United States Air Pollution Training Institute EPA 450/2-90-002
Environmental Protection MD20 August 1990
Agency Environmental Research Center
Research Triangle Park, NC 27711
Air
APTI
Course SL476B
Continuous Emission
Monitoring Systems:
Operation and Maintenance
of Gas Monitors
Self Instructional
Handbook
Prepared By:
ABB Environmental Services, Inc.
Suite 100
6320 Quadrangle Dr.
Chapel Hill, NC 27514
Under Purchase Order No.
BC0061
EPA Project Officer
Charlie Pratt
United States Environmental Prtectfon Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
-------
Notice
This is not an official policy and standards document. The opinions and selections are those of the authors
and not necessarily those of the Environmental Protection Agency. Every attempt has been made to represent the
present state of the an as well as subject areas still under evaluation. Any mention of products or organizations
does not constitute endorsement by the United States Environmental Protection Agency.
The authors request that any material abstracted from this manual be appropriately referenced as a matter of
professional courtesy in the following manner.
U.S. Environmental Protection Agency. APTI Course SI:476B. Continuous Emission Monitoring Systems:
Operation and Maintenance of Gas Monitors, EPA-450/2-90-002.
Availability
This document is issued by the Manpower and Technical Information Branch, Air Quality Management
Division, Office of Air Quality Planning and Standards, USEPA. It was developed for use in training courses
presented by the EPA Air Pollution Training Institute and others receiving contractual or grant support from the
Institute. Other organizations are welcome to use the document
This publication is available, free of charge, to schools or governmental air pollution control agencies
intending to conduct a training course on the subject convened. Submit a written request to the Air Pollution
Training Institute, USEPA, MD17, Research Triangle Park, NC 27711. Others may obtain copies, for a fee, from
the National Technical Information Service (NTTS), 5825 Port Royal Road, Springfield, VA 22161.
Cover Design
The cover design depicts the basic components of the Dynatron TM 401 Oxygen Monitoring System designed
and marketed by Lear Siegler Measurement Controls Corporation, Englewood, Colorado. This system is
described in Lesson 6.
u
-------
TABLE OF CONTENTS
Page
COURSE INTRODUCTION v
UNTT1 INTRODUCTION TO THE ANALYTICAL METHODS
Lesson 1: Types of Emission Monitoring Systems 1-1
Lesson 2: General Principles of Detection in Monitoring Systems 2-1
Lesson 3: Specific Analytical Methods used by Analyzers 3-1
UNIT 2 OPERATION OF COMMERCIALLY AVAILABLE GAS MONITORS
Lesson 4: Operation of Two Spectroscopic Absorption Analyzers 4-1
Lesson 5: Operation of a Chemiluminescence Analyzer 5-1
Lesson 6: Operation of an Hectrocatalytic Analyzer 6-1
Lesson?: Operation of General Purpose analyzers 7-1
UNIT 3 SYSTEM DESIGN
Lesson 8: Extractive Systems Design 8-1
Lesson 9: In-situ Systems Design 9-1
Lesson 10: Applications of Systems 10-1
UNIT 4 REGULATIONS
Lesson 11: Regulatory Requirements for Gas Emission Monitoring Systems. . . 11-1
Lesson 12: Performance Specification Test Procedures 12-1
Lesson 13: The Requirements of Appendix F - Quality Assurance
Requirements for Continuous Gas Emission Monitors 13-1
UNITS CONTINUING OPERATIONS
Lesson 14: Quality Assurance /Quality Control and Audit Programs
for Continuous Gas Emission Monitors 14-1
Lesson 15: Maintenance Procedures - Problems and Troubleshooting
for Continuous Gas Emission Monitoring Systems ..... .15=1
-------
CC:414
Quiz 1
• This quiz is designed to measure whether you have mastered the objectives of
Lessons A through D of your guidebook.
• It is intended that this quiz be an open-book exam. Since you will be using
Volume III as a reference document in your work, you may also use it here,
along with your guidebook.
• You will have thirty minutes to complete this quiz.
• On the answer sheet, circle the letter that corresponds to the best answer to each
test question. There is only one "best" answer for each question.
1. Quality assurance programs are a vital pan of the Environmental Protection
Agency's air monitoring programs because
a. two EPA administrators have endorsed QA programs.
b. agency actions and decisions need to be based on reliable data.
c. QA programs allow extramural projects to increase in cost.
d. accurate data is never obtained without QA programs.
2. EPA quality assurance requirements could be applicable to which of the
following?
a. a source test team performing an efficiency test on a new ESP for Acme
Power
b. a source test team performing a Method 6 test for a compliance determina-
tion at Coalpower Utilities
c. a source test team performing a combustion evaluation on a new incinerator
for Quad Cities Municipal Public Works
d. a source test team developing an opacity-mass correlation for Timberline
Paper Company
S. Quality control
a. means the same thing as quality assurance.
b. is the system of activities performed to provide assurance that the control
system is performing adequately.
c. is a system of checklists and data forms used to assure data quality.
d. is the system of activities performed to provide a quality product.
6/82
Ql-1
-------
4. The figure below shows the results of an audit program using an EPA
calibrated orifice for checking Method 5 trains.
20
15
1
S.
a
I
°0
'£
«
0.
10
I I I 1
Median =1.8-
I I I
Samples, n = 52S
__ Outliers removed
I I 1
Mean =1.7
Ln
J I
Outliers removed
<-13 -12 -10 -8 -6-4-2024 6 8 10 10<
Difference from EPA value, percent
If the EPA value is considered the true value, this figure shows that the par-
ticipants in the program
a. exhibit a negative bias of about 2%.
b. exhibit a positive bias of about 10%.
c. exhibit a negative bias of about 10%.
d. exhibit a positive bias of about 2%.
5. A beginning source tester
a. needs only "on-the-job" training since the reference methods are so simple
they can be learned in a week.
b. would benefit most from short-course training combined with on-the-job
training.
c. would need to obtain a two-year AA degree in source sampling before being
allowed on a stack.
d. would be an expert sampler after completing a two-day short course.
6. A quality assurance report to management should
a. be as detailed as possible so that all activities can be exactly reproduced.
b. always present raw data and tables instead of graphs.
c. be understandable at a glance.
d. include the forms given in Volume III for each reference method.
Ql-2
-------
7. In the Volume III documentation scheme, the second digit of a three-digit
section number, 3 , is associated with
t
a. the number of pages in the section.
b. procurement, calibration, calculations, etc.
c. Reference Method 2, 3, 4, 5, 6, 7, or 8.
d. pre-sampling operations.
8. Section S.6.4 of Volume III would address
a. on-site measurements.
b. calculations.
c. presampling operations.
d. maintenance.
9. In Volume III, where in each method discussion can clean data forms be
found?
a. Method Highlights section only
b. Section 12.0 only
c. both the Method Highlights section and Section 12.0
d. Sections 1 through 12
10. Which sections of Volume III could be abstracted to develop a calculations
reference manual?
a. 3.1.1, 3.1.2, 3.1.3, 3.1.4 3.1.12
b. 3.1.7, 3.2.7, 3.3.7 3.7.7
c. 6.1, 6.2, 6.3 6.7
d. 3.1.6, 3.2.6, 3.3.6 3.7.6
11. The purpose of emission testing is
a. to extract from the stack or duct, a sample that is representative of emis-
sions from that source during a time period in which the process is under a
specified operating condition.
b. to extract from the stack or duct a sample that is representative of emissions
from that source under all possible operating conditions.
c. to extract from the stack or duct a sample that can be analyzed in the
laboratory for materials hazardous to human health and welfare.
d. to get rich quick.
12. A source test presurvey should be performed
a. only when requested.
b. in all but the most routine cases.
c. almost never, since it adds to the sampling costs.
d. always by telephone.
13. Gas flow conditions are not acceptable at a sampling site if the average of the
traverse point angles of rotation is greater than
a. 5°.
b. 1°.
c. 0°.
d. 10°.
-------
14. The figure below illustrates an example of a Volume III
a. data form.
b. checklist.
c. calculation sheet.
d. sample label.
Container No.
Plant City
Site Pollutant
Date Run No.
Front half Front filter no.
Back half Back filter no
Rinse %
re
Volume: Initial Final «.
C*
Cleanup hy Field Chief
Filter
15. Form MD-4.1 is titled
a. Pretest Preparations.
b. On-site Measurements Checklist.
c. Pitot Tube Calibration Data.
d. Method 2 Gas Velocity and Data Form.
16. Method 1 presents guidelines for the selection of a sampling site and minimum
number of sampling points for a paniculate traverse for a stack diameter
greater than 24 inches. The criterion for using 12 sampling points in the duct
states that the sampling site is at least
a. 8 duct diameters downstream and 2 duct diameters upstream of a flow
disturbance.
b. 2 duct diameters downstream and 8 duct diameters upstream of a flow
disturbance.
c. 4 duct diameters downstream and 8 duct diameters upstream of a flow
disturbance.
d. 6 duct diameters downstream and 2 duct diameters upstream of a flow
disturbance.
17. A procurement log
a. is a quality assurance tool used to track purchased sampling equipment.
b. is used to record calibration data of newly purchased sampling equipment.
c. should be used only when purchasing your first sampling train.
d. is a record of equipment to be included in compliance test reports.
QJ-4
-------
18. The principal action(s) suggested in the activity matrices for "Procurement of
Apparatus", if acceptance limits are not met, is(are)
a. adjust or repair.
b. return to supplier or repair!
c. take corrective action.
d. recalibrate.
19. To leak check a differential pressure gauge,
a. a wet test meter is required.
b. the vendor's certification that the system is leak-free is the only thing that
must be checked.
c. both positive and negative sides are checked.
d. a mercury barometer is connected in parallel.
20. A well-designed Orsat analyzer
a. provides for the easy reading of liquid levels.
b. is as compact as possible.
c. has a large connecting manifold volume.
d. has a minimum of glassware.
21. A probe liner should
a. be Teflon® to reduce gas absorption.
b. be stainless steel to reduce breakage.
c. meet the requirements of the sampling job.
d. not soften or melt at 820 °C (1508°F).
22. Graduated cylinders used in source sampling should have subdivisions
a. S4ml.
b. S>2 ml.
c. <4ml.
d. <2ml.
23. Isopropanol used for SO2 determinations should
a. be checked for peroxide impurities using the barium-thorin titration
method.
b. be checked for sulfate impurities using a spectrometer.
c. be checked for peroxide impurities using a spectrometer.
d. be checked for sulfate impurities using the PDS method.
24. Method 5 sampling nozzles
a. should be measured regularly to the nearest 0.025 mm (0.001 in.).
b. must not exceed % in. in diameter.
c. should be tapered to <40°.
d. should be replaced at specific intervals.
Ql-b
-------
25. Given the probe configuration shown below,
Thermocouple
1 in.
Type S pilot tube
) *tov
Nozzle
Kin.
Sample probe
a. a value of Cp = 0.84 can be assigned to the pitot tube.
b. the probe configuration is unacceptable for testing.
c. the Cp for the configuration should be determined in a wind tunnel.
d. a value of Cp = 0.84 can be assigned to the pitot tube if the thermocouple is
removed.
26. The proper relationship for obtaining the calibration coefficient of a Type S
pitot tube by wind tunnel measurements is
*
a.
A i*
d. s(A) =
27. When calibrating a stack temperature sensor for Method 2, the absolute value
of the sensor reading should be within of the reference thermometer
reading.
a. ±1°C(2°F)
b. ±10%
c. ±5°C(10°F)
d. ±1.5%
28. Data obtained from the Orsat analysis of replicate air samples should
a. agree to within ±0.1% of 20.8% O2.
b. be recorded on an X and R chart.
c. be correlated with the O2 readings of a continuous paramagnetic O*
analyzer.
d. be used in an F factor calculation before comparison.
Ql-6
-------
29. When performing a positive leak check on a Method 4 dry gas metering
system, a leak is indicated by
a. a gurgling noise in the vacuum pump.
b. a loss of pressure on the oil manometer connected to the orifice meter.
c. an increase in pressure on the oil manometer connected to the orifice
meter.
d. movement on the dial of the wet test meter.
30. What is the arrangement of equipment in the initial calibration of a dry gas
meter?
a. orifice meter, dry gas meter, wet test meter, pump
b. orifice meter, pump, dry gas meter, wet test meter
c. orifice meter, wet test meter, pump, dry gas meter
d. orifice meter, dry gas meter, pump, wet test meter
-------
Name.
Date_
CC:414
Quiz 1
Answer Sheet
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
SO.
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
I certify that this test was administered in accordance with the specified test
instructions.
Quiz Supervisor
6/82
-------
CC-.414
Quiz 2
• This quiz is designed to measure whether you have mastered the objectives of
Lessons £ through G of your guidebook. :.'
• It is intended that this quiz be an opcn~book exam. Since you will be using
Volume HI as a reference document in your work, you may also use it here,
along with your guidebook.
• You will have thirty minutes to complete this quiz.
• On the answer sheet circle the letter that corresponds to the best answer to each
test question. There is only one "best" answer for each question.
1. What is the most common acceptance criteria given in the Method 5 activity
matrix fot packing equipment for shipment. •
a. all equipment labeled
b. no visual sign of breakage
c. rigid container lined with polyethylene foam
d. follow specified packing instructions
2. When shipping a meter box,
a. the oiler jar should be removed and packed separately to prevent breakage.
b. the oiler jar should be drained to prevent fouling of die components during
shipment.
c. the dry gas meter should be removed and packed in a separate shipping
container.
d. a shipping container is never needed for commercial models.
S. A description of checks that should be made on a Type S pitot tube and
inclined manometer assembly before a field test, can be found in Volume III in
a. Section S.I.I, Paragraph 1.1.
b. Section 3.1.2, Paragraph 2.1.1.
c. Section 5.1.3, Paragraph 3.1.1.
d. Section 3.1.4, Paragraph 4.2.1.
4. The integrated gas sampling tram used in conjunction with an Orsat deter-
mination of %Ot and %COS should
a. be cleaned and leak checked before each field test.
b. be cleaned and leak checked once every three months.
c. be leak checked only, before each field test.
d. be leak checked only, once every three months.
6/82
Q2-1
-------
5. When should the absorbing solutions of an Oriat gas analyzer be changed?
a. before each field test
b. if more than 10 passes are needed to obtain a constant reading
c. only at the request of the administrator
d. if more than 5 passes are needed to obtain a constant reading
6. A probe liner
a. should be sealed inside the metal sheath, since most stacks are under
negative pressure.
b. does not have to be leak free, since it is in the stack.
c. does not have to be cleaned between runs, since any residues will subtract out.
d. should be constructed of stainless steel to prevent corrosion by sulfuric acid.
7. A Method 7 collection flask should not be cleaned with which of the following?
a. strong detergent and hot water
b. decahydronaphthalene (CuHu)
c. dichromate cleaning solution
d. nitric acid
8. A Method 7 flask should be leak checked at 75 mm (S in.) Hg vacuum before
the test. What level of fluctuation would be acceptable?
a. £ 75 mm (S in.) Hg for at least 1 min
b. £ 10 mm (0.4 in.) Hg for at least 1 min
c. £ 75 mm (3 in.) Hg for at least 1 min
d. £ 10 mm (0.4 in.) Hg for at least 1 min
9. Before a test, a Method 5 filter must be
a. desiccated for at least 24 hours at 20 °C (68°F) and weighed to constant
weight.
b. desiccated for at least 6 hours at 20 °C (68°F) and weighed to constant
weight.
c. oven dried for at least 24 hours at 105°C (220°F) and weighed to constant
weight.
d. oven dried for at least 6 hours at 105 °C (220 °F) and weighed to constant
weight.
10. How many Method 5 filter blanks should be weighed for one field test?
a. 0
b. 1
c. S
d. 9
11. A Type S pitot tube was used to determine the velocity of an exhaust gas
stream containing a high concentration of paniculate matter. What precaution
should have been taken during the test?
a. The thermocouple should have been removed from the tube.
b. Both legs of the pitot tube should have been blown out frequently during
. the test.
c. A Magnehelic* gauge should have been used instead of an oil manometer.
d. A standard tube should have been used instead of a Type S tube.
Q2-2
-------
12. Three methods can be used to obtain a sample for Onat analysis (single-point
grab sample, single-point integrated sample, multipoint integrated sample).
What determines the procedure that should be followed?
a. the stack temperature
b. the capability of the stack test team
c. the presence or absence of stratification
d. the applicable standard
IS. The impinger water volume of a Method 4 test should be measured to the
nearest
a. 1ml.
b. 1 g.
c. 0.1 ml.
d. 0.02 ft'.
14. A Method 4 analysis indicated that the moisture content of an exhaust gas
stream at 82 °C (180°F) was 85% H,O. b there a problem here?
a. Yes. The ice probably melted and the impinger temperatures were too high.
b. Yes. The sampling rate was probably not held constant within ± 10%.
c. Yes. Entrained water droplets were probably collected and calculated as
being in the form of water vapor.
d. No problem.
15. All leak checks for the sample train should be conducted
a. from the nozzle inlet with all train components at operating temperature.
b. from the filter inlet at room temperature.
c. from the probe inlet at ambient temperature.
d. from the nozzle inlet at ambient temperature.
16. How would you correct the "C" factor of your nomograph if your pitot tube
had a coefficient of C,* 0.79?
