United States
Environmental Protection
Agency
Environmental Monitoring Systems
Laboratory
Research Triangle Park NC 2771 1
EPA-600/4-83-049
November 1983
&EPA
Research and Development
A Summary of the
1982 National
Performance Audit
Program on Source
Measurements
-------
EPA-600A-83-049
November 1983
A SUMMARY OF THE 1982 EPA NATIONAL PERFORMANCE AUDIT PROGRAM
ON SOURCE MEASUREMENTS
by
E. W. Streib and M. R. Midgett
Quality Assurance Division
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
-------
NOTICE
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.
11
-------
FOREWORD
Measurement and monitoring research efforts are designed to anticipate
potential environmental problems, to support regulatory actions by developing
an in-depth understanding of the nature and processes that impact health
and the ecology, to provide innovative means of monitoring compliance with
regulations, and to evaluate the effectiveness of health and environmental
protection efforts through the monitoring of long-term trends. The Environ-
mental Monitoring Systems Laboratory, Research Triangle Park, North Carolina,
has responsibility for: assessment of environmental monitoring technology
and systems; implementation of agency-wide quality assurance programs for
air pollution measurement systems; and supplying technical support to other
groups in the Agency, including the Office of Air, Noise and Radiation, the
Office of Toxic Substances, and the Office of Enforcement.
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 EPA Method 5 (dry gas meter only), Method 6 (sulfur
dioxide), Method 7 (nitrogen oxides), Method 19 (coal), and Method 3 (carbon
dioxide and oxygen). Statistical analysis was used to characterize the data.
Thomas R. Hauser, Ph.D.
Director
Environmental Monitoring Systems Laboratory
Research Triangle Park, North Carolina
iii
-------
ABSTRACT
In the spring and fall of 1982, the Quality Assurance Division conducted
the National Audits for Stationary Source Test Methods. The audit materials
consist of: a calibrated orifice for Method 5 (dry gas meter only), five
simulated liquid samples each for Method 6 (S02) and Method 7 (NOX), two coal
samples for Method 19A, and a disposable gas cylinder for Method 3 (Orsat
analyzer). Laboratories participating in the audits sent their data to the
Source Branch and later received a written report comparing their results to
EPA's.
In the Method 5 spring audit, the mean for all participants differed by
3.5% from the true (EPA) value. For the fall audit, the participants' mean
was 8.5% from the true value. In the two Method 6 audits, the median values
measured for all 10 samples differed by less than 2% from the true value,
whereas the median values for all 10 samples used in the two Method 7 audits
were within 3% of the true value.
In the two coal audits, the participants' accuracy on all four parameters
was consistently better for the higher concentration samples than for the
lower concentration samples. The mean values for sulfur, ash, and BTU content
were within 5%, whereas the mean values for moisture differed by as much as
9%.
This was the first Method 3 audit conducted by the Quality Assurance
Division. Each parameter had only one concentration. The median values of
CO2 and C>2 were within 12% of the true value, whereas the median values for CO
were within 20% of the true value.
This report covers a period from January 1982 to December 1982, and work
was completed as of December 1982.
iv
-------
CONTENTS
Foreword .......... . ....................
Abstract ............................... iv
Figures ............................... v*-
Tables ................................ vii
Acknowledgments .... ....................... viit
1 . Introduction ....... ............ ..... 1
2. Summary .......................... 2
3. Recommendations ...................... 6
4. Method 5 Dry Gas Meter Audit ................ 7
5. Method 6 Audit ....................... 13
6. Method 7 Audit ....................... 16
7. Method 19A Coal Audit ................... 19
8. Method 3 Audit ....................... 23
References .............................. 25
Appendices
A. Frequency distributions .................. 26
B. Method 3 audit statistics ................. 34
C. Instructions for EPA Audit Materials ............ 40
-------
FIGURES
Number Page
1 Cumulative accuracy for participants in the
Method 5 audits, 0382 and 0982 8
2 Previous results of Method 5 audit ... 9
3 Results of the Method 5 Audit 0382 11
4 Results of the Method 5 Audit 0982 12
5 Previous results of Method 6 audits 15
6 Previous results of Method 7 audit 18
vi
-------
TABLES
Number Page
1 Participants' Results for Method 5 Audit
(All Data - No Outliers Removed) 2
2 Participants' Results for Method 6 and 7 Audits
(All Data (No Outliers Removed) 3
3 Participants' Results for Method 19A Coal Audit
(All Data - No Outliers Removed) 4
4 Participants' Results for Method 3 Audit
(All Data - No Outliers Removed) 5
5 Method 5 Audit Participants 7
6 Method 6 Audit Participants 13
7 Summary of Source SO2 Audits .14
8 Method 7 Audit Participants 16
9 Summary of Source NOX Audits 17
10 Coal Audit Participants 20
11 Source Coal Audit - 0382 21
12 Source Coal Audit - 0982 22
13 Method 3 Audit Participants 23
14 Source Method 3 Audit - 0682 24
vii
-------
ACKNOWLEDGMENTS
We wish to express our appreciation to the laboratories that participated
in our audits. Thanks also to the Standards Laboratory (EPA/EMSL/Performance
Evaluation Branch) who did our Acceptance Testing on the audit samples. For
providing the data systems for storing and evaluating the data, we wish to
thank the programmers of the Data Management and Analysis Division/EMSL.
viii
-------
SECTION 1
INTRODUCTION
In 1977 the Environmental Monitoring Systems Laboratory (EMSL) of EPA
established a performance audit program to evaluate the performance of
companies that conduct compliance testing using EPA Reference Methods. The
audits verify the analytical accuracy of EPA Reference Methods 3, 6, 7, and
19A and the calibration accuracy of the Method 5 control console (1). By
participating in this free and voluntary program, testing companies are able
to compare their performance to other laboratories conducting similar
measurements.
In 1982 two audits each were conducted for Methods 5, 6, 7, and 19A, and
one audit was conducted for Method 3. 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 was also included for
the Method 5 audits for returning the audit device and for the Method 3 audit
for returning the Tedlar bag. Participants had 8 weeks to return data to
EPA. At the end of this period all data received were statistically analyzed
to determine the precision and accuracy obtained by the participants.
This report summarizes the results obtained in the 1982 source audits.
Individual Method 3 results reported by each participant are contained in the
appendices to this report.
-------
SECTION 2
SUMMARY
In the spring and fall of 1982, EPA's Environmental Monitoring Systems
Laboratory, Research Triangle Park, North Carolina, conducted National Quality
Assurance Audits for Stationary Source Test Methods 5 (dry gas meter only),
6 (S02), 7 (NOX), 19A (coal), and 3 (Orsat analyzer). Industrial laboratories,
contractors, foreign countries, and local, state, and Federal agencies
participated.
Two Method 5 audits were conducted in 1982. The overall results (no
outliers removed) are summarized in Table 1 . In the first audit, the mean
for all participants was 3.5% from the true value and in the second audit it
was 8.5%. After correcting for outliers, the means for 0382 and 0982 audits
were 0.2% and 1.3% from the true value. The participants' performances were
not statistically significantly different from previous audits (2,3).
