<>EPA
United States
Environmental Protection
Agency
Environmental Monitoring and Support
Laboratory
Research Triangle Park NC 27711
EPA-600/4-80-029
May 1980
Research and Development
A Summary of the
Intel-laboratory Source
Performance Surveys
for EPA Reference
Methods 5, 6, and 7
1978
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U S Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are
1 Environmental Health Effects Research
2 Environmental Protection Technology
3 Ecological Research
4 Environmental Monitoring
5 Socioeconomic Environmental Studies
6 Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8 "Special" Reports
9 Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL MONITORING series
This series describes research conducted to develop new or improved methods
and instrumentation for the identification and quantification of environmental
pollutants at the lowest conceivably significant concentrations. It also includes
studies to determine the ambient concentrations of pollutants in the environment
and/or the variance of pollutants as a function of time or meteorological factors.
This document is available to the public through the National Technical Informa-
tion Service. Springfield, Virginia 22161
-------
A SUMMARY OF THE INTERLABORATORY SOURCE PERFORMANCE SURVEYS
FOR EPA REFERENCE METHODS 5, 6, AND 7-1978
by
R. G. Fuerst and M. R. Midgett
Quality Assurance Division
Environmental Monitoring Systems Laboratory
U. S. Environmental Protection Agency
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 report has been reviewed by the Environmental Monitoring Systems
Laboratory, U.S. Environmental Protection Agency, and approved for publication
Mention of trade names or commercial products does not constitute endorsement
or recommendation for use.
-------
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 regula-
tions and to evaluate the effectiveness of health and environmental protection
efforts through the monitoring of long-term trends. The Environmental Moni-
toring Systems Laboratory, Research Triangle Park, North Carolina, has respon-
sibility 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 concern of this study is to report the results of surveys in
the national quality assurance program for stationary source tests. Surveys
were designed to estimate the analytical and computational accuracy that can
be expected with EPA Method 5 (dry gas meter only), Method 6 (sulfur dioxide)
and Method 7 (nitrogen oxides). Statistical analysis was used to characterize
the data.
Thomas R. Hauser, Ph.D.
Director
Environmental Monitoring Systems Laboratory
Research Triangle Park, North Carolina
m
-------
ABSTRACT
A national survey of methods in stationary source tests was conducted in
1978 by the Quality Assurance Division of the Environmental Monitoring Systems
Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina. In this program, quality assurance samples were sent to interested
participants for the measurement of a gas volume (Method 5, dry gas meter
only) or the analysis of liquid samples simulating collected sulfur dioxide
and nitrogen oxides (Method 6 and 7, respectively). Each participant returned
the analytical results to the Source Branch, Quality Assurance Division, for
evaluation. An individual report was returned to each participant after
processing.
This report summarizes the survey results for those three source test
methods.
-------
CONTENTS
Foreword iii
Abstract iv
Tables vi
Acknowledgments vii
1. Introduction 1
2. Summary 3
3. Recommendations 5
4. Survey Design 6
Survey procedures 6
Prospective participants 7
Preparation and distribution of survey materials
for methods 5, 6, and 7 7
5. Statistical Data Handling 9
6. Discussion of Method 5 Results 10
7. Discussion of Method 6 Results 15
8. Discussion of Method 7 Results 20
References 25
Appendices
A. Method 5 DGM data summary 26
B. Method 6 S02 data summary 28
C. Method 7 NO data summary 38
X
-------
TABLES
Number Page
1 Method 5 Survey 0578 - Laboratory Distribution 10
2 Method 5 Survey 1078 - Laboratory Distribution 11
3 Method 5 - Percent Difference from True Value .......... 11
4 Method 5 Survey 0578 - Frequency Distribution
of Percent Difference 12
5 Method 5 Survey 0578 - Summary Statistics 12
6 Method 5 Survey 1078 - Frequency Distribution
of Percent Difference 12
7 Method 5 Survey 1078 - Summary Statistics 13
8 Method 6 Survey 0578 - Laboratory Distribution 15
9 Method 6 Survey 1078 - Laboratory Distribution .... 15
10 Method 6 Survey 0578 - Frequency Distribution
of Absolute Percent Difference 17
11 Method 6 Survey 1078 - Frequency Distribution
of Absolute Percent Difference 17
12 Method 6 Survey 0578 - Summary Statistics 18
13 Method 6 Survey 0578 - Summary Statistics (Outliers Removed) ... 18
14 Method 6 Survey 1078 - Summary Statistics 19
15 Method 6 Survey 1078 - Summary Statistics (Outliers Removed) ... 19
16 Method 7 Survey 0578 - Laboratory Distribution 20
17 Method 7 Survey 1078 - Laboratory Distribution 20
18 Method 7 Survey 0578 - Frequency Distribution
of Absolute Percent Difference 22
19 Method 7 Survey 1078 - Frequency Distribution
of Absolute Percent Difference 22
20 Method 7 Survey 0578 - Summary Statistics 23
21 Method 7 Survey 0578 - Summary Statistics (Outliers Removed) ... 23
22 Method 7 Survey 1078 - Summary Statistics 24
23 Method 7 Survey 1078 - Summary Statistics (Outliers Removed) ... 24
VI
-------
ACKNOWLEDGMENTS
Our deep appreciation is extended to each of the survey participants. In
addition, we also express thanks to Ms. Ellen Streib who analyzed the survey
samples under our Acceptance Testing Program, and the programmers of the Data
Management and Analysis Division for providing the data management systems
necessary to store and summarize the survey data.
vn
-------
SECTION 1
INTRODUCTION
One of the responsibilities of the Environmental Protection Agency (EPA)
is to provide adequate methodology as a means to monitor compliance with
emission regulations. But to insure consistent results using EPA methodology,
a good quality assurance program must be maintained.
The Source Branch (SB) of the Quality Assurance Division (QAD) of the
Environmental Monitoring Systems Laboratory (EMSL), EPA provides a nationwide
quality assurance program for air pollution measurement systems. As part of
this overall program, the QAD began in 1977 to periodically provide reference
samples for analysis by any contractor, industrial, or governmental laboratory
that wished to participate in its program (1). This program had three main
purposes:
• to verify that the analytical and computational parts of the specific
reference methods were being properly used,
• to assist wherever possible to improve the quality of the measurement
being made,
• to aid the participating laboratories in assessing their analytical
performance relative to that of other laboratories conducting similar
analyses.
These goals were realized by sending specific performance materials to
interested laboratories for analysis.
-------
In the two source test method surveys conducted in May and October of
1978, the technique of volume measurement by a dry gas meter was examined.
This method is essential to the Method 5 source sampling train (2). Also
examined were the analytical and computational parts of Method 6 for sulfur
dioxide (S09) (3) and Method 7 for nitrogen oxides (NO ) (4). This report
L. ' X
describes the preparation and evaluation of these tests.
-------
SECTION 2
SUMMARY
These quality assurance surveys were conducted in May and October of 1978
by the Quality Assurance Division of EPA's Environmental Monitoring Systems
Laboratory. They included participants from industry, contracting firms,
universities, foreign countries, and governmental agencies. Comparative data
from past surveys (1977) are also contained in this report.
In examining the results of the Method 5 surveys, the investigators found
that an average of 55% of the laboratories requesting samples actually returned
data. The reported results from responding laboratories showed that 55% in
survey 0578 (May 1978) came within 5% of the true value for the requested
volume measurements, while in survey 1078 (October 1978) 70% were within 5%.
For Method 6, an average of 56% of those laboratories requesting samples
returned data for both surveys. Of those laboratories returning data for
survey 0578, 50% came within 1.28% of the true value in the analysis of all
sample concentration levels, while in survey 1078, 50% were within 1.71%.
For Method 7, an average of 53% of those laboratories requesting samples
returned data for both surveys. Fifty percent of the responding laboratories
in survey 0578 came within 6.18 percent of the true value on all sample con-
centration levels while in survey 1078, 7.98 percent were able to do this.
(Outliers, i.e., anomolous values, were removed from the above summary figures.)
Comparing the results of the two Method 6 surveys of 1978 with two source
surveys conducted in 1977, the authors found that in 1977, 50% of the partici-
pants analyzing samples were 2.15 and 1.69% or less from the true value when
the results of all sample concentration levels were combined, while in 1978
-------
this same group showed results containing differences of 1.28 and 1.71%.
Calculating a weighted value for all four surveys based on the number of
samples taken in each survey gives a 50% value of 1.7%.
For the Method 7 study, in two surveys conducted in 1977, 50% of the par-
ticipants came within 15.14 and 7.41% of the true value, respectively, while in
the two surveys conducted in 1978, 50% came within 6.18 and 7.98%, respectively
Except for the Method 7 result from the first survey (15.1%), the last three
surveys gave a weighted value of 7.2% for 50% of the participants.
Whether these percent responses indicate a definite trend in the analyti-
cal abilities of users of these two source methods will be judged from future
surveys.
-------
SECTION 3
RECOMMENDATIONS
To create a sample repository, the Quality Assurance Division of the
Environmental Monitoring Systems Laboratory intentionally produced an over-
supply of samples for the surveys of EPA Methods 6 and 7 discussed in this
report. These samples are available to any laboratory having a legitimate
need for them, such as training new analysts and conducting periodic external
quality control checks of the laboratory. Included with these practice
samples is a statement of true concentration with no requirement for return
of data to EPA. We recommend that all participants make use of this sample
repository, as it may help laboratories to increase their overall analytical
skills with these particular EPA reference methods.
-------
SECTION 4
SURVEY DESIGN
The source sample surveys discussed in this report incorporate the experi-
ence gained from previous source surveys in such areas as survey procedures,
prospective participants, categorization and preparation of survey materials,
and data handling.
SURVEY PROCEDURES
All surveys began with a master list of prospective laboratories which
had in the past participated or indicated a wish to take part in such a program.
Prospective participants were sent a description of the survey methods
and instructions for participation. Through a response card, each laboratory
indicated if it wished to participate. Response cards were returned to the
appropriate EPA Regional Quality Control Coordinator (RQCC) who collected,
logged, and forwarded them to the EPA contractor preparing the survey materials
for QAD. Participating laboratories were assigned an identification number to
facilitate storage of their data in the computer's data bank and to maintain
the confidentiality of each participant's results. At a prearranged date,
requested survey materials were shipped to the participants with the instruc-
tions for sample analysis, a blank data card to report the completed analysis
values, and a mailing label for return of the data card to QAD. When the sur-
vey was completed, the participants received a computer data sheet containing
the results of their performances. At the completion of all the studies, a
summary of the total results will be published without reference to any specific
laboratory.
-------
PROSPECTIVE PARTICIPANTS
Using a previously compiled master list of laboratories from past surveys,
invitations to participate in the upcoming source surveys 0578 (May) and 1078
(October) were sent to all volunteers who had previously participated in one or
more of the source surveys. Other laboratories were added to the master list
through their direct contact with the SB/QAD or the RQCC.
