United States
           Environmental Protection
           Agency
           Environmental Monitoring Systems EPA/600/4-86/005
           Laboratory         February 1986
           Research Triangle Park, NC 27711
           Research and Development
vvEPA
A Summary of the
1984 EPA National
Performance Audit
Program on Source
Measurements

-------
                                                 EPA/600/4-86/005
A SUMMARY OF THE 1984 EPA NATIONAL PERFORMANCE AUDIT PROGRAM
                   ON SOURCE MEASUREMENTS
        E. W. Streib, T. J. Logan and M. R. Midgett
                 Quality Assurance Division
        ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
             OFFICE OF RESEARCH AND DEVELOPMENT
            U.S. ENVIRONMENTAL PROTECTION AGENCY
        RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711

-------
                     Disclaimer

This document has been  reviewed  in accordance with
U.S. Environmental  Protection  Agency  policy  and
approved for  publication.   Mention of  trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
                         11

-------
                                  FOREWORD
     Measurement and monitoring research efforts are designed to anticipate
potential environmental problems, to support regulatory actions by develop-
ing an in-depth understanding of  the nature and processes that impact health
and the environment, to  provide  innovative means of monitoring compliance
with regulations, and to  evaluate the  effectiveness  of health and environ-
mental protection regulations through  the monitoring  of  long-term trends.
The Environmental Monitoring Systems  Laboratory,  Research  Triangle  Park,
North Carolina, has responsibility for:   assessment  of environmental  moni-
toring technology and systems; implementation of Agency-wide quality assur-
ance programs for air pollution measurement systems; and supplying technical
support to  other  groups  in  the  Environmental  Protection  Agency, including
the Office  of  Air  and  Radiation,  and  the  Office  of Toxic  Substances.

     The major task of this study was to report the results of the national
quality assurance audit program for stationary source test methods.  Audits
were designed to estimate the minimal analytical and computational accuracy
that can be expected with EPA Method 5 (dry gas meter only), Method 6 (sul-
fur 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

-------
                                  ABSTRACT
     In 1984, the Quality Assurance Division  conducted the National Audits
for Stationary  Source  Test  Methods.   The audit materials  consisted of:  a
calibrated orifice for Method 5 (dry gas meter only), five simulated liquid
samples each for Method  6  ($02)  and Method  7 (NOX), two  coal  samples for
Method 19A,  and a  disposable gas  cylinder  for Method  3 (Orsat analyzer).
Participating laboratories  sent  their  data  to  the  Source  Branch  and in
return received a written report comparing their results to EPA's.

     In the Method 5 audit,  the mean value for all participants differed by
3.1% from the  expected (EPA) value.   For the Method  6  audit,  the average
mean differed  by  3.8% from  the  expected value.   The  average  mean in the
Method 7 audit was 3.2% from the expected value.

     In the two coal audits, the parameters measured were sulfur, moisture,
ash, and Btu content.  On the average for the  sulfur  analysis, 82% of the
participants measured within 10% of the expected value; for Btu, 97% of the
participants measured within 10% of the expected value.

     In the Method 3 audit,  each parameter had only one concentration.  The
mean for C02 was 1.5% from the expected value.  In this audit, the mean for
02 was 2.1% from the expected value.

     This report includes the results  of the performance audits done during
the period from January through December 1984.
                                     iv

-------
                                  CONTENTS

                                                                       Page
Foreword ...............................
Abstract ...............................   iv
Figures  ...............................   vi
Tables ................................  vii
Acknowledgment ............................ viii
     1.  Introduction  ........................    1
     2.  Summary ...........................    1
     3.  Method 3 Audit  .......................    4
     4.  Method 5 Dry Gas Meter Audit  ................    6
     5.  Method 6 Audit  .......................   12
     6.  Method 7 Audit  .......................   13
     7.  Method 19A Audit  ......................   15
References ..............................   19
Appendices
     A.  Frequency Distributions ...................   20
     B.  Instructions for EPA Audit Materials  ............   26

-------
                                  FIGURES

Number                                                                Page
   1          Cumulative accuracy of participants in                    9
              Method 5 audit, 0684
   2          Laboratories within 2 percent accuracy                   10
   3          Results of Method 5 audit, 0684                          11
                                     vi

-------
                                   TABLES

Number                                                                Page
   1          Participants'  Results from Method 3 Audit                  2
              (All Data - No Outliers Removed)
   2          Participants'  Results from Method 5 Audit                  2
              (All Data - No Outliers Removed)
   3          Participants'  Results from Methods 6 and 7 Audits          3
              (All Data - No Outliers Removed)
   4          Participants'  Results from Method 19A Coal Audits          4
              (All Data - No Outliers Removed)
   5          Method 3 Audit Participants                                5
   6          Source Method  3 Audit • 0784                               6
   7          Method 5 Audit Participants                                7
   8          Previous Results of Method 5 Audit                         7
   9          Method 6 Audit Participants                               12
  10          Summary of Source S02 Audit                               13
  11          Method 7 Audit Participants                               14
  12          Summary of Source NOX Audit                               14
  13          Coal Audit Participants                                   16
  14          Source Coal  Audit - 0384                                  17
  15          Source Coal  Audit - 0984                                  18
                                   vii

-------
                               ACKNOWLEDGMENT
     We express our  appreciation to the  laboratories  that  participated in
the National Performance Audit Program for Stationary Sources.  Thanks also
to the  Standards  Laboratory/Performance  Evaluation  Branch/Environmental
Monitoring Systems Laboratory (EMSL), who did the acceptance testing of the
audit samples,  and  to  the programmers  of the  Monitoring  and  Assessment
Division/EMSL for providing the  data systems  to store and evaluate the data.
                                    viii

-------
                                 Section 1

                                INTRODUCTION

     The Environmental Protection Agency's Environmental Monitoring Systems
Laboratory (EMSL)  at  Research  Triangle  Park,   NC,   established  an  audit
program in  1977  to  evaluate the  performance  of  companies that  conduct
compliance testing  using  EPA  reference  methods.   The  audits  check  the
participants' analytical  accuracy  in  applying  the  analytical  phase  of
EPA Reference Methods 3,  6,  7,  and  19A and  the  calibration accuracy of the
Method 5 control  console (1).  Accuracy is defined as the percent difference
between the  participant's  analytical   results and the EPA  expected value.
By participating in this  free and voluntary program,  testing companies  can
compare their performance to other laboratories  conducting similar measure-
ments.