TABLE 1 . PARTICIPANTS' RESULTS FOR METHOD 5 AUDIT
(ALL DATA - NO OUTLIERS REMOVED)
Type of
sample Parameter
Orifice Volume
Audit
date
0382
0982
No. of
analyses
827
769
Mean
( % from
3.5
8.5
Median
EPA values )
0.4
1 .5
Std.
dev.
40.1
81 .9
Table 2 presents the data (no outliers removed) from the two 1982
Method 6 audits. This audit procedure requires the participants to determine
the sulfate content in five aqueous solutions using the titration procedure
of Method 6. For each concentration in the 0282 audit, the means of the
participants were 43% higher than the true value, but in contrast the median
value differed by less than 1%. In the 0882 audit, the mean and median
values of all concentrations differed by less than 2%. In both audits, 45% to
68% of the participants achieved an accuracy within 2% for 8 out of 10 samples.
However, on the other two samples only 20% achieved this accuracy.
-------
TABLE 2. PARTICIPANTS' RESULTS FOR METHOD 6 AND 7 AUDITS
ALL DATA (NO OUTLIERS REMOVED)
Type of Audit No. of EPA true
sample Parameter date analyses value
Participant results
Mean
Median
Std. dev.
Aqueous S02
Sulfate
(Method 6)
0282
0882
0282
0882
115
102
116
103
381 .3
268.0
877.0
653.0
555.2
266.6
1256.1
645.3
381 .9
267.0
882.3
647.5
1420.3
41 .4
3155.3
102.3
0282
0882
116
103
1410.8
1740.0
2022.6
1738.6
1423.4
1710.0
5133.6
280.7
0282
0882
0282
0882
116
103
115
103
1906.5
1187.0
2325.9
2224.0
2721 .4
1165.4
3322.7
2244.2
1903.2
1176.0
2320.0
2210.0
6863.1
165.3
8480.8
354.6
Aqueous
Nitrate
(Method 7)
NOX
0482
1082
0482
1082
75
64
79
64
179.2
139.4
358.3
318.5
236.1
152.7
467.9
342.7
179.5
141 .7
366.0
324.0
424.9
66.32
795.8
161 .4
0482
1082
0482
1082
79
64
78
64
537.5
477.8
736.6
756.5
706.4
534.3
952.3
818.7
550.0
485.8
747.4
770.9
1227.8
267.4
1679.5
396.9
0482
1082
78
62
975.5
935.7
1274.5
1034.6
986.6
948.0
2233.4
475.4
Table 2 also presents the data (no outliers removed) from the two
Method 7 audits in 1982. This audit procedure requires the participants to
determine the nitrate content in five aqueous solutions. For each concen-
tration in the 0482 audit, the means of the participants were 30% higher than
the true value, but in contrast the median value differed by less than 3%.
In the 1082 audit the means were 10% higher than the true value; however, the
median values differed only by 2%. The participants' level of accuracy was
consistent for all 5 samples in both audits; i.e., 40% were within 3% on all
samples.
-------
Table 3 summarizes the results of the two coal audits. Participants
analyzed each coal sample in duplicate for percent sulfur, moisture, and
ash, and for gross calorific value (BTU/lb). The means of the ash and BTU
content were within 2% of the expected value on both concentrations. An
accuracy of within 5% was achieved on the sulfur content. However, the
mean value for moisture on the low-concentration was as high as 9% from the
expected value.
TABLE 3. PARTICIPANTS' RESULTS FOR METHOD 19A COAL AUDIT
(ALL DATA - NO OUTLIERS REMOVED)
Type of
Sample
Coal
Audit
Date Parameter
0382 %S
0982
0382
0982
No. of
analyses
73
81
73
81
EPA
( true )
value
1 .30
1 .22
2.85
3.22
Participants' results
StdT
Mean Median dev.
1 .25
1 .28
2.85
3.17
1 .24
1 .25
2.83
3.20
0.17
0.31
0.39
0.22
0382
0982
0382
0982
%H2O
73
82
73
82
0.96
2.11
1 .73
2.48
1 .03
2.29
1 .74
2.57
1 .10
2.36
1 .81
2.66
0.28
0.55
0.35
0.54
0382
0982
0382
0982
%Ash
73
81
72
81
21 .48
11 .43
35.40
16.34
21 .60
11 .24
35.36
16.11
21 .63
11 .25
35.32
16.16
0.41
0.27
0.47
0.59
0382
0982
BTU/lb
71
78
9322.0
12277.0
9409.5
12252.9
9420.0
12293.0
71 .96
184.07
0382
0982
71
78
11278.0
12932.0
11408.6
12894.5
11421 .0
12915.0
98.95
159.75
-------
The results of the first Method 3 audit conducted by the Quality
Assurance Division (QAD) are summarized in Table 4. Participants analyzed
the gas sample twice for percent carbon dioxide, oxygen, and carbon monoxide.
The median values of carbon dioxide and oxygen were within 12% of the
expected value. However, the median values for carbon monoxide differed as
much as 20% from the expected value.
TABLE 4. PARTICIPANTS' RESULTS FOR METHOD 3 AUDIT
(ALL DATA - NO OUTLIERS REMOVED)
Type
of
sample
Audit
date Parameter
EPA Participants' results
No. of Repli- (true) Std.
analyses cate value Mean Median Dev.
Small 0682 %C02
cylinder
59
56
57
54
1
2
1
2
14.8
14.8
5.2
5.2
13,
13,
6,
6,
25
29
11
20
13.90
13.75
70
80
1 .83
1 .80
1 .35
1 .42
%CO
50
47
1
2
7.0
7.0
5.06
4.96
5.65
5.70
1 .96
2.05
-------
SECTION 3
RECOMMENDATIONS
The Quality Assurance Division of the Environmental Monitoring Systems
Laboratory maintains a repository of audit samples for EPA Methods 3, 6, 7,
and for coal. These stable samples are available to any laboratory having a
need for them, such as for training new personnel and conducting quality
control checks of the laboratory. Since 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 participants make use of this
sample repository to help improve their overall analytical skills.
-------
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.
Using Equation 5-1, they convert each of the three volumes to cubic meters at
standard conditions, record them on the data card, and mail the orifice and
the data card to EPA for statistical analysis.
In the spring audit (0382), 75% of the 165 laboratories that received
the audit package returned data. In the fall audit (0982), 73% of the 164
laboratories returned data. These percentages are similar to those encoun-
tered in previous audits (2,3). Table 5, which classifies the participants
into general categories, shows the number of participants who requested to
participate in the Method 5 audit and the number who actually returned data.
TABLE 5. METHOD 5 AUDIT PARTICIPANTS __
No. requesting samples No. returning data
Contractors
Industry
Foreign
Federal
State
Local
TOTAL
0382
92
45
3
3
15
7
165
0982
90
44
3
4
16
7
164
0382
68
32
3
2
14
5
124
0982
62
34
3
3
12
6
120
Figure 1, a cumulative histogram, shows the accuracy obtained by
participants in the 0382 and 0982 Method 5 audits, expressed as the percent
difference from the true (EPA) value at various levels of accuracy. The Code
of Federal Regulations (1) requires that the dry gas meter be calibrated with
an accuracy of + 2 percent. Figure 1 shows that 44% of the reporting labora-
tories in the 0382 audit and 43% in the 0982 audit obtained this accuracy.