PREPARATION AND DISTRIBUTION OF SURVEY MATERIALS FOR METHODS 5, 6, AND 7
To provide a check on the calibration of the dry gas meter used in the
Method 5 stack sampling train, a critical orifice device was developed to pass
a certain volume of air through the dry gas meter when the measured vacuum on
the orifice was at least 16 inches of mercury. This device allows an analyst to
compare a volume measured at his location with one measured at an EPA location.
Volumes measured at both locations are compared to the original calibration of
the device, compensated for the effect of ambient temperature and pressure on the
measurement at both locations. After initial calibration by an EPA contractor,
it is recalibrated by an EPA laboratory which rejects any device whose volume
measurement does not fall within ± 2% of the original calibration. This process
of verification of the original calibration is known as Acceptance Testing.
Participants in the Method 5 survey were instructed to insert the critical
orifice device in the probe connection of their gas sampling meter box, and,
after a warmup period, to take three 15 min volume measurements. Using equation
5-1 of Method 5, they were told to calculate each of the three volumes in cubic
meters and record them on the data card along with other pertinent information
concerning sampling conditions. They were then to mail the device and data
card back to EPA for comparison of volumes. Some meter boxes were equipped
with diaphragm pumps that cannot pull 16 inches of mercury vacuum with these
audit devices. Since a certain vacuum is necessary to produce critical flow,
new devices were constructed with smaller orifice openings that allow the
pumps to pull the required vacuum.
-------
For all surveys of Methods 6 and 7, five different concentration levels
of simulated source sulfur dioxide (SCL) and nitrogen oxides (NO ) samples were
^- )\
prepared. These solutions enabled the participants to analyze and calculate
different concentration levels of S09 and NO , using Methods 6 and 7. The true
L. X
values of these samples were based on theoretical concentrations calculated
from gravimetric preparations and certain assumed volume measurements. After
sample solutions were made, their concentrations were verified with the appro-
priate methods. This step was initially conducted by contractor personnel and
then by EPA personnel, via Acceptance Testing.
Each sample solution, approximately 20 ml, was sealed in a 25 ml glass
ampoule, and five different concentration levels were shipped to the partici-
pating laboratories. The ampoules containing NO samples were autoclaved to
/\
destroy bacteria that might possibly attack the solutions.
Instructions for the Method 6 samples prescribed that 5 ml of the test
solution be diluted to 100 ml through the addition of 30 ml of 3% hydrogen
peroxide (H-O-) and distilled water. An aliquot of this solution was then
titrated with barium perchlorate (BaCClO.]^) in the presence of thorin
indicator to a characteristic peach color endpoint. To complete Method 6
calculations, the participants assumed they had an original sample volume
of 100-ml,
stack gas.
of 100-ml, and had sampled 21 x 10 3 DSCM (dry standard cubic meter) of
The analysis of Method 7 samples involved dilution of a 5-ml aliquot of
the original test sample with 25 ml of absorbing reagent, adjustment of the pH
to approximately 9 to 12, and dilution to 50 ml with distilled water. After a
digestion procedure, a colorimetric analysis followed. To complete Method 7
calculations, the participants assumed they had sampled 2000 ml of stack gas.
In each of our surveys, the samples were number coded. The key for the
five concentration levels was based on the first digit of the sample number.
-------
SECTION 5
STATISTICAL DATA HANDLING
Establishing performance criteria in order that participants could evalu-
ate their reported data was a major concern of the survey program. The ideal
approach would have been to develop statistics on a large number of analyses
of the same sample made by laboratories across the nation at different times.
Since this was not possible at the start of our survey program, we initially
developed performance ranges. These performance ranges, based on the results
of collaborative test studies, defined an acceptable variability around the
known concentration of each sample. Construction of these ranges was based
on the 1977 data summary of the Method fi and 7 surveys (1).
This definition of performance ranges is arbitrary, however, because the
ranges are not based on the same statistical population of volunteers as was
used in the surveys.
The 1978 summary report also used information collected from the previous
surveys of 1977. From the participants' reported data, a frequency distribution
of percent difference was next devised showing how well all participants dH in
those surveys when their results were compared against the EPA true value for
each concentration level. We chose to develop frequency distributions to aid
the participant in his self-evaluation instead of the performance ranges we used
initially. This method allowed comparisons and self-evaluations to be based on
results taken from the same participants.
Although statistical comparisons are made between different surveys for
the same pollutant, each laboratory participating in the surveys can use
different analysts for each survey. Thus, any increase in overall survey
accuracy could mean the analyst is becoming more familiar with the methods,
or that better analysts are being used.
-------
SECTION 6
DISCUSSION OF METHOD 5 RESULTS
The distribution of the types of laboratories responding to surveys 0578
and 1078 about Method 5 (dry gas meter only) is shown in Tables 1 and 2 below.
TABLE 1. METHOD 5 SURVEY 0578
Laboratory Distribution
Contractor Industrial Foreign Federal State Local Total
Laboratories
requesting 48 18 1 2 9 3 81
samples
Laboratories
returning 25 6 1 1 8 2 43
data
The fact that a large percentage of participants request survey samples
but do not analyze and return their data in this and the other source surveys
is puzzling. Probable causes for not returning data are either conflicts
with scheduled work or oversights -- failure to schedule the analysis of the
QA samples during the alloted time period. These two reasons would not effect
the overall statistics of the survey. However, if the sample results were not
returned due to suspected inaccuracy of the data the survey statistics would
be biased. In case analytical problems do arise, the survey participant may
obtain a set of practice survey samples with their specific concentrations
listed. These samples come from previous surveys, of which concentrations
have been released to the public.
1C
-------
TABLE 2. METHOD 5 SURVEY 1078
Laboratory Distribution
Contractor Industrial Foreign Federal State Local Total
Laboratories
requesting 55 28 6 5 14 6 114
samples
Laboratories
returning 30 14 4 2 11 4 65
data
Participants in this survey were instructed to take three 15 min volume
3
measurements, calculate each volume at standard conditions in m using equa-
tion 5-1 of Method 5, and report their results on the blank data card provided.
Since the compared results have been reported in percents, the participant
can readily discover his standing in the overall group of participants. Percent
difference has been calculated as follows:
PD = RVT" TV x 100 (1)
where: PD = absolute percent difference
RV = reported value
TV = true value
100 = factor to change decimal to percent
Table 3 (below) describes the participants' degree of accuracy when their
reported values were compared against the true value determined by EPA and
contractor measurements.
TABLE 3. METHOD 5 - ABSOLUTE PERCENT DIFFERENCE
Survey
0578
1078
1%
14
15
2%
21
31
3%
31
50
4%
43
61
5%
52
68
6%
64
78
7%
71
84
8%
79
88
9%
84
90
10%
89
94
11
-------
This table shows that in survey 0578, 52% of the participants were able
to measure within 5% of the EPA value, while in survey 1078, 68% were able to
do so.
A summary of all individual measurements received from survey 0578 for
Method 5 is tabulated in Tables 4 and 5 (below). All outliers in this report
were removed according to Chauvenet's Criterion (5). All outlier tests are
used to remove suspected anomalous values from the various groups of survey
data.
TABLE 4. METHOD 5 SURVEY 0578 - FREQUENCY DISTRIBUTION OF PERCENT DIFFERENCE
All data
Outliers removed
10%
-9.6
-9.2
20%
-7
-7
.9
.9
30%
-6.8
-6.3
40%
-5.2
-5.2
50%
-4.1
-4.1
60%
-3.5
-3.5
70%
-2
-2
.1
.3
80%
-0.
-0.
1
7
90%
1.0
0.3
TABLE 5. METHOD 5 SURVEY 0578 - SUMMARY STATISTICS
MIN MAX MEDIAN MEAN STD DEV
SKEWNESS
All data 180 -18.3 86.3 -4.1 -3.1 ±12.6 5.86
Outliers removed 172 -15.1 7.2 -4.1 -4.2 ± 4.2 0.29
A summary of all individual measurements received from survey 1078 for
Method 5 is tabulated in Tables 6 and 7 below.
TABLE 6. METHOD 5 SURVEY 1078 - FREQUENCY DISTRIBUTION OF PERCENT DIFFERENCE
10% 20% 30% 40% 50% 60% 70% 80% 90%
All data -9.0 -6.2 -5.2 -3.7 -3.0 -2.4 -1.7 -1.0 0.1
Outliers removed -8.9 -6.2 -5.2 -3.7 -3.0 -2.4 -1.7 -1.0 0.0
12
-------
TABLE 7. METHOD 5 SURVEY 1078 - SUMMARY STATISTICS
All data
Outliers removed
n
255
249
MIN
MAX
MEDIAN
MEAN
STD
DEV
(o/\ fo/\ f o/\ fo/\ ( o/\ Ql^FLJNF^
\ /o J \ lo J \ '° j \ i° } \'° I ^ *^~ « 1 ' ^ J J
-27.2
-12.4
45.
4.
2
2
-3.
-3.
0
0
-3.2
-3.5
± 6
± 3
.3
.2
3
0
.44
.14
In Tables 4 and 5 for survey 0578, 70% of the reported data were found to
lie in a range of -15.1 and -2.3% difference from the accepted EPA value. For
survey 1078 the 70% range was -12.4 to -1.7%. In both cases when outliers
were removed -- 8 for survey 0578 and 6 for survey 1078 -- no large effect was
noted on this value. The overall means of both studies were within 0.7% when
outliers were removed.
Since the skewness values were near zero and the median and mean values of
each study were close to each other, both studies may be assumed to be normally
distributed.
There was, however, a negative mean percent difference calculated for both
studies (Tables 5 and 7). The weighted mean of both studies was -3.4% with
outliers removed. When calibrated by an EPA laboratory, the orifices obtained
an overall -0.5 ± 1.2% difference when compared to the original calibration.
All percent difference values are based on the contractor's original calibration.
A negative bias can be caused by leakage in the wet test, dry test meter,
or by infrequent calibration of both meters. There are several helpful publi-
cations on troubleshooting the dry test meter (6-8). During calibration of
each device before each study, these devices must maintain ± 2% of the original
contractor calibration value, or they are sent for cleaning and recalioration.
This method follows EPA's Acceptance Testing Program, in which the EPA contrac-
tor's determinations are verified by an independent EPA laboratory. Although
the volume determinations made with the meter boxes are expected to be within
± 2%, all participants' values have been calculated back to the original con-
tractor calibration. Our laboratory plans to continue examining the results
obtained from use of these critical orifices as calibration checks.
13
-------
All results of these two Method 5 surveys are grouped according to the
Increasing concentration levels reported in Appendix A, so that individuals
may note their exact placement in the survey results.
