EPA-600/2-76-080
March 1976
Environmental Protection Technology Series
A QUALITY ASSURANCE PROGRAM FOR
THE EPA/SHAWNEE WET LIMESTONE
SCRUBBER DEMONSTRATION PROGRAM
SEZ
UJ
O
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U. S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield. Virginia 22161.
-------
EPA-600/2-76-080
Mirch 1976
A QUALITY ASSURANCE PROGRAM
FOR THE EPA/SHAWNEE WET LIMESTONE
SCRUBBER DEMONSTRATION PROGRAM
by
James Buchanan
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, NC 27709
Contract No. 68-02-1398, Task 20
ROAP No. ABA-011
Program Element No. EHB-557
EPA Project Officer: Larry D. Johnson
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ACKNOWLEDGMENTS
The work on this project was performed by the Systems and Measure-
ments Division of the Research Triangle Institute. Mr. Frank Smith,
Supervisor, Quality Assurance Section, served as the Project Leader.
Dr. James Buchanan of the Quality Assurance Section was responsible for
the coordination of the program. Institute staff members Dr. D. E.
Wagoner and Mr. Larry Hackworth, analytical chemists, Mr. Leon Bissette,
an electrical engineer, and Dr. Buchanan, a physical chemist, were
major contributors to the program. Project Officer for the Environ-
mental Protection Agency was Dr. L. D. Johnson of the Process Measure-
ments Branch of the Industrial Environmental Research Laboratory. The
Research Triangle Institute acknowledges the cooperation and assistance
of the Project Officer and Dr. R. Statnick of the Process Measurements
Branch. The Institute also appreciates the assistance and guidance
provided by Mr. John Williams, the EPA Project Officer for the Shawnee
wet limestone scrubber demonstration. Finally, gratitude is extended
to Mr. Joe Barkley and Mr. Ken Metcalf of TVA and Mr. Dewey Burbank of
Bechtel Corporation for their cooperation at the Shawnee test site.
iii
-------
TABLE OF CONTENTS
SECTION EA££
1.0 INTRODUCTION ]
2,0 THE CONTROL LABORATORY *
2-1 MEASUREMENT OF pH 2
2-2 ANALYSIS OF SLURRY 5
2-3 OVERALL LABORATORY EVALUATION 8
3.0 EFFLUENT GAS STREAM SAMPLING 10
3-1 PARTICULATE MASS LOADING 10
3-1.1 PITOT TUBE COMPARISON 10
3.1.2 TEMPERATURE MEASUREMENT 10
3.1-3 MOISTURE MEASUREMENT 10
3.1-4 VOLUME MEASUREMENT 13
3-2 SULFUR DIOXIDE CONCENTRATION DETERMINATIONS 14
4,0 PROCESSOfjSTRUMENTATION 16
5,0 RECOMMENDATIONS 21
APPENDIX A COMPARISON OF ANALYSES OF LIMESTONE SLURRY 24
APPENDIX B QUALITATIVE SYSTEMS REVIEW FOR THE CONTROL 42
LABORATORY
APPENDIX C RTI SULFUR DIOXIDE ANALYTICAL PROCEDURES 69
iv
-------
LIST OF TABLES
TABLE NO. PAGE
1 COMPARISONS FOR pH 4
2 ANALYSIS OF SLURRY SAMPLE 7
3 ANALYSIS OF XRF STANDARD 8
, 4 PARTICULATE EMISSION EVALUATION CHECKLIST 10
5 COMPARISON OF SOX DETERMINATIONS 15
6 CALIBRATION OF LIQUID LEVEL SENSOR 18
LISTS OF TABLES IN APPENDIX A
1 ANALYSIS FOR CALCIUM IN SLURRY SOLID 25
2 ANALYSIS FOR MAGNESIUM IN SLURRY SOLID 26
3 ANALYSIS FOR TOTAL SULFUR IN SLURRY SOLID 2?
4 ANALYSIS FOR CALCIUM IN SLURRY FILTRATE 28
5 ANALYSIS FOR MAGNESIUM IN SLURRY FILTRATE 29
6 ANALYSIS FOR SODIUM IN SLURRY FILTRATE 30
7 ANALYSIS FOR POTASSIUM IN SLURRY FILTRATE 31
8 ANALYSIS FOR CHLORIDE IN SLURRY FILTRATE 32
9 LABORATORY: SHAWNEE (TVA) 33
1° LABORATORY: RTI (SHAWNEE FILTERED) 34
11 LABORATORY: RTI 35
12 LABORATORY: MUSCLE SHOALS (TVA) 56
13 LABORATORY: CHATTANOOGA (TVA) 37
14 TOTAL SULFUR DETERMINATIONS, SHAWNEE FILTERED SAMPLES 38
ANALYZED AT RTI
-------
TABLES (CON,)
TABLE NO. PAGE
15 TOTAL SULFUR DETERMINATIONS, RTI FILTERED AND 39
ANALYZED SAMPLES
16 CALCIUM DETERMINATIONS, SHAWNEE FILTERED SAMPLES 40
ANALYZED AT RTI
17 CALCIUM DETERMINATIONS, RTI FILTERED AND ANALYZED 41
SAMPLES
vi
-------
1.0 INTRODUCTION
General aspects of demonstration program quality assurance (QA) programs
have been treated in the report, "Guidelines for Demonstration Project Quality
Assurance Programs."* The present report contains results of an external audit
program carried out by the Research Triangle Institute** during the week of
November 17-21, 1975, at the Shawnee project. The cooperation of TVA, EPA,
and Bechtel Corporation personnel is acknowledged and gratitude for such coop-
eration is hereby expressed.
The approach taken here is similar to that used in the Interim Report
(Subtask 2) for this project; namely, the program is arbitrarily divided into
three major areas. These are the control laboratory, gas stream sampling and
process instrumentation. Sections 2.0, 3.0 and 4.0 treat these areas, in the
order mentioned. Section 5.0 presents recommendations for an external quality
assurance program at the Shawnee Scrubber facility.
The guidelines report was prepared just prior to planning and implementation
of the short-term quality assurance program at the Shawnee project.
**
Research Triangle Park, N.C.
-------
2.0 THE CONTROL LABORATORY
2.1 Measurement of pH
A major element for process control of the scrubber is pH measurement.
For this reason a significant part of the RTI effort was spent in observing
and verifying the TVA techniques for pH determination at the scrubber inlet
and outlet. For audit purposes, a Fisher Research Grade pH Meter (Acumet
Model 320 with expanded scale) was equipped with a Markson #1888 Polymark
Combination pH electrode fitted with*a 20-foot lead. The long lead allowed
measurement in various slurry "pots" on the scrubber without moving the meter.
TVA technicians routinely use the same type probe, but on a less sensitive
meter. In spite of the claimed ruggedness of the probe, it was said by TVA
personnel that some thirty had been purchased, used and discarded over a period
of a year. The probes are necessarily subjected to much physical abuse.
A portable pH system is used several times a day by TVA operators to check
the control room readings obtained from Universal Interloc, Inc., Model 321
Submersion pH sensors. These devices are permanently mounted in pots through
which slurry continuously circulates when the scrubber is operating. Four
such sensors monitor the inlet and outlet pH for the TCA* and Venturi scrubbers.
The system for checking and recording pH is as follows: at least six times a
day (twice each 8-hour shift) an operator goes out to the scrubber for sample
collection and pH reading. A small metal shed on the scrubber holds the pH
meter, buffer solutions, and associated gear. The operator calibrates the
meter by means of a buffer solution of known pH. The buffer pH is selected so
as to be within one unit or less of the actual slurry pH. The temperatures
of both the buffer and the slurry are checked, and the temperature compensating
potentiometer on the meter is adjusted accordingly.** A study by Bechtel
personnel, underway at the time of the RTI audit and independently verified
*
Turbulent Contact Absorber.
A*
It was observed that these temperatures were not regularly measured, some-
times being estimated. This is not likely a significant error source, since
the pH should be relatively insensitive to a temperature variance of a few
degrees centigrade.
-------
by RTI personnel, showed that the combination probe took a finite time to
respond to solution temperature differences. Over the range from 20° to 50° C
(typical for buffer-slurry temperature differences) the response time was
estimated at 4 to 5 minutes for heating and 10 to 15 minutes for cooling. A
constant temperature bath operated by Bechtel was also used by RTI in making
this study. As a result of this study, a directive was issued by TVA that
operators wait at least 5 minutes before recording slurry pH, to allow for
probe warmup after standardization with ambient temperature buffer. Also,
rapid switching from slurry to buffer or buffer to slurry was forbidden. This
may well improve the accuracy of the technique, since over a 30-degree range
the pH error is about 0.1 unit, which is significantly large.
After recording the indicated slurry pH, the operator uses a paging
telephone to call down to the control room. Technically, the operator is not
allowed to know the pH being read by the inline probe. A control room operator
records the pH called down and compares it with the inline probe measurement.
The inline probe reading is adjusted if the difference is considered too large.
The portable pH system is used as a standardization system for the permanently
mounted probes.
In practice it was found that the reverse was sometimes true; i.e., the
portable system was adjusted to deliver the "expected" pH of the Uni-Loc probe.
This was the case particularly at certain test points where the Markson probe
was inserted in a horizontal pipe fitted into the side of the slurry outlet
pot. Measurement of pH by means of these pipes (equipped with a faucet valve)
has now been discontinued, so a detailed discussion of the problems associated
with such measurements is unwarranted.
A series of direct comparison pH measurements were made on November 19-20,
1975. RTI personnel set up a Fisher Acumet meter beside the TVA (Orion) meter
and made simultaneous measurements of slurry pH. The operators were instructed
to carry out their measurements routinely, from standardization to cleanup.