These results are similar to those in previous audits (Figure 2). Ninety-two
of the laboratories participated in both audits.
-------
LABORATORIES ACHIEVING SPECIFIED ACCURACY, percent
UD
c
n
c
3
c
Q)
O
n
c
-^
Q)
O
oo
-a
QJ
3
fTJ
r-f
rr
o
Q.
ai
cu
o
00
CO
N)
O
CD
00
N)
o
-n
-n
3D
m
n
m
Z JS.
5
m
CO
3D
m
Z
CO
:" o,
3D
O
g
m
-o
J>
^
r—
C
•a
CD CO
3
re
3
'"*'
r
^
1 1
n
>
a
-i
o
::::::::::::::::::::::::::::::::::::::::XvX::::::::;:::-:::::::;:::::;:I 5
::::::::::::::::::x:::::::x::::::::::::v:::::-:;::x::'X-:-:':':-:':j ,M
3
»J
en
CO
SjSiv^
'^^^^^^X^^^^^^i^^i^
^^^^^^MH^^iMii^Mi^ii^^
\
^^^^^fi^m^^^^^MimMMmm^
&fft8$8$8^^
\ \ II
c
o
H
C9
00
3
00
ro
~^
-------
50
45
§ 40
oc
o
CO
35
30
DEGREE OF ACCURACY=2PERCENT
I
0379/342 0879/523 0280/622 0880/725 0381/738 0981/723
AUDIT/NUMBER OF SAMPLES
0382/827
0932/769
Figure 2. Previous results of method 5 audit.
-------
The histograms in Figures 3 and 4 show how the individual results of
the 0382 and 0982 audits compared to the mean and the median values for all
participants. The majority of the laboratories reported values higher than
the EPA value. The standard deviation of the triplicate analyses (precision)
by each laboratory indicated that for the 0382 audit, 70% of the standard
deviations for each set were within 0.3%. For the 0982 audit, 74% of the
standard deviations were within 0.3%. Six percent of the 0382 data and 8% of
the 0982 data were identified as outliers using Chauvenet's Criterion (4).
Before the outliers were removed, the mean values for the 0382 and 0982 data
differed by 3.5 and 8.5% from the true value, respectively. After deletion
of outliers, these values were reduced to 0.2 and 1.3%, respectively.
10
-------
LABORATORIES, percent
c
—(
CD
0>
C/l
c
3
to
o
Q.
cn
Q)
C
Q.
O
00
00
NJ
m 30
O cn
-------
20
I I I I I I I
NUMBER OF VOLUMES=769
1 I T
15
g 10
oc
o
00
MEAN=1.3 || MEDIAN=1.5
1
<10 -8 -6
4-20 246 8
DIFFERENCE FROM EPA VALUE, percent
10 >12
Figure 4. Results of method 5 audit 0982.
12
-------
SECTION 5
METHOD 6 AUDIT
This audit checks the participant's ability to analyze a Method 6 sample
for sulfate. The audit set consists of five dilutions of 10 N sulfuric acid
in 25-ml sealed glass ampoules. These five ampoules contain the equivalent
of approximately 0 to 3000 mg S02/DSCM (dry standard cubic meter). The
analyst withdraws 5.0 ml from each sample, adds 30 ml of 3% hydrogen
peroxide, and dilutes the sample to 100 ml with distilled water. A 20-ml
aliquot is then withdrawn from the diluted sample, 80 ml of 100% isopropanol
and thorin indicator are added, and the sample is titrated with barium
perchlorate (Ba[ClO4]2) to a pink end point. In calculating the results, the
participants assume they had an original sample volume of 100 ml, and that they
sampled 21 x 10~3 DSCM (dry standard cubic meter) of stack gas.
In the spring audit (0282), 74% of the 156 laboratories that received
the audit package returned data. In the fall audit (0882), 67% of the 156
laboratories returned data. These percentages are similar to those encoun-
tered in previous audits (2,3). Table 6, which classifies the participants
into general categories, shows the total number of participants requesting
participation and the number that returned data. Eighty laboratories partici-
pated in both audits and returned data.
TABLE 6. METHOD 6 AUDIT PARTICIPANTS
No. requesting samples No. returning data
Contractors
Industry
Foreign
Federal
State
Local
TOTAL
0282
90
42
2
2
11
9
156
0882
86
42
4
2
13
9
156
0282
67
29
2
1
10
7
116
0882
48
29
4
2
11
9
103
13
-------
Table 7 presents the percent of laboratories that achieved 2% and 5%
accuracy for each of the five different concentrations in the two 1982
Method 6 audits. The table shows that 68% of the reporting laboratories
in the 0282 audit achieved an accuracy within 2% for the higher concen-
tration and that in the 0282 audit, 24% of the laboratories achieved an
accuracy within 2%. However, in both audits, approximately 45% of the
participants achieved an accuracy within 2% for the lowest concentration
samples. Approximately 75% of the laboratories were able to achieve an
accuracy level of within 5% on 8 of the 10 samples.
TABLE 7. SUMMARY OF SOURCE SO2 AUDITS
0282 0882
Concentration +2% +5% +2% +5%
0-500 mg/DSCM
501-1000 mg/DSCM
1001-1500 mg/DSCM
1501-2000 mg/DSCM
2001-3000 mg/DSCM
n
48%
59%
55%
68%
68%
75%
80%
85%
88%
91%
116
45%
62%
60%
20%
24%
75%
82%
83%
51%
54%
103
The results obtained in the Fall 1982 Method 6 audit did differ from
those obtained in previous audits (Figure 5).
14
-------
cc
o
90
80
70
60
50
I
30
20
10
SAMPLE RANGE,
mgS02/DSCM
• 0-500
A 1001-1500
• 2001-3000
DEGREE OF ACCURACY=2 PERCENT
0380/102 0980/101 0281/121 0881/98 0282/116
AUDIT/NUMBER OF SAMPLES
Figure 5. Previous results of method 6 audit.
0882/103
15
-------
SECTION 6
METHOD 7 AUDIT
This audit checks the participant's ability to analyze a Method 7 sample
for nitrate. The NOX audit set consists of five dilutions of potassium
nitrate (KNO3) stock solution in 25-ml glass ampoules that are autoclaved
after sealing to destroy bacteria that might attack the nitrate. The five
samples in the set simulate source samples ranging from 0 to 1000 mg NOX/DSCM.
The analyst withdraws 5.0 ml from an ampoule, adds this and 25 ml of the
Method 7 absorbing solution to a flask, adjusts the pH to 9 to 12 with NaOH,
and dilutes to 50 ml with distilled water. He then withdraws a 25-ml aliquot
from the diluted sample, places it in an evaporating dish, and treats it as
described in Section 4.3 of Method 7. After the treatment is completed, he
measures the absorbance at 410 nm. In calculating the concentrations present,
the participant asumes that 2000 ml of stack gas has been sampled.