     Source Test Methods  3,  5,  6,  and  7 were each  audited once and Method
19A twice in 1984.   Each participating  laboratory received an audit package
consisting of the audit  sample,  a data card, instructions, and an envelope
for returning the data to EPA.  A label for returning the audit device  was
included with the Method  5  audit package.   Participants  had eight weeks to
use the audit material  and  return their data to EPA.  At the  end  of this
period all  data received were  statistically  analyzed  to  determine  the
precision (repeatability  or  reproducibility) and accuracy  with respect to
the EPA expected  value  obtained  by  the  participants.    (See Appendix  A.)

     The Quality Assurance Division of  EMSL maintains a repository of audit
samples for the EPA Methods 3,  6,  7,  and 19A.   These samples are available
to source testing laboratories  for  such purposes as training new personnel
or conducting quality control checks.   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  source  testing
laboratories use  this  sample  repository  to help   improve  their  overall
analytical performance.

     This report summarizes the results obtained in the 1984 source audits.

                                 Section 2

                                  SUMMARY

     In 1984, EPA's EMSL, 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   laboratories,  as  well  as
local, State, and federal agencies participated.

     The results of  the  1984 Method   3 audit   are  summarized  in Table  1.
Participants analyzed the gas  sample twice for percentage  of carbon dioxide,
oxygen, and carbon monoxide.  The mean  values of carbon dioxide and oxygen
differed by 2* from  the expected value.  In contrast, the  mean values  for
carbon monoxide differed as much as  21% from the expected value.

-------
TABLE 1.  PARTICIPANTS' RESULTS FROM METHOD 3 AUDIT
          (ALL DATA - NO OUTLIERS REMOVED)

Type of
Sample

Small
Cylinder
(Gas)





Audit
Date

0784








Parameter

X CO?


X 02

% CO


No. of
Analyses

64
62

64
62
51
50

Repli-
cate

1
2

1
2
1
2
EPA
(True)
Value

8.00
8.00

9.90
9.90
0.70
0.70

Participant Results

Mean
7.88
7,91

10.11
10.06
0.55
0.57

Median
7.80
7.80

10.00
10.00
0.60
0.60

Std. Dev.
0.65
0.59

1.22
1.21
0.19
0.19
     One Method 5  audit was  conducted  in  1984.   The overall  results (no
outliers removed) are summarized in Table 2.  The mean for all participants
was 3.1% from the expected  value.  After  correcting for outliers, the mean
was 2.2%.  The participants'  performances did  improve from previous audits
in that the standard deviation greatly decreased (2, 3, 4).

TABLE 2.  PARTICIPANTS' RESULTS FROM METHOD 5 AUDITS
          (ALL DATA - NO OUTLIERS REMOVED)
Type of                   Audit      No. of
Sample      Parameter     Date      Analyses
                                    Mean        Median    Std.
                                    (X from EPA Value)    Dev.
Orifice
Volume
0684
631
3.1
1.9
4.9
     Table 3 presents the data (no outliers removed) from the Methods 6 and
7 audits.  In the Method 6 audit, the procedure requires the participants to
determine the sulfate content in five aqueous solutions using the titration
procedure.  For  each  concentration, the  mean of  the participants  was 5%
higher tharr the expected value.   In contrast, the median differed by IX.  In
the Method  6 audit, 50 to   70X  of the  participants achieved  an accuracy
within 2X Of the expected  value.  Also, the standard  deviation is propor-
tional (*30X) to the mean level of the sulfate samples.

-------
TABLE 3.  PARTICIPANTS' RESULTS FROM METHODS 6 AND 7 AUDITS
          (ALL DATA - NO OUTLIERS REMOVED)
Type of
Sample
Aqueous
Sulfate



Aqueous
Nitrate



Audit
Date
0584




0484




No. of
Analyses
92
91
92
92
92
71
70
70
71
70
EPA True
Value
(mg/DSCM)
282.2
297.4
899.9
968.5
1197.3
91.0
102.4
284.4
466.4
523.3
Participant Results
Mean
296.3
312.7
924.3
999.2
1234.8
96.3
111.1
288.0
464.9
523.2
Median
284.0
299.8
890,3
966.8
1192.5
90.4
100.5
280.0
457.9
517.5
Std. Dev.
83.74
94.59
277.30
301.92
377.38
36.07
49.50
64.95
117.78
104.37
     The Method 7 audit  procedure requires that the participants determine
the nitrate content in five aqueous solutions.  For each concentration, the
mean of the participants  was not  higher  than 8% from  the  expected value;
but in contrast, the  median  differed by  less than 2%.   In  this  audit, 3li
to 46% of the  participants  achieved an accuracy within 3%  of the expected
value for 4 out of 5  samples.   However,  on the lowest concentration sample
only 21* achieved this accuracy.

     Table 4 summarizes the  results of the two coal  audits that were con-
ducted in 1984.  Participants  analyzed each  coal  sample in  duplicate for
percentage of  sulfur,  moisture,  and ash,  and  for  gross  calorific  value
(Btu/lb).  The means of  the  ash,  moisture, and sulfur  content  were within
5% of the expected  value on  both  concentrations.   An  accuracy  of 1% was
achieved on the Btu content.