0.85
a. use
d. The nomograph can't be corrected for a different C*.
17. During a Method 5 test, the isokinetic sampling rate determined for a sam-
pling point
a. should be adjusted if the stack gas velocity increases 5%.
b. must remain constant at that point.
c. must be adjusted if the absolute stack temperature changes 15%.
d. must be adjusted if the silica gel exit temperature increases to 25 °C (77 °F).
£2-3
-------
18. The absorbing medium used in the collection of NO, is
a. ascarite/silica gel.
b. acidified hydrogen peroxide.
c. 80% isopropanol.
d. phenoldisulfonic acid.
19. The absolute internal pressure in the sampling flask used in the Federal
Reference Method 7 procedure for oxides of nitrogen is equal to
a. atmospheric pressure (mm Hg) plus the positive pressure (mm Hg) exerted
by the sampling flask.
b. atmospheric pressure (mm Hg) at the moment sampling begins.
c. stack pressure (mm Hg) plus barometric pressure.
d. barometric pressure minus internal pressure of the flask.
20. If a pitot tube is damaged during a run. what should be done?
a. The tube should be repaired immediately.
b. The tube should be recalibrated and all of the data taken during the day
should be corrected.
c. The tube should be recalibrated and the new coefficient used to correct all
data collected while the tube was damaged.
d. A value of 0.79 instead of 0.84 should be used to correct all data collected
while the tube was damaged.
21. According to Reference Method 4, the posttest leak check is acceptable if
a. the leak rate does not exceed the maximum vacuum rate occurring during
sampling.
b. the percent volume of gas leaked during sampling does not exceed 5%.
c. the pretest leak check demonstrated a leak rate of 2% of the maximum
sampling rate obtained during testing.
d. a leak rate of 4% or less of the average sampling rate is obtained.
22. In the clean-up procedures of an EPA paniculate train, acetone is used to
wash all internal surfaces of the
a. nozzle, probe, and front half of filter holder.
b. Answer "a," except the probe is rinsed only if the liner is glass.
c. probe and filter holder only.
d. Acetone is not used because it is highly volatile.
23. After a routine Reference Method 6 for SOS, a posttest leak check is man-
datory. If the posttest leak check indicates a leak rate of 0.1 liter per minute.
the test should be
a. accepted, since it is 10% of the sampling rate of 1 liter per minute.
b. accepted, with the final volume adjusted to account for the leak rate.
c. accepted, since the leak rate is less than the ± 0.5% of sampling rate
allowed under the procedure.
d. rejected.
-------
24. A yellow complex is formed when
a. ammonium hydroxide is added to an NO, sample.
b. thorin is added to an S0t sample.
c. starch is added to an iodine solution.
d. phenolphthalein is added to a KHP solution.
25. As pan of the postsampling quality assurance activities, the stack temperature
sensor
a. should agree within 1.0 °C with each of three temperature readings obtained
with a reference thermometer.
b. should agree within 1.5% with the ambient temperature reading of a
reference thermometer. »•'
c. should automatically be recalibrated. i
d. should agree within 1.5% with the stack temperature reading determined
by an ASTM reference thermometer.
26. Where in Volume III can you find a description of posttest calibration checks
for a dry gas meter?
a. Section 5.5.2 only
b. Section 5.4.S only
c. Sections 5.5.2. 5.4.2, and 5.5.2
d. Sections 5.5.5, 5.4.5, and 5.5.5
27. A calculated Method 5 acetone blank residue
a. cannot be subtracted from the sample weight if it is <0.01 mg/g of the
total acetone rinse weight.
b. should never be subtracted from the sample weight.
c. should be multiplied by the number of rinse bottles and added to the sam-
ple weight.
d. cannot be subtracted from the sample weight if it is > 0.001% of the
acetone rinse weight.
28. The 0.0100 N barium perchlorate solution used in Method 6
a. must be exactly 0.0100 N in order for the method to work.
b. must be standardized with sulfuric acid.
c. must be standardized with an ammonium sulfate audit sample.
d. must be standardized with three different sulfuric acid solutions to develop
a calibration curve.
29. Method 6 control samples should be
a. analyzed only by the most proficient chemist in the laboratory.
b. analyzed before and after the actual collected source samples are analyzed.
c. analyzed immediately after the Ba(ClO4)t solution is made up.
d. analyzed to an accuracy of within 0.5% of the true value.
50. Method 7 control samples are solutions of
a. nitric acid and water.
b. PDS and water.
c. pure NOS bubbled through distilled water.
d. potassium nitrate and water.
Q2-5
-------
CC:414
Examination 1
• This test is designed to measure whether you have mastered the objectives of the
course.
• This is an open-book exam. Since you will be using Volume III as a reference
document in your work, you may also use it here, along with your guidebook.
• You will have 75 minutes to complete the test.
• On the answer sheet, circle the letter that corresponds to the best answer to each
question. There is only one "best" answer for each'question. Each correct answer
is worth two points.
1. Quality assurance techniques and practices provide the best means of
a. increasing the costs of extramural projects.
b. developing source sampling reference methods.
c. ensuring reliable source test data.
d. bringing stationary sources in compliance with EPA standards.
2. Which of the. following is an(are) objectives) of quality assurance activities to
produce data that meet user requirements?
a. completeness
b. accuracy
c. precision
d. all of the above
S. Performance audits are a appraisal of quality.
a. qualitative
b. quantitative
c. system audit-type
d. all of the above
4. In which section of Volume III can a discussion of EPA Reference Method 1
be found?
a. Section S.I
b. Section 3.0.2
c. Section S.O.I
d. Section S.2.1
•
5. What is the heading for paragraph S.1.2 of Section S.2.S of Volume III?
a. Integrated Sampling Train
b. Presampling Operations
c. Calibration of Apparatus
d. Apparatus and Calibration Checks
6/82
EM
-------
6. Where in Volume III can a listing of source sampling tools and equipment be
found?
a. Section S.O.I, pages S of 19 through 5 of 19
b. Section S.5.1, pages IS of 15 through 15 of 15
c. Section S.4.S, pages S of 20 .through 15 of 20
d. Section 3.0.12, pages S of 15 through 5 of 15
7. Section 5.4.5 of Volume III would address postsaxnpling operations for Method
a. 2.
b. S.
c. 4.
d. 5. '"•;
"t
8. A sampling location 8 duct diameters downstream from a flow disturbance and
2 duct diameters upstream from a flow disturbance
a. guarantees that the sample will be representative of emissions.
b. has been scientifically determined as the location where gas stratification
will not occur.
c. allows the source tester to sample at the minimum number of EPA-specified
points.
d. will not exhibit cyclonic flow.
9. Chain-of-custody procedures
a. are a critical pan of source test procedures.
b. are not necessary in most sampling programs.
c. are important only in Method 5 compliance tests.
d. are used in CEM performance specification tests.
10. A Type S pitot tube calibration coefficient, C,,
a. can always be assumed to have a value of 0.84.
b. can be assumed to have a value of 0.84 if it meets EPA design criteria.
c. can be assigned a value of 0.84 only if it is calibrated in a wind tunnel.
d. will not change if the pitot tube is attached to a Method 5 sampling probe.
11. Type S pitot tube misalignment angles can be determined by using a(n)
a. degree-indicating level.
b. standard pitot tube.
c. oil-filled manometer.
d. Fecheimer probe.
12. The gas-confining solution in the leveling bottle of an Orsat analyzer should
contain
a. anti-freeze and alkaline pyrogallic acid.
b. distilled water only.
c. potassium or sodium hydroxide, and water.
d. sodium sulfate, sulfuric acid, methyl orange, and water.
£1-2
-------
IS. The use of continuous Ot and CO, analyzers
a. is an approved alternative to Reference Method S.
b. is valid only for the determination of molecular weight.
c. is acceptable as an alternative procedure upon approval by the EPA
Administrator.
d. is acceptable only for F factor calculations.
14. A heated sampling probe
a. should be leak free.
b. should be capable of attaining a maximum temperature of 100°C (212 °F).
c. should have an outside diameter of 5 mm (0.196 in.).
d. should always be constructed out of quartz glass.
15. A glass impinger should be checked for «
a. breaks.
b. cracks.
c. poorly shaped connections.
d. all of the above
16. A newly purchased aneroid barometer should be checked against a(n)
a. U-tube oil manometer.
b. mercury-in-glass barometer.
c. orifice meter.
d. Magnehelic* gauge.
17. When should an Orsat analyzer be calibrated?
a. upon receipt
b. before every third field test
c. if it has not been checked during the previous three months
d. all of the above
18. What grade of acetone should be used in the Method 5 probe washing
procedure?
a. commercial grade
b. ACS reagent grade
c. primary standard grade
d. chemically pure grade
19. Hydrogen peroxide used in source testing
a. should be* prepared fresh daily.
b. can be stored for up to two months in a brown bottle.
c. should be boiled to remove ozone before testing.
d. should be checked for sulfate before testing.
20. A Magnehelic9 gauge can be calibrated using
a. an orifice meter.
b. a mercury barometer.
c. an inclined manometer.
d. a phot tube.
El-S
-------
21 . When calibrating a Type S phot tube in a wind tunnel,
a. only one reading is obtained using the itandard pitot tube..
b. a standard pitot tube reading is obtained only before each Type S "A" side
measurement.
c. a standard pitot tube reading is obtained only before each Type S "B" side
measurement.
d. a standard pitot tube reading is obtained before each Type S "A" and "B"
tide measurement.
22. A metering system is unacceptable if an individual Y value deviates from (Y).,,
by more than
a. 1%.
b. 5%. .
c. 4%.
d.
25. A rate meter, such as a rotameter, can be calibrated against
a. an oil manometer.
b. a pitot tube.
c. a wet test meter.
d. a mercury barometer.
24. An Orsat analyzer should be leak checked
a. before packing and shipping.
b. before each sample measurement.
c. on-site, before the test.
d. on-site, after the test.
25. The glass fiber filter used in Method 5 paniculate sampling must
a. exhibit at least 96.5% collection efficiency.
b. be desiccated 24 hours and weighed to a constant weight.
c. be desiccated 24 hours and weighed to the nearest 1.0 mg.
d. be desiccated 6 hours and weighed.
26. If a %Of value determined from an Orsat measurement is to be used in an
F factor calculation, how close should three analyses be before the average is
judged acceptable?
a. They should differ by £0.5% if the O, is S15.0%.
b. They should differ by fc 0.3% if the O, is >15.0%.
c. They should differ by SO. 2% if the O, is Sl5.0%.
d. They should differ by 2:0.5% if the O, is <15.0%.
27. An estimate of the stack gas moisture content is necessary for a Method 5 test
because
a. the moisture content is used in the F factor calculations.
b. the moisture content is a parameter needed to calculate the isokinetic sam-
pling rate.
c. the probe temperature needs to be adjusted accordingly.
d. the number of sampling points are determined using this parameter.
£1-4
-------
28. In Method 5, a pretest leak check is
a. recommended.
b. required.
c. ill-advised.
d. a waste of time.
29. What assumptions does the nomograph make about the stack gas molecular
weight?
a. The molecular weight can be corrected for %COS and %OS.
b. The dry stack gas molecular weight is measured to be 29.
c. The molecular weight (wet) is assumed to be 29.
d. The stack gas molecular weight is directly related to v., the stack gas velocity.
SO. The leak-check procedure for Reference Methott 7 requires evacuation of the
collection flask to S in. Hg, then observation of the leak rate over a period of
time. If there is a leak,
a. the rate cannot be any greater than 0.4 in. Hg per minute.
b. the flask cannot be used and must be replaced.
c. the rate will be accounted for in the final volume of sample acquired.
d. the leak rate will be adjusted to ±10% of sampling rate.
SI. Federal Reference Method 7 requires a pretest leak check. During a recent
test, a pretest leak rate of I in. Hg per minute was observed. The sampling
team should
a. continue sampling by adjusting the sampling rate so that it is proportional
to the leak rate.
b. continue with the test.
c. accept the leak rate by adjusting final volume sampled.
d. stop the test and correct the leak problem.
32. The actual color-forming step in Reference Method 7 occurs
a. during addition of PDS with the sodium salt.
b. after the 16-hour waiting period and upon addition of 50 ml of TCM.
c. during the addition of concentrated NH«OH which attacks the PDS ring.
d. Color development does not take place because the procedure is an
iodometric thration.
SS. As pan of the postsampling quality assurance operations, what should be
checked on the dry gas meter after the test?
a. the calibration factor (Y) at three orifice meter settings
b. the calibration factor (Y) at one intermediate orifice meter setting and the
thermometer at room temperature
c. the thermometer at ice point, boiling point, and room temperature
d. leak tightness, only
£1-5
-------
54. A balance used to weigh Method 5 filters
a. should be zeroed and checked with at least five Class-S weights in order to
develop a calibration curve.
b. should be zeroed and checked with a 0.500-g Class-S weight.
c. does not need to be checked with standard weights.
d. must be an electronic balance.
55. Control samples for Method 6 are solutions of
a. NaOH and water.
b. Ba(ClO4) and Isopropanol.
c. HSSO« and water.
d. (NH*), SO4 and water. v
56. If fibers from a filter adhere to the gasket pan of the filter assembly, a proper
procedure to follow would be to
a. wash the gasket in an acetone/water rinse.
b. retain the fibers on the gasket for the next run.
c. scrape off the fibers into the filter recovery dish.
d. wipe the fibers off with a Kimwipe®.
57. In Method 5, the percent isokinetic should be 100%. and if it is
a. it ensures sampling accuracy.
b. it means only that, based on the volumetric and velocity data, the proper
sampling rates were used.
c. it means that the source is in compliance with regulations.
d. it means that only the pollutant mass rate will be accurate.
58. The addition of NaOH to the recovered sample in the reference method for
oxides cf nitrogen serves to
a. fix the nitrate to a sodium salt for additional analysis.
b. evaporate excess NHt, which interferes with the analysis, from the solution.
c. adjust the pH of the solution so interfering metals precipitate out as sodium
salts.
d. NaOH does not have to be added.
59. The molecular weight of a stack gas can be determined, on a wet basis, by
which of the following equations?
a. M,«0.44(%CO,)+O.S2(%0,)+O.S8(%N,+ %CO)
b. M,«=18(%H,O)+M<
c.
d M -F %0«-0-5%CO 1
' * |p.264%N,(%0,-0.5%CO)J
40. Calculations performed for a reference method should be checked by
a. the plant engineer who contracted the test team.
. b. the agency test observer.
c. a test team member other than the one who performed them originally.
d. all of the above
£1-6
-------
41. Intermediate calculations should be carried out to at least
a. one significant figure.
b. one decimal figure beyond that of the acquired data.
c. seven significant figures.
d. three decimal figures beyond that of the acquired data.
42. The units associated with each symbol in reference method calculations are
important because
a. many of the equations have constants expressed in specific units.
b. all work performed for governmental agencies must follow the metric
system.
c. the equations can be solved using only English units.
d. source sampling data is generated by computer systems.
45. A Method 7 calculation for ftg NOS per sample, m«2 K.AF, was performed
using the following data:
K.-519
A = 0.612
F=1.00.
The value of m should be reported as
a. 635.256 jig of NO,.
b. 655.26 Mg of NO,.
c. 655.3 Mg of NO,.
d. 635 Mg of NO,.
44. In the reference methods, pressure drop across a phot tube or orifice meter is
expressed in units of
a. in. H,O or mm H,O.
b. ft H,O or m H,O. ,.
c. mm Hg or in. Hg.
d. ft/s or m/s.
45. A programmed calculator or computer
a. cannot be used for compliance testing calculations.
b. can be advantageous in reducing calculation errors.
c. can be used for compliance tests only when calculations are checked with a
slide rule.
: d. cannot be used for reference method calculations because programs are not
available.
46. In the normal use of sampling equipment,
a. maintenance should never be required.
b. routine maintenance should be conducted at most once a year.
c. routine maintenance should be conducted at least quarterly.
d. maintenance should be conducted only when something breaks.
£1-7
-------
47. A fiber vane pump requires a periodic check of
a. the oil and oiler jar.
b. the rotor speed.
c. the magnetic flux.
d. the power factor.
48. Quick disconnects should be replaced
a. after every 1000 ft1 of operation.
b. once every quarter.
c. when the sampling system doesn't leak check.
d. when damaged.
% i
49. Volume III recommends that a source test observer perform a system audit
a. once for every four tests conducted by an experienced test team.
b. once for every four tests conducted by an inexperienced test team.
c. for every test conducted by an experienced test team.
d. only when the source is out of compliance.
50. An audit is
a. used to correct source test data.
b. a pan of the Federal reference method procedures.
c. conducted by the test team leader.
d. an independent assessment of data quality.
£1-8
-------
Name.
Date_
CC:414
Examination 1
Answer Sheet
1.
2.
S.
4.
5.
6.
7.
8.
9.
10.
11.
12.
IS.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
(Please Print)
Address:
£1-9
-------
26.
27.
28.
29.
SO.
SI.
S2.
S3.
54.
85.
36.
57.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
a
.a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b,
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
I certify that this
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
test
d
d
d
d
d*
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
was administered in accordance with the specified test
instructions.
Test Supervisor
6/82
El-10
-------
REGISTRAR-
AIR POLLUTION TRAINING INSTITUTE
US EPA MD-17
RESEARCH TRIANGLE PARK NC 27711
-------
* ERRATA *
Requirements for Successful Completion of Course CC:414
In order to receive 3.5 Continuing Education Units (CEU's) and a certificate of
course completion you must:
* complete and submit a final exam to the APTI
* achieve a final course grade of at least 70%.