14
-------
SECTION 7
DISCUSSION OF METHOD 6 RESULTS
The distribution of participants in the Method 6 surveys, 0578 and 1078,
is shown in Tables 8 and 9 below.
TABLE 3. METHOD 6 SURVEY 0578
i Laboratory Distribution
Contractor Industrial Foreign Federal State Local University Total
Laboratories
requesting 72 19 2 4 15 8 0 120
samples
Laboratories
returning 34 12 2 1 86 0 63
data
TABLE 9. METHOD 6 SURVEY 1078
Laboratory Distribution
Contractor Industrial Foreign Federal State Local University Total
Laboratories
requesting 56 29 5 3 15 7 1 116
samples
Laboratories
returning 26 20 2 1 14 5 1 69
data
Participants were instructed to use Method 6 for all analyses and report
their results, based on equation 6-2 of Method 6 (mg S02/DSCM), on a blank data
card.
15
-------
Tables 10 and 11 are frequency distributions of the absolute percent
differences between the participant's reported values and EPA values for each
concentration level.
Table 10 reveals that 50% of the reported results for all sample concentra-
tion levels of Method 6 survey 0578 were less than or equal to an absolute per-
cent difference of 1.28. The bottom line of this table compiles all the data
regardless of concentration. Table 10 is also useful for self-evaluation. For
instance if a participant reported a value for sample 4 that was more than 1.81%
from the true value, he would see that results from 70% of the participants were
closer to the true value than his. The Min and Max values listed in Tables 10
and 11 show the lowest and highest individual percent differences reported in
the survey.
To allow individuals to note their exact placement in the survey, all re-
sults are grouped in Appendix B according to increasing order of concentration
levels reported.
Tables 12 and 13 list summary statistics on survey 0578 about Method 6,
with and without outliers. Tables 14 and 15 list summary statistics for survey
1078 (Method 6) with and without outliers. Equations 3, 4, and 5 were used to
calculate the statistics in these tables:
% Coefficient of variation = -5- x 100; (2)
X
Z(X- - X)3
Skewness = 5 i (3)
n(s)J
Accuracy = Ml ~& 5 x 100 . (4)
where: s = one standard deviation
X = mean value
X. = individual value
Mi = median value
6 = true value
n = number of values
16
-------
TABLE 10. METHOD 6 SURVEY 0578 - FREQUENCY DISTRIBUTION OF ABSOLUTE PERCENT DIFFERENCE
NO.
Sample 3 63
Sample 4 63
Sample 5 63
Sample 7 63
Sample 9 63
All Samples 315
MIN
0.01
0.03
0.09
0.03
0.05
0.01
10%
0.22
0.20
0.17
0.15
0.37
0.21
20%
0.41
0.44
0.34
0.24
0.79
0.42
30%
0.67
0.61
0.60
0.49
1.21
0.69
40%
0.99
0.83
0.73
0.79
1.68
1.00
50%
1.28
1.08
1.09
1.00
1.94
1.28
60%
1.65
1.47
1.45
1.27
2.99
1.72
70%
2.30
1.81
1.85
1.77
4.35
2.41
80%
3.45
3.75
2.73
2.58
6.03
3.77
90%
4.11
4.50
5.20
4.54
13.5
6.03
MAX
96.9
96.9
96.9
96.9
96.9
96.9
MEAN
3.82
3.79
3.69
3.57
6.47
4.27
TABLE 11.
METHODS
6 SURVEY
1078
- FREQUENCY DISTRIBUTION OF
ABSOLUTE
PERCENT DIFFERENCE
NO.
Sample 1 69
Sample 3 69
Sample 5 68
Sample 6 69
Sample 9 69
All Samples 344
MIN
0.00
0.10
0.00
0.00
0.00
0.00
10%
0.20
0.25
0.37
0.52
0.23
0.30
20%
0.43
0.52
0.56
1.25
0.49
0.58
30%
0.67
0.76
0.80
1.97
0.68
0.86
40%
0.88
1.00
1.19
2.43
0.80
1.19
50%
1.25
1.20
1.77
3.93
1.25
1.71
60%
1.58
1.53
2.38
4.52
2.05
2.39
70%
2.01
2.55
2.76
6.10
2.57
3.11
80%
3.33
3.30
4.00
9.97
3.45
4.47
90%
4.94
4.78
6.42
17.4
5.22
7.12
MAX
23.5
23.4
867.0
48.9
48.1
867.0
MEAN
2.36
2.40
16.0
6.60
3.35
6.12
-------
TABLE 12. METHOD 6 SURVEY 0578 - SUMMARY STATISTICS
Parameter
Samples (n)
True value*
Mean*
Median*
Std. dev.*
% Coef. var.
Skewness
Accuracy
3
63
686.30
672.
683.
87.7
13.1
-6.43
-0.50
4
63
2478.
2427.
2460.
318.
13.
-6
-0
Sample No.
5
63
50 1258.30
1229.
1250.
160.
1 13.0
.36 -6.48
.72 -0.67
7
63
1906.00
1866.
1894.
242.
13.0
-6.56
-0.62
9
61
190.70
194.
192.
11.9
6.1
0.58
0.58
*A11 sample
TABLE 13.
concentrations are
METHOD 6 SURVEY
in mg
0578 -
S02/DSCM.
SUMMARY STATISTICS
(OUTLIERS
REMOVED)
Parameter
Samples (n)
True value*
Mean*
Median*
Std. dev.*
% Coef. var.
Skewness
Accuracy
3
62
686.30
683.
683.
27.6
4.04
0.83
-0.47
4
62
2478.
2465.
2461.
102.
4.
1 .
-0.
Sample No.
5
62
50 1258.30
1249.
1250.
49.0
15 3.93
40 0.81
69 -0.66
7
62
1906.00
1895.
1894.
71.4
3.77
0.76
-0.62
9
63
190.70
193.
192.
29.9
15.5
-3.04
0.58
*A11 sample concentrations are in mg SO-/DSCM.
As previously stated, participants in survey 0578 and 1078 each received
a set of five samples. These samples represented five different concentration
levels of S0?. The sample numbers were randomized from the numbers 0 through 9
18
-------
TABLE 14. METHOD 6 SURVEY 1078 - SUMMARY STATISTICS
Parameter
Samples (n)
True value*
Mean*
Median*
Std. dev.*
% Coef. var.
Skewness
Accuracy
1
69
2555.00
2551.
2550.
109.
4.28
1.64
-0.22
3
69
1335.00
1332.
1329.
57.1
4.28
1.66
-0.47
Sample No.
5
68
572.00
645.
573.
603.
93.5
7.83
0.10
6
69
152.50
158.
156.
15.0
9.45
1.13
2.30
9
69
1754.00
1737.
1751.
132.
7.59
-3.23
-0.14
*A11 sample
TABLE 15.
concentrations are
METHOD 6 SURVEY
in mg S02
/DSCM.
1078 - SUMMARY STATISTICS
(OUTLIERS
REMOVED)
Parameter
Samples (n)
True value*
Mean*
Median*
Std. dev.*
% Coef. var.
Skewness
Accuracy
1
67
2555.00
2547.
2550.
67.2
2.64
-0.78
-0.22
3
67
1335.00
1331.
1329.
34.5
2.59
-0.13
-0.47
Sample No.
5
67
572.00
572.
572.
42.5
7.43
-4.02
0.00
6
67
152.50
158.
156.
11.4
7.23
0.30
2.30
9
67
1754.00
1743.
1752.
69.2
3.97
-1.21
-0.14
*A11 sample concentrations are in mg S02/DSCM.
From an examination of Tables 13 and 15, no bias is evident, as reflected
by the low skewness value. Lack of bias is also suggested by the small diff-
erence between the mean and the true value and the low accuracy values. The
low skewness value, and closeness of the median and true value also suggest a
normally distributed sample population.
19
-------
SECTION 8
DISCUSSION OF METHOD 7 RESULTS
Tables 16 and 17 (below) show the distribution of samples in surveys 0578
and 1078 for the NO method.
X
TABLE 16. METHOD 7 SURVEY 0578
Contractor
Laboratories
requesting 62
samples
Laboratories
returning 29
data
Laboratory
Distribution
Industrial Foreign Federal State Local
9 2
7 2
3 76
1 3 4
University Total
0 89
0 46
TABLE 17. METHOD 7
SURVEY 1078
Contractor
Laboratories
requesting 49
samples
Laboratories
returning 21
data
Laboratory
Distribution
Industrial Foreign Federal State Local
23 5
13 3
4 11 7
1 8 6
University Total
0 99
0 52
Participants were instructed to use Method 7 for the analysis and report
their results based on equation 7-4 of the method as mg NO /DSCM. Under the
X
section called Calculations in Method 7, the analyst was instructed to report
the concentration of the NO samples as N09.
A £-
20
-------
Tables 18 and 19 are frequency distributions of the absolute percent
differences between the participant's reported values and the EPA values for
each concentration level.
Table 18 reveals that 50% of the reported results in surveys 0578 and 1078
for all sample concentration levels of Method 7 were less than or equal to an
absolute percent difference of 6.18 and 7.91, respectively. Like Method 6,
Tables 18 and 19 can be used for self-evaluation. For example, if a participant
reported a value for sample 8 (Table 18) that was more than 15.7% from the true
value, he would see that results from 70% of the participants were closer to the
true value than his.
To allow individuals to note their exact placement in the survey, all re-
sults are grouped in Appendix C according to increasing order of concentration
levels reported.
Tables 20 and 21 list summary statistics for Method 7, survey 0578, with
and without outliers.
Tables 22 and 23 present summary statistics for survey 1078, Method 7.
Examining Tables 21 and 23, no bias is evident, as reflected by a low
skewness value and the closeness of the median and mean value.
21
-------
TABLE 18. METHOD 7 SURVEY 0578 - FREQUENCY DISTRIBUTION OF ABSOLUTE PERCENT DIFFERENCE
,
Sample 1
Sample 2
Sample 6
Sample 8
Sample 9
All Samples
NO.
46
46
46
46
46
230
WIN
0.27
0.21
0.08
0.05
0.12
0.05
10%
0.35
1.07
1.11
0.40
0.55
1.07
20%
1.89
1.60
1.90
2.31
1.65
1.89
30%
3.03
2.67
2.71
3.06
2.13
2.73
40%
4.08
3.94
3.72
5.51
2.68
4.08
50% 60%
5 . 46 6.13
7.46 9.06
5.86 6.52
7.48 10.6
4.11 7.56
6.18 8.29
70%
9.14
12.1
9.27
15.7
10.5
11.8
80%
18.7
21.5
13.6
21.3
14.9
21.3
90%
45.7
65.7
52.2
49.6
48.9
64.3
MAX
142.
123.7
107.
101.
107.8
142.
MEAN
17.3
21.3
17.3
18.3
15.7
18.0
no
ro
TABLE
19.