Also during this period the Uni-Loc probes were removed from their pots and
immersed in pH 5 and 6 buffers. Both RTI and TVA long-lead probes were put
into these same buffers, after independent standardization (RTI standardization
was against Fisher Certified Buffer Solution). The values obtained in the
buffer and in slurry are summarized in table 1. Values of pH were read to the
-------
Table 1. Comparisons* for pH
TEST POINT
18161
18252
281 fi3
2S254
nH
RTI (portable)
5.25
5.13
c**
5.05s
5.24
5.28
5.34
5.06
4.92
4.83
4.77
5.03:1
6.026
5.03JJ
6.016
nil
TVA (portable)
5.3
5.2
4.9
5.2
5.2
5.3
5.1
4.9
4.9
4.8
"6.0
5.0
6.0
: pH
TVA (inline)
5.16
5.29
5.04
5.34
5.38
5.34
5.25
5.17
5.02
5.08
4.82
5.77
4.99
6.12
Temperature
(°Centrigrade)
54
54
16
54
50
51
50
53
54
50
21
21 '
21
21
Date
(mo/day/yr)
11/19/75
11/19/75
11/20/75
11/20/75
11/20/75
11/20/75
11/20/75
11/19/75
11/19/75
11/20/75
11/19/75
11/19/75
11/19/75
11/19/75
Time
(hours)
11:15
15:30
08:45
09:10
09:45
11:15
15:15
11:30
15:30
15:15
10:00
10:30
10:50
11:00
1 Venturi effluent hold tank.
2 Venturi outlet.
3 TCA effluent hold tank.
4 TCA outlet.
n
Unless otherwise noted, measurements are on
slurry in inline probe pots.
*
Superscript number indicates measurement of
a buffer solution of pH 5 or 6.
-------
nearest 0.01 unit using the Acumet expanded scale, but readings of such
precision were not possible with the Orion meter, where estimations to 0.1 pH
unit were made. Actual markings on the Orion meter scale were at 0.2 pH unit
intervals.
Two observations are in order, after study of table 1:
1) The RTI and TVA portable (long-lead) pH systems agreed within 0.1 pH
unit or better in every comparison made. This verifies the accuracy of the
TVA portable system, since the RTI system was standardized against certified
buffer. Operator reading errors are probably the largest error source in the
pH measurement process.
2) The Dni-Loc system readings differed from RTI readings by 0.0 to 0.3
pH unit, with the mean difference being 0.14 unit over 14 readings. Certainly
it would be unwise to dwell on the significance of the statistics of such a
brief study. One point can be made, however, with respect to the confidence
placed in the Uni-Loc pH readings. It appears unlikely that, using the present
system of inline probes, pH measurements on the slurry can be made to better
than 0.1 unit. There are a number of factors which militate against greater
accuracy, the major one probably being the nature of the slurry itself. This
viscous, highly abrasive suspension tends to clog lines and coat out on probe
surfaces, thus making reproducible measurements quite difficult. The non-
equilibrium mixture of reactive chemicals has a pH which will change on removal
from the scrubber proper; i.e., as it flows into the pots within which mea-
surements are made. Another factor is the difficulty of standardization of
inline probes. The present system calls for probe removal, cleaning and
standardization roughly each 2 days. The accuracy of the pH measurement is
surely dependent on the condition of the probe surface, and restandardizing
is ideally done shortly before each measurement.
2.2 Analysis of Slurry
As a check on the reliability of the chemical analysis phase of the
scrubber operation, a series of slurry samples was collected* and sent to
All samples were taken from the venturi effluent hold tank.
-------
several other laboratories for independent analysis of both the liquid phase
and suspended solids.
The laboratories originally selected for participation in this phase of
the audit were two TVA laboratories (Chattanooga and Muscle Shoals), EMSL
(EPA-RTP) and RTI. The EMSL laboratory later declined to participate in the
project.
Each laboratory was given five 1-liter samples of slurry, taken concur-
rently with control laboratory samples. After filtering and drying the solid,
it was analyzed for calcium, magnesium, and total sulfur. The filtrate was
analyzed for calcium, magnesium, solium, potassium, and chloride. As a check
on filtration technique the RTI laboratory was given, in addition to samples
of slurry, samples which were filtered in the control laboratory. The filtrate
and dried solid were then analyzed in the same way as the unfiltered slurry
samples.
Complete results of these analyses are given in appendix A. Results for
one sample, collected at 3 p.m. on November 18, are summarized in table 2.
No statistical analyses have been performed on these data, in view of the
short time period for data collection and the small number of samples analyzed.
The data serves merely as representative of the type obtainable by means of
split samples. One observation is pertinent at this time; namely, that each
cooperating laboratory was strikingly consistent in its analysis of each ele-
ment, with large average differences in between-laboratory results. Part of
this can be attributed to use of different analytical techniques by different
laboratories. In an ongoing program involving several laboratories it would
be necessary to evaluate each analytical technique as to bias. The comprehen-
sive final report for this project will attempt to evaluate each individual
technique used, for comparative purposes.
Liquid analysis at the control laboratory was as follows: calcium mag-
nesium, sodium, and potassium were done by atomic absorption (AA) and chloride
by potentiometric titration. Caldium, magnesium, and total sulfur analysis in
the solid were done by X-ray fluorescence (XRF) . The XRF standard was a sample
of slurry solids sent out to three other TVA laboratories by Mr. Barkley
chief chemist at the control laboratory. Results of the independent laboratory
analyses were averaged and assigned as standard values for calcium (as CaO)
-------
Table 2. Analysis of slurry sample
Shawnee RTI RTI
control (Shawnee (RTI Muscle
Element laboratory filtered) filtered) Chattanooga shoals
Liquid
(in ppm)
Ca
Mg
Na
K
Cl
Solid
(in wt %)
CaO
MgO
TS(SO,)
O
1710
699
57
121
3580
24.50
0.29
34.16
1825
945
69
101
3700
XRF
1 7 "\
AA SS^ RTr
19.01 25.41 22.06
0.48
32.00 21.22
1810
805
161
102
3638
XRF
AA SS RTI
17.96 25.52 22.16
0.52
26.96 30.14
1731
742
62
96
3543
23.24
0.56
30.7
1787
724
37
54
3700
23.1
0.25
30.9
lAtomic absorption, on acid-dissolved solid.
oUsing values given by Shawnee control laboratory for XRF standard.
Using values obtained by RTI for Shawnee XRF standard.
-------
Table 3. Analysis of XRF standard
Element
Ca
Mg
S
Shawnee
value
(wt %)
18.15
0.185
11.38
RTI
value
(wt 35)
15. 761
0.2751
11. 352
Analyzed by atomic absorption
after acid dissolution and dilu-
From precipitation with barium
chloride.
magnesium (as MgO) and total sulfur (as S0_) . A portion of this standard was
taken for analysis by RTI and for use as the RTI XRF standard. Analysis
results for the standard are presented in table 3.
Use of the same XRF standard by both Shawnee and RTI precluded a dis-
crepancy in results due to a matrix effect. Agreement was excellent for sulfur,
fair for calcium, but poor for magnesium. Because of differences in standard
assignment values for calcium, two XRF results are given, as explained in the
footnote to table 2. Magnesium could not be determined with the XRF instru-
mentation used by RTI.
2.3 Overall Laboratory Evaluation
In summation, the control laboratory operation appears to be adequate
for the routine analytical work it performs. It has no formal Quality Control
program, but bad data may be flagged by either TVA or Bechtel personnel. Ac-
ceptance limits on data are not formalized, but "reasonableness" is the
experience-based criterion.
There are problems associated with the lack of operator training programs
incentives for superior performance and the like, but so long as the laboratory
operations remain strictly routine these problems are not likely to seriously
hamper the program.
-------
Equipment and instrumentation are approprite for the type of work done,
and it is maintained on a regular basis (largely by service contracts).
Results of the qualitative systems review for the laboratory are included
as appendix B of this report.
-------
3.0 EFFLUENT GAS STREAM SAMPLING
3.1 Particulate Mass Loading
Side-by-side duplicate runs were not attempted. The entire sampling
procedure was observed, with critical techniques observed repeatedly, during
the site visits. Overall performance was evaluated using the checklist given
as table 4. On a scale of 1 to 5, ranging from unacceptable to excellent, the
Shawnee particulate loading technique was rated 3 (acceptable). Specific
comments are given below.
3.1.1 Pitot Tube Comparison
A comparison of TVA and RTI pitot tubes was performed at the Venturi
inlet only. A check of the outlet tube was not carried out due to the outlet
tube misalignment (>_ 30°) along its roll axis.
Side-by-side measurements were performed. Based upon comparison with the
RTI (NBS calibrated) pitot tube, the TVA tube Cp factor was 0.879. The assumed
value was 0.850. The difference was considered to be negligible.
3.1.2 Temperature Measurement
A system capable of measuring the stack gas temperature to within 1.5
percent of the minimum absolute stack temperature is required. The temperature-
measuring system (inlet sampler) was checked versus a calibrated thermocouple
and was found to be within 1 percent.
3.1.3 Moisture Measurement
The impinger section of the EPA sampling train is intended to collect
moisture from the sample gases for determination of moisture content. The
last impinger contains silica gel to adsorb the water vapor not condensed in
the first two impingers. The moisture content of the sample gas leaving the
silica gel impinger increases as the exit gas temperature rises. Also the
exit gas moisture content will increase as the sample train vacuum increases
at any one sample temperature. Moisture not collected by the condensation
system is incorrectly measured as dry gas by the dry test meter and the error
10
-------
Table 4. Particulate emission evaluation checklist
ES
NO
OPERATION
*1
*2
*2
EQUIPMENT PREPARATION AND CHECK
1. Sampling train assembled and leak-checked.
2. Probe and filter box heaters checked and set for proper
temperatures.
3. Stack gas temperature measuring system assembled and
checked for proper operation by comparing to a mercury
in glass thermometer.
4. Stack gas velocity measuring system assembled and checked
for proper operation.