The quizzes associated with the course are for review purposes. The answers
are enclosed for the student to use to correct his or her own quizzes. This way
the student can assses his/her understanding of the material before taking the
final exam. Do not send in the answer sheets to the quizzes to be graded. The
final exam counts for 100% of the course grade.
If you have questions please contact:
Registrar
Air Pollution Training Institute
US EPA MD-17
Research Triangle Park, NC 27711
-------
CC:414
Quiz 1
Grading Key
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
IS.
14.
15.
a
a
a
C*)
a
a
a
©
• ^^
a
a
©
a
a
a
a
(b)
0
b
b
©
b
b
b
b
b
b
©
b
b
(!)
c
c
c
c
c
©
©
c
©
c
c
c
c
c
c
d
d
<
*^s
a
a
r-*
a
<^
a
a
a
a
a
a
a
b
b
£
b
b
b
b
b
b
b
b
b
T
I
b
c
c
c
^^
c
^_^
c
f->
S_>
c
r^
£
c
r^«
^
c
c
c
c
c
d
d
d
d
d
d
S
d
d
d
d
S
d
d
r?
6/82
-------
10/3/90
ERRATA SHEET
CC 414: Quality Assurance for Source Emission Measurement Methods
There are now Reference Methods 3A and 3B. For Methods 3A and 3B respectively,
see 40 CFR 60, App. A, Meth. 3A and Federal Register notice Vol 55, p. 05211,
Feb. 14, 1990.
For cyclonic flow, an average 20 degree angle of rotation 1s now acceptable.
See 40 CFR 60, App. A, Meth. 1.
p. 196, Reading Assignment 16 Review Exercises. The correct answer to question
l.a., "How many significant figures are 1n 0.007?" Is 1 not 3. This 1s supported
by Rule 1 on p. 194 which states, "Disregard all Initial zeros."
-------
CC:414
Quiz 2
Grading Key
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
IS.
14.
15.
a
a
a
©
a
C*)
a
a
0^
a
a
a
(a)
a
(7T)
b
©
b
b
©
b
b
©
b
b
©
b
b
b
b
©
c
©
c
c
c
c
c
c
©
c
c
c
©
c
d
d
d
d
d
d
©
d
d
d
d
©
d
d
d
16.
17.
18.
19.
20.
21.
22.
2S.
24.
25.
26.
27.
28.
29.
SO.
a
a
a
a
a
a
C?)
a
C?)
a
a
a
a
a
a
b
b
b
b
b
b
b
b
©
b
b
©
b
©
c
c
c
c
c
c
c
©
c
c
c
c
d
d
d
d
©
d
V d
d
d
d
(T
d
d
(d
6/82
-------
AIR POLLUTION TRAINING INSTITUTE
EN VJWNMENTAL RISUflCH CEKTW • ID VI • UttUJOH TRl»>^K*.»C 77711. (Ill) M14C7.
ATTENTION
: You ire now enrolled wHh the A3r Pollution Training Institute. The enclosed materials
ere for your usage. You may keep the printed materials, however, any slides, vldto
tapes and/or audio tapes are LOANED materials and MUST be returned to the
Institute.
To restive Continuing Education Units (CEU's), and a certificate of completion:
1. Fill out this form and mail h in along with any slides or tapes that are
being returned and the f nal exam to be graded.
2. No exam will be graded unless accompanied by the loaned materials.
We encourage you to complete the course within 30 days.
3. r! you choose not to take the fina! exam, please Indicate that on this
form, and return ft along with the loaned materials. This will ensure your
cancellation from the course.
Direct all correspondence to: REGISTRAR
APTI MD-17
US EPA
RTF, NC 27711
We are happy to provide you wfth this training.
Name: _______-_____«_^^ _
Address: _ _ __
Date:
Enclosed please find the loaned materials for Course
I choose not to take the exam for Course #
• UNTTSfi SI ATtt INVRDNUEKUL WOTSCTON AOEMCY*
-------
* ERRATA *
Requirements for Successful Completion of Course SI:476B
In order to receive 3.0 Continuing Education Units (CEU's) and a certificate of
course completion you must:
* complete and submit a final exam to the APTI
* achieve a final course grade of at least 70%.
The quizzes associated with the course are for review purposes. The answers
are enclosed for the student to use to correct his or her own quizzes. This way
the student can assses his/her understanding of the material before taking the
final exam. Do not send in the answer sheets to the quizzes to be graded. The
final exam counts for 100% of the course grade.
If you have questions please contact:
Registrar
Air Pollution Training Institute
US EPA MD-17
Research Triangle Park, NC 27711
-------
Name:
Date:
Course SI: 476B
Quiz Number 1
Units 1-3
Answer Sheet
1.
2:
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
a
a
©
a
a
a
a
a
a
a
©
©
a
a
Q
a
a
a
a
a
a
a
a
@)
a.
Qy
b
b
©
b
b
b
b
©
b
b
b
b
b
b
b
b
©
©
b
©
b
b
b
fa
c
©
c
c
©
©
c
©
c
c
c
c
c
0
c
©
0
c
c
c
c
©
c
c
c
•d
d
d
d
d
d
@
d
d
a
d
d
(2
d
d
d
d
d
d
(d]
d
d
(2
d
d
I certify that this quiz was administered in accordance with the specified test
instructions.
Quiz Supervisor
-------
Name:
Date:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
a
a
[a]
fa]
sS
a
a
a
a
a
(a
a
a
a
a
L^
a
a
a
a
b
b
b
b
b
b
b
b
©
b
©
©
©
b
b
b
b
b
b
£
b
S
©
££
b
c
c
c
fir
c
c
c
Course SI: 476B
Quiz Number 2
Units 4-5
Answer Sheet
d
d
d
d
d
d
E
©
d
I certify that this quiz was administered in accordance with the specified test
instruction • '.'
Quiz Supervisor
-------
Continuous Emission Monitoring Systems
Operation and Maintenance of Gas Monitors
Quiz Number 1
Units 1-3
Directions: This is a closed book test, so you may not use your manual and notes. On your
answer sheet, circle the letter that corresponds to the best answer or write the answer in
the space provided. You will have minutes to complete the test. When you have
completed it, turn in only the Answer Sheet.
1. CEMSs are used on a basis.
a. semi-annual
b. continuous
c. nominal
d. yearly
2. A typical monitoring system contains all of the following except
a. a readout instrument.
b. an analyzer.
c. a concentration enhancer.
d. a data recorder.
3. and are the two monitoring methods that are
primarily dependent of the characteristics of light.
a. Spectroscopic absorption ... luminescence
b. Electroanalysis ... chemical reaction
c. Paramagnetic ... diatomic
d. Absorption ... dilution
4. The sample gas must be conditioned in systems.
a. in-the-stack
b. extractive
c. all
d. in-situ
-------
5. The spectropic absorption techniques used in many systems are based on
a. a gas' ability to absorb water.
b. paramagnetic phenomenon.
c. the absorption of a unique wavelength of light by the target gas.
d. the attraction between gases.
6. Light energy released from excited atoms is the principle behind
a. magnetopneumatic techniques.
b. electroanalytical techniques.
c. luminescence techniques.
d. paramagnetic techniques.
7. In polarographic analyzers, the gas diffuses through a in order to
generate a .
a. beam splitter ... high bias reading
b. reference cell ...light wave
c. large valve ... low bias reading
d. thin-film membrane ... reaction and current flow
8. As a gas absorbs some light energy, the intensity of the light beam will
from its original value.
9.
a. increase
b. not change
c. decrease
d. fluctuate
Nearly all spectrophotometers found in CEMSs use a broad band of wavelengths that
are centered on the of the target gas molecules.
a. luminescence
b. absorption peak
c. molecular weight
d. zero point
-------
10. A flame is used in flame photometric analyzers in order to excite the
gas molecules.
a. luminescent
b. coal-fired
c. natural gas-fired
d. hydrogen
11. The Bodenseewerk Model 677 forces high pressure air through the secondary filter
assembly in order to
a. alleviate blockage.
b. keep the system from decompressing.
c. compress the sample into a smaller volume.
d. force more HC1 sample through the analyzer.
12. The Teledyne Analytical Instruments Model 691 is an example of a
a. differential absorption analyzer.
b. gas filter correlation analyzer.
c. paniculate analyzer.
d. density analyzer.
13. A heat traced sampling system, as in the Teledyne Analytical Instruments Model 691,
prevents during analysis.
a. corrosion
b. leaks
c. cracking
d. water condensation
14. The electrons of excited molecules
a. stop moving
b. interfere with the monitoring equipment
c. are temporarily moved to higher energy levels
d. are temporarily moved to lower energy levels
-------
15. The LSI Dynatron 401 uses the principle of as the basis for the
measurement of Q concentration.
a. electrocatalysis
b. electrolysis
c. absorbance
d. opacity
16. The equation relates voltage to oxygen concentration.
a. Newton
b. Reynolds
c. Nernst
d. Faraday
17. Magnetopneumatic analyzers are generally by temperature fluctuations.
a. damaged
b. high biased
c. unaffected
d. low biased
18. NDIR cross-flow modulation analysis requires the sample gas and the reference gas
to be introduced into the measurement cell.
a. simultaneously
b. alternately
c. randomly
d. pseudo-randomly
19. The moveable membrane in the NDIR cross-flow modulation's detector senses
a. ion concentration
b. pressure differential
c. temperature changes
d. porosity
-------
20. Sample gas from a potentially challenging stack environment is conditioned by the
application of a (n) design.
a. in-the-stack
b. streamline
c. in-situ
d. extractive
21. In-situ systems do all of the following tasks except
a. minimize interference from paniculate matter.
b. remove moisture from the stack.
c. analyze a representative portion of stack gas.
d. operate in high temperature combustion sources.
22. A constant flow of purge air prevents and.
a. sample loss ... corrosion
b. vibration ... water condensation
c. water condensation ... paniculate buildup
d. pressure loss ... cracking
23. Anti-vibration equipment is used to help maintain
a. sample purity.
b. sample integrity.
c. constant pressure.
d. optical alignment of the transmitter and receiver.
24. The purpose of surface and depth filters is to
a. collect paniculate matter greater than 1 micron.
b. protect probe from paniculate buildup.
c. clean the purge air.
d. remove water.
25. Heat tolerance, chemical stability, and corrosion resistance should be considered
when selecting
a. which gas to monitor.
b. sample transport tubing.
c. a signal processor.
d. the analyzer to be used.
-------
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
Name:
Date:
Course SI: 476B
Quiz Number 1
Units 1-3
Answer Sheet
I certify that this quiz was administered in accordance with the specified test instructions.
Quiz Supervisor
-------
Continuous Emission Monitoring Systems
Operation and Maintenance of Gas Monitors
Quiz Number 2
Units 4-5
Directions: This is a closed book test, so you may not use your manual and notes. On your
answer sheet, circle the letter that corresponds to the best answer or write the answer in
the space provided. You will have minutes to complete the test. When you have
completed it, turn in only the Answer Sheet.
1. The Federal Register Index and the LSA Index are the two federally published
indices which can be revised to keep track of changes in the code of federal
regulations.
a. True
b. False
2. Types of regulatory procedures which are used by state and local agencies for
assuring compliance with air quality standards include:
a. operating permits.
b. the promulgation of regulations.
c. the issuing of variances.
d. all of the above.
3. Performance specification tests are used to evaluate a CEM's performance:
a. over an extended period of time.
b. every thirty days.
c. every sixty days.
d. upon, or soon after the initial installation of the system.
4. "Span value" refers to the upper limit of a gas concentration measurement range
specified for affected source categories in the applicable subpart of the regulations.
a. True
b. False
-------
5. Stratification is a term which refers to the amount of variation in pollutant emission
concentrations at the Reference Method measurement locations.
a. True
b. False
6. The Relative Accuracy Audit (RAA) test requires that only sets of
measurements be taken.
a. 9
b. 12
c. 3
d. 6
7. Appendix F states that if excessive inaccuracies in data occur for consecutive
quarters, then the source owner shall modify or replace the CEMs.
a. two
b. four
c. three
d. none of the above
8. Quality control refers to activities which will provide a quality product and quality
assurance refers to activities which provide assurance that the quality control
program is satisfactory.
a. True
b. False
9. The temperature compensation circuit on an in-situ monitoring system can be
bypassed whenever:
a. the instrument is being calibrated.
b. the instrument is calibrated manually using calibration gases.
c. whenever the temperature of the CEMs falls below 2(f C.
d. this circuit should never be by passed.
10. Which of the following has not been shown by recent surveys as a problem area
encountered in CEM maintenance programs?
a. data handling systems.
b. emergency service.
c. locating the CEMs.
d. spare parts.
-------
11. The recommended period for performing routine maintenance checks on both
extractive and in-situ monitoring systems is every 45 days.
a. True
b. False
12. The submittal of a Data Accuracy Report (DAR) is an option which a source
subject to TNG requirements of Appendix F may or may not choose to abide by.
a. True
b. False
13. Regulations for New Source Performance Standards (NSPS) are contained in 40
CFRPart .
a. 51
b. 60
c. 61
d. 65
14. The calibration drift (CD) test involves the comparison of the monitor's response
to a known concentration of gas.
a. True
b. False
15. The issuance of operating permits are a type of regulatory procedure which state
and local air pollution control agencies use for assuring compliance with air quality
standards.
a. True
b. False
16. In performing a daily calibration drift (CD) test of the CEM, the instrument is
considered to be out-of-control if the CD value is times the applicable
performance specification contained in Appendix B of 40 CFR Part 60.
a. 6
b. 2
c. 4
d. 10
-------
17. A Cylinder Gas Audit (CGA) tests both pollutant and diluent (if applicable)
monitors for accuracy three times at points specified in Appendix F, Procedure
1.
a, 5
b. 4
c. 2
d. 3
18. Which of the following criteria should be considered in the development of a
quality assurance program?
a. How the CEMs works.
b. The criteria for determining CEMs deficiencies.
c. The possible things which can go wrong with the CEM.
d. All of the above.
19. Two stages of quality control involved in a CEMs program are operation checks
and routine and corrective maintenance. The third stages of a CEMs quality
control program is
a. schedule system regulatory inspections.
b. management system audits.
c. level system regulatory inspections.
d. performance audits.
20. A performance audit:
a. provides an evaluation of the role of management in a QA program.
b. provides a quantitative assessment of the monitoring system in order to verify
the adequacy of the quality assurance procedures used to determine adequate
CEM performance.
c. evaluates the documentation which is used to show that the data is of a known
quality.
d. all of the above.
21. Three types of maintenance programs are preventive maintenance, routine
maintenance and corrective maintenance.
a. True
b. False
-------
22. Gas cells used to calibrate in-situ monitoring systems are always certified by
independent laboratories.
a. True
b. False
23. The relative accuracy of a gas CEM is defined as:
a. A comparison of the monitor's response to a known concentration of gas.
b. The percent difference in the continuous monitor's measurement of the
pollutant versus the value of the measurement as a result of using an EPA
Reference Method.
c. The accuracy of the instrument's strip recorder.
d. None of the above.
24. Both the Relative Accuracy Audit Test (RATA) and the Relative Accuracy Audit
(RAA) need not be calculated in terms of the applicable standard (e.g/ 1 b./million
BTU).
a. True
b. False
25. Continuous emission monitoring requirements for fossil fuel fired steam generators
and electric utility generating units are contained in 40 CFR Part 60 subparts
and , respectively.
a. G, H
b. BB, J
c. D, Da
d. Db, G
-------
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
c
c
c
c
c
d
d
d
d
d
d
d
d
d
d
d
d
d
d
Name:
Date:
Course SI: 476B
Quiz Number 2
Units 4-5
Answer Sheet
I certify that this quiz was administered in accordance with the specified test instructions.
Quiz Supervisor
-------
Continuous Emission Monitoring Systems
Operation and Maintenance of Gas Monitors
Final Exam 1
Directions: This is a closed book test, so you may not use your manual and notes. On your
answer sheet, circle the letter that corresponds to the best answer or write the answer in
the space provided. You will have minutes to complete the test. When you have
completed it, turn in only the Answer Sheet.
1-1. The two monitoring techniques that are based on the characteristics of light
waves are and .
a. luminescence ... spectroscopic absorption
b. absorption ... adsorption
c. paramagnetism ... polarography
d. electrolysis ... luminescence
1-2. systems condition the sample gas prior to its analysis.
a. Extractive
b. In-situ
c. Opacity
d. Ambient air
1-3. Sample vessels used for spectroscopic analysis should be
a. metal
b. ceramic
c. transparent
d. magnetic
1-4. Differential absorption spectroscopy analyzers may use a diffraction grating to
distinguish between and .
a. measuring gas ... reference gas
b. measuring wavelength ... reference wavelength
c. SO, ... NOx
d. moisture ... paniculate matter
-------
1-5. Which of the following gases cannot be measured by a flame photometric
analyzer?
a. SQj
b. HC1
c.
d.