METHODS 7
SURVEY
1078
- FREQUENCY DISTRIBUTION OF
ABSOLUTE
PERCENT
DIFFERENCE
Sample 1
Sample 2
Sample 3
Sample 5
Sample 7
All Samples
NO.
52
52
52
52
52
260
MIN
0.20
0.56
0.55
0.24
0.26
0.20
10%
0.77
1.78
1.53
0.87
1.70
1.45
20%
1.74
3.74
2.82
2.60
3.30
3.04
30%
3.12
4.79
3.56
3.90
4.72
4.34
40%
4.54
6.08
4.49
5.93
6.18
5.91
50% 60%
6.69 7.91
7.82 9.23
6.33 9.26
7.95 11.4
8.59 10.8
7.91 9.83
70%
11.5
13.3
13.3
14.6
13.8
13.9
80%
15.6
17.6
18.7
18.8
17.2
20.4
90%
54.6
47.3
32.6
26.5
31.5
47.3
MAX
320.
131.
130.
260.
135.
320.
MEAN
22.6
18.9
15.9
19.5
19.0
19.2
-------
TABLE 20. METHOD 7 SURVEY 0578 - SUMMARY STATISTICS
Parameter
Samples (n)
True value*
Mean*
Median*
Std. dev.*
% Coef. var.
Skewness
Accuracy
1
46
703.30
729.
701.
251.
34.4
1.61
-0.27
2
46
93.80
103.
95.2
36.7
35.7
1.68
1.44
Sample No.
6
46
515.70
524.
524
171.
32.6
0.90
1.66
8
46
937.70
958.
941.
301.
31.4
0.89
0.39
9
46
328.20
341
334.
101.
29.7
1.48
1.72
*A11 sample
TABLE 21.
concentrations are
METHOD 7 SURVEY
in mg NO
/DSCM.
X
0578 - SUMMARY STATISTICS
(OUTLIERS
REMOVED)
Parameter
Samples (n)
True value*
Mean*
Median*
Std. dev.*
% Coef. var.
Skewness
Accuracy
1
43
703.30
674.5
697.
140.
20.7
-2.14
-0.85
2
42
93.80
93.0
94.2
19.3
20.8
0.29
0.37
Sample No.
6
42
515.70
498.3
522.
91.9
18.5
-2.16
1.12
8
42
937.70
916.
940.
162.
17.7
-1.28
0.20
9
43
328.20
319.
333.
56.9
17.9
-2.13
1.49
*A11 sample concentrations are in mg NOX/DSCM.
23
-------
TABLE 22. METHOD 7 SURVEY 1078 - SUMMARY STATISTICS
Parameter
Samples (n)
True value*
Mean*
Modian*
Std. dev.*
% Coef. var.
Skewness
Accuracy
1
52
246.50
274.
257.
133.
43.6
3.55
4.14
2
52
730.
776.
764.
251.
32.
0.
4.
Sample No.
3
52
60 880.30
925.
910.
256.
4 27.7
98 1.82
61 3.33
5
52
123.20
137.
129.
52.8
38.4
3.88
4.30
7
52
457.70
491.
481.
161.
32.7
1.76
5.17
*A11 sample
TABLE 23.
concentrations are
METHOD 7 SURVEY 1
in mg
078 -
NO/DSCM.
A
SUMMARY STATISTICS
(OUTLIERS
REMOVED)
Parameter
Samples (n)
True value*
Mean*
Median*
Std. dev.*
% Coef. var.
Skewness
Accuracy
1
51
246.50
259.
256.
79.1
30.5
0.34
3.81
2
50
730.
740.
761.
177.
23.
-1.
4.
Sample No.
3
50
60 880.30
885.
908.
161.
9 18.2
75 -1.29
21 3.10
5
51
123.20
131.
128.
30.5
23.2
1.26
3.90
7
49
457.70
457.
479.
87.9
19.2
-2.14
4.57
*A11 sample concentrations are in mg NO/DSCM.
X
24
-------
REFERENCES
Fuerst, R. G., R. L. Denny, and M. R. Midgett. A Summary of the Inter-
laboratory Source Performance Surveys for EPA Reference Methods 6
and 7 - 1977. EPA-600/4-79-045, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711, August 1979. 50 pp.
Standards of Performance for New Stationary Sources. Revision to Reference
Method 1-8. Federal Register, 42060):41776-41782, August 18, 1977.
Standards of Performance for New Stationary Sources. Revision to Reference
Method 1-8. Federal Register, 42_( 160): 41782-41784, August 18, 1977.
Standards of Performance for New Stationary Sources. Revision to Reference
Method 1-8. Federal Register, 42_( 160): 41784-41786, August 18, 1977.
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, Pennsylvania, 1863. pp. 558-565.
Rom, J. J. Maintenance, Calibration, and Operation of Isokinetic Source-
Sampling Equipment. EPA/APTD-0576, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711, March 1972.
Hordo, L. A. Errors in the Calibration of EPA Method 5 Dry Gas Meters.
J. of the Air Pollut. Control Assoc., 2_7_(8) :776-778.
U.S. Environmental Protection Agency. Stack Sampling Technical Information
A Collection of Monographs and Papers: Volume III. EPA-450/2-78-042c.
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina 27711, October 1978.
25
-------
INTERLABORATORY STUDY 0578
Pollutant: DGM
Units: Percent Difference
(DATA IN ASCENDING ORDER)
APPENDIX A
METHOD 5 DGM SUMMARY DATA
rv>
CTi
18.3
18.1
17.8
17.5
17.0
15.1
14.8
14.1
13.6
12.3
11. S
11.7
11.7
11.4
11.2
10.7
-9.7
-9.6
-9.5
-9.2
-9.2
-9.2
-9.2
-9.1
-9.1
-9.0
-9.0
-8.8
-8.8
-8.J
-8.4
-8.6
-(.4
-8.1
-7.9
-7.9
-7.9
-7.9
-7.9
-7.9
-7.4
-7.5
-7.4
-7.4
-7.4
-7.4
-7.3
•7.2
•7.C
-7.C
-6.9
-6.8
•6.3
6.3
•6.2
•6.1
•6.1
6.0
5.7
5.6
5.6
5.6
5.5
5.4
5.4
5.2
5.2
5.2
5.2
;.2
-5.2
-5.2
-5.1
-5.1
-5.1
-5.0
-4.9
-4.7
-4.6
-4.5
-4.5
-4.4
-4.3
-4.J
-4.3
-4.2
-4.2
-4.1
4.1
4.1
4.1
4.0
3.9
3.9
3.8
3.7
3.7
3.7
3.6
3.6
3.6
3.6
3.6
3.6
3.5
3.5
—
-
-
-
-
.
-
-;
-i
).5
.4
.4
.4
.2
.1
.0
.0
.9
'.8
.8
.8
.5
.5
.4
.3
.3
.1
2.1
2.1
.9
.7
.5
.5
.4
.4
.3
1.2
1.1
1.0
1.0
-.8
-.8
-.7
-.7
-.T
7
7
6
6
6
6
5
5
3
3
2
1
1
1
3
3
4
0
1.
1.
1.
1.
2.
2.
2.
3.0
4.2
5.7
6.0
6.0
6.2
6.5
7.2
83.}
83.6
«6.3
DATA SUMMARY
180
Pi Its
-18.J
5.84
-9.4
20*
301
-6.8
-4.1
40X 50X 60S
-5.2 -4.1 -3.5
70S
80X
-.7
90!
1.0
Hftt
94.3
WEAN
-3.1
STDEV
12.3
DATA SUMMARY (OUTLIERS REMOVED)
fSARP
172
SKIVNESS •
SIK
-15.1
.29
101
-9.2
MEDIAN
J01
-7.9
301
-6.3
-4.1
40X
-5.2
5 OS
-4.1
602
-1.5
70t 80S 90S MAI MEAN SIDE*
-Z.3 -.7 .3 7.2 -4.J 4.2
-------
INTERLABORATORY STUDY 1078
Pollutant: DGM
Units: Percent Difference
(DATA IN ASCENDING ORDER)
ro
27.2
27.0
24. V '
12.4
11.7
11.7
11. *
10.6
10.5
10.1
10.2
-S.5
-«.4
-.o
.9
.9
.9
.9
.9
.8
.8
.8
.7
.7
.7
.6
.6
.6
1.
1.
1.
1.
1.
2.
2.
2.
3.
3.
3.
3.
3.
4.
37.7
41.0
45.2
DATA SUMMARY
255
-27.2
3.44
102
-9.0
201
301
-5.2
-3.0
40X
-3.7
;ox
-3.0
60S
-2.4
7 OX
-1.7
80 X
-1.0
90X
.1
MAX
45.2
MEAN
-J.2
STOEV
6.3
DATA SUMMARY (OUTLIERS REMOVED)
*1»HP HIM
249 -12.4
,U
102
20X
-8.9 -6.2 -5.2
MEDIAN * -3.Q
4CX
-3.7
50X
-3.0
6 OX
-2.4
7 OX
-1.7
60 X
-1.0
90X
.0
HAM
4.2
HEAN STOEV
-3.S 3.2
-------
APPENDIX B
METHOD 6
S02 SUMMARY DATA
INTEB-LABOBAT08T STUB?
POLLUTAWT - S02
SAHPLI WUH6E8 - 3
TRUC-VALUE 686.30
1EAM 672.31
9EDIAM 6S2.9Q
BATA IN ASCENBINS OSDEB
057S
UNITS
HIilI£R*HS P£» 88* STB CUBIC HETER
8AW6E
WAR I A MCE
STD.
COEf.
770.00
7696.96
87.73
13.05
C.I.(UPPER) 693.97
C.I.(LOWE ill 650.64
SKEHWESS -6.43
ACCUBAC? -.50
21.20
573.60
645.20
657.40
657.*0
658.10
659.00
660.40
660.30
661.00
662.60
664 .40
666.00
667.00
668. 10
672.00
672.40
674.90
675.00
676.00
676.20
676.20
678.00
679.00
679.30
679.40
6F9,
681,
681,
682,
6E2,
683,
6E3,
6E4,
6E4,
6E4
664
60
10
20
40
10
90
30
90
CO
30
70
80
685.30
685.50
686.40
686.70
686.90
688.80
689.10
689.20
689.40
690.90
692.60
692.90
693.1Q
693.20
695.10
696.00
696.50
700.30
700.50
701.60
7C2.10
707.30
711.50
789.40
791.20
I NTEH-LABOR AtOB? STUD*
»OLLUT*NT - S02
SAHPLI WUHBES - 3
V 62
THUE-WSLUE 686.30
3 iff. 'n 682.81
683.10
0578
UNITS - MILLIGRAHS PE8 OR? STB CUBIC
••«* WITH OUTLIERS BEHOWEO •»»*
RANGE
VAS
STB, DEU.
COEF. VftR.