PRELIMINARY MEASUREMENTS
5. Selection of traverse points according to Method 1.
6. Moisture content by Method 4, or equivalent.
7. Molecular weight by Method 3, or equivalent.
8. Measurement of stack dimensions.
9. Mark probe for sampling at traverse points.
SAMPLE COLLECTION
10. Equal sampling time at each traverse point
11. Probe temperature satisfactory throughout the test.
12. Filter box temperature 120° C + 15° (250° F + 25°) through
the test.
13. Sample gas temperature at last impinger -v 21° C (70° F)
throughout the test.
14. Isokinetic sampling checked and adjusted if necessary at
least every 5 minutes.
15. Leak-check
SAMPLE RECOVERY
16. Satisfactory handling and movement of probe and filter
to sample recovery area.
*1
*2
Not observed.
Not monitored.
11
-------
Table 4. Particulate emission evaluation checklist
(continued)
YES
/
/
/
;
NO
/
/
*3
*3
*1
OPERATION
17. Recovery area satisfactory (i.e., space, cleanliness,
etc.).
18. Sample recovery procedure adequate.
19. Proper labeling of sample containers.
20. Determination of moisture content procedure adequate.
ANALYSIS
21. Proper equilibration of (1) filter, (2) probe wash
residue, and (3) acetone blank residue.
22. Correct collected particulates for acetone blank.
23. Analytical balance checked before weighings.
DOCUMENTATION
24. All information recorded on data sheet as obtained.
25. All unusual conditions recorded.
COMMENTS
*3
^-.Probe wash and acetone blank residue not measured.
li|» rf^L^rf^ u«m • ^^ ^J
Not observed.
12
-------
is carried through the isokinetic and grain loading calculations. However, if
the exit gas temperature is held below 25p C and the rain vacuum is held below
380 mm of Hg, the resulting error in the sample volume will be less than 2 per-
cent. A single RTI reading of exit gas temperature was 22° C.
There was evidence of significant moisture accumulation in the silica gel,
indicating the presence of some water vapor in the total gas volume measured.
This does not likely introduce a large error into the technique, although it
would be advisable to make quantitative or semiquantitative checks on the
actual water volume collected versus water content of the stack gas. This is
not presently being done at Shawnee.
3.1.4 Volume Measurement
The sampling train was checked for accuracy of volumetric measurement
with a calibrated dry test meter (1 cf/revolution) which had been previously
calibrated versus a 1 cf wet test meter. The RTI meter was connected directly
to the Shawnee probe tip, so that the actual volume intake at the probe was
measured. RTI volume was 15.8 percent lower than TVA volume, indicating a
rather large positive bias in the TVA measurement. Critical examination of
the TVA sampling system led to the conclusion that the bias could be attributed
to leaks in the system (broken or cracked polycarbonate impinger tubes, loose
probe tip, etc.). Leakage rate was estimated to be 0.36 cfm at 380 mm of Hg
vacuum.* Inaccuracies in volume measurements appear directly in the concen-
tration and particulate mass emission rate determinations.
A probe tip diameter check was made with a micrometer. The range of the
diameter measurements was 0.7 mm, indicating a severely out-of-round nozzle
which should be repaired or replaced. The estimated nozzle area was calculated
2
to be roughly 20 percent lower than the assumed area (0.583 cm calculated,
0.7125 cm2 assumed). An error in the nozzle diameter is quadrupled in the
process of determining isokinetic sampling rates and is doubled in the percent
of isokinetic sampling calculation. The percent isokinetic, as calculated with
As indicated in table 4, TVA leak-checking was not observed by the audit team.
A thorough leak-check would surely have detected such a significant leak-
rate.
13
-------
respect to the above errors in volume measurement and nozzle diameter could
result in either a positive or negative bias, depending upon which factor
predominates.
3.2 Sulfur Dioxide Concentration Determinations
Sulfur dioxide concentrations at the wet limestones scrubber facility are
determined by means of du Pont Model 400 Photometric Analyzers. The analyzers
continuously monitor inlet and outlet gas streams of the venturi and TCA units.
They constitute a critically important measurement system, and as such a con-
centrated effort was made to assess their performance.
The RTI audit team collected a total of 23 gas samples, all collected at
the venturi inlet. Thirteen of these samples were analyzed by the barium
chloranilate (colorimetric) method, the remaining ten by sodium hydroxide
titration. Results are given in table 5. The average bias of the photometric
method with respect to the wet chemical methods was +6.9 percent, with a stand-
ard deviation of 8.7 percent. These results indicate that the du Pont analyzer
at the venturi inlet is yielding data of high validity. At the 95-percent
confidence level, an individual photometric determination should be within
+18 percent of the true SO mean concentration, biased 7 percent high on the
average, based on the audit data. Due to the time limitation of the audit
team it was not possible to run checks on the other three analyzers. Details
of the RTI analytical procedures are given in appendix C.
14
-------
Table 5. Comparison of SO determinations
Sampling Train Mo.l
Sampling Train No.2
Date
11/18/75
11/18/75
11/19/75
11/19/75
11/19/75
11/19/75
11/19/75
11/19/75
11/19/75
11/19/75
11/19/75
11/19/75
11/19/75
11/19/75
11/20/75
11/20/75
11/20/75
11/20/75
11/20/75
11/20/75
11/20/75
11/20/75
11/20/75
*
fHf\ "I -,
Sample
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Sample
Time
17:15-17:25
17:35-17:45
10:15-10:23
10:42-10:50
11:07-11:21
12:08-12:17
12:43-12:49
12:53-13:02
13:39-13:47
13:50-13:59
14:22-14:31
14:32-14:41
15:06-15:15
15:17-15:27
09:16-09:27
09:30-09:40
09:53-10:04
10:05-10:15
10:22-10:33
10:41-10:52
10:58-11:09
11:11-11:22
11:37-11:48
f\n \\ac-t G nf el ttl
SO by
Barium Chloranilate
(ppm)
4187
2736
Sample Voided
Sample Voided
2261
2172
2309
1958
1582
1827
1750
1724
1690
ital acid determination (TO
S0x Contained
in Isopropranol
Scrubber
-------
4.0 PROCESS INSTRUMENTATION
In a continuing effort to establish guidelines for a comprehensive quality
assurance program to be implemented at the EPA/TVA Shawnee Scrubber demonstra-
tion project and to further the overall QA program for demonstration projects,
a set of quantitative data was scheduled to be taken* onsite by monitoring
actual instrumentation working conditions. A series of calibration tests were
planned using in-house calibrating equipment and an RTI precision digital
multimeter (DMM) as a voltage and current monitor.
Three types of sensors (temperature, differential pressure, and flow rate)
are the primary sources of measurement information being recorded and used
for the scrubber's mechanical operation control. Each sensor's output is
translated by its associated transmitter into a direct current, which passes
through the series-connected set of control room monitoring instruments. This
scheme has the advantage of having the same information (current value) applied
to all readout devices alike, making wiring from the transmitter less critical
than in a potential system (which requires parallel connections for the read-
out devices).
Four readout devices are employed for visual display of the sensor's
output signal. Three of the devices, the Data Acquisition System (DAS), the
Foxboro Trend Recorders, and Foxboro Process Controllers convert the transmitted
current to a potential which is developed across a precision input resistance.
The resistance is an integral part of the transmitter's current loop. The
potential is electronically amplified for further processing. The stability
of the resistance values of these input resistors is critical to the accuracy
of measurement. The fourth device, a Foxboro strip chart recorder, uses a
galvanometer movement (pen motor) to directly convert current into a propor-
tional movement of the recorder pen. Both the strip chart units and the trend
recorder units produce a continuous real-time record of the transmitted current
value.
*
Daily work schedules of TVA process personnel would not allow completion of
the entire testing program over the short on-site time period. Enough was
accomplished for a judgment to be made as to the quality of the instrumenta-
tion facilities.
16
-------
The methods of test and calibration are simple, using rudimentary sources
of stimuli for sensor examination. Straightforward electrical current measur-
ing instruments are used to monitor currents produced by the transmitters.
The tests as performed are sufficient to maintain the quality of measurement
to the degree established by the manufacturers in their design specifications.
Calibration of each measurement channel (consisting of a sensor, a transmitter,
and one or more readout devices) is performed by the instrument technicians.
First, the sensor is subjected to a known stimulus under TVA instrument shop
conditions (with essentially no temperature or humidity controls). The trans-
mitter current level is monitored with a Foxboro Test Set/Calibrator Model
8121. Second, sensor and transmitter are connected in their normal operation
positions and the DAS readout is used to monitor the transmitter's current
level while stimulating the sensor with a known signal. One important part of
any calibration procedure that is not presently being done at Shawnee is the
recording of data so that a calibration curve can be generated. The calibra-
tion curve is useful in checking the linearity of the system as well as out-
of-tolerance points between zero and full scale. Calibration curves made at
a later time can be monitored for changes in instrument performance that could
signal an impending instrument failure. For flow measurements a Foxboro Model
8120 Magnetic Flow Calibration is used in place of the sensor for preliminary
transmitter/indicator calibration. Final flow measurements are calibrated by
pumping known quantities of water for a measured time interval through the
sensor and comparing the DAS reading with the water measurements. The above
calibration measurements are performed using methods described in the Foxboro
instruction manuals. Methods found necessary for setup and calibration of
sensors, transmitters, and indicators for inline operation have not been
written and referenced. The lack of written instructions does not necessarily
indicate that the calibration work done at the Shawnee scrubber has been
inferior. The crux of the matter, as far as a quality assurance program is
concerned, is the complete lack of records pertaining to what has been done,
when it was done and how accurately.
Due to the TVA work schedule only one laboratory calibration procedure
was observed in the maintenance shop. Calibration of a Foxboro liquid level
sensor and transmitter Model 617FM was accomplished using the Foxboro Model
17
-------
8121 calibrator and a mercury-filled manometer. The sensor was subjected to
an air pressure monitored by the manometer. The air pressure was varied from
0 inches Hg to 16.5 inches Hg, representing zero and full scale values re-
spectively. The table below lists the current output for measured values of
air pressure being applied. No recordings of instrument performance prior to
readjustment were available. The mercury manometer used in the above calibra-
tion was readable to 0.1 inch of Hg, which is not sufficient for calibrating
instruments requiring less than 10 inches of Hg pressure for full-scale opera-
tion.