-------
TABLE 4.  PARTICIPANTS' RESULTS FROM METHOD 19A COAL AUDITS
          (ALL DATA - NO OUTLIERS REMOVED)
Type of
Sample
Coal







Audit
Date
0384
0984
0384
0984
0384
0984
0384
0984
0384
0984
0384
0984
0384
0984
0384
0984
Parameter
% S

% H20

% Ash

Btu/lb

No. of
Analyses
113
100
113
99
114
102
114
101
112
101
112
100
110
95
110
96
EPA
Value
0.45
1.17
3.28
2.93
2.94
1.86
12.01
10.65
9.45
9.74
13.13
12.87
12115.0
12595.0
13135.0
13710.0
Participant Results
Mean
0.45
1.18
3.04
2.84
2.92
1.90
11.85
10.44
9.46
9.90
13.04
12.75
12020.3
12504.3
13068.9
13850.7
Median
0.46
1.17
3.06
2.83
2.95
1.97
12.16
10.79
9.50
9.67
13.13
12.80
12133.0
12631.0
13087.0
13860.0
Std. Dev.
0.23
0.11
0.29
0.22
0.44
0.33
1.52
1.50
0.47
2.32
1.00
1.07
385.99
409.65
105.50
224.92
                                 Section 3

                               METHOD 3 AUDIT

     This audit checks  the  participants'  ability to  analyze a  gas sample
using an  Orsat  analyzer.   The  audit  package  consists  of a  disposable
cylinder that  contains  a 4-liter  sample  of  oxygen,  carbon dioxide,  and
carbon monoxide.  The analyst expels the  gas into the Orsat analyzer using
the positive pressure of  the  cylinder.   The gas sample  is quantitatively
analyzed for percentage  of carbon  dioxide,  oxygen,  and  carbon monoxide.

     In the  1984  audit,  50%  of  the 109  laboratories receiving  the audit
package returned  data.   Table  5 shows  the  total  number  of  laboratories
requesting participation versus  the  number that returned  data for the Method
3 audit.

-------
TABLE 5.  METHOD 3 AUDIT PARTICIPANTS
                         No. Receiving Samples         No. Returning Data
Category                         0784                         0784
Contractor
Industry
Foreign
Federal
State
Local
57
32
1
2
14
3
21
20
1
1
8
3
Total                            109                           54


     Table 6  summarizes  the Method  3  audit results.   Each  laboratory was
asked to  analyze  the sample  in  duplicate.   Five  and  ten percent accuracy
were chosen for the precision reporting criteria for each of the parameters.
Each parameter had only one concentration.

     In the  1984   audit,  82%  of  the   reporting  laboratories  achieved  an
accuracy within 5% for the  C02, which  was a 34% increase  from  the last
audit (4).  Eighty percent  of the  laboratories  achieved an accuracy within
5% of the  expected value for the  03  analysis.   For the  CO  analysis, only
18% of  the laboratories  achieved  an  accuracy  within  5%,  and 19%  of the
laboratories did not report a value for CO.

-------
 TABLE  6.   SOURCE METHOD  3  AUDIT  - 0784
 Expected        No.  of           Laboratories              Laboratories
  Value        Analyses      Accurate within ± 5%      Accurate within ±  10%
                                    CO
  8.00         (1)   64               80%                        92%
               (2)   62               85%                        94%

                                     02

  9.90         (1)   64               78%                        94%
               (2)   62               81%                        95%

                                     CO

  0.70         (1)   51               18%                        18%
               (2)   50               10%                        10%


                                 Section 4

                        METHOD 5 DRY GAS METER AUDIT

     In the  Method 5  audit  procedure, the  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'minute volume measure-
ments.  The participants convert each  of  the three  volumes to  cubic meters
at standard conditions  using the formula  specified in Eq.  5.1  of Method 5
(Appendix A 40 CFR 60)  and  record them  on  the data card.   Finally, they
return the  orifice and  the data  card to  EPA,  where  the data  undergoes
statistical analysis.

     In the 1984 audit, 62% of the  155 laboratories that received the audit
package returned  data.   This  percentage  was  slightly  lower than  those
responding in previous  audits  (2,  3,  4).   Table  7  shows the  categories of
the participants  and   compares  the number  of  participants who  requested
participation in the Method 5  audit  with  the number  who actually returned
data.

-------
TABLE 7.  METHOD 5 AUDIT PARTICIPANTS
Category
Contractor
Industry
Foreign
Federal
State
Local
Total
No. Receiving Samples
0684
78
51
4
2
15
5
155
No. Returning Data
0684
45
31
3
2
12
3
96
     Figure 1,  a  cumulative histogram,  shows  the  accuracy  obtained  by
participants in  the  Method  5 audit,  expressed  as the  percent  difference
from the expected  (EPA)  value at various levels of accuracy.  The Code of
Federal Regulations (1) requires that the dry gas meter be calibrated within
an accuracy of ±2%.  Figure  2 shows that 54* of the reporting laboratories
attained this accuracy.  These results are an improvement compared to three
years ago, but may now be leveling off.  Table 8  is  a summary of previous
audits which, like Figure  2, shows improvement over  the past  three years.

TABLE 8.  PREVIOUS RESULTS OF METHOD 5 AUDIT
Audit
0381
0981
0382
0982
0383
0983
0684
Number of
Analyses
738
723
827
769
763
614
631
Mean
(% from
13.6
4.3
7.6
12.5
5.7
4.1
3.1
Median
EPA Value)
2.7
2.5
2.5
2.5
2.2
1.9
1.9
Std Dev
110.2
9.7
39.5
81.4
32.7
21.3
4.9

-------
     The histogram in  Figure 3  compares the  individual  results  from all
participants of  the  1984  audit  with  the  mean  and  median   values.   The
majority of the laboratories reported values lower than the EPA value.  The
standard deviation  of  the  triplicate   analysis  (repeatability)  by  each
laboratory indicated that 75% of the  standard  deviations  for  each set were
within 0.3%.  Four  percent of  the  1984 data  were   identified  as outliers
using Chauvenet's Criterion  (5).    Before  the  outliers  were  removed, the
mean value  differed  by 3.1% from  the  expected  value.  After  deletion of
outliers, this value was reduced to 2.2%.
                                     8

-------
      WikOlO>vlOO(OO
                                   ooooooooooo
        m M
        3)
        m
        Z
        O

        11 fa)
        z

        2
        m

        w *
        c
        31
        m

        m tn
                              to
                              •n
                              31
                              O
                              2
                              m
        m
        n
        »
           09
                                               i—i—i—r
                                                     T
T
T
T
                                                                             O)
                                                                             w

-------
  55
  50
*«

»
u

i

(A
HI

£ 45
O
oc
O
m
  40
  35
       DEGREE OF ACCURACY = 2 percent
                                                                         >  *
           0381/738    0981/723
 0382/827    0982/769    0383/763    0983/614