1-6. Electroanalyzers can relate subject gas concentration to and.
a. voltage ... capacitance
b. current... capacitance
c. Planck's constant... Faraday's constant
d. current... voltage
1-7. Paramagnetic molecules, like oxygen, are by magnetic fields.
a. repelled
b. attracted
c. condensed
d. diluted
1-8. High pressure purge air is forced through the secondary filter assembly of the
Bodenseewerk Model 677 in order to
a. compress the sample volume.
b. purge the filter, alleviating blockage.
c. decompress the system.
d. remove HC1 from the gas stream.
1-9. A protects the ceramic diffuser from erosion caused by paniculate
matter.
a. dispersive spectrophotometer
b. compact design
c. platinum electrode
d. paniculate deflector
-------
1-10. The Horiba MPA-311 analyzer uses the magneto-pneumatic method to measure
b. diamagnetism
c. velocity profile
d. temperature fluctuations
1-11. Analyzers that use a_
between maintenance.
sampling volume may provide longer intervals
a. spectroscopic
b. small
c. proportional
d. porous
1-12. The NDIR cross-flow modulation process requires the sample gas and the
reference gas to be introduced alternately into the measurement cell.
a. True
b. False
1-13. Extractive CEMSs require _
compared to in-situ CEMSs.
a. more
b. less
c. the same amount of
equipment for a single sample point
1-14.
allows gases to permeate and prevents paniculate matter from
entering the in-situ analyzer.
a. diaphragm pump
b. high pressure blowback system
c. splitter
d. ceramic thimble
1-15. In-situ monitors which measure across a distance greater than 10% of the
equivalent stack diameter are classified as point monitors.
a. True
b. False
-------
1-16. In-situ analyzers use anti-vibration equipment to help maintain
a. sample integrity.
b. constant temperature.
c. optical alignment of the transmitter and receiver.
d. a clean sample that is free of all paniculate matter greater than one micron.
1-17. CEMSs data has generally been collected on.
a. a yearly basis
b. a weekly basis
c. computers and semi-log paper
d. computers and operator log sheets
1-18. The initial cost of an in-situ CEMS is the initial cost of an extractive
CEMS when monitoring at a single location.
a. less than
b. about the same as
c. greater than
1-19. A sample transport system contains.
a. a retroreflector
b. depth filters
c. a torsion pendulum
d. tubing
1-20. System calibration should be performed under the same conditions encountered
during normal operation.
a. True
b. False
1-21. In extractive designs, the separation of the probe and analyzer
a. increases analytical error.
b. decreases paniculate buildup.
c. causes a time lag between extraction and analysis.
d. decreases sampling volume.
-------
1-22. Moisture content of stack gas should be determined in order to
a. assess the effect of stratification.
b. estimate the amount of water recycled per hour.
c. purge the filters.
d. calculate the water removal needs of the system.
1-23. Moisture reduction is not accomplished by.
a. heating
b. permeation
c. dilution
d. refrigeration
1-24. In-situ CEMSs can be designed in a manner to accommodate time-sharing of a
single analyzer.
a. True
b. False
1-25. Reliable CEMSs data can be used to enhance process control systems.
a. True
b. False
1-26. The use of heat traced sample lines prevents and during
transport.
a. water condensation ... sample loss
b. corrosion ... cracking
c. dilution ... corrosion
d. quenching ...luminescence
1-27. In the Horiba ENHA-C900 Series, the reduction catalyst assists in the reaction
of ammonia with sulfur dioxide.
a. True
b. False
-------
1-28. A constant sample flow rate can be more easily maintained with the use of a
a. primary filter
b. secondary filter
c. pressure trap
d. diffraction grating
1-29. The pressure differential in the fuel type cell sensor in the Dynatron analyzer
results in a .
a. high amplitude
b. voltage
c. low temperature
d. zero setting
1-30. The release of excitation energy in the form of light is the basis for the
method of analysis.
a. catalytic
b. inductive
c. reduction
d. luminescence
1-31. The process of relaxing an excited gas molecule without the release of light
energy is called and is caused by fluctuations in diluent gas
concentration.
a. quenching
b. cleansing
c. modulation
d. filtering
1-32. The intensity of a light beam will
that light energy.
a. remain the same
b. become ultraviolet
c. increase
d. decrease
as a subject gas absorbs some of
-------
1-33. The purpose of a gas filter cell is to
a. filter gas before going to the analysis unit.
b. remove all the sample gas from the stack.
c. absorb all light energy from the absorption peak
associated with a subject gas.
d. purify air.
1-34. An order is usually issued as a result of public hearings and is an authorized
method for assuring compliance with applicable air quality standards.
a. True
b. False
1-35. Continuous emissions monitoring requirements for nitric acid plants are
contained in which subpart to 40 CFR Part 60?
a. R
b. G
c. Da
d. BB
1-36. The National Emissions Standards for Hazardous Air Pollutants (NESHAPS) are
contained in 40 CFR Part 50.
a. True
b. False
1-37. Excess Emission Monitoring Data are used whenever the use of continuous
emission monitoring data:
a. is the state's approved compliance method.
b. is not the state's approved compliance method.
c. must be reported to the EPA headquarters in Washington, D.C.
d. none of the above.
-------
1-38. In situations where a source's emissions are less than 50% of the applicable
standard, the source may use an alternative mathematical expression in
determining the relative accuracy of the source monitor. This alternative method
involves dividing the absolute difference between the monitor value and
reference method value plus the 2.5 percent confidence internal by the
applicable standard.
a. True
b. False
1-39. Performance Specification Tests 3-5 (Appendix B) use none of the same
procedures outlined in Performance Specification Test 2.
a. True
b. False
1-40. The EPA Reference Methods are contained in Appendix to 40 CFR Part
60.
a. B
b. A
c. F
d. P
1-41. A Cylinder Gas Audit (CGA) test may be performed in any of calendar
quarters.
a. Four
b. Three
c. Two
d. May not be performed at any time during the fiscal year.
1-42. The audit gas used in performing a CGA test is used to check the CEM at four
points within the specified ranges.
a. True
b. False
1-43. Procedures for maintaining a spare parts inventory is a QC requirement as stated
in Appendix F to 40 CFR Part 60.
a. True
b. False
-------
1-44. The level I function in a quality assurance program involves:
a. duties of the CEM operator.
b. the implementation and support of the QA program.
c. writing the monthly and quarterly reports.
d. none of the above.
1-45. Routine and corrective maintenance is one of the three stages of a CEMS
quality control program.
a. True
b. False
1-46. The "level" approach involved in a regulatory inspection audit program involves:
a. source records review and stack testing to determine compliance.
b. issuing source permits.
c. observing the initial CEMS performance specification test.
d. all of the above.
1-47. Problems encountered with extractive monitoring systems generally occur:
a. in the system electronics.
b. in the gas transport system and gas conditioning components.
c. because the calibration gases are usually not NIST certified.
d. none of the above.
1-48. Daily operation checks for an extractive monitoring system should include:
a. system indicator lights.
b. strip charts / other data recording devices.
c. conditioning system.
d. a daily calibration check.
e. all of the above.
1-49. Recent surveys have identified emergency service as the only problem to be
encountered in CEM maintenance programs.
a. True
b. False
-------
1-50. A performance specification test will determine if a continuous emission
monitoring system is:
a. accurate.
b. durable.
c. reliable.
d. all of the above.
-------
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
Name:
Date:
Course SI: 476B
Final Exam 1
Answer Sheet
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
ardance with the specified
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
test
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
instructions.
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d e
d
Test Supervisor
-------
-------
Course Introduction
Description
This course is a self-instructional study program designed to help you develop a working knowledge of the
operation and maintenance of continuous gas emission monitoring systems. It presents the detailed operating
characteristics of commercially available continuous emission monitors and the common maintenance techniques
use to provide for continuing operation. A comprehensive discussion of regulatory specifications in terms of
instrument design, installation, and performance testing is also included. Major topics include the following:
• Analytical methods
• Operation of commercially available monitors
• System design
• Regulations
• Continuing operations (quality assurance / quality control and audits)
Course Goal and Objectives
Goal
To present basic operational principles of continuous gas emission monitors and to develop a comprehensive
understanding of maintenance requirements and procedures.
Objectives
Upon completing this course, you should be able to:
1. describe the types of emission monitoring systems and identify the scientific principles .associated with
the various analytical techniques;
2. explain the operation and describe the special features of specific analyzers in the these categories:
spectroscopic, luminescence, electroanalytical, and paramagnetic;
-------
3. identify the analytical techniques of and gases monitored by general purpose analyzers;
4. describe the primary subsystems and list the criteria to be evaluated in the selection of extractive systems
and in-situ systems;
5. identify source categories required to install continuous emission monitoring systems and explain how
to find the applicable requirements;
6. explain performance specification test procedures, quality assurance/quality control requirements, audit
programs, and maintenance procedures for CEMs.
Identification of Constraints I Prerequisite Skills
Before studying this self-instructional course, you should have some knowledge of continuous emission
monitoring systems. This knowledge should be at the level of material presented in APT! Course 474. Continuous
Emission Monitoring, or in equivalent introductory workshops presented by the EPA Stationary Source
Compliance Division.
If you are not able to attend such course offerings, but wish to study this self-instructional course on continuous
gas emission monitoring, the handbook, Continuous Air Pollution Source Monitoring Systems, EPA 625/6-79-
005 should first be reviewed. This document is available at no cost as Handbook No. 6005 from:
Center for Environmental Research Information
U.S. EPA Cincinnati, Ohio 45628
Description of Intended Trainees
This course is designed for persons in regulatory agencies and in industry, who, as pan of their job function,
will evaluate continuous gas emission monitoring systems and system data or who will be responsible for the actual
systems. The intended trainees are either engineers, physical scientists, or supervisory engineers involved in
environmental compliance programs. Trainees should have bachelor's degrees in some technical field of study.
Technicians who are responsible for emissions monitoring systems, and who have either an associate degree or
several years experience in instrumental monitoring, should experience little difficulty with the course.
Requirements for Successful Completion of this Course
In order to receive 2.0 Continuing Education Units (CEUs) and a certificate of course completion you must:
• take two supervised quizzes and a supervised final examination.
• achieve a final course grade of at least 70 (out of 100) determined as follows:
• Quiz 1 is 25% of the final grade.
VI
-------
Quiz 2 is 25% of the final grade.
The final examination is 50% of the final grade.
Materials
Reading
This text - supplementary reading materials are not required.
Using the Handbook
This handbook contains fifteen lessons separated into five units. Each lesson contains the following:
• lesson learning goal and objectives,
• text of the lesson,
• review exercises,
• answers to review exercises, and
• references
Complete the review exercises immediately after reading the assigned materials. You may find it helpful to
look over the review questions before reading. By having an idea of what to look for in the reading materials, your
attention will be better focused and your study will be more efficiently directed.
To complete an exercise, place a piece of paper across the page, covering the questions below the one you are
answering. After answering die question, slide the paper down to uncover the next question. The answer for the
first question will be given on the right side of the page, as shown here. All answers to review questions will appear
below and to the right of their respective questions. The answer will be numbered to match the question.
REVIEW EXERCISES
1. Question
2. Question
3. Question
1. Answer
2. Answer
Complete each review exercise in the lessons. If you are unsure abut a question or answer, review the material
preceding the question. Then proceed to the next section.
NOTE: If more man one peisonwfll be using these materials, we recommend mat you use a separate sheet of paper
to record your answers to the review exercises.
Vll
-------
Instructions for Completing the Quizzes and the Final Examination
• You should have received, along with this guidebook, a separate sealed envelope containing two quizzes
and a final examination.
• You must arrange to have someone serve as your test supervisor.
• You must give the sealed envelope containing the quizzes and the final examination to your test supervisor.
• Atdesigiiatedtimesduringrnecouise,uiKlerthesupem
the final exam.
• After you have completed a quiz or the exam, your test supervisor must sign a statement on the quiz/exam
answer sheet certifying that the quiz or exam was administered in accordance with the specified test
instructions.
• Aftersignmgmeqiuz/examanswersheet,yourtestsupeivisormustmaUmequizorexam
to the following address:
Air Pollution Training Institute
MD17 Research Triangle Park, NC 27711
• After completing a quiz, continue with the course. Do not wait for quiz results.
• Quiz£xam and course grade results wffl be mailed to you.
If you have questions, contact:
Air Pollution Training Institute
MD17
Research Triangle Park, NC 27711
Telephone numbers
Commercial: (919) 541-2401
FTS: 629-2401
vui
-------
UNIT1
INTRODUCTION
TO THE
ANALYTICAL METHODS
-------
LESSON 1
Types of Emissions Monitoring Systems
Lesson Goal and Objectives
Goal
To introduce some general information and terminology associated with continuous emissions monitoring
systems (CEMSs).
Objectives
At the end of this lesson, you should be able to —
1. list primary reasons for having a continuous emissions monitoring program;
2. describe, in general terms, the means by which gases can be analyzed; and
3. describe the primary differences among the various designs of CEMSs.
Introduction
Though there are many reasons why a facility might install a continuous emissions monitoring system,
perhaps the most pervasive reasons are to optimize process and control systems and to satisfy regulatory
requirements. CEMSs are required to be installed at certain facilities which are subject to the New Source
Performance Standards (NSPSs) and other Clean Air Act regulations promulgated by the United States
Environmental Protection Agency (USEPA). Other federal and state regulatory agencies might also have
requirements for the installation of CEMSs. These systems are used on a continuous basis for several purposes.
For the plant, properly installed and operating CEMSs can provide information about the operation of key
processes and describe the effectiveness of the air pollution control techniques. Having the ability to examine
various aspects of the facility's operation can provide the user the opportunity to make critical adjustments for
1-1
-------
process optimization and cost reduction. Additionally, CEM records can prove to be important documentation
of compliance status in this time of heightened public concern about air pollution.
For regulatory agencies, the data provided by a CEMS are used to supplement on-site visits, to identify
"problem" sources, and to determine if source emission limits and proper operation and maintenance requirements
are being met With a better picture of actual emissions, the agency's ability to develop strategies for future
programs directed at emissions reduction can be enhanced.
The main goal of both the plant and the regulatory agency is to obtain accurate and repeatable information
about stack emissions. The EPA standards require continuous emissions monitoring data to be obtained from a
representative location. Meeting this requirement entails studying the applicable regulations with respect to the
gases to be monitored.
Analytical Methods
A system, which is defined as the total equipment necessary for the accurate determination and recording of
flue gas concentration, typically consists of a probe, an analyzer, and some form of data readout/recorder. It is
used to generate emissions data which are representative of the total actual emissions from the facility. CEMSs
have been developed to monitor gases including, but not limited to, sulfur dioxide (SO2), nitrogen oxide (NO),
nitrogen dioxide (NO2), carbon monoxide (CO), carbon dioxide (COa), oxygen (O2), hydrogen chloride (Hd), and
ammonia (NH,).
In general, monitoring techniques are based on both physical and chemical properties of gases. The method
selected for gas analysis is primarily dependent on characteristics of the subject gas, but it can also be affected by
other parameters such as regulatory requirements and stack conditions. Brief discussions of the commonly used
analytical techniques are given in the following sections.
Spectroscopic Absorption Techniques
Some of the most common analytical methods used in CEMSs are spectroscopic absorption techniques. These
methods are able to monitor specific gases because various types of gaseous molecules are known to behave
differently in the presence of electromagnetic radiation
Luminescence Techniques
Luminescence techniques can selectively monitor gases by examining the emission of light from an atom or
molecule that has been "excited" in some manner. Because the spectral emissions occur at different wavelengths.
depending on the gas that is excited, these monitors can be used for very specific analysis.
Electroanalytical Techniques
Another technique used for continuous emissions monitoring is electroanalysis. Analyzers mat use this
method can quantify a target gas by measuring the current (or voltage) that is produced in the electrochemical cell.
Different types of electrodes and electrolytes are required depending on which gas is selected for monitoring.
1-2
-------
Paramagnetic Techniques
Paramagnetic methods, used primarily for oxygen monitoring, are based on the fact that molecules behave
differently when placed in magnetic fields. The behavior is classified as either diamagnetic or paramagnetic.
Paramagnetic molecules, like O2, are attracted to and interact with a magnetic field. Measurement of these
interactions can be related to O2 concentration.
Systems Design
CEMSs can be divided into two general classifications based on the means by which sample gas is delivered
to the analyzer. These two categories are extractive systems and in-situ systems.
Extractive CEMSs
Extractive systems withdraw flue gas from the stack, condition it, and transport the gas to the analysis Section
which is usually at some convenient site within the plant The separation of the probe and the analyzer imposes
a time lapse between extraction of the sample gas and its analysis. Consequently, the output reading (concentration
value) will lag behind the process conditions.
The first generation of extractive CEMSs were essentially ambient air analyzers which were modified to
incorporate a simple dilution system suitable for reducing the pollutant concentration to a level which could be
measured by the analyzer. Such systems were used mostly at combustion sources. Because of the relative lack
of sophistication of these early systems, there were significant operational problems associated with them. By
using the specially designed transport and conditioning subsystems, subsequent monitors have been able to avoid
the difficulties posed by paniculate matter, moisture, and corrosive gases.
In-situ CEMSs
In-situ systems have at least some pan of their analysis subsystem mounted in the stack in direct contact with
the flue gas. Because they perform analyses at the stack, in-situ systems neither condition the sample gas nor
transport it to a remote analytical unit The lack of conditioning and transport subsystems indicates mat in-situ
CEMSs designs require substantially less equipment than do extractive CEMSs designs. However, the analytical
subsystems of these CEMSs are generally exposed to a more harsh environment than those of extractive systems.
