PP 1*1782-U178U. This amended method should be
adhered to in all details in the analysis of these reference stand-
ards.
1. Prepare 3 percent hydrogen peroxide according to Section 3.1.3 of the
method (30 ml is required for each sample and each blank).
2. Prepare each reference sample for analysis as follows: Wrap a paper
towel around the ampoule and with the ampoule in an upright position
break off the top at the prescored mark by exerting pressure sideways.
From the ampoule pipette exactly 5 ml of the reference sample into a
100 ml volumetric flask. Add 30 ml of 3 percent hydrogen peroxide
solution. Dilute exactly to the mark with deionized, distilled water.
Analyze the sample in accordance with the procedure detailed in Section
1».3 of the method, beginning with "Pipette a 20 ml aliquot of this
solution..." (Note: If more than 50 ml of barium perchlorate titrant
is required for any sample analysis, a smaller aliquot should be
selected to allow titration with less than 50 ml titrant.)
3. Calculate the concentration, Cso2 (concentration of sulfur dioxide,
dry basis, corrected to standard conditions, mg/dscm), using Equation
6-2. A value of 21 x 10"^ dscm should be used for Vm(std), in the
equation. A value of 100 ml should be used for Vsoinin the equation.
U. Record the reference standard sample numbers and their corresponding
SOg concentrations in mg/dscm on the enclosed data form. Return the
form to:
Quality Assurance Division (MD-77A)
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
ATTN: Ellen W. Streib
If other than EPA Method 6 is used for your analyses, please explain in
detail your analytical procedure on the back of the enclosed data form.
35
-------
INSTRUCTIONS FOR USE OF ENVIRONMENTAL PROTECTION AGENCY STATIONARY
SOURCE QUALITY ASSURANCE NOX REFERENCE SAMPLES
Note: All Method 7 procedures referred to are from the amended method
published in the Federal Register Vol. 42, No. 160, Part 11,
Thursday, August 18, 1977, pp 41784-41786. This amended method
should be adhered to in all details in the analysis of these
reference standards.
1. Prepare absorbing solution according to Section 3.1 of the method.
2. Prepare each reference sample for analysis as follows: Wrap a paper
towel around the ampule and with the ampule in an upright position
break off the top at the prescored mark by exerting pressure sideways.
From the ampule pipette exactly 5 ml of the reference sample into a
100-ml beaker. Add 25 ml absorbing solution to the beaker; adjust
the pH to 9-12 (using pH paper as indicated in Section 4.2 of the
method) by dropwise addition of sodium hydroxide (1N). Quantitatively
transfer the contents of the beaker to a 50-ml volumetric flask and
dilute exactly to the mark with deionized, distilled water. Mix
thoroughly and pipette a 25-ml aliquot of the diluted sample into
a porcelain evaporating dish. Beginning with the evaporation step
in Section 4.3, complete the sample analysis.
3. Calculate total jjg N02 per sample using Equation 7-3. Calculate
the sample concentration, C (concentration of NOX as NC>2» dry basis,
corrected to standard conditions, mg/dscm), using Equation 7-4.
A value of 2000 ml should be used for Vsc in Equation 7-4.
4. Record the reference sample numbers and their corresponding concentrations,
C, in mg/dscm on the enclosed data form. Return the form to:
Quality Assurance Division (MD-77A)
Environmental Monitoring Systems Laboratory
Environmental Protection Agency
Research Triangle Park, NC 27711
ATTN: Ellen W. Streib
If other than EPA Method 7 is used for your analyses, please explain in detail
your analytical procedure on the back of the enclosed data form.
3G
-------
COAL AUDIT PROGRAM INFORMATION
1. There 1s approximately 50 grains of 60 mesh coal per bottle.
2. Analyze the coal samples for moisture and on a dry basis for ash,
sulfur and gross calorific value. Report moisture, ash, and sulfur
1n weight percent with gross calorific value reported as BTU/lb.
3. All methods used In the analysis of these coal samples should follow
American Society for Testing and Materials (ASTM) recomnended procedures
or an accepted automatic analytical device.
4. Suggested procedures are:
Moisture D-3173
Ash D-3174
Sulfur D-3177
Gross Calorific Value ... D-2015
Please note on the data card (columns 17-32) the ASTM method number.
If an ASTM method was not used for analysis note that on the back of
the data card. Be parameter specific.
5. If you cannot analyze the coal sample for all four parameters, analyze
for what you can. Analysis of moisture 1s necessary to calculate on
a dry basis any of the other three parameters. Analysis of sulfur 1s
also necessary for the calculation of gross calorific value.
6. Analyze each sample In duplicate (If possible) and record results as
analysis 1 and analysis 2 for each parameter.
7. Most laboratories will use site number 001. Multiple site numbers
are used by laboratories that receive more than one set of samples.
These central laboratories have requested auditing of their satellite
laboratories.
8. After recording the requested data on the enclosed data card, return the
data card to:
Ms. Ellen W. Strelb
Quality Assurance Division (MD-77)
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
A postpaid return envelope Is enclosed for this purpose.
9. If you have any questions concerning this or any source nethod audit,
please call (919/541-7834).
?7
0' *U.S.Government Printing Office: 1987 - 748-121/40689
------- |