In the spring audit (0482), 61% of the 129 laboratories that received
the audit package returned data. In the fall audit (1082), 52% of the labora-
tories returned data. Fifty-one laboratories participated in both audits and
returned data. Table 8 shows the total number of laboratories requesting
participation and the number that returned data for Method 7 audits 0482 and
1082.
TABLE 8. METHOD 7 AUDIT PARTICIPANTS
No. requesting samples No. returning data
Contractors
Industry
Foreign
Federal
State
Local
TOTAL
0482
80
31
2
1
8
7
129
1082
76
30
2
1
7
7
123
0482
46
19
1
1
6
6
79
1082
34
19
2
1
4
4
64
16
-------
Table 9 shows the percent of laboratories that can achieve 3 and 7%
accuracy for each of the five concentrations. The table shows that 36% of
the reporting laboratories in the 0482 audit achieved an accuracy within 3%
on the lowest concentration. In the 1082 audit, 31% of the laboratories
achieved an accuracy within 3%. Sixty percent of the laboratories were able
to achieve 7% accuracy on all samples in both audits.
TABLE 9. SUMMARY OF SOURCE NOX AUDITS
0482 1082
Concentration +3% +7% +3% +7%
0-200 mg/DSCM
201-400 mg/DSCM
401-600 mg/DSCM
601-800 mg/DSCM
801-1000 mg/DSCM
n
36%
41%
41%
40%
36%
79
59%
66%
65%
65%
65%
31%
38%
42%
36%
37%
64
55%
61%
64%
67%
61%
Figure 6 compares the results of the 1982 audit to those of the past
four audits (2,3). The results have improved and the percent of laboratories
obtaining an accuracy of 3% has held steady, between 30 to 40%.
17
-------
50
40
CO
UJ
E
30
CC
O
CO
20
10
r
SAMPLE RANGE
mgNOx/DSCM
• 0200
• 401-600
A 801-1000
DEGREE OF ACCURACY=3PERCENT
0480/71
1080/71 0481/89 1081/76
AUDIT/NUMBER OF SAMPLES
0482/79
1082/64
Figure 6. Previous results of method 7 audit.
18
-------
SECTION 7
METHOD 19A COAL AUDIT
A proposed revision in Part 60, Title 40 of the Code of Federal Regula-
tions will allow coal sampling and analysis to serve as an acceptable method
for demonstrating compliance with the SO2 emissions standard for large Subpart
Da coal-fired power plants. The coal audit checks the participant's ability
to analyze a coal sample for sulfur, ash, moisture, and BTU content. Accept-
ance Testing on the commercially obtained coal was done by EPA contractors.
Each set of coal samples consisted of two bottles containing 50 g of
60-mesh coal. Participants measured each sample for sulfur, moisture, ash,
and gross calorific content. The following American Society for Testing
and Materials (ASTM) procedures were recommended but not necessarily
mandated (5).
• 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); and
• ASTM D-2015 (Standard Test Method for Gross Calorific Value of Solid
Fuel by the Adiabatic Bomb Method) (5).
The participants measured the four 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 the spring audit (0382), 73% of the 77 laboratories that received
the audit package returned data. In the fall audit (0982), 77% of the 90
laboratories returned data. Forty-seven laboratories participated in both
audits and returned data. Table 10 shows the total number of laboratories
requesting participation and the number that returned data for coal audits
0382 and 0982.
19
-------
TABLE 10. COAL AUDIT PARTICIPANTS
No. requesting samples No. returning data
0382
33
30
2
7
5
77
0982
36
39
1
10
4
90
0382
19
27
0
7
3
56
0982
22
36
0
7
4
69
Contractors
Industry
Federal
State
Local
TOTAL
Tables 11 and 12 summarize the coal audit results. The N value 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. Each laboratory was asked to analyze the samples in
duplicate. Accuracies of 5% and 10% were chosen for the precision criteria
for each of the four parameters.
In both audits, 50% of the laboratories were able to analyze the sulfur
content of the low level samples within 5% of the expected value. Seventy
percent of the laboratories achieved 5% accuracy for the high level sulfur
concentrations. Only 10% of the laboratories achieved 5% on both of the
moisture concentrations. For the ash analysis and BTU content, 95% to 100%
of the reporting laboratories were able to achieve an accuracy within 5% for
both sample concentrations.
The participants' accuracy improved with higher concentrations on all
four parameters. For those that did duplicate analyses, the intra-laboratory
precision showed no correlation with concentration. Therefore, the standard
deviation (precision) was independent of the sample concentration for all
four parameters.
20
-------
TABLE 11. SOURCE COAL AUDIT - 0382
Expected No.
value anal}
1.30% (1)
(2)
2.85% (1)
(2)
0.96% (1)
(2)
1.73% (1)
(2)
21.48% (1)
(2)
35.40% (1)
(2)
9322 BTU/lb (1)
(2)
1 1 278 BTU/lb { 1 )
(2)
of
fses
73
64
73
64
73
63
73
63
73
64
73
64
71
62
71
62
Laboratories
accurate within +5%
48%
44%
68%
69%
10%
6%
22%
19%
97%
95%
99%
100%
100%
100%
100%
100%
Laboratories
accurate within +10%
85%
84%
88%
86%
23% .
19%
42%
41%
100%
98%
100%
100%
100%
100%
100%
100%
21
-------
TABLE 12. SOURCE COAL AUDIT - 0982
Expected No.
value anal;
1.22% (1)
(2)
3.22% (1)
(2)
2.11% (1)
(2)
2.48% (1)
(2)
11.43% (1)
(2)
16.34% (1)
(2)
12277 BTU/lb (1 )
(2)
12932 BTU/lb (1 )
(2)
of
yses
81
78
81
78
82
76
82
76
81
78
81
78
78
76
78
76
Laboratories
accurate within +5%
57%
53%
73%
78%
17%
12%
17%
20%
98%
94%
99%
97%
97%
96%
100%
99%
Laboratories
accurate within +10%
79%
76%
90%
92%
32%
29%
40%
39%
100%
100%
99%
99%
100%
100%
100%
100%
22
-------
SECTION 8
METHOD 3 AUDIT
This audit checks the participant's ability to analyze a gas sample
using an Orsat analyzer. The audit package consists of a disposable 4-liter
cylinder and a Tedlar sample bag. The analyst expels the gas from the cylinder
into the sample bag. Then a gas sample is extracted from the bag into the
Orsat analyzer. The gas sample is analyzed for percent carbon dioxide, oxygen,
and carbon monoxide. This was the first Method 3 audit conducted by QAD.
In this audit, 59% of the 96 laboratories that received the audit package
returned data. Table 13 shows the total number of laboratories requesting
participation and the number that returned data for the Method 3 audit.
TABLE 13. METHOD 3 AUDIT PARTICIPANTS
No. requesting samples
No. returning data
Contractors
Industry
Foreign
Federal
State
Local
TOTAL
0682
57
25
1
2
8
3
96
0682
29
17
1
1
6
3
57
Table 14 summarizes the Method 3 audit. Each laboratory was asked
to analyze the sample in duplicate. Some laboratories exhausted their gas
sample after the first analysis. Five and ten percent accuracy were chosen
for the precision criteria for each of the three parameters. Each parameter
had only one concentration.