AUDIT/NUMBER OF SAMPLES
0684/631
                  Figure 2.  Laboratories within 2 percent accuracy.
                                         10

-------
                           LABORATORIES, percent
                                       ^            .4
                         01             o            01
    A
(O
C

(D
30
0)
(A
C_
r+
V)
I
I
01
c
Q.
*•+"

O
O)
00
          00
    o>
m   »
3D
m
Z
%   b

3D
O
S   o
m
•u
—   M
m   ,jj,
          oo
           o
           V

           N)
                                       1
                                              1
                                                        S.
                                                        00
                                                        m
                                                        3D'

                                                        S-

                                                        §•


                                                    m   m'

                                                    I   "-
                                                    S   O>

                                                    ,\
                                                    m
                                                    >
                                                    2
                                                    ii
                                                    •
                                                    O
                                                    NJ

-------
                                 Section 5

                               METHOD 6 AUDIT

     This audit  checks  the participants'  ability to quantitatively  analyze
Method 6  samples  for  sulfate  content.   The  audit  set  consists  of five
aqueous dilutions  of 10  N sulfuric acid  in 25 ml  sealed  glass ampoules.
The analyst  withdraws  5  ml  from each ampoule,  adds 30 ml  of 3% hydrogen
peroxide, and dilutes this  sample to  100 ml  with distilled  water.  A 20-ml
aliquot is then withdrawn from the diluted  sample,  80 ml of 100% isopropanol
and thorin  indicator are added, and the  sample  is titrated  with   barium
perchlorate  (BaCCIO^) to a pink endpoint.   In  calculating the results, the
participants assume  an original sample volume of  100  ml, and  a  sample  volume
of 0.021 dry standard cubic meter of stack gas.

     Table 9  shows  the  categories  of  the  participants  and  compares the
total number  of  participants  requesting   participation  with the   number
returning data.  In the  1984 audit, 59% of the 157 laboratories  that received
the audit package returned data.

TABLE 9.  METHOD 6 AUDIT  PARTICIPANTS
                         No. Receiving Samples         No. Returning Data
Category                         0684                         0684
Contractor
Industry
Foreign
Federal
State
Local
79
49
5
1
15
8
40
31
3
1
11
6
Total                            157                           92


     Table 10 shows the percentage  of  laboratories  that  achieved 2% and 5%
accuracy for  each of  the five  different concentrations  in the  Method  6
audit.  Also, at least  50% of the reporting laboratories achieved  an accuracy
within 2% for the three lower  concentrations,  but showed an improvement of
65% for the higher concentrations.   Seventy-five percent of the laboratories
were able to  achieve an  accuracy level  of  within 5%  on  all five samples.
                                     12

-------
TABLE 10.  SUMMARY OF SOURCE S02 AUDIT

Concentration
282.2 mg/DSCM
297.4 mg/DSCM
899.9 mg/DSCM
968.5 mg/DSCM
1197.3 mg/DSCM
n
0584
± 2%
50%
48%
55%
63%
70%
± 5%
79%
76%
88%
88%
91%
92
                                 Section 6

                               METHOD 7 AUDIT

     This audit checks the participants1  ability to quantitatively analyze
Method 7 samples for  nitrate  content.   The NOX  audit  set  consists of five
aqueous dilutions of  a potassium  nitrate (KN03) solution  in 25  ml  glass
ampoules which are  autoclaved after sealing  so that  bacteria  which  might
attack the nitrate are destroyed.  The analyst withdraws 5.0 ml of solution
from an ampoule,  adds this with 25 ml of the Method 7 absorbing solution to
a flask, adjusts the  pH  with  sodium hydroxide,  and dilutes to 50 ml  with
distilled water.   A  25-ml  aliquot  is  withdrawn  from  the  diluted sample,
placed in an  evaporating dish,  and analyzed  as  described in Section 4.3 of
Method 7 (1).  After this  treatment is  completed,  the absorbance is measured
at 410 nm with a calibrated spectrophotometer.   In  calculating the concen-
trations present, the  participant assumes that  2000  ml  of stack  gas  was
sampled.

     Table 11 shows the total  number of laboratories requesting participa-
ting in and  the number that returned  data for the 1984 Method 7 audit.  Sixty
percent of the 118  laboratories  receiving the  audit package returned data.
                                     13

-------
TABLE 11.  METHOD 7 AUDIT PARTICIPANTS
Category
Contractor
Industry
Foreign
Federal
State
Local
Total
No. Receiving Samples
0484
66
37
3
1
7
4
118
No. Returning Data
0484
36
23
2
1
5
4
71
     The percentage of  laboratories  that achieved  5 and  10% accuracy for
each of the five concentrations are shown in Table 12.  In addition, 40% of
the reporting  laboratories  achieved  an  accuracy within  5%  on  the lowest
concentrations and 55%  achieved  within 5% on  the  highest three concentra-
tions.  Seventy percent  of  the  71  laboratories  were  able  to  achieve an
accuracy within 10% on all five samples.

TABLE 12.  SUMMARY OF SOURCE NOX AUDIT

Concentration
91.0 mg/DSCM
102.4 mg/DSCM
284.4 mg/DSCM
466.4 mg/DSCM
523.3 mg/DSCM
n
0484
± 5%
38%
49%
53%
54%
63%
± 10%
68%
66%
74%
77%
79%
71
                                     14

-------
                                 Section 7

                           METHOD 19A COAL AUDIT

     Standards of performance  for  newer electric utility  steam generators
(Subpart Da of 40 CFR  60)  allow coal sampling  and  analysis to serve as an
acceptable method for  determining  scrubber  inlet  flue gas  sulfur concen-
tration.  The  coal  audit checks the participants'  ability to analyze coal
samples for sulfur,  ash, moisture,  and Btu content.