2 17.60
761.19
27.59
4.04
C.I.(UPPER) 689.68
C.I.(LOUE 85 675.94
SKEyNESS .83
«CCUBAC¥ -047
BATA IN ASCE^DINS OBDfA
573.60
645.20
657.40
657.40
658.10
659.00
660 .40
660.8 0
661.00
662 .60
664.40
666 .00
667.00
.10
672.00
672.40
674.90
675.00
676.00
676.20
676.20
678.00
679.00
679.30
679.40
679.60
661.10
681.20
6E1.40
6EZ.10
6EZ.90
66!
663
684
6E4
30
90
00
30
6B4.70
664.80
6E5.70
6B5.50
686.40
686.70
686.90
688 .80
689.10
689.20
689.40
690.90
692.60
692.90
693.10
693 .20
695.10
696.00
696.50
70C.30
700.50
701.60
702.10
7Q7.
711,
789 .40
791.20
.30
.50
28
-------
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 4
1 63
TRUE-VALUE 2478.50
SEAN 2427.12
OED1AN 2460.70
DATA IN ASCENDIN6 ORDER
0578
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
RANGE 2866.80
VARIANCE 1D0799.29
STD. DEV. 317.49
COEF. VAR. 13.08
C.I.(UPPER>2505.52
C.I.(LOWE RJ2348.72
SKEWNESS -6.36
ACCURACY -.72
76.90
2085.20
2344.30
2355.60
2367.CO
2374.CO
2374.20
2379.70
2380.00
2385.50
2421.30
2430.00
2431.60
2433.70
2434.40
2437.00
2437.00
2437.20
2439.00
2439.70
2442.10
2442.40
2445.CO
2451.00
2451.80
2452.00
2452.60
2453.00
2455.00
2458.40
2460.30
2460.70
2462.00
2462.50
2463.10
2463.30
2463.80
2464.00
2465.60
2467.20
2467.60
2468.00
2470.00
2473.90
2477.00
2477.00
2479.30
2479.60
2483.40
2485.00
2487.30
2488.70
2489.50
2499.00
2501.40
2509.10
2509.20
2561.60
2569.70
2580.30
2605.40
2803.80
2943.70
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 4
N 62
TRUE-VALUE 2478.50
4EAN 2465.03
MEDIAN 2461.35
DATA IN ASCENDING ORDER
057E
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
•••• WITH OUTLIERS REMOVED *••*
RANGE
VARIANCE
STD. DEV.
COEF. VAR,
858.50
10441.51
102.18
4.15
C.I.CUPPER)2490.46
C.I.(LOWE R>2439.59
SKEWNESS 1.4C
ACCURACY -.69
2085.20
2344.30
2355.60
2367.CO
2374.00
2374.20
2379.70
2380.CO
2385.50
2421.30
2430.00
2431.60
2433.70
2434.40
2437.00
2437.00
2437.20
2439.00
2439.70
2442.10
2442.40
2445.00
2451.00
2451.80
2452.00
2452.60
2453.CO
2455.CO
2458.40
2460.30
246D.70
2462.00
2462.50
24tl.10
2413.3C
246!.EC
2464.CO
2465.60
2467.20
2467.60
2468.00
2470.00
2473.90
2477.00
2477.OC
2479.30
2479.60
2483.40
24P5.0C
2487.30
2488.70
2489.50
2499.CO
2501.40
2509.10
2509.20
2561.60
2569.7C
2580.30
2605.40
2803.80
2943.70
29
-------
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 5
H 63
TRUE-VALUE 1258.30
^EAN 1229.44
SEB1AN 1249.90
»ATA IN ASCEWD1N6 OR8ER
0578
UNITS - H1LLIGRAHS PER
STC CUBIC
RANGE 1417.40
VARIANCE 25591.84
ST6. DEW* 159.97
COEF. VAH. 13.01
C.I.«UPPER>1268.94
C.I.CLOWE RM189.94
SKEWWESS -6.48
ACCURACY -.67
39.00
1CS3.80
1170.00
1190.10
1191.20
1192.90
1202.90
1216.40
1218.70
1220.00
1224.00
1225.20
1225.50
1227.70
1230.30
1230.SO
1235.00
1235.40
1235.70
1239.40
1240.00
1240.00
1240.80
1243.60
1244.00
1244.00
60
30
70
SO
12*4
1245
1245
1247
1249.50
1249.90
1250
125C
125C
125*
1254
10
50
70
CO
CO
1255.CO
1255.60
1256.00
1256.20
1256.80
1256.90
1257..20
1259.40
1259.80
1261.60
1262.10
1262.80
1263.40
1263.40
1264.10
1266.00
1267.00
1267.50
1268.30
1268.60
1273.70
1278.60
1280.00
1318.50
1421.40
1456.40
I NTER-LABOBATORY STUDY
POLLUTANT - S02
SAMPLE MUHBEB - 5
S 62
TRUE-VALUE 1258.30
*EAN 1248.64
lEBIAfc 1250.00
DATA }N ASCEN01ME OfiDER
0578
***«
UNITS - BILLI6RAHS f»E8 C«Y STB CUBIC H£T£R
OUHLIE8S BEHOVED **«*
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
4C2.60
2404.77
49.04
3.93
C.1.CUPPEHJ1260.85
C.I.UOtiE 8)1236.«,3
SKEUNESS .81
ACCURACY -.66
1C53.?0
1170.00
1190.10
1191.20
1192.90
1202.90
1216.40
1218.70
1220.00
1224.00
1225.20
1225.50
1227.70
123C.30
123C.50
1235.00
1235.40
1235.70
1239.40
124C.OO
1240.00
1240.80
1243.60
1244.00
1244.00
1244.60
1245.30
1245.70
1247.80
1249.50
1249.90
1250.10
125C.50
12IC.70
1254.00
1254.CO
1255.CO
1255.60
1256.00
1256.20
1256.80
1256.90
1257.20
1259.40
1259.80
1261.60
1262.10
1262.80
1263.40
1263.40
1264.10
1266.00
1267.CO
1267.50
1268.30
1268.80
1273.70
1278.60
1280.00
1318.50
1421.40
1456.40
30
-------
INTER-LAB OftATONY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 7
N 63
TRUE-VALUE 1906.00
4EAN 1865.95
HEDIAN 1894.10
DATA IN ASCENDIN6 ORDER
0578
UNITS - MILLIGRAMS PER DRY STD CUBIC M£T£R
RANGE 2137.40
VARIANCE 58547.23
STD. DEV. 241.97
COEF. VAR. 12.97
C.I.(UPPER)192S.7C
C.I.CLOWE RM806.20
SKEUNESS -6.56
ACCURACY -.62
58.60
1612.40
1787.80
1792.CO
1799.40
1820.40
1827.10
1840.00
1851.60
1852.80
1860.00
1868.50
1870.90
1871.70
1876.10
1879.70
1879.90
1881.40
1881.80
1882.10
1884.30
1884.50
1884.60
1885.10
1887.00
1867.30
1868.00
1890.CO
189C.90
1892.30
1892.80
1894.10
1894.30
1895.00
1896.60
1898.30
1899.50
19C0.70
19C1.50
1902.00
1902.50
19C4.10
19C6.50
1907.00
1907.40
1908.30
1908.80
19C8.80
19C9.80
1910.00
1914.40
1915.00
1921.00
1923.00
1923.30
1936.40
1936.60
1939.80
1942.20
1955.20
1992.60
2143.00
2196.00
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 7
t 62
TRUE-VALUE 1906.00
JEAN 1895.10
MEDIAN 1894.20
DATA IN ASCENDING ORDER
0578
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
**** WITH OUTLIERS REMOVED *••*
OAN6E
VARIANCE
STD. DEV.
COEF. VAR
583.60
5094.03
71.37
3.77
C.I.1912.86
C.I.tLOWE RM877.33
SKEUNESS .76
ACCURACY -.62
1612.40
1787.80
1792.CO
1799.40
1820.40
1827.10
1840.CO
1851.60
1852.80
1860.00
1868.50
1873.90
1871.70
1876.10
1879.70
1879.90
1881.40
1881.80
1882.10
1884.30
1884.50
1884.60
1885.10
1887.00
1887.30
1888.00
189C.00
1690.90
1892.30
1892.80
1894.10
1894.30
1895.00
1896.60
1898.30
1899.50
19C0.70
1901.50
19C2.CO
1902.50
1904.10
19C6.50
1907.00
1907.40
19C8.30
19C8.80
1908.80
19C9.80
1910.00
1914.40
1915.00
1921.00
1923.00
1923.30
1936.40
1936.60
1939.80
1942.20
1955.20
1992.60
2143.CO
2196.00
31
-------
I NTER-LAB ORATORT STUDY
POLLUTANT - S02
SABPLE NUMES - 9
* 61
TRUE-VALUE 190.70
*E«W 193.73
«E01Af» 191. 10
BATA IN ASCENDIWS ORDER
0578
UNITS - MILLIGRAMS PER CRT STO CUBIC METER
•«•• WITH OUTLIERS REMOVED ••••
RANGE
VARIANCE
STB. 6EV.
CCEF. VAB.
65.80
141.13
11.88
6.13
C.I.(UPPER) 196.71
C.I.(LOWE R) 190.75
SKEMHESS .58
ACCURACY .58
160.70
165.CO
179.20
179.?C
182.40
184.60
184.90
165.00
185.90
187.00
187.00
187.10
187.10
187.30
18 8 . DO
188.30
188.40
188.40
188.40
186.80
189.00
189.30
189.90
189.90
190.00
190.00
150.40
190.60
191.30
191.SO
192=00
1S2.20
192.30
19J.40
193.3C
193.40
193.50
193.90
194*00
194.30
194.50
195.30
196.00
196.90
197.60
197.60
197.90
199.90
SCO.10
201.30
2C2.00
203.40
208.60
211.00
212.30
213.50
219.60
221.20
226.00
226.50
I NTER-LA80RAT081T STUOT
POLLUTANT - 502
SABPLE NUWBEB - 9
<( 63
THUE-WALUE 190.70
<*E«« 192.46
BED IAN 191 .80
IN ASCENDING ORDER
0578
UNITS - B1LLI6S8HS PER 687 ST6 CUBIC M£T£B
BANGE
VARIANCE
STS. 6EV.
COEF. VAR.
295.80
892.24
29.87
15.52
C.I.(UPPER) 199.83
C.I.(LOME R> 185.08
SKEWKESS -3«04
ACCURACY .58
5
160
165
179
179
182
184.60
184.90
.CO
.90
.CO
70
CO
20
20
40
185.
185,
187,
187.CO
U7.10
187.10
187.30
188.00
188.30
188.40
188.40
188.40
188.80
189.00
189.30
189.90
189.90
19C.OO
19C.
150.
190,
190.
191.
191.
192,
192.
192,
192,
193.