23
-------
Sixty-five percent of the reporting laboratories were able to analyze
the percent C02 within an accuracy of 10%. For the percent O2 analysis, only
45% of the laboratories achieved an accuracy within 10%. Since CO was
analyzed last and the concentration was higher than for normal conditions,
only 31% of the laboratories achieved an accuracy within 10%, and 15% of the
participants did not report a value for CO.
TABLE 14. SOURCE METHOD 3 AUDIT - 0682
Expected No. of Laboratories Laboratories
value analyses accurate within +5% accurate within +10%
14.8% (1)
(2)
5.2% (1)
(2)
7.0% (1)
(2)
59
56
57
54
50
47
C02
34% 68%
34% 66%
°2 ~
28% 49%
28% 43%
CO
14% 30%
13% 32%
24
-------
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. Fuerst, R. G., E. W. Streib, and M. R. Midgett. A Summary of the EPA
National Source Performance Audit Program - 1979. EPA-600/4-81-029,
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711,
April 1981 . 53 pp.
3. Fuerst, R. G. E. W. Streib, and M. R. Midgett. A Summary of the EPA
National Source Performance Audit Program - 1980. EPA-600/4-81-077,
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711,
December 1981 .
4. Chauvenet, W. Manual of Spherical and Practical Astronomy: Volume II -
Theory and Use of Astronomical Instruments (Method of Least Squares).
J, B. Lippincott and Co., Philadelphia, PA 1863(7). pp. 558-565.
5. American Society for Testing and Materials. Annual Book of ASTM
Standards - 1979. Part 26. 01-026079-13, American Society for Testing
and Materials, Philadelphia, PA.
25
-------
APPENDIX A
FREQUENCY DISTRIBUTIONS
26
-------
DGM FREQUENCY DISTRIBUTION OF ABSOLUTE PERCENT DIFFERENCE - 0382
#SAMP
827
820
803
786
775
771
769
766
764
762
759
755
752
751
HIN
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
10%
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
20%
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
30%
1 .2
1 .2
1 .2
1 .2
1 .2
1 .2
1 .2
1 .2
1 .2
1 .2
1 .2
1 .1
1 .1
1 .1
40%
1 .8
1 .8
1 .8
1 .7
1 .7
1 .6
1 .6
1 .6
1 .6
1 .6
1 .6
1 .6
1 .6
1 .6
50%
2.5
2.4
2.4
2.3
2.3
2.3
2.2
2.2
2.2
2.2
2.2
2.2
2.2
2.2
60%
3.1
3.1
3.0
3.0
3.0
2.9
2.9
2.9
2.9
2.9
2.8
2.8
2.8
2.8
70%
3.9
3.8
3.7
3.6
3.5
3.5
3.5
3.4
3.4
3.4
3.4
3.4
3.4
3.4
80%
5.1
5.0
4.7
4.5
4.3
4.3
4.3
4.2
4.2
4.2
4.2
4.1
4.1
4.1
90%
7.6
7.1
6.5
6.3
6.1
6.0
6.0
5.9
5.8
5.7
5.6
5.5
5.5
5.5
MAX
596.0
51 .6
25.3
15.2
11 .1
10.2
10.0
9.8
9.7
9.6
9.4
9.3
9.0
8.9
MEAN
7.6
4.1
3.3
2.9
2.8
2.7
2.7
2.7
2.7
2.6
2.6
2.6
2.6
2.6
STD. DEV.
39.5
6.6
3.7
2.6
2.3
2.2
2.2
2.1
2.1
2.1
2.0
2.0
1 .9
1 .9
SKEWNESS
10.82
-.00
-.02
-.03
-.04
-.05
-.06
-.08
-.10
-.11
-.13
-.15
-.17
-.19
MEDIAN
2.5
2.4
2.4
2.3
2.3
2.3
2.3
2.2
2.2
2.2
2.2
2.2
2.2
2.2
-------
DGM FREQUENCY DISTRIBUTION OF A ,_,..'JTE PERCENT DIFFERENCE - 0982
#SAMP
769
763
748
728
715
710
702
694
688
682
00
680
677
676
675
674
.672
670
669
667
666
665
MIN
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
10%
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
.4
20%
1 .0
.9
.9
.9
.9
.9
.9
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
.8
30%
1 .5
1 .5
1 .4
1 .4
1 .4
1 .4
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
1 .3
40%
1 .9
1 .9
1 .9
1 .8
1 .8
1 .8
1 .8
1 .7
1 .7
1 .7
1 .7
1 .7
1.7
1 .7
1 .7
1 .7
1 .7
1 .7
1 .7
1 .7
1 .7
50%
2.5
2.5
2.4
2.3
2.3
2.3
2.2
2.2
2.2
2.2
2.2
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
60%
3.2
3.1
3.0
2.9
2.8
2.8
2.8
2.8
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.6
2.6
2.6
2.6
70%
4.0
4.0
3.8
3.6
3.6
3.5
3.5
3.4
3.4
3.3
3.3
3.3
3.3
3.3
3.3
3.3
3.2
3.2
3.2
3.2
3.2
80'-
5.4
5.3
4.9
4.5
4.3
4.3
4.2
4.1
4.1
4.0
4.0
4.0
4.0
4.0
4.0
4.0 -
4.0
4.0
4.0
3.9
3.9
90%
10.5
10.0
8.9
7.3
6.9
6.8
5.8
5.6
5.5
5.4
5.4
5.3
5.2
5.2
5.1
5.1 s
5.0
5.0
5.0
4.9
4.9
MAX
933.8
96.4
29.1
17.8
12.9
11 .9
11 .4
10.6
10.1
9.8
9.4
9.1
9.0
9.0
8.9
8.9
8.7
8.7
8.5
8.5
8.3
MEAN
12.5
5.3
3.7
3.2
3.0
2.9
2.8
2.7
2.7
2.6
2.6
2.6
2.6
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
STD. DEV.