     Coal audit  samples  consisted  of  two  bottles,   each   containing  50
grams of  60-mesh coal.  The following  American Society  for  Testing  and
Materials (ASTM) procedures  (6) were  recommended,  but not  required,  for
participants' use in analyzing the coal samples:

     ' ASTM D-3177  (Standard Test  Method for Total  Sulfur in the Analysis
       of Coal and Coke);

     0 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

     0 ASTM D-2015 (Standard Test Method for Gross Calorific Value of Solid
       Fuel by the Adiabatic Bomb Method).

The participants measured  the  parameters  and  reported their  results  for
moisture (%)  on  an  as-received  basis,  and  their  results for  sulfur (X),
ash (%), and gross calorific value (Btu/16) on a dry basis.

     In both audits, 83% of the  laboratories  that received  the audit package
returned data.   Seventy-two  of the same laboratories  participated in both
audits and returned data.  Table 13  shows  the total number of  laboratories
requesting participation versus the  number  that  returned  data   for  coal
audits 0384 and 0984.
                                     15

-------
TABLE 13.  COAL AUDIT PARTICIPANTS
                        No. Receiving Samples           No. Returning Data
Category                0384             0984           0384          0984
Contractors
Industry
Federal
State
Local
59
40
0
12
5
41
47
1
13
4
42
38
0
11
5
35
38
0
11
4
Total                   116              106             96            88


     Tables 14  and  15 summarize  the coal  audit results.   The number  of
analyses is greater than the number  of  participants  because  some companies
had more than one laboratory participating.   In  this  case,  each laboratory
received its  own  set  of  samples  and was  asked  to analyze  the  samples  in
duplicate.  Accuracies of 5% and 10% were  chosen as  the  reporting  criteria
for each of the four parameters (sulfur, moisture, ash, and gross calorific
content).

     In the 0384  audit,  60% of the  laboratories were able  to  analyze the
sulfur content  of the low level sample  within 10% of the  expected value;
however, 86% were able to analyze the high level sample within  10% of the
expected value.    In the  0984  audit, 90%  of the   laboratories were  able  to
analyze both  sulfur samples  within  10%  of the  expected value.  Only 50%
achieved 10% on the low moisture  concentration in the 0984 audit;  whereas,
75% in the 0384 audit achieved  10% on the  low moisture concentration.   For
the ash analysis and Btu  content,  92% to 100% of  the reporting laboratories
were able to achieve an accuracy within  10% for both sample concentrations.

     The participants' accuracy improved with higher  concentrations on all
parameters except ash  in  the 0984  audit, where there was  a slight decrease.
For those laboratories that reported duplicate analyses, the intra-labora-
tory precision  (repeatability)  was  not  dependent on  concentration level.
                                     16

-------
TABLE 14.  SOURCE COAL AUDIT - 0384
Expected
Value
0.45
0.45
3.28
3.28
2.94
2.94
12.01
12.01

9.45
9.45
13.13
13.13
12115.
12115.
13135.
13135.
No. of
Analyses
(D*
(2)*
(D
(2)
(D
(2)
(D
(2)

(D
(2)
(D
(2)
(D
(2)
(D
(2)
113
100
113
99
114
100
114
101

112
100
112
98
110
99
110
97
Laboratories
Accurate within ± 5%
SULFUR
42*
37%
23%
28%
MOISTURE
48%
51%
79%
81%
ASH
91%
93%
96%
96%
GROSS CALORIFIC
87%
94%
100%
100%
Laboratories
Accurate within ± 10%
60%
60%
86%
86%
75%
78%
92%
93%

94%
94%
98%
99%
89%
95%
100%
100%
* Indicates first and second analyses
                                      17

-------
TABLE 15.  SOURCE COAL AUDIT - 0984
Expected
Value
1.17
1.17
2.93
2.93
1.86
1.86
10.65
10.65

9.74
9.74
12.87
12.87
12595.
12595.
13710.
13710.
No. of
Analyses
(D*
(2)*
(D
(2)
(D
(2)
(1)
(2)

(D
(2)
(D
(2)
(D
(2)
(D
(2)
100
95
99
95
102
96
101
96

101
95
100
95
95
91
96
91
Laboratories
Accurate within ± 5%
SULFUR
77%
77%
64%
63%
MOISTURE
25%
20%
79%
84%
ASH
95%
92%
87%
86%
GROSS CALORIFIC
89%
90%
99%
99%
Laboratories
Accurate within ± 10%
95%
93%
90%
89%
48%
51%
92%
93%

97%
97%
92%
92%
96%
96%
99%
99%
* Indicates first and second analyses
                                     18

-------
                                 REFERENCES
1.  U.S. Environmental Protection Agency.  Standards of Performance for New
    Stationary Sources -  Appendix A.   Title 40, Part  60, Code  of Federal
    Regulations.

2.  Streib, E.  W.,  R.  6.  Fuerst  and M.  R.  Midgett.   A Summary of the 1981
    EPA National  Performance  Audit Program  on Source  Measurements.   EPA-
    600/4-83-026, U.S. Environmental  Protection Agency,  Research  Triangle
    Park, North Carolina 27711, June 1983.

3.  Streib, E.  W. and M.  R. Midgett.  A Summary of the  1982  EPA National
    Performance Audit  Program  on  Source Measurements.   EPA-600/4-83-049,
    U.S. Environmental Protection Agency,  Research  Triangle  Park,  North
    Carolina 27711, April  1984.

4.  Streib, E.  W.,  T.  0.  Logan and M. R. Midgett.  A  Summary of  the 1983
    EPA National  Performance  Audit Program  on Source  Measurements.   EPA-
    600/4-85-004, U.S. Environmental  Protection Agency,  Research  Triangle
    Park, North Carolina 27711, January 1985.

5.  Chauvenet, W.  Manual  of Spherical and Practical Astronomy:  Volume II -
    Theory and Use  of  Astronomical Instruments (Method of  Least Squares).
    J. P. Lippincott and Co.,  Philadelphia,  Pennsylvania,  1863 (?).