The Fluke DMM was calibrated at an authorized factory service center and
certified to be within the 0.005 percent accuracy specifications for that
model instrument. The 100-ohm resistor was verified at the above facility to
be 100.000 ohms 40.005 percent. Based on the above data, the Foxboro liquid
level sensor and transmitter were calibrated to an accuracy exceeding the
+1.0 percent value specified by the Foxboro Company. A significant part of
the instrument's calibration is the ability of the readout device to indicate
the proper numbers. The Data Acquisition System digital readout is capable of
a +0.1 percent (of full scale) accuracy. Twelve of the 25-ohm precision re-
sistors used by the DAS to monitor loop current were measured by the Fluke
DMM. All were within 0.2 percent of the nominal value. Continuous operation
of the DAS prevented an accurate check of the system. It was noted, however,
Table 6. Calibration of liquid level sensor
Manometer reading Foxboro indicator Fluke DMM reading
(inches Hg) current (milliamps) across 100.0 ohms (volts)
0
4.1
8.2
12.3
16.5
10.0
20.01
30.02
40.01
50.00
1.001
2.002
3.004
4.003
5.001
18
-------
that no data was recorded by the technician (in the form of a table) during
his calibration procedure. The only record made was a note on the instrument's
record card that the instrument had been checked and found to be satisfactory
for service. Such records are not proof of the instrument's ability to per-
form within specifications. Since no previous calibrations performed on this
instrument were recorded, it was impossible to determine what general behavior
the instrument had followed. No trends of performance fluctuations could be
derived. This information is vital to a good quality assurance program.
Records are valuable in determining the nominal rate at which instruments must
be removed from service, repaired, and recalibrated in order to keep their
performance within tolerance requirements. The above test also demonstrated
the ability of the Foxboro Model 8121 Calibrator to measure the transmitter
current to an accuracy of better than +1.0 percent. There were no records
available for this calibrator which would indicate the instrument's accuracy
or shift in calibration from month to month. Since this instrument is used in
the scrubber environment, making it subject to much abuse, it should be cali-
brated against a laboratory standard at least once a month. The same procedure
should be followed for all other instruments used as general purpose calibrators,
Calibration procedures for these instruments need to be developed and performed
by one technician who is qualified for the work. That technician should be
the only person having access to the laboratory standard instruments used to
calibrate all other shop instruments, such as the Foxboro Model 8121 Calibra-
tor. The laboratory standard instruments should be confined to a specific
area in the shop and kept secure from use by other personnel.
The performance of equipment can only be judged by the review of accurate
records which clearly show a life history of each item having a functional
part in the operation of a system. An individual should be made responsible
for keeping all records in a restricted area. Another function of record-
keeping is to maintain up-to-date instruction manuals for each instrument being
used. This would include the manufacturer's manuals, in-house use manuals,
and calibration instructions as well as a history of performance data. An
up-to-date log of the disposition and condition of each instrument on inven-
tory should be kept, for the determination of overall accuracy of data
describing the system's performance. This log would keep management informed
19
-------
about the availability of spares for any changes or modifications that become
necessary. It would also be useful to personnel in the instrumentation group
for making necessary changes in equipment, should their supervisor not be
available during an emergency.
A significant part of any quality assurance program is a formal audit of
the work being performed by a qualified auditor. There was no evidence to show
that an auditing program exists at the scrubber, nor has any attempt been made
to have competent personnel make checks of the work being done by operating
personnel. An audit program requires careful study and analysis of records
being kept, and over-the-shoulder observations of how the work is being done,
to assure management of the quality of work necessary to complete the project
as required. Data submitted by the operating staff without having the benefits
of an auditing program can only be judged as having a confidence level equiva-
lent to the skills of the least trained person performing work on the project.
Management must actively pursue a course consistent with the needed confidence
level of data.
In spite of records being unavailable, it is felt that the electronic
devices used for physical measurements are being maintained sufficiently to
provide pressure, level and flow information to a +2 percent tolerance of
desired nominal values, and temperature information to a +10 percent tolerance
of desired information (temperature sensors can be calibrated to a +2 percent
tolerance of a known temperature—the inaccuracies are estimated to be high
because of the lack of knowledge of the thermodynamics of the stack gases
being measured).
20
-------
5.0 RECOMMENDATIONS
Specifications of a quality control program for the Shawnee scrubber
project does not fall within the scope of the project. The recommendations
made in the following paragraphs will apply to the implementation of an external
(systems review and performance audit) QA program.* This type of program
normally should be carried out by an organization which has no special interest
in the data; i.e., no self-interest to protect and no preconceptions as to the
quality of the information forthcoming. On the other hand, the organization
should be reputable and well qualified to carry out the type of auditing
program desired.
In the case of EPA demonstration projects, EPA may wish to contract a
third party to handle the audit program, or it may handle the program by means
of its own QA staff. In either case, it is quite Important that the auditing
be done competently and objectively.
It is recommended that the wet limestone scrubber operation located at the
Shawnee steam-electric plant, Paducah, Kentucky, be externally audited twice
each calendar year. Timing of the audit program, which normally should take
1 calendar week, should be coordinated among the auditing team, EPA, TVA, and
the Bechtel Corporation. Some advance notice is necessary in order to insure
cooperation of operational personnel. It is not recommended that the audits
be scheduled on a regular hasis, since by definition an audit is conducted
without extensive "preparation" at the project being audited. Advance notice
to EPA and Bechtel supervisory staff should be at least 2 weeks, so that the
audit team can be apprised of special test and analysis schedules which may
alter its audit procedure or cause postponement of the audit itself. Advance
notice to TVA staff (senior chemist, instrumentation foreman) should be at
least 1 week.
It is recommended that, for a facility such as the Shawnee wet limestone
scrubber unit, the audit team concentrate its efforts in the following major
areas:
*It is important that the Shawnee project develop its own internal QA program,
which might well be along the lines suggested here for the external audit
and review.
21
-------
1. Verification of pH measurements at inlet and outlet, on both TCA and
venturi scrubbers. An accurate pH meter brought in by the audit
team, with appropriate buffer solutions, should be used. Measure-
ment of pH is critical to efficient process control at this facility.
2. Independent chemical analysis of slurry samples, by several labora-
tories. A continuing external audit program can aid in establishing
acceptance limits and method biases.
3. Verification of particulate mass loading and sulfur dioxide measure-
ment systems. If possible, side-by-side operation of TVA and audit
team sampling trains should be carried out, with independent analyses
of the collected samples. A wet chemical technique such as total
acid titration should be used to check the S02 analyzer response.
If a duplicate sampling train could be used by the audit team, then
critical measurement parameters should be identified and checked.
For stack sampling procedure, this includes (at a minimum) :
a. Sample volume measurement check by means of a calibrated wet
test or dry gas meter;*
b. Pitot tube (Cp factor) check by means of an NBS calibrated pitot
tube;
c. Thermometer and thermocouple checks with a calibrated tempera-
ture measurement system;
d. Stack gas moisture content check by means of an absorbing im-
pinger train.
4. Monitoring of process control instrumentation with appropriate elec-
tronic devices (precision resistances, calibrated voltage and current
meters with digital readout, signal generators).
It is premature to set acceptance limits based on audit data from the
Shawnee scrubber. Several observations can be made, however, as follows:
1. pH measurement as currently being made in the scrubber pots should
be accurate within 0.2 pH unit, possibly within 0.1 unit. The in-
line systems suffer from bias due to flow rate variance, probe sur-
face modifications and the like. A much more sophisticated (and
costly) measurement system, employing in-situ self-cleaning probes,
could possibly improve the accuracy and precision of the measurement.
It should be kept in mind, however, that because of the nature of
the medium being measured there is a "built-in" uncertainty in the
pH which no measurement system can overcome.
2. Sulfur dioxide concentrations in the stack gas, as read from the
du Pont Photometric Analyzer, are probably reliable to H-20 percent.
For suggestions as to techniques available, see "Process Stream Volumetric
Flow Measurement and Gas Sample Extraction Methodology," by Brooks & Williams
This manual (TRW Document No. 24916-6028-RU-OO) was prepared under EPA Con-
tract No. 68-02-1412, for the Process Measurements Branch of IERL.
22
-------
3. Control laboratory slurry analysis acceptance limits will vary
depending on the particular technique but should be controllable
generally to +20 percent. Particular attention should be devoted to
defining and quantitating matrix effects in the X-ray fluorescence
method, with appropriate corrections for concentration differences.
This appears to be a sizable error source.
4. Process instrumentation and physical measurement techniques do not
represent sizable error sources at the Shawnee facility, although
recordkeeping by instrumentation personnel has been minimal. Gener-
ally, instrumentation is reliable and well maintained.
23
-------
APPENDIX A COMPARISON ON ANALYSIS OF LIMESTONE SLURRY
Cooperating laboratories were:
1. TVA Power Service Center Laboratory, Chattanooga, Tennessee - Mr.
John Rose, contact
2. TVA Power Service Center Laboratory, Muscle Shoals, Alabama - Dr.
Guerry McClellon, contact
3. Research Triangle Institute, Research Triangle Park, North Carolina -
Dr. D. E. Wagoner, contact
Results from RTI laboratories are presented in two sections. One set of
data was obtained on slurry which was filtered at the Shawnee Laboratory. The
second set of data results from analysis of samples filtered in the RTI labora-
tory.
The first eight matrixes present results for each element: calcium, mag-
nesium, and total sulfur in the solid; and calcium, magnesium, sodium, potas-
sium, and chloride in the liquid. The next five matrixes give results of all
analyses for each laboratory, with Shawnee results listed first. The last
four matrixes break down the total sulfur and calcium analyses into results by
a standard "wet" technique, by X-ray fluorescence using Shawnee standard values,
and by X-ray fluorescence using RTI-derived standard values on the Shawnee
standard material.