193,
CO
10
40
60
30
80
00
20
30
40
30
40
193.50
193.90
194.00
194.30
194.50
195.30
196.00
196.80
197.60
197.60
197.90
199.90
200.10
201.30
202.00
203.40
208.60
211.00
212.30
213.50
219.60
221.20
226.00
226.50
301.60
32
-------
INTER-LABORATORY STUDY
POLLUTANT - 502
SAMPLE NUMBER - 1
* 69
TRUE-VALUE 2555.30
«EAN 2550.51
MEDIAN 2549.50
DATA IN ASCENOINE ORDER
1078
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
RANGE 984.30
VARIANCE 11690.82
STD. DEV. 1C9.C5
COEF. VAR. 4.28
C.I.(UPPER)2576.24
C.I.(LOWE R)2524.78
SKEWNESS 1.64
ACCURACY -.22
2170.70
2289.70
2385.10
2426.90
2428.70
2440.40
2452. 80
2455.90
2486.10
2489.CO
2503.70
2503.70
2505.20
2509.00
2512.50
2514.30
2514.70
2515.20
2518.00
2519.00
2522.90
2525.00
2527.50
2528.50
2532.20
2533.00
2533.60
2537.00
2539.70
2542.80
2543.CO
2544.2C
2547.80
254B.50
2549.50
255:.oo
2552.40
2554.90
2555.PO
2559.40
2559.80
2563.70
2565.00
2565.90
2568.40
2569.70
2571.40
2572.10
2573.40
2574.60
2575.00
2577.50
2577.70
2583.00
2587.00
25"2.60
2595.CO
2605.80
2628.00
2630.CO
2630.20
2633.00
2640.20
2645.CO
2654.00
2654.60
2684.40
2687.80
3155.00
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 1
-------
INTER-LABORATQR1 STUDY
POLLUTANT - 502
SABPLE NUMBER - 3
* 69
TRUE-VALUE 1335.03
*EAN 1332.31
MEDIAN 1328.70
DATA IN ASCENDING ORDER
1C78
UNITS - MILLIGRAMS PER DRY STD CUBIC M£TER
R ASGE 5 23.70
VAR I ANC E 32 55 .68
STD. DE V. 57.06
COE F. VAR. 4.28
C.I. (UPPER J1345.67
C.I.(LOWE RM318.75
SK£yNESS 1.66
ACCURACY -.47
1123
1216
1264
1270
1272
1275
1277
1288
1294
1295
1297
13CO
1310
.1
.3
.5
.9
.4
.3
.4
.8
.2
.2
.9
0
0
0
n
u
/^
>j
0
0
0
0
0
0
.00
.5
0
1314.
1314.
1315.
1316.
1316.
1317.
1319.
1319.
131V.
1319.
1320.
1321.
1321.
00
60
00
00
90
40
00
00
40
90
70
20
60
1311.CO
1322.80
1IZ3.2G
1323.40
1724.20
1324. PC
1325.40
1326. CC
1325.70
1T2KZC
1333.60
1335.70
1336.70
1317.70
1T35.CC
133S.3C
1338.40
1340.00
1340.20
1340.60
1342.00
1342.20
1342.20
1342.80
1344.00
1346.00
1346.60
1348.30
1350.90
1352.00
1357.CO
1360.80
1362.00
1369.10
1372.50
1376.00
1379.00
1381.CO
1389.40
1398.80
1400.90
1430.00
1646.80
INTER-LABORATOR V STUDY
POLLUTANT - S02
SAMPLE NUMBER - 3
V 67
TRUE-VALU E 1335.00
*EAN 133C.63
MEDIAN 132E .70
DATA IM ASCENDING ORDER
1C78
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
>*** WITH OUTLIERS REMOVED ****
RANGE 213.70
VARIANCE 1189.77
STD. DEV. 34.49
COEF. VAR. 2.59
C.I.(UPPER)1338.89
C.I.(LOWE RJ1322.37
SKEUNESS -.13
ACCURACY -.47
1216.30
1264.50
127C.9C
1272.40
1275.30
1277.40
1288. f 0
1294.20
1295.20
1297.90
1300. CO
1310.50
13 11 .CO
1314. CC
1314.60
1315. CO
1316.00
1316.90
1317.40
1319.00
1319.00
1319.40
1319.90
132C.70
1321. 2C
1321.60
1322.60
1323.20
1323.40
132*.. 20
1324.80
1325.40
1326. CC
1328.70
1331.20
1333.60
1334.70
1T36.7Q
1T37.7C
1338. CO
1333. 3C
1338.40
1340.00
1340.20
1340.60
1342.00
1342.20
1342.20
1342.80
1344.00
1346.00
1346.60
1348.30
1350.90
1352.00
1357.00
136C.80
1362.00
1369.10
1372.50
1376.CO
1379.CO
1381 .00
1369.40
1398.80
1400.90
1430.00
34
-------
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 5
1 68
TRUE-VALUE 572.DO
SEAN 644.64
MEDIAN 572.55
DATA IN ASCENDING ORDER
1078
UNITS - MILLIGRAMS PER DRY STD CUBIC M£T£R
RANGE 5244.10
VARIANCE 3i3461.84
STD. DEV. 6C2.E8
COEF. VAR. 93.52
C.I.(UPPER ) 787.94
C.I.(LOWE R) 5C1.35
SKEUNESS 7.83
ACCURACY .1C
286.90
529.20
537.6C
547.CO
549.10
549.10
552.00
553.10
556.70
556.70
557.20
558.40
560.CO
560.50
561.90
562.00
562.10
565.00
565.00
565.20
566.20
566.60
566.90
568.00
568.CO
566.30
569.00
569.00
569,
5t9.
57D,
57C,
573,
572,
573,
574,
574,
575.
575,
576,
576,
40
70
30
40
50
CO
10
1C
9C
10
20
CO
20
576.30
5^6.60
577.00
577.70
577.90
579.60
580.30
583.40
584.10
584.80
586.00
586.00
586.00
587.80
589.80
591.80
592.00
596.00
598.CO
599.30
6C1.10
608.70
611.00
611.50
639.80
698.50
5531.00
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 5
1 67
TRUE-VALUE 572.30
*EAN 571.71
MEDIAN 572.3D
DATA IN ASCENDING ORDER
1C78
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
**** WITH OUTLIERS REMOVED *»**
RANGE
VARIANCE
STD. DEV.
CCEF. VAR.
411.60
1 E C4 . 16
42.48
7.43
C.I.(UPPER) 581.88
C.I.(LOWE R) 561.54
SKEWNESS -4.02
ACCURACY .00
266.90
529.20
537.60
547.00
5*9.1C
549.10
552.CO
553.10
556.70
556.70
557.20
558.40
560.00
560.50
561.90
562.00
56Z.10
565.00
565.00
565.20
566.20
566.60
566.90
568.00
566.00
568.30
569.00
569.00
5t?.40
569.70
57D.70
57?.40
57D.50
572.CO
57!.10
574.1C
574.90
575.10
575.20
576.CO
576.20
57i.3c
576.60
577.00
577.70
577.90
579.60
580.30
5P3.40
5P4.1Q
584.80
586.00
586.00
586.00
587.80
589.80
591.80
592.00
596.CO
598.00
599.30
601.10
606.70
611.00
611.50
639.80
698.50
35
-------
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 6
1 69
TRUE-VALUE 152.53
*EAN 158.29
1EDIAN 156.DO
DATA IN ASCENDING ORDER
1 C78
UNITS - MILLIGRAMS PER DRY STD CUBIC WETER
RANGE
VARIANCE
STD. DEV.
COE F. VAR.
1 12.20
2 23.96
14.97
9.45
C.I.(UPPER ) 161 .83
C.I.UOHE R) 154.76
SKEUNESS 1.13
ACCURACY 2.30
114.80
120.00
136.CO
142.60
143.20
144.90
145.60
147,
147,
148
10
60
70
149.50
149.50
149.50
150.40
151.30
151.70
152.50
152.50
153.CO
153.00
153.03
153.90
153.90
154.CO
154.10
154.10
154.^,0
154.40
154,
154,
155,
155,
156,
1 56.
15£,
156,
US.
156.
157,
1 57.
15P.
5C
"C
30
6C
CO
CC
CO
CO
20
9C
rc
10
50
155.50
158.50
159.00
159.10
159.20
159.30
159,
160.
160.
160,
161 ,
163.00
163.00
163.90
164.70
.50
,00
.40
.80
.40
164.80
167.70
167.80
173.70
175.60
176.40
178.10
179.00
181.CO
183.00
183.00
189.70
227.00
INTER-LABORATORY STUDY
9QLLUTANT - S02
SAMPLE NUMBER - i
-------
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 9
M 69
TRUE-VALUE 1754.30
JEAN 1737.41
MEDIAN 1751.50
DATA IN ASCENDING ORDER
1C78
UNITS - MILLIGRAMS PER DRV STD CUBIC M£T£R
RANGE 1251.40
VARIANCE 17395.91
STD. DEV. 131.89
COEF. VAR. 7.59
C.I.(UPPER)1768.56
C.I.(LOWE R>1706.32
SKEUNFSS -3.23
ACCURACY -.14
910.CO
1447.70
1515.50
1599.90
1632.00
1667.40
1677.70
1680.90
1699.80
17C0.6C
1701.40
17C7.60
17C8.10
1709.CO
1709.
1717.
1717,
1718.
1726,
1728.
1731.
1732.
1733.80
1734.30
1736.00
1737.30
174C.80
1740.60
,50
,00
,60
.30
.00
.00
.60
.GO
1741.40
1743.CO
1745.90
1748.60
1749.90
1751.
1751.