81 .4
12.3
4.4
3.2
2.7
2.6
2.4
2.3
2.2
2.1
2.0
2.0
2.0
1 .9
1 .9
1 .9
1 .9
1 .9
1 .8
1 .8
1 .8
SKEWNESS
10.81
.00
-.01
-.01
-.01
-.02
-.02
-.03
-.03
-.04
-.04
-.05
-.06
-.06
-.07
-.08
-.09
-.09
-.10
-.11
-.12
MEDIAN
2.5
2.5
2.4
2.3
2.3
2.3
2.2
2.2
2.2
2.2
2.2
2.2
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
-------
S02 FREQUENCY DISTRIBUTION OF PERCENT DIFFERENCE WITH OUTLIERS REMOVED
AUDIT 0282
IO
LEVEL (mq/DSCM) NOBS
.1- 500.0 103
500.1-1000.0 89
1000.0-1500.0 105
1500.1-2000.0 101
2000.1-3000.0 105
LEVEL (mg/DSCM) NOBS
.1- 500.0 87
500.1-1000.0 83
1000.1-1500.0 81
1500.1-2000.0 100
2000.1-3000.0 101
MIN
.00
.01
.01
.00
.03
SO 2
MIN
.00
.02
.07
.06
.59
10%
.16
.22
.30
.16
.18
20%
.60
.57
.73
.29
.35
30%
1 .00
.74
1 .03
.40
.62
FREQUENCY DISTRIBUTION
10%
.11
.23
.25
1 .15
1 .17
20%
.45
.46
.42
1 .84
1 .58
30%
1 .12
.72
.59
2.92
2.20
40%
1 .49
1 .08
1 .30
.81
.74
50%
1 .91
1 .32
1 .47
1 .08
1.11
60%
2.54
1 .53
1 .93
1 .32
1 .46
OF PERCENT DIFFERENCE
AUDIT
40%
1 .42
.96
.85
3.45
2.77
0882
50%
1 .83
1 .23
1 .06
4.55
3.64
60%
2.46
1 .45
1 .36
5.95
6.12
70%
3.33
1 .74
2.22
1 .59
1 .80
80%
4.59
2.18
2.89
2.03
2.45
WITH OUTLIERS
70%
2.80
1 .68
1 .68
7.30
7.04
80%
3.54
2.05
2.18
8.81
8.13
90% MAX
6.43 9.89
2.78 3.92
4.12 6.00
2.81 4.41
3.49 4.93
REMOVED
90% MAX
6.19 7.87
2.85 4.03
2.63 3.16
11.49 14.94
9.26 1 1 .96
AVE
2.73
1 .45
1 .95
1 .30
1 .48
AVE
2.40
1 .41
1 .29
5.58
4.85
STD
2.48
.96
1 .46
1.11
1 .25
STD
2.03
1 .01
.89
3.91
3.22
-------
FREQUENCY DISTRIBUTION OF PERCENT DIFFERENCE WITH OUTLIERS REMOVED
AUDIT 0482
LEVEL (mg/DSCM)
.1- 200.0
200.1- 400.0
400.1- 600.0
600.1- 800.0
800.1-1000.0
NOBS
61
62
61
66
68
MIN
.11
.06
.04
.01
.09
NOX
10%
.45
.36
.45
.62
.24
20%
1 .12
.89
1 .25
1 .38
.87
30%
1 .79
1 .73
1 .99
2.05
1 .63
FREQUENCY DISTRIBUTION
40%
2.62
2.20
2.29
2.77
2.82
50%
3.40
2.82
2.51
3.11
3.64
60%
4.24
3.82
3.48
4.19
4.14
OF PERCENT DIFFERENCE
70%
6.25
4.61
4.37
5.47
5.28
80%
7.53
5.44
6.57
7.56
7.88
90%
11 .33
8.18
8.67
1 .19
12.02
MAX
16.07
10.52
12.09
15.15
16.86
AVE STD
4.84 4.23
3.70 2.89
3.87 3.13
4.77 4.02
4.95 4.57
WITH OUTLIERS REMOVED
AUDIT 1082
LEVEL (mg/DSCM)
.1- 200.0
200.1- 400.0
400.1- 600.0
600.1- 800.0
800.1-1000.0
NOBS
54
51
45
45
52
MIN
.29
.03
.15
.07
.12
10%
1 .00
.44
.27
.20
.43
20%
1 .36
1 .22
.63
.79
.98
30%
2.44
1 .73
1 .07
1 .44
1 .54
40%
3.16
2.17
1 .80
2.22
2.11
50%
3.80
2.9£
2.05
2.71
3.35
60%
5.45
3.92
2.97
4.16
5.06
70%
7.68
5.87
3.52
4.44
5.95
80%
11 .91
7.06
5.06
5.57
7.62
90%
17.65
9.89
6.95
6.15
12.17
MAX
20.52
13.03
9.29
9.29
17.99
AVE STD
6.82 6.30
4.31 3.51
3.00 2.53
3.33 2.39
5.27 4.97
-------
NATIONAL COAL AUDIT FREQUENCY DISTRIBUTION OF ABSOLUTE PERCENT DIFFERENCES OF EXPECTED AND REPORTED VALUES
AUDIT 0382
Sample
Sample
Sample
Sample
Sample
Sample
Sample
Sample
3000
8000
3000
8000
3000
8000
3000
8000
NO.
137
137
136
136
137
137
133
133
WIN
.00
.00
1 .04
1 .16
.00
.00
.01
.01
10%
.35
.77
5.21
3.47
.08
.23
.39
.43
20%
1 .05
3.08
9.38
4.62
.17
.47
.68
.71
30%
1 .75
3.85
14.58
6.94
.28
.79
.89
.99
40%
2.11
4.62
18.75
9.25
.34
1 .02
.99
1 .15
50%
2.81 3.
5.38 6.
20.83 23.
12.72 14.
.51
1 .40 1 .
1 .08 1 .
1 .29 1 .
60%
86
15
96
45
62
68
15
51
70%
5.26
7.69
28.13
16.76
.79
2.00
1 .27
1 .66
80%
7.02
9.23
34.38
20.23
.99
2.33
1 .45
1 .91
90%
12.28
12.31
50.00
32.95
1.53
2.75
1 .74
2.17
MAX
96.84
93.08
72.92
61 .27
8.11
13.83
3.14
2.66
MEAN
5.97
7.77
24.34
15.29
.72
1 .64
1.11
1 .32
STD DEV
12.22
11 .59
16.99
12.87
.89
1 .60
.55
.63
-------
NATIONAL COAL AUDIT FREQUENCY DISTRIBUTION OF ABSOLUTE PERCENT DIFFERENCES OF EXPECTED AND REPORTED VALUES
AUDIT 0982
Sample
Sample
Sample
Sample
Sample
Sample
Sample
Sample
4000
5000
4000
5000
4000
5000
4000
5000
NO.
159
159
158
158
159
159
154
154
MIN
.00
.00
.47
.00
.17
.00
.00
.01
10%
.31
.82
3.32
3.23
.52
.18
.06
.09
20%
.62
1 .64
6.16
5.24
.79
.37
.12
.18
30%
1 .24
2.46
9.95
7.66
1 .05
.61
.21
.27
40%
1 .86
3.28
13.74
9.68
1 .40
.98
.29
.37
50%
2 . 48 3 .
4.92 6.
16.11 18.
12.90 15.
1.66 1.
1.16 1 .