24
-------
Table 1. Analysis for calcium in slurry solid
Ul
Sample;
Ca (as CaO)
wt %
Laboratory
Shawnee
RTI
(Shawnee filtered)
11/18/75
11/19/75
1100
1500
2300
24.73
22.12
24.50
19.03
22.78
1100
21.05
19.60
18.40
2300
22.87
19.46
RTI
(RTI filtered)
17.99
17.96
18.63
18.13
19.32
Chattanooga
(TVA)
23.45
23.24
23.02
22.46
Muacle Shoals
CTVA)
23.6
23.1
23.6
22.6
22.46
22.9
-------
Table 2. Analysis for magnesium in slurry solid
S3
Sample;
Mg (as MgO)
wt %
Laboratory
Shawnee
RTI
(Shawnee filtered)
RTI
(RTI filtered)
11/18/75
1100
1500
2300
0.30
0.43
0.29
0.48
0.61
0.52
0.28
0.39
0.34
11/19/75
1100
0.27
0.36
0.39
2300
0.29
0.42
0.35
Chattanooga
(TVA)
0.65
0.56
Muscle Shoals
CTVA)
0.25
0.25
0.56
0.24
0.61
0.25
0.61
0.24
-------
Table 3. Analysis for total sulfur in slurry solid
Sample;
TS (as SO )
wt % J
Laboratory
Shawnee
RTI
(Shawnee filtered)
11/18/75
1100
1500
2300
34.78
36.70
34.16
32.00
31.22
33.03
11/19/75
1100
28.17
30.53
2300
31.64
32.80
RTI
(RTI filtered)
28.60
27.00 I 31.13
28.78
32.03
Chattanooga
(TVA)
30.5
30.7
30.0
Muscle Shoals
(TVA)
31.3
30.9
31.4
29.2
29.7
29.6
30.4
-------
Table 4. Analysis for calcium in slurry filtrate
^•N. Sample;
X. Ca
Laboratory ^v.
Shawnee
RTI
(Shawnee filtered)
RTI
(RTI filtered)
Chattanooga
(TVA)
Muscle Shoals
CTVA)
11/18/75 11/19/75
1100 1500 23.00 1100 2300
1720
1775
1700
1756
1787
1710
1825
1810
1740
1787
1810
1810
1730
1676
1716
2090
1708
1720
1596
1787
2315
1885
1825
1732
1716
oo
-------
Table 5. Analysis for magnesium in slurry filtrate
ro
vo
Sample:
Mg
Laboratory
Shawnee
RTI
(Shawnee filtered)
RTI
(RTI filtered)
11/18/75
11/19/75
1100
1500
733
699
785
730
945
805
2300
662
1100
691
813
1000
2300
698
1115
805
795
770
Chattanooga
(TVA)
768
734
763
780
816
Muscle Shoals
(TVA)
724
724
724
724
784
*i ^j- Ir . __ ^ ,, _ _.___— --
-------
Table 6. Analysis for sodium in slurry filtrate
LO
o
Sample:
Na
(ppm)
Laboratory
Shawnee
RTI
(Shawnee filtered)
RTI
(RTI filtered)
11/18/75
11/19/75
1100
1500
71
71
57
69
153
161
2300
70
1100
73
79
75
2300
82
77
180
176
145
Chattanooga
(TVA)
66
62
64
66
69
Muscle Shoals
(TVA)
41
37
37
41
41
-------
Table 7. Analysis for potassium in slurry filtrate
Sample:
K
Laboratory
Shawnee
RTI
(Shavmee filtered)
RTI
(RTI filtered)
11/18/75
11/19/75
1100
1500
2300
118
103
116
121
101
102
123
105
114
1100
126
2300
153
108
111
118
116
Chattanooga
(TVA)
107
96
107
116
116
Muscle Shoals
CTVA)
58
54
50
58
58
-------
Table 8. Analysis for chloride in slurry filtrate
^X. Sample:
\\ Cl
Laboratory ^X^^
Shawn ee
RTI
(Shawnee filtered)
RTI
(RTI filtered)
Chattanooga
(TVA)
Muscle Shoals
OIVA)
11/18/75 11/19/75
1100 1500 2300 1100 2300
3651
3697
3621
3692
3800
3580
3700
3638
3543
3700
3545
3855
3754
3571
3600
F
3545
3660
3519
3571
3600
3580
3987
3566
3628
3700
CO
to
-------
Table 9. Laboratory: Shawnee (TVA)
1
i
i
SOLID (Wt. %)
Ca (CaO)
Mg (MgO)
TS (S03)
' ' '"" ' " *•"" •"••*• "in mummm •• »l !!•••.. • .• . nn.^ai,, mmm^,m », „„,,, , „,, , ,,, „„
11/18/75 11/19/75 {
----- — ,_ (
1100 1500 2300 1100 2300 I
24.73
0.30
34.78
19.03
0.48
32.00
22.78
0.28
31.22
- . — ..
21.05
0.27
28.17
22.87 1
.29 |
i
31.64 !
1
OJ
LIQUID (ppra)
Ca
Mg
Na
K
Cl
1720
733
71
118
1710
699
57
121
3651
3580
1810
662
70
123
2090
691
73
126
!
3545 i 3545
2315
698
82
153
I
• -1
1
3580
-------
Table 10. Laboratory: RTI (Shawnee filtered)
' •
SOLID (Wt. %)
Ca (CaO)
Mg (MgO)
i
.11/18/75
•'" -
1100 1500
22.12
0.43
•T1 ~ ' "" ' •"•
1 TS (S03) j 36*?
19.03
0.48
32.00
11/19/75
2300 1100 2300
19.10
0.39
33.03
18.41
0.36
30.53
19.46
0.42
32.80
••'
f
i LIQUID (ppia)
'
Ca
llg
Na
K
1775
785
71
103
1825
K .
945
69
101
1810
913
79
105
„
1708
1000
75
108
1885
1115
77
111
J
]
Cl
3697
3700
3855
3660
3987
-------
Table 11. Laboratory: RTI
i
SOLID (Wt. %)
Ca (CaO)
!
Mg (MgO)
TS (SO,)
I -3
11/18/75 11/19/75
1100 1500 2300 1100 2300
17.99
0.61
28.6
i
LIQUID (ppn)
1 Ca
M«
I ' i-t-j
-,- .. ._
Na
K
1700
r ' '
730
153
116
19.96 18.63
0.52 0.34
27.0 31.13
1810 1730
805 i 805
1
101 180
102 114
18.13
0.39
28.78
19.32
0.35 1
i
32.03
i
t
1
1720
795
176
118
1825
770
145
I
116
i
- —.....—,_ - , , 1
Cl
3621 I 3638 3754
3519
3566
-------
Table 12. Laboratory: Muscle shoals (TVA)
T ' — — •
SOLID (Wt. %)
Ca (CaO)
Mg (MgO)
TS (S03)
.11/18/75 11/19/75
HOG 1500 2300 1100 2300
23.6:
0,25
31.3
LIQUID (ppra)
Ca
Mg
J
Na
K
Cl
1
1787
724
41
58
38QO
23.1
0.25
30.9
23.6
0.24
31.4
22.6
0.25
29.7
1787
724
37
54
1716
724
37
1787
724
41
i
50 i 58
3700
3600
3600
22.9
0.24
|
36.4
•
1716
784
41
58
i
3700 |
-------
Table 13. Laboratory: Chattanooga (TVA)
.
i
SOLID (Wt. %)
Ca (CaO)
Mg (MgO)
| TS ;
— — — " — - . ,,-T
11/18/75 11/19/75
1100 1500 2300 1100 2300
23.45
0.65
30.5
23.24
0.56
30.7
23.02
0.56
30.0
22.46
0.61
29.2
i
22.46 !
0..61
! t
29.6
CO
LIQUID (ppra)
Ca
Mg
Na
i
K
Cl
1756
768
66
107
3692
1740
r 7M "
62
96
3543
j. _ ..
1676
763
64
107
3571
1596
780
66
116
3571
_ _.,.
•
1732
816
67
(
116
3628 1
-------
Table 14. Total sulfur determinations, Shawnee filtered samples analyzed at RTI
wt % (as SCO
i 3
;
1
By BaCl2 precipitation
11/18/75 11/19/75
1100 1500 2300 1100 2300
36.7
By X-Ray fluroescence
Using Shawnee X-Ray
standard number for TS
Using RTI determined
number for TS
Shawnee X-Ray Standard (as SO^)*
Shawnee given
RTI determined
11 fiR*
— — £.£e\JO
32.00
33.03
ZJL.Z.J
2J.J.D
30.53
32.80
a
*>7 7«
£* f » t J
z/.ys
28.45
28.38
CO
oo
*Shawnee and RTI TS determinations on the XRF standard were virtually idential, so the Shawnee TS value
only was used in calculating wt % TS in each sample.
-------
Table 15. Total sulfur determinations, RTI filtered and analyzed samples
LO
1 *
BaCl? precipitation
X-Ray fluorescence
1
Using Shawnee X-Ray
standard number for TS
Using RTI-determined
number for TS
11/18/75 11/19/75 \
HOC 1500 2300 1100 2300
28.6
35,13
27.00
31.13
30 03
29 8
28.78
34 45
32.03
29 18 —
9
T
-------
Table 16. Calcium determinations, Shawnee filtered samples analyzed at RTI
wt % (as CaO)
By AA
By X-Ray fluorescence
Using Shawnee X-Ray
standard number for CaO
Using RTI determined
number for CaO
Shawnee X-^Ray Standard (as CaO]
Shawnee given
RTI determined
H/18/75 11/19/75
1100 1500 2300 1100 2300
22.12
19.03
19.60
23.13
20.09
25.41
22.06
22.25
19.32
18.41
19.46
25.76
22.37
22.89
19.88
25.41
22.06
-------
Table 17. Calcium determinations, RTI filtered and analyzed samples
wt % (as CaO)
i
By AA
By X-Ray fluorescence
Using Shawnee X-Ray
standard number for CaO
Using RTI-determined
number for CaO
U/18/75 11/19/75
1100 1500 2300 HOT) 2300
17.99
28.35
24.61
17.96
18.63
25.52
22.16
26.17
22.72
18.13
27.68
24.04
19.32
26.32
22.85
-------
APPENDIX B QUALITATIVE SYSTEMS REVIEW FOR THE CONTROL LABORATORY
This checklist is designed to:
1. Identify existing system deocumentation; i.e., maintenance manuals,
organizational structure, operating procedures, etc.