1754.CO
1754.9C
175ft.50
1758.CO
1758.CO
1759.CO
1761.1C
.00
,50
1762.60
1763.10
1763.40
1763.50
1764.20
1765.00
1766.00
1767.10
1767.9C
1767.90
1768.00
1768.00
1770.50
1784.00
17°0.00
179C.CO
1795.CO
1796,00
1798.00
18C8.50
1814.60
1816.00
1817.70
1840.30
1845.50
1975.00
2161.40
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 9
N 67
TRUE-VALUE 1754.30
1EAN 1743.46
MEDIAN 1751.50
DATA IN ASCENDING ORDER
1078
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
**** WITH OUTLIERS REMOVED »*•*
RANGE 527.30
VARIANCE 4789.27
STD. DEV. 69.20
COEf. VAR. 3.97
C.I.(UPPER)1760.03
C.I.(LOWE R)1726.89
SKEUNESS -1.21
ACCURACY -.14
1447.70
1515.50
1599.90
1632.CO
1667.40
1677.70
1660.90
1699.8C
17CC.60
1701.40
1707.60
1708.10
1709.00
1709.50
1717.00
1717.60
1716.30
1726.00
1726.00
1731.60
1732.CO
1733.80
1734.30
1736.00
1737.30
1740.at)
174C.SO
1741.40
1743. CO
1748.60
1749.90
1751. CC
1751.50
1754. CO
1754. 9C
1756. 50
1758. CO
1753.00
1759. CO
1761.90
17J2.60
1763.10
1763.40
1763.50
1764.20
1765.DC
1766.00
1767.10
1767.90
1767.90
1768.00
1768.00
1770.50
1784.00
1790.00
1790.00
1795.CO
1796.00
1798.CO
1808.50
1814.60
1816.00
1817.70
1840.30
1845.50
1975.00
37
-------
APPENDIX C
METHOD 7
NOX SUMMARY DATA
INTER-tABORATORY STUDY
POLLUTANT - NOX
SAHPLE BUBSER - 1
0578
« 46
TBUE-VALUE 7C3.30
728. 9t
70i.*o
BATS 1M SSCEND1N6 ORDER
UNITS - HILLI6RAHS PER DRY STB CUBIC METER
RANGE 1577.10
VARIANCE S2760.93
STB. DEV. 250.52
COEF. VAB. 34.37
C.I.JUPPER) 8C1.31
C.I.(LOWE »> 656.52
SKEUNESS 1.61
ACCURACY -.27
123.30
239.40
381.90
533.00
565.00
590.60
639.00
640.CO
645.00
655.00
660.80
668.30
669.80
675.00
675.60
677.50
678
682
686
60
00
30
690.00
693.CO
697.30
7U.4C
7C1.40
7C8.50
712.10
713.80
716.CO
718.20
718.80
732.00
733.50
740.OC
741.70
741.80
746.00
746.40
764.40
769.00
833.70
855.30
865 .50
878.CO
1406.90
1418.80
1700.40
INTER-LABORATOR Y STUDY
POLLUTANT - NOX
SABPLE BURBER - 1
« 43
TRUE-VALUE 703.30
-------
INTER-LABORATORY STUDY
POLLUTANT - NOX
SAMPLE NUMBER - 2
1 46
TRUE-VALUE 93.80
1EAN 102.70
MEDIAN 95.15
DATA IN ASCENDING ORDER
0578
UNITS - MILLIGRAMS PER DRV STD CUBIC METER
RANGE 177.00
VARIANCE 1343.66
STD. DEV. 36.66
COEF. VAR. 35.69
C.I.(UPPER) 113.29
C.I.(LOWE R) 92.10
SKEWNESS 1.68
ACCURACY 1.44
32.20
45.50
71.40
72.90
73.60
80.00
82.50
85.10
85.30
86.00
88.00
88.50
91.00
91.00
91.30
92.00
92.30
92.80
92.80
93.10
97.60
94.70
95.10
95.20
53.30
95.30
95.60
96.80
97.50
57.50
99.50
100.00
100.80
102.00
1C2.00
1C2.80
104.80
105.00
110.50
113.00
115.50
168.00
195.30
204.00
207.80
209.20
INTER-LABORATORV STUDY
POLLUTANT - NOX
SAMPLE NUMBER - 2
TRUE-VALUE
4EAN
MEDIAN
42
93.80
93.04
94.15
DATA IN ASCENDING ORDER
0578
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
*•** WITH OUTLIERS REMOVED ***•
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
135.80
373.73
19.33
20.78
C.I.(UPPER) 98.89
C.I.(LOWE R) 87.20
SKEWNESS .29
ACCURACY .37
32.20
45.50
71.40
72.90
73.60
80.00
82.50
85.10
85.30
86.00
88.00
88.50
91.00
91.00
91.30
92.00
92.30
92.80
92.80
93.10
93.60
94.70
95.10
95.20
55.30
95.30
95.60
96.80
97.50
97.50
99.50
1CO.OO
1C0.8C
1C2.00
102.00
102.80
104.80
105.00
110.50
113.00
115.50
168.00
39
-------
INTER-LABORATORY STUDY
POLLUTANT - NOX
SAWPLE WUHBEH - 6
0 46
TRUE-VALUE 515.70
JEAN 523.73
1ED1AW 524.25
8ATA IN ASCEN01NS ORDER
0578
UNITS - BIH.I6RABS PER DRY STD CUBIC HETER
RANGE
VARIANCE
STD. DEV.
COEF, WAR,
971.90
29195.07
170.87
32.63
C.1. 573.10
C.I.(LOWE R) 474.35
SKEWWESS .90
ACCURACY 1.66
93.60
184.10
246.70
256.00
262.40
429.60
479.00
480.00
482.10
484.1 0
496.50
499.00
499.00
501.30
505.20
507.20
508.30
508.80
510.CO
515.30
520.CO
520.00
523.00
525.50
529,00
529.00
529.70
534.00
535.00
537.20
541.40
541.80
545.90
548.00
548.60
555.00
558.00
563.50
564.00
573.30
576.00
5S5.80
589.50
959.00
1045.50
1065.50
INTER-LABORATORY STUDY
POLLUTANT - NO*
SAHPLE WUHBER - 6
TRUE-VALUE
JEAN
MEDIAN
42
515.70
498.28
521.50
0578
UNITS - HILLI6RAHS PER DRY STD CUBIC WEIER
«••• U1TH OUTLIERS REMOVED ****
RANGE
VAR1ANCE
STD. DEV.
COEF. VAR,
4C5.40
8447.47
91.91
18.45
C.I.(UPPEHJ 526.08
C.I.(LOVE R) 470.48
SKEWNESS -2.16
ACCURACY 1.12
BAT A IN ASCENDING ORDER
184.10
246.70
256.00
262.40
429.60
479.00
480.00
482.10
484.10
496.50
499.00
499.00
501.30
505.20
507.20
508.30
508.80
510.00
515.
520.
52:.
523.
525.
529.
529.
529.
30
CO
00
00
50
00
00
70
534.00
535.00
537.20
541.40
541 .80
545.90
548.00
548.60
555.00
558.00
563.50
564.00
573.30
576.00
585.80
589.50
40
-------
INTER-LABORATORY STUDY
POLLUTANT - NOX
SAMPLE NUMBER - 8
1 46
TRUE-VALUE 937.70
4EAN 958.46
4EDIAN 941.40
DATA IN ASCENDING ORDER
0578
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
RANGE 1726.50
VARIANCE 90641.87
STD. DEV. 3C1.07
COEF. VAR. 31.41
C.I.(UPPER)1045.47
C.I.(LOWE R) 871.46
SKEWNESS .89
ACCURACT .39
156.20
340.60
472.80
674.00
715.CO
763.70
765.00
790.60
817.00
861.20
862.40
866.20
872.00
875.00
890.00
909.00
915.00
915.70
916.00
925.60
927.80
938.20
941.00
941.80
942.00
946.00
955.50
963.30
967.00
96B.OO
983.00
989.40
1000.00
1002.50
1007.80
1037.00
1039.50
1040.00
1048.40
1133.00
1137.60
1184.80
1218.20
1717.40
1879.30
1882.70
INTER-LABORATORY STUDY
POLLUTANT - NOX
SAMPLE NUMBER - 8
0578
TRUE-VALUE
«EAN
1ED1AN
42
937.70
915.56
939.60
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
•*•* WITH OUTLIERS REMOVED •*•*
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
DATA IN ASCENDING ORDER
877.60
26336.99
162.29
17.73
C.I.(UPPER) 964.64
C.I.(LOWE R> 866.48
SKEWNESS -1.28
ACCURACY .20
340.60
472.80
674.00
715.00
763.70
765.00
790.60
812.00
861.20
862.40
866.20
872.00
875.00
89C.OO
909.00
915.00
915.70
916.00
925.60
927.80
938.20
941.00
941.80
942.00
946.CO
955.50
963.30
967.00
968.00
983.00
989.40
1000.00
10C2.50
1007.80
1037.00
1039.50
1040.00
1048.40
1133.00
1137.60
1184.80
1218.20
41
-------
INTER-LABORATORT STUOT
POLLUTANT - NOX
SAMPLE NUMBER - 9
*) 46
TRUE-VALUE 328.20
HEAN 340.70
1EPIAN 333.85
»ATA IN ASCENDING ORBER
0578
UNITS - MILLIGRAMS PER DRT STO CUBIC WETER
RANGE
VARIANCE
STC. DEV»
£OEF. VAR,
564. 50
10228.06
1 C1.13
29.68
C.I.
-------
INTER-LABORATORY STUDY
POLLUTANT - NOX
SAMPLE NUMBER - 1
1 52
TRUE-VALUE 246.53
•UAN 274.03
MEDIAN 256.73
DATA IN ASCENDING ORDER
1G78
UNITS - MILLIGRAMS PER DRY STO CUBIC METER
RANGE 1020.90
VARIANCE 17719.72
STD. DEV. 133.12
CCEF. VAR. 43.58
C.I.(UPPER) 310.21
C.I.(LOUE R) 237.85
SKEUNESS 3.55
ACCURACY 4.14
14.60
28.60
68.30
210.CO
225.CO
233.00
233.10
233.80
238.80
239.00
243.60
243.70
244.60
245.00
245.90
247.00
24E.OO
250.00
250.20
250.80
252.70
253.50
254.CO
254.60
255.40
255.9G
257.50
257.70
25P.CO
2t1.tO
263.CO
2t4.0C
264.90
265.00
265.80
266.00
269.10
270.00
274.30
274.80
276.50
280.00
280.00
285.00
296.70
313.2C
316.30
317.80
381.00
439.CO
577.50
1035.50
INTER-LABORATORY STUDY
POLLUTANT - NOX
SAMPLE NUMBER - 1
N 51
TRUE-VALUE 246.50
1EAN 259.13
MEDIAN 255.93
DATA IN ASCENDING ORDER
1C78
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
>*** WITH OUTLIERS REMOVED **•*
RANGE 562.90
VARIANCE 6249.96
STD. DEV. 79.06
COEF. VAR. 20.51
C.I.(UPPER) 280.80
C.I.UCUE R) 237.40
SKEWNESS .34
ACCURACY 3.81
14.60
28.80
88.30
210.CO
225.CO
233.CO
233.10
233.SO
238.80
239.00
243.60
243.70
244.60
245.00
245.90
247.00
248.00
25C.OO
25C.20
25C.80
252.70
253.50
25**CO
254.60
255.40
255.9C
257.50
257.70
25?.CO
2b1.60
2t3.CO
2U.CC
2t4.90
265.00
265.80
266.00
269.1Q
270.00
274.30
2^4.80
276.50
2PO.OO
280.00
285.00
296.70
313.20
316.30
317.80
381.00
439.CO
577.5C
43
-------
INTER-LABORATORY STUDY
POLLUTANT - NOX
SAMPLE NUMBER - 2
TRUE-VALUE 730.60
*EAN 775.51
MEDIAN 7t4 .33
DATA IN ASCENDING ORDER
1C78
UNITS - MILLIGRAMS PEP DRY STD CUBIC M£TER
R ANGE
V ARIANC E
STD. DEV.