6.  American Society for Testing and Materials.  Annual Book of ASTM Stand-
    ards - 1984.  Volume  5.05.  01-050584-13,  American Society for Testing
    and Materials, Philadelphia, Pennsylvania.
                                     19

-------
       APPENDIX A



FREQUENCY DISTRIBUTIONS
           21

-------
ro
ro
                                      OGM FREQUENCY DISTRIBUTION OF ABSOLUTE PERCENT DIFFERENCE - 0684




               f SAHP   MIN   10%   20*   30%   40%   50%   60%   70%   80%   90%    MAX   MEAN   STDEV   SKEWNESS   MEDIAN
631
625
611
603
599
598
596
595
593
590
587
585
.0
,0
.0
.0
,0
.0
.0
.0
.0
.0
.0
.0
.3
.3
.3
.3
.3
.3
.3
.3
.3
.3
.3
.3
.7
,7
.7
.6
,6
.6
.6
.6
.6
.6
.6
.6
1,0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
.9
1.4
1.4
1.3
1,3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.9
1.8
1.8
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
2.5
2.4
2.4
2.3
2.3
2,3
2,2
2.2
2.2
2.2
2.2
2.2
3.1
3.1
3.0
2.9
2.9
2.9
2.9
2. a
2.8
2,8
2.8
2.8
4.4
4.2
4.0
3,8
3,8
3.7
3.7
3.7
3.7
3.6
3.6
3.5
6.1
5.9
5.5
5.2
5.2
5.2
5.2
5.1
5.1
5.1
5.1
5.0
58.8
17,1
12.1
9,6
9.1
8.9
8.9
8,7
8.6
8.5
8.2
8.0
3.1
2.8
2.5
2.4
2,3
2.3
2.3
2.3
2.3
2.2
2.2
2.2
4.9
2.9
2.3
2.1
2.0
2.0
2.0
1.9
1,9
1.9
1.8
1.8
6,97
-.12
-.27
-.38
-,48
-.56
-.64
-.71
-.77
-.83
-.88
-.93
1.9
1.9
1,8
1.7
1.7
1.7
1.7
1.7
1,7
1.7
1.7
1.7

-------
ro
CO
                                  S02 FREQUENCY DISTRIBUTION OF PERCENT DIFFERENCE  - NO OUTLIERS REMOVED


                                                                AUDIT 0584
Sample
No.
2
3
5
7
9

NO.
91
92
92
92
92

MIN
.00
.00
.00
.00
.00
NOX
105!
.27
.43
.36
.26
.20
20%
.74
.67
.54
.61
.45
30%
1.14
1.13
.87
1.09
.61
40%
1.51
1.45
1.09
1.44
.90
FREQUENCY DISTRIBUTION OF
50%
2.15
1.88
1.60
1.77
1.36
PERCENT
60% 70%
2.56 3.56
2.69 3.40
1.87 2.55
2.22 2.63
1.65 1.99
DIFFERENCE -
80%
6.52
4.32
3.25
3.49
3.12
90%
10.26
9.14
5.16
5.01
4.15
MAX
300.13
275.62
296.49
293.38
300.90
MEAN
7.10
7.10
5.69
5.57
5.21
STDEV
31.42
29.24
30.81
30.42
31.24
NO OUTLIERS REMOVED
AUDIT 0484
Sample
No.
1
4
6
8
9
NO.
70
70
71
71
70
MIN
,11
.00
.00
.11
.29 1
10%
.67
.63
.99
.73
.17
20%
1.44
.90
2.75
1.37
2.25
30%
1.83
1.78
4.40
2.79
2.83
40%
2.43
2.35
5.27
3.52
3.81
50%
4.36
3.50
6.92
4.40
5.08
60% 70%
6.12 9.11
4.36 5,33
7.69 10.99
5.66 7.20
7.03 10.64
80%
11.46
10.09
15.27
10.38
15.82
90%
18.74
16.17
31.87
16.66
38.67
MAX
127.74
87.46
278.02
117.13
347.46
MEAN
10.41
9.18
16.83
11.44
18.09
STDEV
20.33
17.67
36,30
22.48
45.59

-------
ro
                                                NATIONAL COAL AUDIT  FREQUENCY  DISTRIBUTION



                               OF ABSOLUTE  PERCENT  DIFFERENCES  OF  EXPECTED  AND REPORTED VALUES - STUDY 0384





                                                                  SULFUR
Sample
No.
4000
7000
NO.
212
213
MIN
.00
.00
10%
3.35
2.22
20%
4.57
2.22
30%
5.18
4.44
40%
6.10
6.67
50%
6.71
8.89
60%
7.32
8.89
70%
8.23
13.33
80%
9.15
15.56
90%
10.37
24.44
MAX
88.11
535.56
MEAN
7.80
16.03
STDEV
8.47
51.99
MOISTURE
Sample
No.
4000
7000
Sample
No.
4000
7000
Sample
No.
4000
7000
NO.
214
215
NO.
210
212
NO.
207
209
MIN
.00
.00
MIN
.00
.00
MIN
.01
.00
10%
1.36
.58
10*
.15
.21
10%
.05
.06
20%
2.38
1.00
20%
.23
.42
20%
.11
.17
30%
3.40
1.33
30%
.38
.74
30%
.20
.26
40%
4.42
1.75
40%
.53
.95
40%
.27
.34
50%
5.44
2.25
ASH
50%
.61
1.16
GROSS
50%
.40
.40
60%
6.46
2.83
60%
.76
1.59
CAL
60%
.55
.54
70%
8.16
3.50
70%
.99
1.80
70%
.73
.67
80%
10.88
5.00
80%
1.37
2.43
80%
.97
.93
90%
14.97
7.16
90%
2.36
3.28
90%
1.36
2.48
MAX
104.76
97.59
MAX
76.69
32.80
MAX
4.59
13.02
MEAN
7.83
4.38
MEAN
1.54
2.40
MEAN
.65
1.31
STDEV
9.97
10.30
STDEV
5.58
4.39
STDEV
.75
2.84