.42
.48
60%
1 1
56
48
73
92
47
60
60
70%
4.66
8.20
21 .80
18.55
2.36
1 .65
.94
.85
80%
6.21
11 .48
24.64
21 .37
2.80
2.20
1 .58
1 .60
90%
9.01
16.39
32.23
26.21
3.50
2.94
2.34
2.36
MAX
39.75
161 .48
141 .23
116.94
15.40
31 .33
5.22
5.78
MEAN
4.14
10.39
19.29
15.58
2.00
1 .74
.85
.95
STD DEV
5.45
25.47
19.05
15.64
1 .72
3.49
.98
1 .23
-------
NATIONAL ORSAT AUDIT FREQUENCY DISTRIBUTION OF ABSOLUTE PERCENT DIFFERENCES OF EXPECTED AND REPORTED VALUES
AUDIT 0682
NO. MIN 10% 20% 30% 40% 50% 60% 70% 80% 90% MAX MEAN STD DEV
C02
Sample 6000 115 .00 1.35 2.70 4.05 5.41 6.76 8.11 10.14 12.16 23.65 58.78 10.44 12.10
Sample 6000 111 .00 1.92 3.85 5.77 7.69 11.54 15.38 19.23 26.92 53.85 134.62 20.06 25.22
CO
OJ Sample 6000 97 .00 2.86 8.57 10.00 12.86 18.57 22.86 30.00 57.14 74.29 97.14 29.47 27.37
OJ
-------
APPENDIX B
NATIONAL ORSAT (METHOD 3) AUDIT STATISTICS
Parameter: CO2
Sample Number: 6000
Analysis: 1
Number of OBS: 59
Expected Value: 14.80
STUDY ID: 0682
Mean: 13.25
Median: 13.90
Range: 8.50
Variance: 3.35
Std. Dev.: 1 .83
Coef. Var.: 13.81
Upper Conf. Int.: 13.72
Lower Conf. Int.: 12.79
Skewness: -2.28
Accuracy: -6.08
6.30
11 .20
12.60
13.20
13.40
13.70
13.90
14.00
14.20
14.40
14.50
14.70
7.20
11 .30
13.00
13.30
13.50
13.70
13.90
14.00
14.20
14.40
14.50
14.80
Lft O.IN ftOV^EjlNL/JLlNlj <-
7.60
11 .60
13.00
13.30
13.50
13.80
13.90
14.00
14.20
14.40
14.60
14.80
8.80
12.00
13.20
13.30
13.60
13.80
14.00
14.00
14.30
14.40
14.60
14.80
10.80
12.00
13.20
13.40
13.60
13.90
14.00
14.20
14.30
14.40
14.70
34
-------
Parameter: CC>2
Sample Number: 6000
Analysis: 2
Number of OBS: 56
Expected Value: 14.80
STUDY ID: 0682
Mean: 13.29
Median: 13.75
Range: 9.70
Variance: 3.23
Std. Dev.: 1 .80
Coef. Var.: 13.52
Upper Conf. Int.: 13.77
Lower Conf. Int.: 12.82
Skewness: —2.19
Accuracy: -7.09
6.10
11 .00
12.80
13.20
13.40
13.70
14.00
14.00
14.30
14.50
14.70
15.80
7.20
11 .40
12.80
13.20
13.40
13.70
14.00
14.10
14.40
14.50
14.70
Ltt J.1N ftOV-.rilNUJ.lNU V.
9.20
12.00
13.00
13.20
13.60
13.70
14.00
14.20
14.40
14.50
14.80
9.50
12.50
13.20
13.30
13.60
13.80
14.00
14.20
14.40
14.50
14.80
10.60
12.50
13.20
13.40
13.60
13.80
14.00
14.30
14.40
14.60
14.80
35
-------
Parameter: 02
Sample Number: 6000
Analysis: 1
Number of OBS: 57
Expected Value: 5.20
STUDY ID: 0682
Mean: 6.11
Median: 5.70
Range: 7.80
Variance: 1.83
Std. Dev.: 1 .35
Coef. Var.: 22.17
Upper Conf. Int.: 6.46
Lower Conf. Int.: 5.76
Skewness: 2.41
Accuracy: 9.62
4.40
5.20
5.30
5.40
5.50
5.60
5.80
6.00
6.20
6.50
7.60
10.50
4.60
5.20
5.30
5.40
5.50
5.60
5.90
6.00
6.20
6.60
8.00
12.20
ft J.IN ftOV^riLMU-HNlj V
4.80
5.20
5.30
5.40
5.50
5.70
5.90
6.00
6.30
6.60
8.00
5.00
5.30
5.30
5.40
5.60
5.70
6.00
6.10
6.40
6.80
8.40
5.00
5.30
5.30
5.50
5.60
5.70
6.00
6.20
6.40
7.00
8.90
36
-------
Parameter: 02
Sample Number: 6000
Analysts: 2
Number of OBS: 54
Expected Value: 5.20
STUDY ID: 0682
Mean: 6.20
Median: 5.80
Range: 6.80
Variance: 2.01
Std. Dev.: 1 .42
Coef. Var.: 22.90
Upper Conf. Int.: 6.58
Lower Conf. Int.: 5.82
Skewness: 1.70
Accuracy: 11.54
4.30
5.10
5.30
5.40
5.60
5.70
6.00
6.20
6.50
7.00
9.00
4.30
5.20
5.30
5.40
5.60
5.80
6.00
6.20
6.60
7.40
9.60
Lrt -LIN AO^JtlNlJXlNlj (.
4.60
5.20
5.30
5.40
5.60
5.80
6.10
6.30
6.60
8.00
10.50
5.00
5.20
5.30
5.50
5.60
5.80
6.10
6.30
6.60
8.40
11 .10
5.00
5.30
5.40
5.50
5.70
5.90
6.20
6.50
6.60
8.80
37
-------
STUDY ID: 0682
Parameter: CO
Sample Number: 6000
Analysis: 1
Number of DBS: 50
Expected Value: 7.00
Mean: 5.06
Median: 5.65
Range: 7.80
Variance: 3.86
Std. Dev.: 1.96
Coef. Var.: 38.82
Upper Conf. Int.: 5.60
Lower Conf. Int.: 4.52
Skewness: -.91
Accuracy: -19.29
.20
2.00
3.10
4.90
5.40
5.70
6.20
6.40
6.60
6.80
.20
2.50
3.40
5.00
5.40
5.80
6.20
6.40
6.70
6.90
rt J.IN ftCH-JMNUXlNVj V_l
1 .20
2.80
4.10
5.00
5.40
6.00
6.20
6.40
6.80
7.10
1 .30
2.80
4.10
5.00
5.50
6.10
6.30
6.40
6.80
8.00
1 .80
3.00
4.60
5.20
5.60
6.10
6.30
6.50
6.80
8.00
38
-------
STUDY ID: 0682
Parameter: CO
Sample Number: 6000
Analysis: 2
Number of DBS: 47
Expected value: 7.00
Mean: 4.96
Median: 5.70
Range: 7.50
Variance: 4.20
Std. Dev.: 2.05
Coef. Var.: 41.27
Upper Conf. Int.: 5.55
Lower Conf. Int.: 4.38
Skewness: -.86
Accuracy: -18.57
.20
1 .60
3.00
5.00
5.50
5.80
6.20
6.40
6.80
7.50
.30
2.20
3.40
5.00
5.60
6.00
6.20
6.40
6.80
7.70
rt ilN rta<_EjlNL/XlNlj U
.90
2.80
3.60
5.10
5.60
6.10
6.30
6.60
6.90
1 .30
3.00
3.90
5.20
5.70
6.20
6.30
6.70
7.00
1 .50
3.00
4.00
5.40
5.80
6.20
6.40
6.80
7.40
39
-------
APPENDIX C
INSTRUCTIONS FOR EPA AUDIT MATERIALS
INSTRUCTIONS FOR USE OF ENVIRONMENTAL PROTECTION AGENCY
METHOD 5 DRY GAS METER PERFORMANCE TEST DEVICE
NOTE All procedures referrred to are from revised Method 5 published in the Federal Register, Vol. 12,
No 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 CIR-
CUMSTANCES. Use these devices at room temperature.