There are generally two approaches taken in in-situ system design. One approach is to monitor across a certain
path of stack gas; the other is to monitor at a single point within the stack. EPA distinguishes between path and
point monitors by the percentage of the stack diameter (or equivalent diameter for non-circular ducts) represented
by the measurement path. An instrument that measures gas concentrations along a path which is greater than 10
percent of the diameter is said to be a path analyzer. If the measurement path is less than or equal to 10 percent
of the diameter, the instrument is considered a point analyzer.
1-3
-------
Summary
Lesson 1 introduced some terminology and general information associated with the different types of
continuous emissions monitoring systems (CEMSs). More specifically, this lesson gave reasons for having
a monitoring program, briefly described gas analysis techniques, and described some differences in systems'
designs. The material in Lesson 1 provides the basis for more detailed discussions in forthcoming lessons.
In addition to meeting federal and state regulations, otherreasons for having amonitoring program include:
1. gaining information about the effects of certain process parameters,
2. describing the effectiveness of installed control devices, and
3. recording information pertaining to non-compliance situations.
Monitoring techniques are based on both chemical and physical properties of gases and include the
following methods:
1. spectroscopy,
2. luminescence,
3. electroanalysis, and
4. paramagnetism.
CEMSs can be divided into two general classifications based on the means by which sample gas is delivered
to the analyzer. Those two categories are extractive systems and in-situ systems. Extractive systems withdraw
flue gas from the stack, condition it, and transport the gas to the analysis Section which is usually at some
convenient site within the plant In-situ systems have at least some part of their analysis subsystem mounted
in the stack in direct contact with the flue gas. Because they perform analyses at the stack, in-situ systems
neither condition the sample gas nor transport it to a remote analytical unit.
1-4
-------
REVIEW EXERCISES
1. True or false. CEMSs arc often used by plant personnel to optimize the
processes and control systems in addition to meeting regulatory require-
mwiK
2. CEMSs are used on a.
a yeariy
b. nominal
c. continuous
basis.
1. True
3. True or false. Inordertodetenninethe reo^riiements for representative
sampling, one should consult federal and state regulations.
2. c
4. The entire monitoring system contains all of the following except
a a data recorder.
b. a concentration enhancer.
c. an analyzer.
d. a readout instrument
3. True
5. True or false. Monitoring techniques are dependent upon the molecular
weight of the gases to be measured
4.. b
6. Trie two monitoring techniques mat are primarily dependent on the
charactenstics of light are and methods.
a paramagnetic, metaphysical
b. absorption, dilution
c. electroanalysis, electrocution
d. spectroscopic absorption, luminescence
5. False
7.
. systems condition the sample gas before analysis.
a In-sim
b. Opacity
c. Ambient air
d. Extractive
6. d
1-5
-------
8. True or false. Oxygen is a diamagnetic gas and is attracted to a magnetic
field.
7. d
9. True or false. Extractive analyzes are usuaDy designed as path and
point monitors.
8. False
10. True or false. A primary objective of continuous emissions monitoring
programs is to obtain accurate and lepeatable emissions data
9. False
1-6
10. True
-------
REFERENCES
1. Jahnke, JA and Aldina,G.J. 1979. ContiMoiu Air PolMon Source Monitoring Systems. 1-PA 625fo-79-Q05.
2. KflkeDy Environmental Assodaffis. 1988. ContinuousEmissionMontioring Guidelines: Update. Electric Power
Research Institute, Inc. Palo Alto, CA. Report CS-5998.
3. U.S. Environmental Protection Agency. 1986. Quality Assurance Handbook for Air Pollution Measurement
Systems, Volume III. Stationary Source Specific Methods. EPA-60CV4-77-a27b.
1-7
-------
LESSON 2
General Principles of Detection in Monitoring Systems
Lesson Goal and Objectives
Goal
To describe the fundamental means by which gases are detected and analyzed within commercially available
continuous emissions monitoring systems (CEMSs).
Objectives
At the end of this lesson, you should be able to —
1. identify the primary analytical techniques for the detection of gases by commercially available
CEMSs.
2. identify the scientific principles which enable specific gases to be monitored and analyzed, and
3. recognize the most common detection methods used within bom extractive and in-situ CEMSs.
Introduction
In Lesson 1, a variety of CEMSs were discussed. Some information on their background and capabilities was
presented in an effort to acquaint you with those systems and the extent of their utility. Using that information
as a foundation, we are now ready to examine the principles of detection in monitoring systems.
Comprehending the principles and techniques used to monitor certain gases win not only enable you to gain
a better understanding of equipment operation, but it will also provide the technical background necessary to
improve troubleshooting strategies for currently operating systems. The knowledge gained in this lesson will also
help you select the appropriate OEMS to be used in future monitoring applications.
2-1
-------
This lesson will identify some analytical techniques, describe their underlying scientific principles, and
discuss their use in specific types of CEMSs.
Monitoring Techniques
The monitoring techniques upon which this lesson focuses are based on both physical and chemical properties
of gases. Detection of individual gases is possible because of the unique characteristics they exhibit when
subjected to specific, analytical methods. These characteristics determine the principles of design and operation
of reliable CEMSs. The four analytical techniques to be discussed herein are spectroscopic absorption,
luminescence, electroanalysis, and paramagnetism.
Spectroscopic Absorption Techniques
Some of the most common analytical methods used in CEMSs are spectroscopic absorption techniques. Using
these methods, CEMSs are able to identify and to measure the amounts of selected gases via the attenuation
(weakening) of a beam of electromagnetic radiation whose wavelength is known. Figure 2-1, below, provides
a pictorial representation of the electromagnetic spectrum and the general categories of electromagnetic radiation
("light").
(Wavelengths in Meters)
10-i710-is10.D
10J 10s 10r 109
i
10" 1018ltf7
i _ i
Gamma Rays Ultraviolet | infrared Radio Waves
X-Rays
Visible Ught
Microwaves Short Wave Radio
Radar Television AM Radio
(Types of Waves)
FIGURE 2-1 ELECTROMAGNETIC SPECTRUM
Identification of a gas is possible because different types of gaseous molecules are known to absorb specific,
unique wavelengths of electromagnetic radiation. The points (wavelengths) in the spectrum where gases absorb
the maximum amount of radiation are termed "peaks." Determining the amount of gas is accomplished by
converting the percent of the beam's absorbance to the concentration of the identified gas. Though there are
several types of absorption spectroscopy which can identify and measure gases, each method employs similar
equipment for the purposes of gas monitoring. The five basic components for such instrumentation include:
• 1. a stable source of radiam energy ("light"),
2. a wavelength selector (filter) for the isolation of the desired region of spectral emissions.
2-2
-------
3. a sample vessel,
4. a radiation detector/transducer used to convert the radiant energy to a usable signal, and
5. a signal processor and readout.
Figure 2-2 depicts a common set-up for the most fundamental instrumentation used in spectroscopic
analyzers. More detailed information about the various methods and analyzers is presented in the upcoming
paragraphs.
\/\s\
W\AT
\
\/\/\
^
Light Wavelength /
Source Selector
Detector
Signal Processor/
Readout
Sample
Vessel
FIGURE 2-2 BASIC SPECTROSCOPIC INSTRUMENTATION
Non-dispersive infrared (NOTR) spectroscopy uses infrared "light" in a limited range of the electromagnetic
spectrum. The light is filtered in order to isolate the wavelengths which will be absorbed by the gaseous molecules
selected for monitoring. (The light is not scanned or "dispersed" as with scanning laboratory spectrometers). The
filtered light passes through a cell that contains flue gas extracted from the stack. Another portion of the light
passes through a cell which contains a reference gas that does not absorb any of the filtered light (reference=0.0%
absorbance). A detector senses the amount of light absorption in the sample cell relative to the signal from the
reference cell The detector responses are then electronically converted to concentration readings forthe specified
gas.
Non-dispersive IR analyzers (like the one shown in Figure 2-3) are commonly used for extractive CEMSs.
Such instruments have been developed to measure SO,, NO, NO,, HQ, CO. and COZ The commercially available
monitors differ primarily in the design of the detector and the level of rejection of interfering gases (i.e., the degree
to which interference from extraneous gases is suppressed or eliminated).
JSUL Wavelength
S0""* Selector
Sample
Vessel
Reference
Vessel
Signal Processor/
Readout
FIGURE 2-3 NDIR ANALYZER
2-3
-------
A variant of this technique, called gas filter correlation (GFO spectroscopv. requires light, usually in the IR
region, to pass through the sample cell and into a receiving unit In this unit, the beam (which is usually split) is
sent to two, parallel cells called the gas filter cell and the neutral density filter cell. The gas filter cell contains a high
concentration of the gas being analyzed, and therefore removes nearly all of the energy of the predetermined
wavelength (reference = 100% absorbance). What remains of that beam (i.e., the wavelengths not absorbed in the
gas filter cell) passes to the detector. The beam from the neutral density filter cell also goes to the detector. The
difference in energy of these two beams can then be related to the concentration of the gas of interest Figure 2-
4 illustrates a GFC Analyzer.
Neutral Density
Filter Cell
Light
Source
Sample
Vessel
Signal Processor/
Readout
Gas Filter
Cell
FIGURE 2-4 GFC ANALYZER
The advantage of this method is that low levels of paniculate matter will not adversely affect the concentration
reading. Paniculate matter can further reduce the intensity of a light beam which is transmitted through a sample
cell. But because the beam goes to two separate cells and the signals from these cells are ratioed, the effect of
paniculate matter will then cancel and the concentration reading of the subject gas will be more accurate.
Gas filter correlation spectroscopy is used in single-pass, in-situ analyzers as well as ambient air analyzers in
order to measure S02, NO, CO, and CO2. This method has also been used by extractive CEMSs for monitoring Hd
Differential absorption spectroscopy is also based on the principle that certain gas molecules absorb unique
wavelengths of light Unlike non-dispersive IR spectroscopy which uses a sample cell and a reference cell,
differential absorption analyzers employ a "measuring" wavelength and a "reference" wavelength. The measuring
wavelength corresponds to a region of the spectrum where the gaseous molecules of interest absorb light energy,
whereas the reference wavelength corresponds to a region of the spectrum in which the gaseous molecules of interest
absorb little or no light energy. The concentration of the identified gas is determined by finding the difference in
energy for the measuring wavelength and the reference wavelength. The detected difference is then processed and
sent to the readout instrument as a concentration value. Figure 2-5 ilustrates a differential absorption analyzer.
2-4
-------
Diffraction „ Measuring
Grating Photomultiplier
Tube
Light
Source
Sample
Vessel
Signal Processor/
Readout
Reference ..
Photomultiplier
Tube
FIGURE 2-5 DIFFERENTIAL ABSORPTION ANALYZER
Some differential absorption systems are single-pass, in-situ, path monitors that use a diffraction grating to
distinguish between measuring and reference wavelengths in the ultra-violet (UV) region of the spectrum. These
monitors measure SO2 and NO, although by changing the optical system it is possible to measure other gases.
Certain filters can also be used to distinguish between measuring and reference wavelengths. This alteration
allows the monitoring of CO by IR light
The differential absorption method can also be used in extractive CEMSs. While most of these systems operate
in the UV region, it is also possible to use IR radiation. When using a UV apparatus, subject gases may be measured
hot without removing water vapor; however, it is generally advisable to dry the sample before taking measure-
ments.
Many extractive CEMSs use differential absorption spectroscopy to monitor S02. This technique has also
been modified in order to measure NO from flue gas. By injecting oxygen into the sample chamber and sealing
it, NO is measured via the production of N02 from NO while monitoring an N02 absorption wavelength.
Second derivative spectroscopv uses a different means of concentration determination than the previously
described absorption techniques. In this case, a scanner moves back and forth across the absorption peak of the
subject gas. The scanning is related to the second derivative of the absorption peak, with respect to wavelength.
The signal from scanning is then used to determine the concentration of the selected gas. A second derivative
spectroscopy analyzer is shown in Figure 2-6. Second derivative spectroscopy is used in a widely marketed in-
situ, point monitor for SO2 and NO.
2-5
-------
Signal Processor/
Readout
Light
Source
Scanner
FIGURE 2-6 SECOND DERIVATIVE SPECTROSCOPIC ANALYZER
Luminescence Techniques
Luminescence is the emission of light from an atom or molecule that has been excited in some manner.
Excitation refers to the temporary placement of electrons into energy levels (orbitals) which are higher than the
ground G.C., relaxed) state energy levels for that atom or molecule. As the electrons return to their normal levels,
excitation energy in the form of light is emitted. Three luminescence techniques are used in the field of source
monitoring. They are UV fluorescence, chemiluminescence, and flame photometry. Since each method differs
primarily with respect to their means of excitation rather than their means of measurement, similar instrumentation
can be employed foreach technique. The basic requirements are the same as for spectroscopic techniques, though
their configuration is somewhat different for luminescence techniques. They are:
1. a stable source of radiant energy.
2. a sample vessel,
3. a wavelength selector for the isolation of the desired region of spectral emissions,
4. a radiation detector/transducer used to convert the radiant energy to a usable signal, and
5. a signal processor and readout
2-6
-------
Shown in Figure 2-7 is a schematic representation of a typical set-up for the basic equipment requirements.
Please note that the equipment is basically the same as that which is used in absorption spectroscopy; however,
its configuration differs slightly.
Energy
Source
-A/W
Sample
Vessel
Wavelength
Selector
Signal Processor/
Readout
FIGURE 2-7 BASIC INSTRUMENTATION FOR LUMINESCENCE TECHNIQUES
The UV fluorescence method of analysis is primarily used to measure S02 present in exhaust gas streams. S02
molecules are excited to S02* by using UV light of wavelength 210 nanometers (run). Excitation is the result of
an atom's absorption of this beam of electromagnetic radiation. The excited molecular state persists for only a
few nanoseconds during which time some of the energy is lost through vibrational transitions of the electrons. As
the molecule returns to its relaxed state, light of a longer wavelength, 350 nm, is released. The released light can
be detected by the instrumentation within the monitor. Its intensity is then related to gas concentration.
Fluorescence monitors are affected by changes in the flue gas composition (%O2, %CO, and %N2). These
fluctuations in diluent gas concentration result in the relaxation of S02* molecules through a process called
quenching. Quenching occurs when excited molecules return to their normal state by losing their extra energy
through interactions with other gases instead of releasing their energy in the form of light; consequently, some of
the excitation energy would not be detected. In order to reduce the problems caused by quenching, the background
gas composition should be held relatively constant
Fluorescence analyzers which employ a dilution system provide the constant background required for
successful measurements. Dilution systems consequently dictate the use of extractive CEMSs for this type of
monitoring for S02. Figure 2-8 depicts a UV fluorescence analyzer.
Light
Source
^1
X^jv/\/\>
Wavelength,
Selector
Sample
Vessel
Photomurtiplier
Tube
Detector
Signal Processor/
Readout
FIGURE 2-8 UV FLUORESCENCE ANALYZER
Chemiluminescence is used in flue gas analysis to measure NO and N02 concentrations. In this application
of Chemiluminescence, excited molecules, N02*. are produced by reacting ozone with the NO present in the
exhaust stream. The excited product, N02*. returns to its ground state with the release of light energy. The emitted
light enters a photomultiplier tube and is later related to the concentrations of NO and NO2.
2-7
-------
Quenching effects can also occur in this method, but dilution of the sample by the introduction of the reactant
ozone gas stream minimizes the effect This technique also requires the use of extractive CEMSs. Chemilumi-
nescence methods for monitoring NO and N02 already exist and are being developed for NHj.
Flame photometry can be used to measure compounds that contain sulfur. In this method, the compounds are
"bumed"in a hydrogen flame which results in the formation of excited diatomic sulfurmolecules, S2*. The conversion
of the high energy S2* molecules to the lower energy ground state. S2, occurs with the emission of light The
intensity of this light is measured and related to the concentration of sulfur species in the sample.
The flame photometric method does not discriminate between sulfur-containing compounds, so scrubbers or
gas chromatographic columns may be required if more than one species is present in the sample. As with the two,
prior luminescence methods, flame photometry involves the use of extractive CEMSs. A flame photometry
analyzer is shown in Figure 2-9.
Sample
Vessel
Wavelength
Selector
Photomuttiplier
Tube
Detector
Signal Processor/
Readout
FIGURE 2-9 FLAME PHOTOMETRIC ANALYZER
Electroanalytical Techniques
Two principal electroanalytical techniques have been developed for the measurement of gases in exhaust
streams. They are polarography and electrocatalysis. These methods can quantify flue gases by measuring the
currents or voltages that are produced in the electrochemical ceHs.
Polarography utilizes analyzers which are basically diffusion-controlled electrochemical cells. The current
across the cell is proportional to the rate of diffusion of the pollutant into the cell and also proportional to the
pollutant concentratioa Being constructed much like batteries, the cells have a sensing electrode, an electrolyte,
and a counter electrode. The main difference from a wet cell battery is that polarographic analyzers also possess
a thin-film membrane through which the pollutant must diffuse in order to initiate the electrochemical reaction
and current flow.
Polarogra hie analyzers (Figure 2-10) have been developed to measure gases such as O2, SO2, NO, NO,. CO,
and CO2. Different types of electrodes and electrolytes are required depending on which gas is selected for
monitoring. As with batteries, the electrolyte will eventually be consumed and the cell must either be replaced or
recharged. Because these types of analyzers require dean, dry gas samples, they can only be used in conjunction
with extractive CEMSs.