-------
ro
en
                                                NATIONAL COAL AUDIT FREQUENCY DISTRIBUTION


                               OF ABSOLUTE PERCENT DIFFERENCES OF EXPECTED AND REPORTED VALUES  -  STUDY  0984




                                                                  SULFUR
Sample
No.
5000
8000
Sample
No.
5000
8000

Sample
No.
5000
8000
Sample
No.
5000
8000
NO.
195
194
NO.
198
197

NO.
196
195
NO.
187
186
MIN
.00
.00
MIN
.00
.00

MIN
.00
.00
MIN
.06
.02
10%
.85
.34
10%
2.15
.66

10%
.21
.08
10*
.50
.21
20%
.85
1.02
20%
4.30
1.22

20%
.31
.23
20%
.74
.37
30%
1.71
1.71
30%
6.99
1.88

30%
.51
.47
30%
.85
.52
40%
1.71
2.73
40%
8.06
2.25

40%
.72
.62
40%
.97
.67
50% 60%
2.56 3.42
3.75 4.78
MOISTURE
50% 60%
10.22 11.29
2.72 3.38
ASH
50% 60%
.92 1.13
.85 1.01
GROSS CAL
50% 60%
1.09 1.20
.78 .91
70%
4.27
5.80
70%
13.44
3.94

70%
1.64
1.40
70%
1.26
1.05
80%
5.13
7.17
80%
16.13
4.79

80%
2.05
2.18
80%
1.39
1.71
90%
8.55
9.90
90%
22.04
6.20

90%
3.90
7.23
90%
1.54
4.92
MAX
101.71
44.71
MAX
89.25
90.61

MAX
250.10
113.68
MAX
15.28
12.18
MEAN
4.36
5.06
MEAN
12.78
5.35

MEAN
4.12
3.18
MEAN
1.22
1.84
STDEV
9.62
5.79
STDEV
12.84
12.94

STDEV
24.51
10.22
STDEV
1.53
2.82

-------
                                               NATIONAL COAL AUDIT FREQUENCY DISTRIBUTION

                              OF ABSOLUTE PERCENT DIFFERENCES OF EXPECTED AND REPORTED VALUES - STUDY 0784


                                                                  CO
            Sample     NO.     MIN    10X    20*    30*    40*    50*    60*    70*    80*    90*      MAX     MEAN     STDEV
               No.	

             4000      126     .00    .00   1.25   2.50   2.50   2.50   3.75   5.00   5.00    7.50    45.00    4.41     6.53



            Sample     NO.     MIN    10X    20*    30*    40*    50*    60*    70*    SOX    90*      MAX     MEAN     STDEV
               No.	
ro                                               '                                                                   '—'	
°"           4000      126     .00   1.01   1.01   1.01   1.01   2.02   3.03   3.03   4.04    5.05    75.76    4.74    11.38

                                                                   CO

            Sample     NO.     MIN    10*    20*    30*    40*    50*    60*    70*    80*    90*      MAX     MEAN     STDEV
               No.	

             4000      101     .00    .00   14.29   14.29   14.29  14.29  28.57  28.57  42.86  57.14    85.71    26.17    21.19

-------
             APPENDIX B



INSTRUCTIONS FOR EPA AUDIT MATERIALS
                 27

-------
               INSTRUCTIONS  FOR USING EPA METHOD 3 AUDIT MATERIALS


 Equipment  Supplied with Audit Kit

     (1)    Small gas cylinder containing four liters of gas

     (2)    Nozzle for cylinder (taped on cylinder)

     (3)    Tygon tubing

 Equipment  to be Supplied by Participant

     (1)    Orsat analyzer

     (2)    Clamp

     (3')    Extra Tygon or surgical tubing


 Procedure

     (1)    Leak-check apparatus by clamping off tubing.
           SEE DIAGRAM.

     (2)    Fill  up burette with gas by using positive pressure.  Vent this sample
           through the manifold to the atmosphere.

     (3)    Repeat Step 2.

     (4)    Fill  up burette past fill mark with the gas and carefully vent out
           excess to the atmosphere, until  the fill mark is reached.

     (5)   Analyze for CO,,, 09 and CO as described in Sections 4.2.5, 4.2.6
           and 4.2.7 of EPA Method 3.

     (6)    Record the results on the data card enclosed with the sample.

     (7)    Repeat Steps 4 through 6.

    CAUTION:  If the tubing is punctured excess times, leakage can occur.
              Replace if necessary.

    Send the data card to the address below. (The cylinder gas can should
not be returned.)


                    Ms. Ellen Streib
                    Quality Assurance Division (MD-77A)
                    Environmental Monitoring Systems Laboratory
                    U.S. Environmental  Protection Agency
                    Research Triangle Park,  NC  27711

                                     28

-------
Note:  Site number will always be 001  except when other Orsat apparatus
       or participants are using the same gas sample.   The extra
       apparatus or participants should be labeled 002, 003,  etc.
APPARATUS SET-UP
                                29

-------
            INSTRUCTIONS FOR USE OF ENVIRONMENTAL PROTECTION AGENCY
                 METHOD 5 DRY GAS METER PERFORMANCE TEST DEVICE


NOTE:  All procedures referred to are from revised Method 5 published in the
       Federal Register, Volume 12, Number 160, Part II, Thursday, August 18,
       1977, pp. 41776-41782 and references contained therein.  This revised
       method should be adhered to in all details in the use of this quality
       assurance performance device.

EQUIPMENT:  The participant in this study should possess the following
            equipment, including the performance test device supplied by EPA.

Quantity	Item	

    1     Method 5/Source Sampling Meter Box
    1     Stopwatch, preferably calibrated in decimal minutes
    1     Thermometer, ambient range
    1     Barometer.  If unavailable,  call nearest National Weather Service and
          request the ABSOLUTE barometric pressure.  (Corrected for temperature
          and acceleration due to gravity, but not corrected for altitude.)
    1     Performance Test Device.  A calibrated flow orifice housed in a quick-
          connect coupling and identified with an engraved three-digit serial
          number.
            WARNING:  THE DEVICE MUST NOT BE DISASSEMBLED UNDER ANY CIRCUMSTANCES.
            Use these devices at room temperature.
PROCEDURE:
          Calibration of Vacuum Gauge — The vacuum pressure gauge on the
          meter box must be calibrated in the range of use (11-22" Hg) against
          a standard (Hg Manometer) to ensure accurate results.

          Remove the performance test device from its case and insert it into
          the gas inlet quick-connect coupling on the source sampling meter box.

          Turn the power to the meter box on and start the pump.

          Adjust the coarse flow rate control valve and the fine flow rate
          control valve to give a reading of 19" Hg (vacuum reading).
            CAUTION:  The vacuum reading must be accurate and stable for the
            test period.