PROCEDURE
1. Remove the performance test device from its case and insert it into the gas inlet quick-
connect coupling on the source sampling meter box.
2. Turn the power to the meter box on and start the pump.
3. Adjust the coarse flow rate control valve and the fine flow rate control valve to give a maxi
mum vacuum reading CAUT ION: A vacuum reading of less than 17 inches Hg will result
in flow rate errors.
4. Allow the orifice and source sampling meter box to warm up for 45 minutes with flow con-
trols adjusted asdescnbed in Step 3 before starting quality assurance runs.
5. 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 participant 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.
6. Calculate the corrected dry gas volume for each run using equation 5.1 of the above-refer
enced Method 5. For each replicate, record the corrected dry gas volume in dry standaH
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 m3 ft3
to obtain the volume in metric units
NOTE 2 If your stopwatch is not in decimal minutes, be sure to convert (eg. 15 minutes 20 seconds is
reported as 15.33 minutes)
7. After recording the requested data on the enclosed data form, return the data form and the
performance test device to:
Quality Assurance Division (MD-77)
Environmental Monitoring Systems Laboratory
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Attention: Ms. Ellen Streib
A postpaid return envelope and label are enclosed for this purpose.
40
-------
INSTRUCTIONS FOR USE OF ENVIRONMENTAL PROTECTION AGENCY STATIONARY
SOURCE QUALITY ASSURANCE SO2 REFERENCE SAMPLES
Note: All Method 6 procedures referred to are from the amended method published in the Federal
Register Vol. 42, No. 160, Part II, Thursday, August 18, 1977, pp 41782-41784. This amend-
ed method should be adhered to in all details in the analysis of these reference standards.
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 ampule and
with the ampule in an upright position break off the top at the prescored mark by exerting pres-
sure sideways. From the ampule pipette exactly 5 ml of the reference sample into a 100-ml volu-
metric 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 4.3 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, CSQ? ' concentration «>' tulfur dioxide, dry basis, corrected to standard con-
ditions, mg/dscm), using Equation 6-2. A value of 21 X 10 3 dscm should be used for Vm(std),
in the equation. A value of 100 ml should be used for VSO|n in the equation.
4. Record the reference standard sample numbers and their corresponding S02 concentrations in
mg/dscm on the enclosed data form. Return the form to:
Quality Assurance Division (MD 77)
Environmental Monitoring Systems Laboratory
Environmental Protection Agency
Research Triangle Park, N.C. 27711
ATTN: Ellen W. Streib
If other than EPA Method 6 is used for your analyses, please explain in detail your analytical pro-
cedure on the back of the enclosed data form.
41
-------
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 Vo\. 42, No. 160, Part 11, Thursday, August 18, 1977, pp 41784-41786. This amend-
ed 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 pres-
sure sideways. From the ampule pipette exactly 5 ml of the reference sample into a 100-ml beak-
er. Add 25 ml absorbing solution to the beaker; adjust the pH to 9-12 (using pH paper as indi-
cated in Section 4.2 of the method) by dropwise addition of sodium hydroxide (1N). Quanti-
tatively 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 N02, dry basis, corrected to standard conditions, mg/dscm), using
Equation 7-4. A value of 2000 ml should be used for VK 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 (MO 77)
Environmental Monitoring Systems Laboratory
Environmental Protection Agency
Research Triangle Park. N.C. 27711
ATTN: Ellen H. Strelb
If other than EPA Method 7 is used for your analyses, please explain in detail your analytical pro-
cedure on the back of the enclosed data form.
42
-------
COAL AUDIT PROGRAM INFORMATION
1. There is approximately 50 grams 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
in weight percent with gross calorific value reported as BTLJ/lb.
3. All methods used in the analysis of these coal samples should follow
American Society for Testing and Materials (ASTM) recommended 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 is necessary to calculate on
a dry basis any of the other three parameters. Analysis of sulfur is
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. Streib
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 method audit,
please call (919/541-7834).
43
-------
INSTRUCTIONS FOR USING EPA METHOD 3 AUDIT MATERIALS
Equipment Supplied with Audit Kit
(1) Small gas cylinder containing four liters of gas
(2) Small needle with rubber septum (inside gas cylinder cap)
(3) Tedlar bag, 0.5 £ with valve and Tygon tubing
Equipment to be Supplied by Participant
(1) Orsat analyzer
(2) Vacuum source vented to hood for evacuating bag
(3) Container filled with water for checking Tedlar bag for leaks
Procedure
CAUTION: Before performing Step 1 below, become familiar with the
operation and performance of the valve on the Tedlar bag. Turning the valve
stem clockwise closes the valve. Turning it counterclockwise opens it. This
type valve leaks at the stem base when NOT fully closed. Thus, when conducting
the leak-check and when the bag is not being filled or emptied, the VALVE MUST
BE FULLY CLOSED. Further, when gas is withdrawn from the bag, the bag wall
opposite the valve must not block the valve opening because air will then leak
into the gas stream at the valve stem base.
(1) Fully open the valve and gently blow into the bag until it is fully
inflated. Do not overpressurize the bag!
(2) Close the valve fully.
(3) Immerse the bag in water and determine if it is leaking (as shown
by the presence of air bubbles). If the bag, itself, is leaking the leak
may be repairable by placing transparent tape over it. If the leak cannot be
repaired, do not proceed further. Contact Ms. Ellen Streib at 919/541-7834
for instructions.
(4) After the bag passes the leak-check, evacuate it completely and
close the valve. (CAUTION: Gas mixture is toxic (carbon monoxide), do not
evacuate by mouth.)
(5) Take the cap from the gas cylinder and remove the plastic tube
inside. Place the plastic tube into hole in push button of can with the
septum end facing outward.
44
-------
(6) Insert the septum into the Tygon tubing on the bag valve, open the
bag valve and carefully fill the bag with cylinder gas. Do not overpressurize!
Fully close the valve and remove the septum/can from the Tygon tubing. Fully
evacuate the bag.
(7) Refill the bag with cylinder gas. Fully close the valve and remove
the septum/can assembly from the Tygon tubing.
(8) Attach the bag to the Orsat analyzer. Open the bag valve fully and
draw 100 cc of gas into the analyzer. Vent the Orsat gas sample to the atmos-
phere and then refill the Orsat with gas from the Tedlar bag.
(9) Analyze for C02, 02 and CO as described in Sections 4.2.5, 4.2.6 and
4.2.7 of EPA Method 3.
(10) Record the results on the data card enclosed with the sample.
(11) Evacuate the bag completely and repeat Steps 6 through 10.
Send the data card and the Tedlar bag to the address below. (The cylinder
gas can should not be returned.)
Ms. Ellen Streib
Quality Assurance Division (MD-77)
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
NOTE: Site number will always be 001 except when other Orsat apparatus or
participants are using the same gas sample. The extra apparatus or partici-
pants should be labeled 002, 003, etc.
•w-U S. GOVERNMENT PRINTING OFFICE: 1983/759-102/0798
------- |