CCE F . VAR
15 81 .60
S2999.C4
2 51.00
32.37
£.I. (UPPER ) £43.73
C.I.UOWE R) 7C7.29
SKEWNESS .98
ACCURACY 4.o1
107.CO
124.60
268.70
546.60
625.CO
664 .CO
673.50
676.7C
6EC.OO
684 .90
693.00
694.00
701.50
712.00
712.50
'57.90
71 7,
72C.
723,
60
oO
20
726.50
741.20
751.70
757.43
75!.
759.
~t!.
765.
76!.
770.
775.
7E :.
7E3.
752.
CO
70
ro
60
CO
60
CO
CO
60
793.80
797.00
798.00
798 .20
802.70
822.30
827.90
828.00
871.80
856 .60
858.00
859.40
878.00
920.20
933.70
1075.80
1118.30
1662 .50
1688.60
INTER-tABORATCRY STUDY
"OLLUTANT - NOX
SAMPLE NUMBER - 2
S 5D
TRUE-VALUE 730.63
SEAN 739.51
MEDIAN 761.35
9ATA IN ASCENDING ORDER
1 C78
UNITS - MILLIGRAMS PER DRY STD CUBIC
*«** WITH OUTLIERS REMOVED *»**
RANGE 1011.30
VARIANCE 31177.CO
STD . DEV. 1 76 .57
COE F. VAR = 23.88
C.I.(UPPER) 788.45
C.I.(LOWE R) 690.57
SKEyNFSS -1.75
ACCURACY 4.21
107.00
124.60
268.70
546.60
625.00
664 .CO
673.50
676 .70
660.00
684 .90
693.00
694.00
701.50
712.00
712.50
717.60
72C.80
723.23
726.50
741.20
751.70
757.40
"57.90
755.00
759.70
7t3.
7t5.
7t3.
77C.
775.
CO
60
CO
60
CO
CO
752.10
793.80
797.00
798.00
798.20
8C2.7C
822.3T
827.90
828 .00
831.80
856.60
858.00
859.40
878.00
920.20
933.70
1075.80
1118.30
44
-------
INTER-LABORATORY STUDY
POLLUTANT - NOX
SAMPLE NUMBER - 3
1 52
TRUE-VALUE 860.50
1EAN 9£4.82
MEDIAN 9C9.6D
DATA IN ASCENDING ORDER
1C78
UNITS - MILLIGRAMS PER DRY STD CUBIC M£T£R
RAJ.GE 1728.30
VARIANCE 65544.65
STD. DEV. 256.02
CCEF. VAR. 27.68
C.I.(UPPER! 994.41
C.I.(LOME R) 855.23
SKEHNESS 1.82
ACCURACY 3.33
298.00
370.40
593.50
687.00
718.70
755.00
773.CO
795.CO
798.SO
603.CO
833.00
635.50
836.30
649.00
655.50
858.00
663.90
867.50
669.60
872.60
675.53
893.60
8 9 B . 2 C
SCO.90
•506.10
9J9.CC
91D.20
911.CO
911.70
912.70
518.CO
918.50
919.8C
920.50
932.30
936.00
950.00
953. 80
960.60
981.10
990.00
997.00
1000.30
1015.20
1045.00
106C.10
1060.80
1089.30
1202.80
1246.00
1805.00
2026.30
INTER-LABORATORY STUDY
POLLUTANT - NOX
SAMPLE NUMBER - 3
1 5C
TRUE-VALUE 880.33
SEAN 865.19
MEDIAN 907.55
DATA IN ASCENDING ORDER
1C78
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
**** WITH OUTLIERS REMOVED •***
RANGE 948.00
VARIANCE 26046.47
STD. DEV. 161.39
CCEF. VAR. 18.23
C.I.(UPPER) 929.92
C.I.(LOWE R) 640.45
SKEUNESS -1.29
ACCURACY 3.10
298.CO
370.40
593.50
687.00
718.70
755.CO
773.CO
795.CO
798.80
803.CO
833.CO
£35.50
836.30
849.00
C55.50
858.00
663.90
867.50
669.60
672.60
875.50
893.80
P5B.20
•5:0.90
90S.10
9C9.CO
51D.20
911.CO
911.70
912.70
91!.CO
918.50
915.80
920.50
932.30
936.00
950.00
953.80
960.60
981.10
99Q.OO
997.00
1000.00
1015.20
1045.00
1060.10
1060.80
1089.30
1202.80
1246.00
45
-------
INTER-LABORATORY STUDY
°OLLUTANT - NOX
SAMPLE NUMBER - 5
M 52
TRUE-VALUE 123.20
1EAW 137,42
1EDIAN 128.50
BAT* IN ASCENDING ORDER
1078
UNITS - WILLIGRAHS PER DRY STD CUBIC
R ANGE
VARIANC E
STD. DEVa
CCEF- VAR,
4 13.30
2787*77
52.80
£.1. (UPPES )
C.I. «IOUE
SKEWNESS
ACCURACY
151 .77
J 123.07
3.68
4.30
30.50
90,5 0
95 .60
105 .6 0
108.00
109.20
111.20
111.50
113.90
115.10
117.30
119.50
121.90
122.00
122
123
123
Ml
124
125
125
,50
,50
.80
,90
,90
, GO
,50
127
12°
123
129
1 *:
1 10
1 30
13C
.CO
,50
. CO
, CO
.CO
.CO
• CO
.50
= 80
126.40
132»CO
1 32.70
133.CO
135.00
176.00
158.00
no.90
141 .30
U1 .40
144.10
145.00
145.50
146.30
151 .00
153.50
155»30
156.CO
165.00
219,50
262.50
443.EO
INTER-IABORATORt STUDY
POLLUTANT - NOX
SAWPLE NUMBER - 5
1 51
TRUE-VALUE 123.20
*
-------
INTER-LABORATORY STUDT
POLLUTANT - NOX
SAMPLE NUMBER - 7
1 52
TRUE-VALUE 457.73
SEAN 490.56
MEDIAN 481.35
DATA IN ASCENDING ORDER
1C78
UNITS - MILLIGRAMS PER DRY STD CUBIC METER
RANGE 952.40
VARIANCE 25784.07
STD. DEV. 160.57
CCEF. VAR. 32.73
C.I.(UPPER) 534.2C
C.I.(LOWE R) 446.91
SKEUNESS 1.76
ACCURACY 5.17
125.10
139.90
305.50
313.40
337.50
379.2C
394.40
400.00
420.50
421.50
43C.CO
445.00
448.00
455.90
456.50
459.20
463.40
465.50
466.EC
468.50
472.80
473.00
474.CO
475.4C
478.6C
479.30
4E3.40
4E4.CO
4E6.CG
4E6.CO
4S0.30
492.6C
454.80
497.00
499.20
SCO.00
5C2.70
5C3.00
5C7.10
513.00
515.00
517.30
521.20
522.10
523.90
53C.10
537.00
552.80
598.70
976.40
1046.90
1077.50
INTER-tABORATORY STUDY
"OLLUTANT - NOX
SAMPLE NUMBER - 7
* 49
TRUE-VALUE 457.73
4EAN 457.31
MEDIAN 478.63
DATA IN ASCENDING ORDER
1078
UNITS - MILL1GOAMS PER DRY STD CUBIC M£TE"
**** UITH OUTLIERS REMOVED *»**
RANGE 473.6C
VARIANCE 7724.08
STD. DEV. 87.89
CCEF. VAR. 19.22
C.I.(UPPER) 481.92
C.I.(LOWE R) 432.70
SKErfNESS -2.14
ACCURACY 4.57
125
139
305
313
337
379
394
400
420
.10
.90
.50
.40
.50
.20
.40
.00
.50
421.50
43C.OO
445.CO
448.00
455.90
456.50
459.20
463.40
465.50
466.80
468.50
472.8C
473.CO
474.CO
475.40
473.60
479.10
4E3.40
4E4.CO
4E«.rO
490.30
492.6C
496.30
497.00
499.20
SCO.00
5C2.70
5C3.00
5C7.10
513.00
515.CO
517.30
521.20
522.10
523.90
53C.10
537.00
552.80
598.70
47
-------
TECHNICAL REPORT DATA
FPA finO/4 80-029
A SUMMARY OF THE INTERLABORATORY SOURCE PE
SURVEYS FOR EPA REFERENCE METHODS 5, 6, AN
R. u. Fuerst and M. . R. Midgett
# PL *•- C ~ Vi'.w C = _- . C- * C .'.-•. i AND -C.L.RE £3
Quality Assurance Division
2 FUCf '-' TSACCi_- _»-&
-. REPORT DATE
May 1980
RFORMANCL PE RFOPV1INC ORGM..,2AT,c-. C..-L
D 7 - 1978"
g p£_f,FQRVilKGORGANi<-~TIC-^t
10 P f= 1 G P A M t. L L V E '- T .N C
tnvironmenta 1 Monitoring Systems Laboratory " cc'- ThACT- GRANT ^o
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
12 SPONSORING AGENCY NAVE fiNDADDRESS
Environmental Monitoring Systems Laborator
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13 TYPE OF P. EPOFiT AND PERIOD COVERED
v Final
"" 14 SPC^SDRi NG AGE NC Y CODE
EPA-600/08
•i SUPPLE. '.-cNTA^Y NOTES
To be published as an Environmental Monitoring Series report.
A national survey of methods in stati
by the Quality Assurance Division of the E
U.S. Environmental Protection Agency, Rese
this program, quality assurance samples we
measurement of a gas volume (Method 5, dry
samples simulating collected sulfur dioxid
respectively). Each participant returned
Quality Assurance Division, for evaluation
each participant after processing.
This report summarizes the survey res
onary source tests was conducted in 1978
nvironmental Monitoring Systems Laboratory,
arch Triangle Park, North Carolina. In
re sent to interested participants for the
gas meter only) or the analysis of liquid
e and nitrogen oxides (Method 6 and 7,
the analytical results to the Source Branch:
An individual report was returned to
ults for those three source test methods.
• 1 KEY/. ^RCS AND DOCUMENT ANALYSIS
Ci-~- ~~ ' --
Performance survey
Reference methods
RELEASE TC PUBLIC
b IDENTIF IE RS.OPEN ENOE D TE RMS C COS A T i I u Li uuup
EPA -Source Method 5 43F
EPA Source Method 6 68A
EPA Source Method 7
Sulfur dioxide
Nitrogen dioxide
10 SECURITY CLA^b, //,, ( /icf)..//, 21 '-\-' Ul >' A _, t .
UNCLASSIFIED 55
70 SECURITY CLASSi//;urJ-'' ^T PRiCf
UNCIASSIFIFD
48
------- |