          Allow the orifice and source sampling meter box to warm up for 45
          minutes with flow controls adjusted as described in Step 3 before
          starting quality assurance runs.
                                      30

-------
PROCEDURE:   (continued)
          Make triplicate quality assurance runs.  Ftor each run, record
          initial and final dry gas meter volumes, dry gas meter inlet and
          outlet temperatures,  internal orifice pressure drop (AH), ambient
          temperature,  and barometric pressure.  Run duration should be
          slightly greater than 15 minutes.  The following procedure is
          recommended.   Fifteen minutes after a run is started,  the partici-
          pant watches  the dry gas meter needle closely.  As the needle reaches
          the zero (12  o'clock) position, the pump and stopwatch are stopped
          simultaneously.  The  dry gas meter volume and time are recorded.

          This complete run procedure is performed three times to provide the
          required triplicate quality assurance runs.

          Calculate the corrected dry gas volume for each run using equation 5.1
          of the above-referenced Method 5.  For each replicate, record the
          corrected dry gas volume in dry standard cubic meters, the sampling
          time in decimal minutes, the barometric pressure in mm Hg, and the
          ambient temperature in degrees Celcius on the enclosed data card.  Be
          sure to record the performance test device serial number on the data
          card in the column headed "Orifice Number."

            NOTE 1:   If you calculate dry gas volume in English  Units, use the
                     conversion factor of 0.02832 m^ft^ to obtain the volume in
                     metric units.

            NOTE 2:   If your stopwatch is not in decimal minutes,  be sure to
                     convert (e.g. 15 minutes 20 seconds is reported as 15.33
                     minutes).

          After recording the requested data on the enclosed data form, return
          the data form and the performance test device to:

          Quality Assurance Division (MD-77A)
          Environmental Monitoring Systems Laboratory
          U.S.  Environmental Protection Agency
          Research Triangle Park,  NC  27711
          ATTN:   Ellen  W.  Streib

          A  postpaid  return envelope and label are enclosed for  this purpose.
                                     31

-------
     INSTRUCTIONS FOR  USE OF  ENVIRONMENTAL  PROTECTION  AGENCY  STATIONARY
               SOURCE QUALITY ASSURANCE S02 REFERENCE SAMPLES


NOTE:  All Method 6 procedures referred to are from the amended method pub-
       lished in the Federal Register  Vol.  1*2,  No.  l6o,  Part II, Thursday,
       August 18,  1977,  PP  1*1782-1*1784.   This  amended  method  should  be
       adhered to in all  details in the analysis  of  these  reference stand-
       ards.

1.   Prepare 3 percent hydrogen peroxide according to Section 3.1.3 of the
     method (30 ml is required for each sample and each blank).

2.   Prepare each reference sample for analysis as follows:  Wrap a paper
     towel around the ampoule  and  with the ampoule  in an  upright  position
     break off the top at the prescored mark by exerting pressure sideways.
     From the ampoule pipette  exactly  5  ml of the reference  sample into a
     100 ml volumetric  flask.   Add 30 ml of  3 percent hydrogen  peroxide
     'solution.  Dilute exactly to the mark with deionized,  distilled water.
     Analyze the sample in accordance with the procedure detailed in Section
     U.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, 0302 (concentration of  sulfur dioxide,
     dry basis, corrected to standard  conditions, mg/dscm),  using Equation
     6-2.  A value  of  21 x  10"^  dscm should be used for Vm(std),  in the
     equation.  A value of 100 ml  should  be used for V8Oinin the equation.

1*.   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-77A)
               Environmental Monitoring Systems Laboratory
               U.S. Environmental Protection Agency
               Research Triangle Park, NC 27711
               ATTN:  Ellen W. Streib

If other than  EPA Method 6  is used  for  your analyses, please  explain  in
detail your analytical  procedure on the  back of the enclosed  data form.
                                    32

-------
    INSTRUCTIONS FOR USE OF ENVIRONMENTAL PROTECTION AGENCY STATIONARY
              SOURCE QUALITY ASSURANCE NOX REFERENCE SAMPLES


 Note:  All Method  7 procedures referred to are from the amended method
       published in the Federal Register Vol. 42, No.  160, Part 11,
       Thursday, August 18, 1977,  pp 41784-41786.  This amended method
       should be adhered to in all details in the analysis of these
       reference standards.

 1.  Prepare absorbing solution according to Section 3.1 of the method.

 2.  Prepare each reference sample for analysis as follows:  Wrap a paper
    towel around the ampule and with the ampule in an upright position
    break off the  top at the prescored mark by exerting pressure sideways.
    From the  ampule pipette exactly 5 ml of the reference sample into a
    100-ml beaker.  Add 25 ml absorbing solution to the beaker; adjust
    the pH to 9-12 (using pH paper as indicated in Section 4.2 of the
    method) by dropwise addition of sodium hydroxide (1N). Quantitatively
    transfer  the contents of the beaker to a 50-ml volumetric flask and
    dilute exactly to the mark with deionized, distilled water.  Mix
    thoroughly and pipette a 25-ml aliquot of the diluted sample into
    a porcelain evaporating dish.  Beginning with the evaporation step
    in Section 4.3, complete the sample analysis.

 3.  Calculate total yg NO2 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 Vsc in Equation 7-4.

 4.  Record the reference sample numbers and their corresponding concentrations,
    C, in mg/dscm on the enclosed data form.  Return the form to:

                       Quality Assurance Division (MD-77A)
                       Environmental Monitoring Systems Laboratory
                       Environmental Protection Agency
                       Research Triangle Park, NC 27711
                       ATTN:   Ellen W. Streib

If other than EPA Method 7 is  used for your analyses,  please explain in detail
your analytical procedure on the back of the enclosed data form.
                                     33

-------
                    COAL AUDIT  PROGRAM  INFORMATION
1.   There 1s 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
     1n weight percent with gross calorific value reported as BTU/lb.

3.   All methods used 1n 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 1s 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 1s enclosed for this purpose.

9.   If you nave any questions concerning this or any source method audit,
     please call (919/541-7834).
                                        34
                                                       frUS GOVERNMENT PRINTING OFFICE 1966 - 646-116/20766

-------