2-8
-------
Thin-Film
Membrane
Counter
Electrode
Signal Processor/
Readout
FIGURE 2-10 POLAROGRAPHIC ANALYZER
Electrocatalysis employs analyzers like the one shown in Figure 2-11 that use a solid electrolyte instead of
a liquid electrolyte often associated with electrochemical cells. A thin film, applied to the solid's surface, catalyzes
a reaction which allows gaseous molecules to migrate through the solid and generate a current
Sensing
Electrode
Thin-Rim
Membrane
/ Counter
Electrode^..
Signal Processor/
Readout
FIGURE 2-11 ELECTROCATALYTIC ANALYZER
In oxygen electrocatalytic analyzers, a zirconium oxide (ZrO2) disc which has been coated with a thin film
of platinum is heated to 850°C. A reference gas of about 21 percent oxygen is maintained on one side of the solid;
the sample gas is on the other side. Oxygen ions are generated at the platinum surface and then migrate through
vacancies in the solid electrolyte. Electrons are released in the process as the system attempts to reach an
equilibrium concentration of oxygen. The electron flow, or current, is then related to the concentration of oxygen
in the sample.
2-9
-------
An electrocatalytic analyzer has also been developed for the measurement of S0r This apparatus uses a
potassium sulfate crystal and requires the simultaneous measurement of oxygen concentration in the sample. This
type of in-situ analyzer is used exclusively as a point monitor and has a ceramic thimble to keep the sob'd electrolyte
free of paniculate matter. Both 02 and S02 are able to be monitored by this technique.
Paramagnetic Techniques
Oxygen is known to exhibit paramagnetic behavior. That is to say that oxygen is attracted by and interacts
with a magnetic field. This behavior is discussed in relation to the following three extractive flue gas analyzers.
The thermomapnetic technique is used to monitor oxygen by the following procedures. Such analyzers use
a magnet to channel the flow of oxygen through a designated tube. A higher concentration of 02 in the gas will
result in a higher velocity of gas through the tube. As illustrated in Figure 2-12, the flow of O2 gas past a resistor
will cause the resistor to cool. Beacuse a constant voltage is maintained across the resistor, a decrease in resistance
due to cooling will be compensated by an increase in current (because V = I x R). The current can then be related
to oxygen concentration.
Signal Processor/
Readout
FIGURE 2-12 THERMOMAGNETIC ANALYZER
The magnetodynamic technique is used in CEMSs to relate oxygen concentration in sample gas to distur-
bances in a magnetic field about a torsion pendulum. As shown in Figure 2-13,02 gas flows through a magnetic
field and disrupts the alignment of the magnetic charges which consequently causes the torsion pendulum to turn.
The degree of this movement can then be related to the amount of O, within the sample.
2-10
-------
Signal Processor/
Readout
FIGURE 2-13 MAGNETODYNAMIC ANALYZER
Paramagnetic pressure is the pressure imbalance which results from the interaction of O2 molecules with a
magnetic field The resulting movement is measured by a device (magnetopneumatic analyzer) which is very
sensitive to this change and is then related to the concentration of O2 in the sample.
Summary
Lesson 2 presented the four primary methods used in commercially available CEMSs for the analysis of flue
gas. These methods are:
1. Spectroscopic Absorption,
2. Luminescence,
3. Hectroanalysis, and
4. Paramagnetism.
The requirements of and limitations to certain analytical methods were presented so that the user can better
understand the real-world utility of gas analyzers.
Additionally, the physical and chemical principles by which these instruments operate were discussed in
relation to specific gases. Lesson 3 will give more information about these general principles as well as their
application in analytical devices.
Tables 2-1 through 2-4 summarize the types of monitors and gases monitored by the analytical techniques
cited in Lesson 2. Full comprehension of this information will aid the reader in the selection of the appropriate
CEMS for his needs and also aid him with maintenance and troubleshooting programs.
2-11
-------
TABLE 2-1
SUMMARY OF SPECTROSCOPIC
ABSORPTION TECHNIQUES
Spectroscopic Method
System Type
Gas Monitored
Nondispersive Infrared
Gas Filter Correlation
Differential Absorption
Second Derivative
Extractive
Extractive
In-situ
Extractive and In-situ
. In-situ
SOg.NO.NOfe.
HCI, CO. CCfe
HCI
SCj.NO.CO.CCfe
SCg.Npg
SCfe.NCj,
TABLE 2-2
SUMMARY OF
LUMINESCENCE TECHNIQUES
Luminescence Method
System Type
Gas Monitored
Ultraviolet Fluorescence
Chemiluminescence
Rama Photometry
Extractive
Extractive
Extractive
SPz
NO, NO2, NK,
Gases containing Sulfur
2-12
-------
TABLE 2-3
SUMMARY OF
ELECTROANALYTICAL TECHNIQUES
Electroanalytical Method
System Type
Gas Monitored
Polarography
Elect rocatalysis
Extractive
Extractive and In-situ
q,,sq,.NO.cq,
^.sq.
TABLE 2-4
SUMMARY OF
PARAMAGNETIC TECHNIQUES
Paramagnetic Method
System Type
Gas Monitored
Thermomagnetic
Magnetodynamics
Magnetopneumatics
Extractive
Extractive
Extractive
02
%
Oz
2-13
-------
REVIEW EXERCISES
1. Detection of specific gases is posible because
a. they are different colors of the electromagnetic spectrum.
b. of their unique characteristics.
c. each has a different smelL
2. Spectroscopic absorption techniques are based on
a. the molecular weight of the gas.
b. a gas' ability to absorb water.
c. the absorption of a unique wavelength of light by the gas.
1. b
3. Spectroscopic methods relate the gas concentration to
a. the amount of light absorbance.
b. the wavelength of the gases.
c. radioactivity.
2. c
4. Sample vessels for Spectroscopic analysis & gases should be
a. transparent.
b. transcendental.
c. transducers.
3. a
5. True or false. Gas filter correlation spectrosoopy uses a cell which
absorbs all of the gas to be analyzed.
4. a
6. In differential absorption spectroscopy, a diffraction grating is used
to distinguish between
a. species of gases emitted from the stack.
b. measuring and reference wavelengths.
c. extractive and in-situ systems
5. False
7. Excitation energy released as light is the principle behind
a. luminescence techniques.
b. electroanalytical techniques.
c. paramagnetic techniques
6. b
2-14
-------
8. Quenching is
a. the altering of a magnetic field.
b. not a problem in luminescence analysis.
c. the relaxation of excited gas molecules by diluent gas fluctua-
tions.
7. a
9. In polarographic analyzers, the gas diffuses through a.
order to generate a
in
8.c
10. Paramagnetic techniques are primarily used to detect
a. S2
b. NO
c. O,
9. a
10. c
2-15
-------
REFERENCES
1. Entropy Environmentalists, Inc. 1988. Strategies for Continuous Monitoring of Hydrogen Chloride
Emissions from Municipal Solid Waste Incinerators. EPA/600/D-88/072.
2. Horiba Instrument, Inc 1989. Stack Gas AnalyzerSystem (Advertising Brochure). Bulletin: HRE-2354B.Horiba
Instruments, Incx, 1021 Duiyea Ave., Irvine, CA 92714-5583.
3. Jahnke, JA. and Aldina, G. J. 1979. Continuous Air Pollution Source Monitoring Systems. EPA 625/6-79-005.
4. KflkeDy Environmental Associates. 1988. Continuous Emission Monitoring Guidelines: Update. Electric
Power Research Institute, Inc. Palo Alto, CA. Report CS-5998.
5. KVB.Inc. 1988.XmTO?maMomwr(SpecShsct).NH3SS 12/6/88. KVB, Inc., 9342 Jeronimo, Suite 101, Irvine,
CA 92718.X
6. KVB, Inc. 1988. Hydrogen Chloride (HCl) Emissions Monitor (Advertising Brochure). HCL PP 1/9/88.
KVB, Inc, 9342 Jeronimo, Suite 101, Irvine, CA 92718.
7.
PA,
8. U. S. Environmental Protection Agency. 1986. Quality Assurance Handbook for Air Pollution Measurement
Systems, Volume EL Stationary Source Specific Methods. EPA-60Q4-77-027b.
9. Winbeny, Jr., W.T. 1985. TechnicalAssistanceDocumentforMonitoring TatalReducedSulfur(TRS) from Kraft
Pulp Mills. EPA-340/l-85-013a.
2-16
-------
LESSON 3
Specific Analytical Methods Used by Analyzers
Lesson Goal and Objectives
Goal
To provide a better understanding of detection principles when applied to certain gases and their measurement
Objectives
At the end of this lesson, you should be able to —
1. describe the process of spectroscopic absorption and analysis,
2. explain the primary differences among spectroscopic techniques,
3. explain how luminescence techniques work,
4. describe electrocatalysis and its means of determining gas concentration, and
5. discuss the fundamental similarities and differences of paramagnetic techniques.
Introduction
In Lesson 2, four primary analytical techniques used in continuous emissions monitoring systems (CEMSs)
were introduced. This lesson will provide more specific information about the four detection methods
(spectroscopic absorption, luminescence, electroanalysis, and paramagnetism) used to measure gaseous emis-
sions. The theories, derivations, and applications of each principle will be discussed in order to provide you with
a sound understanding of the analytical basis of each type of instrument
3-1
-------
Spectroscopic Absorption
Theory and Derivation
CEMSs which use spectroscopic analyzers (spectrophotometers) rely on the interaction of light energy with
gaseous molecules in order to determine the gas concentration. Since light energy and wavelength are related,
analyzers can employ light from different parts of the spectrum for monitoring. The proper spectral region is
selected by using the equations below.
Given that:
E = energy
h = Planck's constant
c = speed of light (constant)
•o = frequency
X = wavelength
we can then determine the relationship of light to wavelength.
Since
(Eg. 3-1) E «
-------
Given that
Ic ss intensity of original light beam
I = intensity of attenuated light beam
A = absorbance
a = absorptivity constant
b = path length through absorbing medium
c = concentration of absorbing species
we can then determine the gas concentration.
Since
(Eq.3-4) A = log a. /I)
and
(Eg. 3-5) A = (a)(b)(c) [Beer's Law],
then
(Eq.3-6) (a)(b)(c) = log (Ie/1), and
therefore
(Eg. 3-7) c = [log(I0/I)]/(a)(b)
Since all the quantities on the right of the equation are known or can be measured, the gas concentration can
be calculated.
Applications
The theories and relationships just presented provide the basis for all spectrophotometric techniques in general
use. Of course, the implementation of these schemes will vary from one analytical device to another. Though
spectrophotometers may be either dispersive or non-dispersive, nearly all which are used in GEMSs are non-
dispersive, as indicated in Lesson2. These analyzers do not separate light into its component wavelengths; instead,
they utilize a broad band of wavelengths which is centered on the absorption peak of the subject gas. Some
absorption bands of common gases are given in Table 3-1.
Non-dispersive IR spectrophotometers use a filtered band of light from the infrared region of the spectrum.
The filtered light passes through a sample cell that contains flue gas extracted from the stack. Another portion
of the light passes through a cell which contains a reference gas that does not absorb any of the filtered light The
beams from each cell enter the detector where the difference in intensity is measured.
3-3
-------
TABLE 3-1
INFRARED BAND CENTERS OF SOME COMMON GASES
Gas
Location of Band Centers (urn) Wave Number (cm"1)
NO
N02
Sq,
H2O
CO
COz
NH3
5.0-5.5
5.5-20
6-14
3.1
5.0-5.5
7.1-10
2.3
4.6
2.7
5.2
8-12
10.5
1800-2000
500-1800
700-1250
1000-1400
1800-2000
3200
2200
4300
850-1250
1900
3700
950
Typically, microphone detectors are used in these analyzers. They are gas-filled cells with two compartments
that are separated by a flexible metal diaphragm. The cells contain the same gas that is being measured. As the IR
light passes through the sample cell and the reference cell, the molecules in the detector are heated. .Because of the
absorption of some of the IR energy in the sample cell, the intensity of the beams is different Consequently, the
gas in the detector cells will be heated to different temperatures which will cause the diaphragm to flex due to the
higher pressure in the warmer cell. This produces a signal which is in turn converted to gas concentration. This
process is illustrated in Figure 3-1.
Sample
Vessel
Light
Source
Wavelength
Selector
Reference
Vessel
Signal Processor/
Readout
Microphone
Detector
FIGURE 3-1 NDIR ANALYZER WITH MICROPHONE DETECTOR
Gas filter correlation fGFO analyzers measure the degree of similarity between the absorption of light by a gas
of known concentration and the absorption of light by a sample gas. In GFC systems, light emitted from the lamp
passes through the sample gas and then into a receiving unit In this unit, the beam (which is usually split) is sent
to two, parallel cells called the gas filter cell and the neutral density filter cell. The gas filter cell contains a high
concentration of the gas being analyzed, and therefore removes nearly all of the energy of the predetermined
3-4
-------
wavelength. What remains of the beam (i.e., the wavelengths not absorbed in the gas filter cell) passes to the
detector. The light which passes through the neutral density filter cell also goes to the detector where its energy
is measured and compared to that of the other beam. The difference in energy of these two beams can then be related
to the concentration of the gas of interest. This process is illustrated in Figure 3-2.
Neutral Density
Filter Cell
Light
Source
Sample
Vessel
Mirror
Signal Processor/
Readout
Gas Filter
Cell
FIGURE 3-2 GFC ANALYZER
Other gases do not usually cause interference since they absorb the reference and measurement beams equally.
Therefore, the system responds only to the gas of interest and provides higher sensitivity when compared to some
other instruments.
Differential absorption spectrophotometers employ a measuring wavelength and a reference wavelength. The
measuring wavelength corresponds to a region of the spectrum where the gaseous molecules of interest absorb
light energy, whereas the reference wavelength corresponds to a region of the spectrum in which the gaseous
molecules of interest absorb little or no light energy. A beam of UV radiation passes through the sample cell and
goes to a diffraction grating. One beam is directed to the sample photomultiplier tube while the other is sent to
the reference photomultipliertube. The concentration of the identified gas is determined by finding the difference
in energy for the measuring wavelength and the reference wavelength. The detected difference is then processed
and sent to the readout instrument as a concentration. A representation of this process is given in Figure 3-3.
3-5
-------
Diffraction .
Gratina Photomultiplier
9 Tube
Light
Source
Sample
Vessel
Signal Processor/
Readout
Reference
Photomultiplier
Tube
FIGURE 3-3 DIFFERENTIAL ABSORPTION ANALYZER
Second derivative spectroscopic analyzers rely on two devices not found in ordinary spectrophotometers. The
first, a diffraction grating, is used to narrow the region of light wavelengths which is initially emitted from the light
source. More specifically, the diffraction grating separates the wavelengths associated with the target gas. The
second device, a scanner (or moving slit), moves back and forth across the central wavelength that corresponds
to the maximum absorption peak of the subject gas. The result of this process is a gradual decrease to a minimum
transmittance value followed by a gradual increase to its maximum value and is seen as a sinusoidal wave. The
resulting signal varies at twice the frequency of the scanner. The signal's amplitude is directly related to the
concentration of the target gas. A diagram of this scanning process is provided in Figure 3-4.
Signal Processor/
Readout
Light
Source
Scanner
FIGURE 3-4 SECOND DERIVATIVE SPECTROSCOPIC ANALYZER
3-6
-------
Luminescence
Theory and Derivation
CEMSs which use luminescence analyzers rely on light emissions from excited molecules. In this process,
energy converts the absorbing atom, molecule, or ion (M) to a more excited form (M*). Excitation refers to the
temporary placement of electrons into higher energy levels (orbitals) and can be represented by the following
equation:
(£0.3-$) M + (h)OU) ». M*
Please recall that the product of h (Planck*s constant) and nu (frequency) is defined as the energy of a particular
wavelength of light
As the electrons return to their normal levels, excitation energy in the form of light is emitted. The release
of this energy is basically the reverse of the prior process and is represented by this equation:
(Eq.3-9) M* *. M + (h) (V)
The difference in the second equation is that the light emitted will have a different frequency and wavelength
than that of the absorbed beam. Measuring the energy of the emitted beam can be used to calculate the gas
concentratioa The methods described below differ primarily with respect to their means of excitation rather than
their means of measurement
Applications
UV fluorescence analyzers are primarily used to measure S02 present in exhaust gas streams. S02 molecules
are excited to SO2* by using UV light of wavelength 210 nanometers (nm). Excitation is the result of an atom's
absorption of this beam of electromagnetic radiation and is written as:
(Eq.3-10) S02+(h)(\>) *. SO*
The excited molecular state persists for only a few nanoseconds during which time some of the energy is lost
through vibrational transitions of the electrons. As the molecule returns to its relaxed state, light of a longer
wavelength, 350 nm, is released.
(Eq.3-11) SO* +. S02-f (h)^1)
The released light can be detected by the instrumentation within the monitor. Its intensity is men related to
gas concentratioa
Chemiluminescence analyzers are primarily used to measure NO and NO2 concentrations. In this application
of Chemiluminescence, excited molecules, N02*. are produced by reacting ozone with the NO present in the
exhaust stream.