2. Evaluate the adequacy of the procedures as documented.
3. Evaluate the degree of use of and adherence to the documented proce-
dures in day-to-day operations based on observed conditions (auditor)
and a review of applicable records on file.
The checklist gives three descriptions to each facet of a quality control
system. In all cases the "5" choice is the most desirable and effective mode
of operation; "3" is marginal and tolerable; "1" is definitely unacceptable
and ineffective as a mode of operation.
It is not always possible to describe accurately all options with only
three choices. Therefore, a "2" or "4" rating may be selected if the evaluator
feels that an in-between score is more descriptive of the actual situation.
After all the applicable questions are answered, an average is computed
to give an overall indication of the quality system effectiveness.
Generally, a rating of 3.8 or better is considered acceptable.
A rating between 2.5 and 3.8 indicates a need for improvement but there
is no imminent threat to project performance as it now stands.
For the control laboratory, the results are as follows:
a. Of 82 check questions, 65 were answered on site;
b. Average score was 3.0 (5.0 maximum), indicating a. satisfactory but
not outstanding program as presently operated;
c. The control laboratory was judged weak in its quality control organi-
zation, procurement and inventory procedures, and personnel training
policy;
d. Strong points were its day-to-day "in-process" quality assurance,
its calibration procedures, and its facilities and equipment.
The completed questionnaire, with indicated judgments in specific areas,
is given herewith. These judgments are for the control laboratory operation
only.
42
-------
QUALITY ORGANIZATION
SCORE
(1) Overall responsibility for quality assurance (or
quality control) for the organization is:
(a) Assigned to one individual by title (e.g.,
Quality Control Coordinator). 5
(b) Assigned to a specific group within the organi-
zation. 3
/ (c) Not specifically assigned but left to the dis-
cretion of the various operation, analytical,
inspection, antf' testing personnel. 1
(2) The Quality Control Coordinator is located in the
organization such that:
(a) He has direct access to the top management level
for the total operation independent of others in-
volved in operational activities. 5
(b) He performs as a peer with others involved in
operational activities with access to top manage-
ment through the normal chain of command. 3
/ (c) His primary resonsibility is in operational ac-
tivities with quality assurance as an extra or
part-time effort. 1
(3) Data reports are distributed to:
*
y (a) All levels of management. ->
(b) One level of management only. 3
(c) The quality control group only. 1
(4) Data Quality Reports contain:
(a) Information of operation trends, required actions,
and danger spots. 5
V (b) Information on suspected data/analyses and their
causes
(c) Percent of valid data per month. 1
Management appropriate levels in all applicable organizations
such as subcontractors, prime contractor, EPA.
43
-------
THE QUALITY SYSTEM
SCORE
(5) The quality control system is:
(a) Formalized and documented by a set of procedures
which clearly describe the activities necessary
and sufficient to achieve desired quality objec-
tives from procurement through to reporting data
to the EPA/RTP.
(b) Contained in methods procedures or is implicit in
those procedures. Experience with the materials,
product, and equipment is needed for continuity
of control.
v (c) Undefined in any procedures and is left to the
current managers or supervisors to determine as
the situation dictates.
(6) Support for quality goals and results is indicated by:
(a) A clear statement of quality objectives by the
top executive with continuing visible evidence
of its sincerity to all levels of the organiza-
tion. 5
-» (b) Periodic meetings among operations personnel and
the individual(s) responsible for quality assur-
ance on quality objectives and progress toward
their achievement. 3
(c) A "one-shot" statement of the desire for product
quality by the top executive after which the
quality assurance staff is on its own. 1
(7) Accountability for quality is:
(a) Clearly defined for all sections and operators/
analysts where their actions have an impact on
quality. 5
(b) Vested with the Quality Control Coordinator who
must use whatever means possible to achieve
quality goals. 3
J (c) Not defined. 1
44
-------
NA
THE QUALITY SYSTEM (continued)
SCORE
(8) The acceptance criteria for the level of quality of
the demonstration projects routine performance are:
(a) Clearly defined in writing for all characteristics. 5
(b) Defined in writing for some characteristics and
some are dependent on experience, memory, and/or
verbal communication. 3
v (c) Only defined by experience and verbal communica-
tion, i
(9) Acceptance criteria for the level of quality of the
project's routine performance are determined by:
(a) Monitoring the performance in a structured pro-
gram of inter- and intralaboratory evaluations.
/ (b) Scientific determination of what is technically
feasible.
(c) Laboratory determination of what can be done
using currently available equipment, techniques,
and manpower.
(10) Decisions on acceptability of questionable results are
made by:
/ (a) A review group consisting of the chief chemist or
engineer, quality control, and others who can ren-
der expert judgment. 5
(b) An informal assessment by quality control. 3
(c) The operator/chemist. 1
(11) The quality control coordinator has the authority to:
(a) Affect the quality of analytical results by in-
serting controls to assure that the methods meet
the requirements for precision, accuracy, sensi-
tivity, and specificity. 5
(b) Reject suspected results and stop any method that
produces high levels of discrepancies. 3
(c) Submit suspected results to management for a de-
cision on disposition. !
45
-------
IN-PROCESS QUALITY ASSURANCE
SCORE
(12) Measurement methods are checked:
/ (a) During operation for conformance to operating
conditions and to specifications; e.g., flow
rates, reasonableness of data, etc. 5
(b) During calibration to determine acceptability
of the results. 3
(c) Only when malfunctions are reported. 1
(13) The capability of the method to produce within
specification limit, is: t
/ (a) Known through method capability analysis (X-R
Charts) to be able to produce consistently
acceptable results. 5
(b) Assumed to be able to produce a reasonably
acceptable result. 3
(c) Unknown. 1
(14) Method determination discrepancies are:
(a) Analyzed immediately to seek out the cause and
apply corrective action. 5
/ (b) Checked out when time permits. 3
(c) Not detectable with present controls and pro-
cedures. 1
(15) The operating conditions (e.g., flow rate, range,
temperature, etc.) of the methods are:
/ (a) Clearly defined in writing in the method for each
significant variable. 5
(b) Controlled by supervision based on general guide-
lines . 3
(c) Left up to the operator/analyst. 1
46
-------
IN-PROCESS QUALITY ASSURANCE (continued)
SCORE
(16) Auxiliary measuring, gaging, and analytical in-
struments are:
(a) Maintained operative, accurate, and precise
by regular checks and calibrations against
stable standards which are traceable to the
U.S. Bureau of Standards.
(b) Periodically checked against a zero point
or other reference and examined for evidence
of physical damage, wear, or inadequate main-
tenance .
(c) Checked only when they stop working or when
excessive defects are experienced which can
be traced to inadequate instrumentation.
47
-------
CONFIGURATION CONTROL
SCORE
NA (17) Procedures for documenting, for the record, any
design change in the system are:
(a) Written down and readily accessible to those
individuals responsible for configuration
control. 5
(b) Written down but not in detail. 3
(c) Not documented. 1
(18) Engineering schematics are:
(a) Maintained current on the system and subsystem
levels. 5
•"< - -t
(b) Maintained current on certain subsystems only. 3
(c) Not maintained current. 1
(19) All computer programs are:
(a) Documented and flow charted. 5
(b) Flow charted. 3
(c) Summarized. 1
? (20) Procedures for transmitting significant design
changes in hardware and/or software to the EPA
project officer are:
(a) Documented in detail sufficient for imple-
mentation. 5
(b) Documented too briefly for implementation. 3
(c) Not documented. 1
48
-------
DOCUMENTATION CONTROL
SCORE
(21) Procedures for making revisions to technical docu-
ments are:
(a) Clearly spelled out in written form with the
line of authority indicated and available to
all involved personnel. 5
J (b) Recorded but not readily available to all per-
sonnel . 3
(c) Left to the discretion of present supervisors/
managers. 1
(22) In revising technical documents, the revisions are:
(a) Clearly spelled out in written form and dis-
tributed to all parties affected on a con-
trolled basis which assures that the change
will be implemented and permanent. 5
/ (b) Communicated through memoranda to key people who
are responsible for effecting the change through
whatever method they choose. 3
(c) Communicated verbally to operating personnel who
then depend on experience to maintain continuity
of the change. 1
(23) Changes to technical documents pertaining to opera-
tional activities are:
i
(a) Analyzed to make sure that any harmful side effects
are known and controlled prior to revision effec-
tivity.
/ (b) Installed on a trial or gradual basis, monitoring
the product to see if the revision has a net bene-
ficial effect.
(c) Installed immediately with action for correcting
side effects taken if they show up in the final
results.
49
-------
DOCUMENTATION CONTROL (continued)
SCORE
(24) Revisions to technical documents are:
/ (a) Recorded as to date, serial number, etc. when
the revision becomes effective. 5
(b) Recorded as to the date the revision was made
on written specifications. 3
(c) Not recorded with any degree of precision. 1
NA (25) Procedures for making revisions to computer software
programs are:
(a) Clearly spelled out in written form with the line
of authority indicated.
(b) Not recorded but changes must be approved by the
present supervisor/manager.
(c) Not recorded and left to the discretion of the
programmer.
NA (26) In revising software program documentation, the re-
visions are:
(a) Clearly spelled out in written form with reasons
for the change and the authority for making the
change distributed to all parties affected by the
change. 5
(b) Incorporated by the programmer and communicated
through memoranda to key people. 3
(c) Incorporated by the programmer at his will. 1
50
-------
DOCUMENTATION CONTROL (continued)
SCORE
NA (27) Changes to software program documentation are:
(a) Analyzed to make sure that any harmful side
effects are known and controlled prior to
revision effectivity.