(Eq.3-12) NO + 05 * N02* + 02
3-7
-------
The excited product, N02*. returns to its ground state with the release of light energy whose wavelength is
in the range of 600 to 900 nm. The emitted light is measured with a photomultiplier tube and is related to the
concentrations of NO.
(Eg. 3-13) N02* * N02+ (h) (D)
In the case of measuring N02, the NO2 is heated in the presence of a catalyst and converted to NO. The equation
is as follows:
heat
(Eq.3-14) NO, + NO + 1/2 02
cat
Upon conversion of the N02 to NO, the analysis proceeds as described for NO measurement and the
concentration is reported as total N0t (i.e., both NO and N02).
Flame photometry analyzers excite a subject gas sample by introduction into a hydrogen flame. This method
is primarily used to analyze gas streams which contain sulfur and sulfur compounds. The sulfur compounds are
"burned" in a hydrogen flame which results in the formation of excited sulfur atoms. The conversion of the high
energy molecules to the lower energy ground state occurs with the emission of light The following equation
illustrates this process.
(Eq.3-15) S2* + S2 + (h)(D)
The intensity of this light is amplified by a photomultiplier tube and is related to the concentration of sulfur
species in the sample. The flame photometric method does not discriminate between sulfur-containing
compounds, so reported concentrations will be for total sulfur content
Electroanalysis
Theory and Derivation
CEMSs that employ electroanalytical equipment for measuring flue gas components use basic electrochem-
istry to determine the amounts of the target gases. Typical electrochemical cells contain electrodes which have
been selected for the specific gas in mind. As the chemical reaction within the cell proceeds, substances will either
gain electrons or lose electrons. The electrons migrate through the medium and create an electrical current
between the electrodes. As a consequence of the reaction, the chemical potential across the electrodes will be
different; this difference is termed electromotive force (EMF) or voltage. These measurable quantities (current
and voltage) can be related to the concentration of the subject gas.
Applications
Polaropaphic analyzers (also called voltametric analyzers or electrochemical transducers) contain an
electrochemical cell where the subject gas is involved in a chemical reaction. In the first part of the process, the
subject gas diffuses through a selective, semi-permeable membrane. In the second part, the gas will participate
in a reaction within the cell during which time a flow of electrons will be generated and detected by the sensing
electrode. For S02, the oxidation reaction will proceed as follows:
3-8
-------
(Eg. 3-16) S02 + 2HjO
This reaction generates two moles of electrons for every one mole of S02 that reacts.
If the rate of this process is controlled by the rate of diffusion, then the current is proportional to the
concentration of the subject gas. The following equation, known as Pick's Law, shows this relationship:
(Eg. 3-17) i = (n) (F) (A) (D) (c) / 8 = (k) (c)
when
i = current
n = # of electrons exchanged per mole of gas
F = Faraday's constant
A = exposed electrode surface area
D = diffusion coefficient (constant)
c = concentration of gas
6 = thickness of diffusion layer
k = proportionality constant
Since i, A,, and 5 are measurable and n, F, and D are known, the equation can be solved for the gas
concentratioa
Electrocatalvtic analyzers use the principle that solutions of two concentrations will proceed to equilibrium
upon coming in contact with each other. The two sections of an electrocatalytic cell have different concentrations
of the subject gas which also means that the chemical potential of those sections is different This difference in
potential across the electrode can be measured as EMF or voltage. The Nemst equation gives the relationship
of voltage to concentration as:
(Eg. 3-18) EMF * [(R) (T) / (4) (F)] nn(c,/c2)]
when
EMF = electromotive force (voltage)
R = gas constant
T = absolute temperature
F = Faraday's constant
c, = concentration of gas in section 1 (stack)
Cj = concentration of gas in section 2 (known)
As the system attempts to reach an equilibrium concentration of gas in both sections, a measurable flow of
electrons is generated. The current can then be related to the concentration of the target gas in the sample.
3-9
-------
Paramagnetism
Theory and Derivations
CEMSs which use paramagnetic analyzers employ the principle that molecules behave differently when
placed in magnetic fields. The behavior is classified as either diamagnetic or paramagnetic. Most molecules are
diamagnetic and are repelled when placed in a magnetic field. A few molecules possess paramagnetic properties
and are attracted by the magnetic field. Though most molecules have paired electrons (i.e., the same number of
electrons spinning clockwise as there are spinning counterclockwise), some molecules have one ormore electrons
spinning in the same direction. This characteristic gives rise to paramagnetic behavior as is the case with oxygen.
Specifically, oxygen has two unpaired electrons that cause O2 molecules to be attracted to and interact with a
magnetic field. Paramagnetic analyzers are primarily used to measure the concentration of 02 molecules based
on their interaction with the magnetic field.
The only other molecules typically found in flue gas which exhibit a significant amount of paramagnetic
behavior are NO and N02. If the concentration of NOX is sufficiently high, some interference in the measurement
of 02 may occur.
Applications
Thermomapietic monitors measure oxygen by the following procedures. A magnet is used to channel the flow
of oxygen through a designated tube. A higher concentration of O2 in the gas will result in a higher velocity of
gas through the tube. As illustrated in Figure 3-5, the flow of O2 gas past a resistor will cause the resistor to cool.
Because a constant voltage is maintained across the resistor, a decrease in resistance due to cooling will be
compensated by an increase in current This relationship is described below. The current can then be related to
oxygen concentration.
(Eq.3-19) V = G)(R)
when
V = voltage
i = current
R = resistance
Magnetodvnamic analyzers use a torsion pendulum to measure disturbances in a magnetic field. In these
analyzers, a dumbbell-shaped, platinum ribbon is suspended in a magnetic field. Since platinum is diamagnetic,
it is slightly repelled. When a sample that contains 02is introduced, it causes the magnetic field to become
reoriented. This process, shown in Figure 3-6, causes the dumbbell to turn and realign itself within the new field.
As the dumbbell repositions, light hitting a small reflector on the pendulum will indicate the degree of the turn.
The degree of the turn is used to calculate the concentration of 02 in the sample.
3-10
-------
Signal Processor/
Readout
FIGURE 3-6 MAGNETODYNAMIC ANALYZER
Paramagnetic pressure fmapnetopneumptic) analyzers measure the pressure imbalance which results from the
migration of O2 molecules in an uneven magnetic field. As the O2 enters the magnetic field, it will be drawn to
the stronger side of the field and cause a pressure rise. This increase in pressure can be represented by the following
equation:
(Eg. 3-20) 5P = 1/2(H)2(X)(O
where
H = strength of magnetic field
X = magnetic susceptibility of Or and
C = concentration of O2
The resulting movement is measured by a device which is very sensitive to this change and is then related to
the concentration of O2 in the sample.
Summary
Lesson 3 provided specific information about the four detection methods (spectroscopic absorption,
luminescence, electrocatalysis, and paramagnetism) used to measure gaseous emissions. The theories, deriva-
tions, and applications of each principle were presented in orderto provide you with a sound understanding of the
analytical basis of each type of instrument.
3-11
-------
Spectroscopic analyzers (spectrophotometers) rely on the interaction of light energy with gaseous molecules
in order to determine the gas concentration. Since light energy and wavelength are related, analyzers can employ
light from different parts of the spectrum for monitoring. The selection of the proper spectral region is dependent
upon the amount of energy absorbed by the gas.
Luminescence analyzers rely on light emissions from excited molecules for monitoring purposes. In this
process, light energy converts the absorbing atom, molecule, or ion (M) to a more excited form (M*). Excitation
refers to the temporary placement of electrons into higher energy levels (orbitals) than the ground state of that
particular atom. As the electrons return to their normal levels, excitation energy, in the form of light, is emitted.
Qectroanalytical equipment is also used for measuring the amount of a target gas. Typical electrochemical
cells contain electrodes which have been selected for the specific gas in mind. As the chemical reaction within
the cell proceeds, substances will either gain electrons or lose electrons. The electrons migrate through the medium
and create an electrical current between the electrodes. As a consequence of the reaction, the chemical potential
across the electrodes wfll be different; this difference is termed voltage. These measurable quantities (current and
voltage) can be related to the concentration of the subject gas.
Paramagnetic analyzers are based on the fact that molecules behave differently when placed in magnetic
fields. The behavior is classified as either diamagnetic orparamagnetic. Most molecules are diamagnetic and are
repelled when placed in a magnetic field. A few molecules possess paramagnetic properties and will be attracted
by the magnetic field. Though most molecules have paired electrons (i.e. the same number of electrons spinning
clockwise as there are spinning counterclockwise), some molecules have one or more electrons spinning in the
same directioa This characteristic gives rise to paramagnetic behavior as is the case with oxygen. Specifically,
oxygen has two unpaired electrons that cause 02 molecules to be attracted to and interact with a magnetic field.
Measurement of these interactions can be related to 02 concentration.
3-12
-------
REVIEW EXERCISES
1. In spectroscopic absorption, the amount oflight energy absorbed
by a subject gas corresponds to
a. the amount of light given off by that same gas.
b. the value of Planck's constant
c. the amount of energy required to cause a change in the gas.
d. its luminescence.
2. As a gas absorbs some light energy, the intensity of the light beam
will from its original value.
a. increase
b. decrease
c. not change
1. c
3. Nearly all spectrophotometers used in CEMSs are and use a
broad band of wavelengths that are centered on the of the target
gas molecules.
a. dispersive... reference
b. dispersive... absorption peak
c. non-dispersive... luminescence
d. non-dispersive... absorption peak
2. b
4. The purpose of a gas filter cell is
a. to remove all the sample gas from the stack.
b. to separate the target gas from interfering gases.
c. to absorb from a light beam the spectral absorption wavelengths
of the gas.
d. to filter gas before going to the reference cell.
3. d
5. The scanning done in a second derivative spectroscopic analyzer
results in
a. a fluctuation in the transmittance value received at the detector.
b. an emission of light from excited molecules.
c. a longer wavelength of light.
d. a diffraction detector.
4. c
3-13
-------
6. Flame photometric analyzers employ a.
excitation of gas molecules.
a. hydrogen
b. luminescent
c. natural gas
flame for the
5. a
7. Flame photometry can be used to measure
a. HjS
b. H2S5.
c. both of the above.
d. none of the above.
6. a
8. Electroanalysis is able to relate gas concentration to
a. current and capacitance.
b. voltage and current
c. resistance and capacitance.
d. Faraday's constant and Planck's constant
7. c
9. Some molecules, especially O2, termed.
are attracted by magnetic fields.
a. diamagnetic
b. magnetic
c. paramagnetic
d. interactive
. since they
8. b
10. The torsion pendulum in magnetodynamic analyzers.
concentration of O2 in the sample changes.
a. cools
b. turns
c. glows
d. heats
as the
9. c
10. b
3-14
-------
REFERENCES
1. Jahnke, JA and Aldina, G. J. 1979. Continuous Air Pollution Source Monitoring Systems. EPA 625/6-79-005.
2. Kflkelly Environmental Associates. 1988. Continuous Emission Monitoring Guidelines: Update. Electric Power
Research Institute, Inc. Palo Alto, CA. Report CS-5998.
3. Skoog.D. A_andWest,D.M. l9S2.FundamentaIsofAnafyticedChemistry.Swi^ColtegpPuVb^
PA
4. U.S.EnvironmentalProtectionAgency. 1986.QuahtyAssw'cwceHandbookforAirPolhitionMeasurementSystems,
Volume III. Stationaiy Source Specific Methods. EPA-60CV4-77-Q27b.
5. Winbeny, Jr., W. T. 1985. Technical Assistance Document for Monitoring Total Reduced Suffur (TOS) from Kraft
^ PulpMn]s.EPA-340/l-85-013a.
3-15
-------
UNIT 2
OPERATION
OF
COMMERICALLY AVAILABLE
GAS MONITORS
-------
LESSON 4
Operation of Two Spectroscopic Absorption Analyzers
Lesson Goal and Objectives
Goal
To explain the operation of both a Bodenseewerk Hd monitor and a Teledyne SO2 monitor.
Objectives
At the end of this lesson, you should be able to —
1. identify which region of the electromagnetic spectrum is employed by each of the analyzers,
2. describe the development and utility of reference signals for each of the analyzers, and
3. describe the "special features" of each instrument which are intended to improve the accuracy
and repeatability of the measurements.
Introduction
In Lessons 2 and 3, four primary analytical techniques used in continuous emissions monitoring systems
(CEMSs) were presented. These techniques are: spectroscopic absorption, luminescence, electroanalysis, and
paramagnetism. A general overview of detection procedures and calculation methods within the various types of
analyzers has already been given. This lesson will more specifically describe the operation of two commercially
available CEMSs which are based on spectroscopic analysis.
Spectroscopic Absorption Analyzers
CEMSs which use spectre-photometers rely on the interaction of light energy with gaseous molecules in order
to determine the gas concentration. Since light energy and wavelength are related, analyzers can employ light from
4-1
-------
different parts of the spectrum for monitoring. When a certain gas absorbs some amount of the energy from a light
beam, the intensity of that beam will decrease. This decrease in intensity can then be related to the concentration
of the gas.
As discussed in previous lessons, there are a number of variations in the application of spectroscopy to the
analysis of flue gases. Monitors which use two of those techniques, gas filter correlation and differential
absorption, will be described in this lesson.
Gas Filter Correlation (GFC) Analyzer
Bodenseewerk Geosystem GmbH of West Germany has developed the Model 677 IR HQ Emissions
Monitoring System for the measurement of hydrogen chloride gas (Hd). (KVB, Inc. is the distributor in the U.S.
for this monitoring system.) This system employs the GFC technique to obtain accurate and repeatable
measurements of HG gas emissions from sources such as waste incinerators. A view of this monitoring system
is shown in Figure 4-1.
FIGURE 4-1 BODENSEEWERK MODEL 677 HCL MONITOR
The handling and conditioning of sample gases is of great importance in all CEMSs, especially those which
monitor HCL The potential for measurement errors due to reaction of the HQ with other components could
threaten the reliability of the data from HQ analyzers. The Bodenseewerk Model 677 uses an extractive sampling
system which is designed to eliminate errors caused by HQ reaction. Figure 4-2 is a diagram of the sampling
system of the Model 677.
4-2
-------
Flue gas is extracted from the stack through a coarse, sintered stainless steel filter which has a typical porosity
of 50 microns. The filter assembly is held in the stack by a 1/2" pipe that is connected to a 150 pound mounting
flange.
SAMPLE
INTERFACE
ENCLOSURE
SPECTRAN677
IR PHOTOMETER
I »EXAUST
COMP'RESSED
ZERO
QAS
FIGURE 4-2 MODEL 677 SAMPLING SYSTEM
. A heated sample interface enclosure is attached to the other side of the mounting flange. Within the enclosure
is a secondary fine filter with a porosity of 5 microns which is to keep smaller paniculate matter from migrating
into the system. Air, at approximately 80-100 psig, is piped through this enclosure and is used to backpurge the
filters in order to alleviate blockage. If paniculate matter were allowed to accumulate on the filters, the entering
HG could react with these deposits and cause the measurement errors already mentioned.
Because it has been shown that HG absorbs on sampling component surfaces of temperatures less man 185°C
(which not only results in measurement errors but also in metal degradation), the sample interface enclosure on
the Bodenseewerk's Model 677 is constantly maintained at 185°C. Since this temperature is above the sample's
dewpoint, HG losses which could occur by its reaction with the surface areas of the sample system are
consequently eliminated.
Next, the sample gas exits the interface enclosure and is transported to the analysis section via a heated
Teflon™ sample line which terminates at the inlet of a heated sample pump. This diaphragm pump, specially
designed on the Bodenseewerk's Model 677 for use at 185°C, maintains a sample flow rate of approximately 10
liters per minute. A heated flowmeter, which measures the flow of gas from the outlet of the sample pump, is
4-3
-------
equipped with an optical sensor that triggers an alaira if the sample flow rate falls below a predetermined limit. It
also automatically places the system in stand-by mode by shutting off the sample gas flow and purging the system
with ultra-pure, "zero" air.
Sample gas exits the flowmeter and enters the heated sample cell which uses infrared (IR) light to determine
the concentration of HQ in the sample gas. The sample cell, like the rest of the sampling system, is heated to 185°C.
Its interior is coated to resist corrosion and to all eviate the possibility of HQ reaction with the walls. After analysis,
and upon exiting the cell, the sample gas exhaust is allowed to cool to ambient temperature; the water vapor, which
is no longer above its dewpoint, will then condense and be removed by the condensate trap.
The gas filter correlation technique, as implemented by the Bodenseewerk Model 677 Hd monitor, is
schematically represented in Figure 4-3.
QAS
FILTER ,
ROTATING
IR
LIGHT
MEASURING
APERTURE
DETECTOR
FIGURE 4-3 GFC INSTRUMENTATION FOR BODENSEEWERK
MODEL 677 MONITOR
I
sSfl^^
An energy source emits a broad band of infrared light that passes through the sample cell which contains a
continuous flow of sample gas. The intensity of the IR light is reduced in the sample cell at the wavelengths
absorbed by the HQ molecules. The grcaterthe concentration of HQ in the sample cell, the greater the absorption.