(b) Incorporated on a trial basis, monitoring the
results to see if the revision has a net bene-
ficial effect.
(c) Incorporated immediately with action for de-
tecting and correcting side effects taken as
necessary.
NA (28) Revisions to software program documentation are:
»
(a) Recorded as to date, program name or number, etc.,
when the revision becomes effective. 5
(b) Recorded as to the date the revision was made. 3
(c) Not recorded with any degree of precision. 1
51
-------
PREVENTIVE MAINTENANCE
SCORE
(29) Preventative maintenance procedures are:
(a) Clearly defined and written for all measure-
ment systems and support equipment. 5
/ (b) Clearly defined and written for most of the
measurement systems and support equipment. 3
(c) Defined and written ifor only a small fraction!
of the total number of systems. 1
(30) Preventative maintenance activities are documented:
(a) On standard forms in station log books. 5
/ (b) Operator/analyst summary in log book. 3
(c) As operator/analyst notes. 1
(31) Preventative maintenance procedures as written appear
adequate to insure proper equipment operation for:
(a) All measurement systems and support equipment.
/ (b) Most of the measurement systems and support
equipment.
(c) Less than half of the measurement systems and
support equipment.
(32) A review of the preventative maintenance records indi-
cates that:
/ (a) Preventative maintenance procedures have been
carried out on schedule and completely documented. 5
(b) The procedures were carried out on schedule but
not completely documented. 3
(c) The procedures were not carried out on schedule
all the time and not always documented. i
52
-------
PREVENTATIVE MAINTENANCE (continued)
SCORE
(33) Preventative maintenance records (histories) are:
(a) Utilized in revising maintenance schedules,
developing an optimum parts/reagents inventory
and development of scheduled replacements to
minimize wear-out failures.
(b) Utilized when specific questions arise and for
estimating future work loads.
(c) Utilized only when unusual problems occur.
53
-------
DATA VALIDATION PROCEDURES
SCORE
(34) Data validation procedures are:
(a) Clearly defined in writing for all measurement
systems.
(b) Defined in writing for some measurement systems,
some dependent on experience, memory, and/or
verbal communication.
/ (c) Only defined by experience and verbal communica-
tion.
(35) Data validation procedures are:
/ (a) A coordinated combination of computerized and
manual checks applied at different levels in
the measurement process.
(b) Applied with a degree of completeness at no
more than two levels of the measurement process.
(c) Applied at only one level of the measurement
process.
(36) Data validation criteria are documented and include:
(a) Limits on: (1) operational parameters such as
as flow rates; (2) calibration data; (3) special
checks unique to each measurement; e.g., succes-
sive values/averages; (4) statistical tests; e.g.,
outliers; (5) manual checks such as hand calcula-
tions.
(b) Limits on the above type checks for most of the
measurement systems.
/ (c) Limits on some of the above type checks for only
the high priority measurements.
54
-------
DATA VALIDATION PROCEDURES (continued)
SCORE
(37) Acceptable limits as set are reasonable and adequate
to insure the detection of invalid data with a high
probability for:
(a) All measurement systems. 5
(b) At least 3/4 of the measurement systems. 3
/ (c) No more than 1/2 of the measurement systems. 1
(38) Data validation activities are:
(a) Recorded on standard forms at all levels of the
measurement process. 5
/ (b) Recorded in the operator's/analyst log book. 3
(c) Not recorded in any prescribed manner. 1
(39) Examination of data validation records indicates that:
/ (a) Data validation activities have been carried out
as specified and completely documented. 5
(b) Data validation activities appear to have been
performed but not completely documented. 3
(c) Data validation activities, if performed, are
not formally documented. 1
(40) Data validation summaries are:
/ (a) Prepared at each level or critical point in the
measurement process and forwarded to the next level
with the applicable block of data. 5
(b) Prepared by and retained at each level. 3
(c) Not prepared at each level nor communicated between
levels. 1
55
-------
DATA VALIDATION PROCEDURES (continued)
SCORE
(41) Procedures for deleting invalidated data are:
(a) Clearly defined in writing for all levels of
the measurement process, and invalid data are
automatically deleted when one of the com-
puterized validation criteria are exceeded.
(b) Programmed for automatic deletion when com-
puterized validation criteria are exceeded
but procedures not defined when manual checks
detect invalid data.
/ (c) Not defined for all levels of the measurement
process.
(42) Quality audits (i.e., both on-site system reviews and/or
quantitative performance audits) independent of the nor-
mal operations are:
(a) Performed on a random but regular basis to insure
and quantify data quality. 5
J (b) Performed whenever a suspicion arises that there
are areas of ineffective performance. 3
(c) Never performed. 1
56
-------
PROCUREMENT AND INVENTORY PROCEDURES
SCORE
(43) Purchasing guidelines are established and documented
for:
(a) All equipment and reagents having an effect
on data quality. 5
(b) Major items of equipment and critical reagents. 3
/ (c) A very few items of equipment and reagents. 1
(44) Performance specifications are:
(a) Documented for all items of equipment which have
/ an effect on data quality. 5
(b) Documented for the most critical items only. 3
(c) Taken from the presently used items of equipment. 1
(45) Reagents and chemicals (critical items) are:
(a) Procured from suppliers who must submit samples
for test: and approval prior to initial shipment. 5
(b) Procured from suppliers who certify they can meet
/ all applicable specifications. 3
Cc) Procured from suppliers on the basis of price and
delivery only. 1
(46) Acceptance testing for incoming equipment is:
(a) An established and documented inspection procedure
to determine if procurements meet the quality assur-
ance and acceptance requirements. Results are docu-
mented. 5
(b) A series of undocumented performance tests performed
by the operator before using the equipment. 3
/ (c) The receiving document is signed by the responsible
individual indicating either acceptance or rejection. 1
57
-------
PROCUREMENT AND INVENTORY PROCEDURES
SCORE
(47) Reagents and chemicals are:
(a) Checked 100 percent against specification, quantity,
and for certification where required and accepted
only if they conform to all specifications. 5
/ (b) Spot-checked for proper quantity and for shipping
damage. 3
(c) Released to analyst by the receiving clerk without
being checked as above. 1
(48) Information on discrepant purchased materials is:
(a) Transmitted to the supplier with a request for
corrective action. 5
/ (b) Filed for future use. 3
(c) Not maintained. 1
? (49) Discrepant purchased materials are:
(a) Submitted to a review by Quality Control and
Chief Chemist for disposition. 5
(b) Submitted to Service Section for determination
on acceptability. 3
(c) Used because of scheduling requirements. 1
(50) Inventories are maintained on:
(a) First-in, first-out basis. 5
/ (b) Random selection in stock room. 3
(c) Last-in, first-out basis. 1
58
-------
PROCUREMENT AND INVENTORY PROCEDURES (continued)
SCORE
(51) Receiving of materials is:
(a) Documented in a receiving record log giving a
description of the material, the date of re-
ceipt, results of acceptance test, and the sig-
nature of the responsible individual. 5
(b) Documented in a receiving record log with material
title, receipt date, and initials of the individual
logging the material in. 3
/ (c) Documented by filing a signed copy of the requisi-
tion. 1
(52) Inventories are:
(a) Identified as to type, age, and acceptance status. 5
(b) Identified as to material only. 3
/ (c) Not identified in writing. 1
(53) Reagents and chemicals which have limited shelf life
are:
(a) Identified as to shelf life expiration date and
systematically issued from stock only if they
are still within that date. 5
(b) Issued on a first-in, first-out basis, expecting
that there is enough safety factor so that the
expiration date is rarely exceeded. 3
/ (c) Issued at random from stock. 1
59
-------
PERSONNEL TRAINING PROCEDURES
SCORE
(54) Training of new employees is accomplished by:
(a) A programmed system of training where elements of
training, including quality standards, are in-
cluded in a training checklist. The employee's
work is immediately rechecked by supervisors fqr
errors or defects and the information is fed back
instantaneously for corrective action. 5
(b) On-the-job training.by the supervisor who gives an
overview of quality standards. Details of quality
standards are learned as normal results are fed
back to the chemist. 3
/ (c) On-the-job learning with training on the rudiments
of the job by senior coworkers. 1
(55) When key personnel changes occur:
(a) Specialized knowledge and skills are retained in
the form of documented methods and descriptions. 5
(b) Replacement people can acquire the knowledge of
their predecessors from coworkers, supervisors,
and detailed study of the specifications and
memoranda. 3
' (c) Knowledge is lost and must be regained through
long experience or trial-and-error. 1
(56) The people who have an impact on quality; e.g., cali-
bration personnel, maintenance personnel, bench chemists,
supervisors, etc., are:
(a) Trained in the reasons for and the benefits of
standards of quality and the methods by which high
quality can be achieved. 5
(b) Told about quality only when their work falls be-
low acceptable levels. 3
/ (c) Are reprimanded when quality deficiencies are
directly traceable to their work. 1
60
-------
PERSONNEL TRAINING PROCEDURES (continued)
SCORE
(57) The employee's history of training accomplishments is
maintained through:
(a) A written record maintained and periodically re-
viewed by the supervisor. 5
(b) A written record maintained by the employee. 3
/ (c) The memory of the supervisor/employee. 1
(58) Employee proficiency is evaluated on a continuing
basis by:
(a) Periodic testing in some planned manner with the
results of such tests recorded. 5
(b) Testing when felt necessary by the supervisor. 3
/ (c) Observation of performance by the supervisor. 1
NA (59) Results of employee proficiency tests are:
(a) Used by management to establish the need for and
type of special training. 5
(b) Used by the employee for self-evaluation of needs. 3
(c) Used mostly during salary reviews. 1
61
-------
FEEDBACK AND CORRECTIVE ACTION
SCORE
(60) A feedback and corrective action mechanism to
assure that problems are reported to those who
can correct them and that a closed loop mecha-
nism is established to assure that appropriate
corrective actions have been taken is:
(a) Clearly defined in writing with individuals
assigned specific areas of responsibility,
(b) Written in general terms with no assignment
of responsibilities.
/ (c) Not formalized but left to the present super-
visors /managers.
(61) Feedback and corrective action activities are:
(a) Documented on standard forms. 5
/ (b) Documented in the station log book. 3
(c) Documented in the operator's/analyst's
notebook. 1
(62) A review of corrective action records indicates that:
(a) Corrective actions were systematic, timely, and
fully documented. 5
(b) Corrective actions were not always systematic,
timely, or fully documented. 3
(c) A closed loop mechanism did not exist. 1
(63) Periodic summary reports on the status of corrective
action are distributed by the responsible individual to:
(a) All levels of management. 5
(b) One level of management only. 3
(c) The group generating the report only. 1
62
-------
FEEDBACK AND CORRECTIVE ACTION (continued)
SCORE
(64) The reports include:
(a) A listing of major problems for the reporting
period; names of persons responsible for cor-
rective actions; criticality of problems; due
dates; present status; trend of quality per-
formance (i.e., response time, etc.); listing
of items still open from previous reports. 5
(b) Most of the above items. 3
/ (c) Present status of problems and corrective actions. 1
63
-------
CALIBRATION PROCEDURES
SCORE
(65) Calibration procedures are:
/ (a) Clearly defined and written out in step-by-step
fashion for each measurement system and support
device. 5
(b) Defined and summarized for each system and device. 3
(c) Defined but operational procedures developed by
the individual. 1
(66) Calibration procedures as written are:
/ (a) Judged to be technically sound and consistent 5
with data quality requirements.
(b) Technically sound but lacking in detail. 3
(c) Technically questionable and lacking in detail. 1
(67) Calibration standards are:
(a) Specified for all systems and measurement devices
with written procedures for assuring, on a con-
tinuing basis, traceability to primary standards. 5
/ (b) Specified for all major systems with written
procedures for assuring traceability to
primary standards, 3
(c) Specified for all major systems but no procedures
for assuring traceability to primary standards. 1
(68) Calibration standards and traceability procedures as
specified and written are:
/ (a) Judged to be technically sound and consistent with
data quality requirements. 5
(b) Standards are satisfactory but traceability is not
verified frequently enough. 3
(c) Standards are questionable. 1
64
-------
CALIBRATION PROCEDURES (continued)
SCORE
(69) Frequency of calibration is:
(a) Established and documented for each measurement
system and support measurement device 5
/ (b) Established and documented for each major
measurement system. 3
(c) Established and documented for each air quality
measurement system. 1
(70) A review of calibration data indicates that the
frequency of calibration as implemented:
(a) Is adequate and consistent with data quality
requirements. 5
/ (b) Results in limits being exceeded a small
fraction of the time, 3
(c) Results in limits being exceeded frequently. 1
(71) A review of calibration history indicates that:
(a) Calibration schedules are adhered to and results
fully documented. 5
/ (b) Schedules are adhered to most of the time. 3
(c) Schedules are frequently not adhered to. 1
? (72) A review of calibration history and data validation
records indicates that:
(a) Data are always invalidated and deleted when
calibration criteria are exceeded. 5
(b) Data are not always invalidate and/or deleted
when criteria are exceeded. 3
(c) Data are frequently not invalidated and/or de-
leted when criteria are exceeded. 1
65
-------
CALIBRATION PROCEDURES (continued)
SCORE
(73) Acceptability requirements for calibration results
are:
(a) Defined for each system and/or device requiring
calibration including elapsed time since the
last calibration as well as maximum allowable
change from the previous calibration. 5
/ (b) Defined for all major measurement systems. 3
(c) Defined for some major measurements systems
only. 1
(74) Acceptability requirements for calibration results as
written are:
(a) Adequate and consistent with data quality require-
ments . 5
(b) Adequate but others should be added. 3
(c) Inadequate to insure data of acceptable quality. 1
(75) Calibration records (histories) are:
(a) Utilized in revising calibration schedules
(i.e., frequency). 5
(b) Utilized when specific questions arise and re-
viewed periodically for trends, completeness, etc. 3
(c) Utilized only when unusual problems occur. 1
66
-------
FACILITIES/EQUIPMENT
SCORE
(76) Facilities/Equipment are:
/ (a) Adequate to obtain acceptable results. 5
(b) Adequate to obtain acceptable results most of
the time. 3
(c) Additional facilities and space are needed. 1
(77) Facilities, equipment, and materials are:
v (a) As specified in appropriate documentation and/or
standards. 5
(b) Generally as specified in appropriate standards. 3
(c) Frequently different from specifications. 1
(78) Housekeeping reflects an orderly, neat, and effective
attitude of, attention to detail in:
/ (a) All of the facilities. 5
(b) Most of the facilities. 3
(c) Some of the facilities. 1
(79) Maintenance Manuals are:
(a) Complete and readily accessible to maintenance
personnel for all systems, components, and de-
vices. 5
(b) Complete and readily accessible to maintenance
personnel for all major systems, components, and
devices. 3
(c) Complete and accessible for only a few of the
systems. 1
67
-------
RELIABILITY
SCORE
(80) Procedures for reliability data collection, processing,
and reporting are:
(a) Clearly defined and written for all system
components. 5
/ (b) Clearly defined and written for major components
of the system. 3
(c) Not defined. 1
(81) Reliability data are:
(a) Recorded on standard forms. 5
/ (b) Recorded as operator/analyst notes. 3
(c) Not recorded. 1
(82) Reliability data are:
/ (a) Utilized in revising maintenance and/or replace-
ment schedules. 5
(b) Utilized to determine optimum parts inventory. 3
(c) Not utilized in any organized fashion. 1
68
-------
APPENDIX C RTI SULFUR DIOXIDE ANALYTICAL PROCEDURES
Total Aid Determination
Sampling
Samples were obtained by absorbing SC>2 from the stack gas in an impinger
train containing 3% ^2°2' A samPle volume of approximately 10 t of sample was
pulled through the impingers per run. The contents of the impingers as well
as a rinse of 3% H^ were transferred to a labelled sample bottle and shipped
to the RTI laboratory for analysis. In addition an appropriate blank was pre-
pared for all absorbing reagents.
Analys is
Each sample was diluted to 100 m£ and a 15-m£ aliquot was titrated to a
bromophenol blue endpoint with 0.01 N sodium hydroxide. A pH meter and probe
were used to detect the final endpoint. Reproducibility of the titrations was
0.1 percent or less.
Determination by Barium Chloranilate
Sampling
Samples were obtained from the stack gas by a sampling train containing
80% IPN* in the first impinger (bubbler) and 3% H-O- in the second and third
impingers. The sample volume was approximately 10 L. All samples, rinses,
and appropriate blanks were shipped to the RTI laboratory for analysis.
Analysis
The 80% IPN samples were titrated as a total acid sample. The 3% H^
samples were diluted to 100 m£ and a 10-m£ aliquot was taken for analysis. The
pH of the aliquot was adjusted to slightly acidic with IN HC1 and buffered with
5 ml of 5.6 buffer. The sample then was diluted with alcohol and solid barium
chloranilate added. The sample was shaken for 20 minutes and centrifuged at
2,800 rpm. The sample absorbance was determined in a 1-cm cell at a wavelength
of 530 nra.
Isopropanol.
69
-------
TECHNICAL REPORT DATA
fPlcase read 'Instructions on the reverse before completing)
. REPORT NO.
EPA-600/2-76-080
2.
3. RECIPIENT'S ACCESSION-NO.
. TITLE AND SUBTITLE
A Quality Assurance Program for the EPA/Shawnee
Wet Limestone Scrubber Demonstration Program
5. REPORT DATE
March 1976
6. PERFORMING ORGANIZATION CODE
7. AUIHOR(S)
James Buchanan
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
P. O. Box 121S4
Research Triangle Park, NC 2770G
10. PROGRAM ELEMENT NO.
EHB-557; ROAP AEA-011
11. CONTRACT/GRANT NO.
68-02-1398, Task 20
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Envii-onmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PE3IOD COVERED
Task Final; 7-12/75
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES project officer for this report is L.D. Johnson, Mail Drop 82,
Ext 2557.
16. ABSTRACT
The report describes a short-term quality assurance program, implemented
at the EPA wet limestone scrubber facility located at the Shawnee steam/electric
plant, Paducah, Kentucky. The program was part of a project to prepare a set of
quality assurance quidelines for demonstration projects. The control laboratory,
the effluent gas streams, and process instrumentation were reviewed and audited.
In the control laboratory, side-by-side pH measurements were made, and limestone
slurry samples were collected. These samples were sent to three independent
laboratories for analysis of selected elements in the solid and liquid phases. Gas
stream work covered both particulate grain loading and analysis of SC2. Particulate
sampling and weighing techniques were observed, and volume calibration checks
were made. SO2 was collected, analyzed by two chemical methods, and compared
with the in-stack photometric measurement system. Process instrumentation was
checked with portable precision electronic equipment carried on-site and inserted
into instrumentation circuitry to verify accuracy of sensors and readout devices.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Air Pollution
Quality Assurance
Quality Control
Scrubbers
Limestone
Flue Gases
Industrial Processes
Instruments
Sulfur Dioxide
Dust
Sampling
Weight Measurement
Air Pollution Control
Stationary Sources
Control Laboratory
Particulate
b.lDENTIFIERS/OPEN ENDED TERMS
CO5ATI Field/Group
13B
13H,14D 14B
07B
07A 11G
08G
21B
3. UI3THIUUTION .STATCMEN1
Unlimited
EPA Form 2220-1 (3-73)
19.SLCUHITY CLASS (This Report)
Unclassified
21. NO. CP PAGES
76
20. SECURITY CLASS (This page)
Unclassified
------- |