The IR light that exits the sample cell is mechanically chopped by a rotating disk containing two apertures. One
of these apertures is actually the gas filter cell which is filled with a high concentration of HQ. When the gas filter
cell comes in line with the light beam, die energy corresponding to the wavelengths absorbed by the HQ molecules
is removed from the beam; the energy of the remaining light going to the detector is independent of the HQ
concentration and is used as a reference signal
The second aperture in the rotating disk is the measuring cell. This cell contains an inert gas, typically nitrogen,
which does not further reduce the energy of the IR beam. The beam passing through the measuring cell will only
have been attenuated while in the sample cell. The IR absorbance of HQ is measured by calculating the log of the
ratio of the signals from the filter cell and the measuring cell Please recall the appropriate discussion in Lesson
3 for further details about this process.
Though the Bodenseewerk Model 677 employs the gas filter correlation (GFC) technique, there are a number
of marketed CEMSs which use other spectroscopic methods for monitoring purposes. Among these other
4-4
-------
techniques is differential absorption spectroscopy. A commercially available S02 monitor that uses mis method
is discussed in the following section.
Differential Absorption Analyzer
Teledyne Analytical Instruments has developed the Model 691 for the continuous monitoring of S02 from
stacks, especially combustion sources. It uses ultraviolet (UV) light for the differential absorption method of
analysis. Figure 4-4 presents an internal view of the Model 691 monitoring system.
FIGURE 4-4 INTERNAL VIEW OF TELEDYNE
MODEL 691 SO2 MONITOR
The
The Model 691 is a fully integrated system which provides extractive sample handling and conditioning. ....
sample is drawn from the stack through a specially designed filter probe which removes entrained paniculate
matter. From the probe, the sample gas passes to the sample handling and conditioning system that effectively
controls the temperature and pressure of the sample gas to prevent water condensation and SO2 loss before and
during analysis. This system also includes a manual selector valve for the introduction of a span gas used for
calibration.
On Teledyne *s Model 691, the electronic zero mode (i.e., the time during which the analyzer's zero
concentration output is reset in order to provide "drift-free" readings) is automatically repeated once each hour
to provide long-term stability of the system. A three-way solenoid valve controls the flow of sample gas and zero
4-5
-------
air while in the zero mode. During the zero mode, zero air (ultra-pure grade air) is fed to the "common" port of
the valve. When the system introduces zero air, the output from the analyzer is automatically adjusted to zero.
Also during the zero mode, the zero gas flows back through the probe, providing a backflush to clear away
blockage due to particulates collecting on the filter. When the zero mode is completed, the valve closes and zero
air is replaced by the sample gas in the cell.
Using the differential absorption method of analysis, the Teledyne Model 691 employs two filters to select
the specific wavelengths to be monitored. The measuring wavelength, 289 nanometers, corresponds to the
absorption peak of S02 molecules while the reference wavelength corresponds to a region of the spectrum where
little or no light energy is absorbed by S02 or other interfering gas molecules.
Since the output of the analyzer is a ratio of the absorbances at the measuring and reference wavelengths,
extraneous interferences (which reduce the signals from both wavelengths equally) are effectively cancelled and,
thus, do not affect the accuracy of the S02 analysis.
Summary
Different systems have been developed which use spectroscopic analysis. The CEMSs examined in this
lesson are summarized in the following paragraphs.
The Bodenseewerk Model 677IR Hd Emissions Monitoring System uses infrared light for the measurement
of hydrogen chloride gas (HQ). This system uses an extractive design and the gas filter correlation (GFQ
technique which allow for the development of a reference signal that is independent of Hd concentration to obtain
accurate measurements of HQ gas emissions from sources such as waste incinerators. This system has also
incorporated some features which are important for Hd monitoring. In order to minimize Hd losses, the entire
sampling system is maintained at a temperature which is above the dew point of HCL A special backpurge
technique is incorporated to eliminate paniculate build-up on the filters which could adversely affect the Hd
measurements.
The Teledyne Analytical Instruments Model 691S02 Analyzer System uses ultraviolet light for the continu-
ous monitoring of S02 from stacks, especially combustion sources. The reference wavelength used in this device
provides a reference signal that is independent of SO2concentratioa The Model 691 extractive system possesses
several features which make accurate analyses possible. The temperature and pressure of the sampling system
are effectively controlled in order to prevent water condensation and S02 loss during analysis. Also, an electronic
zero mode is automatically repeated once each hour to provide long-term stability of the system.
4-6
-------
REVIEW EXERCISES
1. What type of sampling train does the Bodenseewerk Model 677
utilize?
a. Extractive.
b. In-situ
2. High pressure air is forced through the Bodenseewerk Model 677
secondary filter assembly in order to
a. force more HO sample thru the analyzer.
b. compress the sample, so there will be less to measure.
c. purge the filters, alleviating blockage.
d. keep the system from decompressing.
1. a
3. The diaphragm pump maintains a sample flow rate of about.
a. 10 gal/day
b. 1 gal/day
c. 10 liters/rain
d. 100 litersAnin
2. c
4. The Model 677 sample cell uses
tration.
a. IR light
b. UV light
c. an orifice meter
d. a dry gas meter
. to deter mine HC1 concen-
3. c
5. The absorption of IR light increases at.
a. lower
b. higher
. concentrations of HG.
4. a
6. Teledyne Analytical Instruments Model 691 uses to monitor
a. IR light, NO2
b. UV light, NO,
c. IR light, SO,
d. UV light, SO,
5. b
4-7
-------
7. The Teledyne Analytic Instruments Model 691 is a
a. gas filter correlation analyzer.
b. differential absorption analyzer.
6. d
8. The Model 691 employs.
lengths to be monitored.
a. 1
b. 2
c. 3
d. 4
. filters to select the specific wave-
7. b
9. Heat tracing in the TAI Model 691's sampling system prevents
and during analysis.
a. SO2 loss, energy waste
b. cracking, Hd loss
c. corrosion, HG loss
d. water condensation, S02 loss
8. b
10. Model 677 and Model 691 both use
concentration of .
a. spectrophotometers, gaseous molecules
b. spectrophotmeters, S02
c. light energy, Hd
d. light energy, NO2
to determine the
9. d
10. a
4-8
-------
REFERENCES
1. Kilkelly Environmental Associates. 1988. Continuous Emission Monitoring Guidelines: Update. Electric
Power Research Institute, Inc. Palo Alto, CA. Report CS-5998.
2. KVB, Inc. 1988. Hydrogen Chloride (HCl) Emissions Monitor (Advertising Brochure). HQ PP 1/9/88.
KVB, Inc., 9342 Jeronimo, Suite 101, Irvine, CA 92718 or P.O. Box 19518, Irvine, CA
92713.
3. Skoog, D. A. and West, D. M1982. Fundamentals of Analytical Chemistry. Saunders College Publishing,
Philadelphia, PA.
4. Teledyne Analytical Instruments. 1989. 697 Sulfur Dioxide in Stack Gas Analyzer System (Adverising
Brochure). Teledyne Analytical Instruments, 16830 Chestnut St, P.O. Box 1580, City of Industry. CA
91749-1580.
5. U.S.EnvironmentalProtectionAgency. 1986. Quality Assurance HandbookforAir PoUutionMeasurement
Systems, Volume III. Stationary Source Specific Methods. EPA-600/4-T7-027b.
4-9
-------
LESSON 5
Operation of a Chemiluminescence Analyzer
Lesson Goal and Objectives
Goal
To explain the operation of a Horiba Ammonia (NH,) monitoring system.
Objectives
At the end of this lesson, you should be able to —
1. describe the sample probe used for NH, monitoring,
2. explain the significance of NO to this particular technique,
3. explain the use of a pressure trap, and
4. identify the special features of this series of instruments.
Introduction
Lessons 2 and 3 provided a general overview of the detection procedures and calculation methods used for
gas analysis in continuous emissions monitoring systems (CEMSs). Lesson 4 gave more specific information
about the operation of marketed CEMSs which use spectroscopic absorption as their basis of gas analysis.
Similarly, this lesson will give more detailed information about the operation of a commercially available
monitoring system that is based on the principle of luminescence.
5-1
-------
Luminescence Analyzers
CEMSs which use luminescence analyzers rely on light emissions from excited molecules for monitoring
purposes. In this process, light energy converts the absorbing atom, molecule, or ion (M) to an excited form
(M*). Excitation refers to the temporary placement of electrons into energy higher levels (orbitals). As the
electrons return to theirnormal levels, excitation energy in the form of light is emitted. The amount of this energy
can then be related to the gas concentration.
Chemiluminescence Analyzers
Horiba Instruments, Incorporated has developed two stack gas analyzer systems for monitoring ammonia
(NH,). The ENHA-C900 Series is comprised of the ENHA-C900 and the ENHA-C901 models. The ENHA-
C900 is designed for the continuous monitoring of residual NH, in exhaust gas, while the ENHA-C901 also
offers the independent measurement of nitrogen oxides (NO,). These systems use the principle of Chemilumi-
nescence detection for gas analysis. Figure 5-1 gives an internal view of this series of monitors.
FIGURE 5-1 INTERNAL VIEW OF
ENHA-C6900 SERIES NHo MONITOR
FROM HORIBA
5-2
-------
The Horiba ENHA-C900 Series monitors use an extractive sampling system to draw flue gas from the stack.
The double-tube sample probe, with integral reduction catalyst, is designed for direct insertion into the stack. A
diagram of the probe and heated catalyzer section is shown in Figure 5-2. An optional blowback system can be
added to the probe assembly to prevent blockage caused by dust and paniculate matter.
CATALYZER
MH,
FIGURE 5-2 ENHA-C900 SERIES SAMPLE PROBE
From the probe assembly, the gas travels to the analyzer through sample lines which are maintained at
approximately 100- 120aCinordertopreventmoisturecondensation. A pressure trap is included to provide stable
control of the sample flow from the stack and through the system. The sample gas will pass through a mist catcher
that efficiently removes only SO,, thus minimizing constituent changes in the sample preconditioning unit
The NHj is reduced by the reduction catalyst as the sample gas passes through the inner tube of the probe, and
an amount of NO proportional to the volume of NH, is consumed as described by the following equation:
(Eq.S-1)
NO + NH, + 1/4 O,
N
Meanwhile, the sample gas passing through the outer tube of the probe undergoes no change. These two
samples travel through heat-traced lines to an N02-to-NO converter. They are then introduced alternately at 1
Hertz (Hz) into the chemiluminescence analyzer by the flow chopper. The analyzer determines the concentration
of NO in each sample.
Horiba Instruments notes that the NO content of the sample which passes through the inner tube is reduced
due to the aforementioned reduction reaction by an amount proportional to the NH, in the sample. The sample
in the-other line has the original content of NO in the flue gas. Accordingly, the difference between the two values
represents the NH, content in the gas from the stack. The integral microprocessor calculates the difference.
The techniques just described for the analysis of NH3 in exhaust gas streams are the same for the ENHA-C900
and the ENHA-C901. However, the ENHA-C901 incorporates twin chemiluminescence analyzers. One
5-3
-------
determines the NH, content and the other simultaneously and independently determines the NO, of the flue gas
stream by a similar procedure involving the flow chopping of the original sample and a zero gas.
Some of the special features of both models include the automatic calibration system and flow chopping
technique which are intended to provide "drift-free" readings. Furthermore, the ammonia monitoring procedures
are unaffected by co-existing S0t, NO,, or other substances.
Many of the functions are easily selected via the control panel. Power, gas species, range, calibration, and other
switches are provided. There is also a digital display and alarm lamps on the face of the panel. Figure 5-3 is an
illustration of the control panel.
Digital display
Display switch
Mode selection switch
AIC start/reset switches
Digit setting switches
AIC indication lamp
Alarm lamps — —
Range selection switch,
analyzer
Range selection switch,
NOX analyzer
Manual control
switches for
calibration service
Power switch
Maintenance switch
FIGURE 5-3 EHNA-C900 SERIES CONTROL PANEL
Summary
Horiba Instruments, Inc. has developed two stack gas analysis systems for monitoring ammonia (NH,) via
chenuluminescencedetectioaTheENHA-C900 is designed forthecontinuousmonitonBgofreddualNHjinexhaust
gas, while the ENHA-C901 also offers the independent measurement of nitrogen oxides (NO,). Both systems use
an extractive design and a double-tube sample probe, with integral reduction catalyst, for gas analysis.
The Niys reduced by the reduction catalyst as the sample gas passes through the inner tube of the probe, and
an amount of NO proportional to the volume of NH, is consumed. The sample gas passing through the outer tube
of the probe undergoes no change. Since the NO content of the inner tube has been reduced and the NO content of
5-4
-------
the outer tube is that of the stack gas, the difference between the two values represents the NHj content in the gas
from the stack.
The special features include heat-traced sample lines which are used to deliver the sample gas from the probe
while preventing moisture condensation. A pressure trap enables a constant sample flow rate to be maintained
more easily. Also, an optional blowback system can be added to the probe assembly to prevent blockage caused
by dust and paniculate matter.
5-5
-------
REVIEW EXERCISES
1. The electrons of excited molecules.
a. interfere with the monitoring equipment
b. stop moving
c. are temporarily moved to lower energy levels
d. are temporarily moved to higher energy levels
2. The ENHA C900 Series' sample lines are heated in order to
a. prevent leaks.
b. prevent dew formation.
c. condense NH, vapor.
d. condense NOX vapor
1. d
3. The mist catch efficiently removes
a. SOt
b. N0r
c.
d.
2. b
4. True or false. The sample probe cannot be inserted into the stack due to
its double tube construction.
3. a
5. True or false. The optional blowback system prevents blockage caused 4 False
by NO, and SO,.
6. True or false. The reduction catalyst reduces the NH, with NO and O2.
5. False
7. The ENHA-C901 measures the concentration of
a. NO, only.
b. NH,only.
c. NH3andNOI.
d. NH,andSO3.
6. True
5-6
-------
8. Luminescence analyzers relate the light energy emitted from excited 7. c
molecules to the .
a. gas temperature
b. gas concentration
c. gas pressure
d. gas opacity
9. True or false. The ammonia monitoring procedures are affected by SO,
8. b
and NOX.
10. True or false. A constant sample flow rate is maintained more easily due
to the pressure trap.
9. False
10. True
5-7
-------
REFERENCES
1. Horiba Instruments, Inc. 19^9. StackGas Analyzer System (Advertising Brochure). Bulletin: HRE-2354B.
Horiba Instruments, Inc., 1021 Duiyea Ave., Irvine, CA 92714-5583.
2. KflkeUy Environmental Associates. 1988. Continuous Emission Monitoring Guidelines: Update. Electric
Power Research Institute, Inc. Palo Alto, CA. Report CS-5998.
3. Skoog, D. A. and West, D. M. 1982. Fundamentals of Analytical Chemistry. Saundere College Publishing,
Philadelphia, PA.
4. U.S.EnvironmentalProtectionAgency. \9%6.QiuduyAssuranceHcmdbookforAirPollutionMeasurcrnent
Systems, Volume III. Stationary Source Specie Methods. EPA-600/4-77-027b.
5-8
-------
LESSON 6
Operation of an Electrocatalytic Analyzer
Lesson Goal and Objectives
Goal
To explain the operation of an Lear Siegler Measurement Controls Corporation (LSI) oxygen (02) monitoring
system.
Objectives
At the end of this lesson, you should be able to —
1. describe the LSI sample probe used for 02 monitoring,
2. explain how the fuel cell sensor measures the concentration of O2 in the flue gas, and
3. identify the special features of the LSI O2 monitoring instrument.
Introduction
Lessons 2 and 3 provided a general overview of the detection procedures and calculation methods used for
gas analysis in continuous emissions monitoring systems (CEMSs). Lessons 4 and 5 described the operation of
analytical systems which use spectroscopic and luminescence techniques. Similarly, this lesson will explain the
operation of a commercially available CEMS which is based on the principle of electrocatalysis.
Electroanalyzers
CEMSs that use electroanalytical equipment for measuring flue gas components use basic electrochemistry
to determine the concentrations of the target gases. Typical electrochemical cells contain electrodes which have
6-1
-------
been selected forthe specific gas in mind. As the chemical reaction within the cell proceeds, substances will either
gain electrons or lose electrons. The ions migrate through the medium and create an electrical current between
the electrodes. As a consequence of the reaction, the chemical potential across the electrodes will be different;
this difference is termed electromotive force (EMF) or voltage. These measurable quantities can be related to the
concentration of the subject gas.
Electrocatalytic Analyzers
Lear Siegler Measurement Controls Corporation (LSI) has designed and marketed the Dynatron™ 401
Oxygen Monitoring System for the measurement of O2 in stack gas from sources such as oil-, coal-, and gas-fired
boilers, packaged boilers, incinerators, furnaces, dryers, process heaters, kilns, and other industrial combustion
efficiency monitoring applications. This system uses the principle of electrocatalysis as the basis for 0%
measurement of 02 in flue gas. Figure 6-1 depicts the basic components of the Dynatron1* 401 Oxygen Moni-
toring System.
FIGURE 6-1 DYNATRON™401 OXYGEN
MONITORING SYSTEM BY LSI
LSI indicates that the in-situ design of the 401 System provides rapid response, analytical accuracy, and low
m aintenance. This model takes measurements at temperatures which are at or above the dewpoint of the flue gases,
so sample conditioning systems and their associated maintenance are not required.
The 2000A sample probe, used in the Dynatron™ 401 System, is available in a variety of lengths and can
withstand process temperatures of 760°C or, with the use of a high temperature bypass, up to 1093"C. Parts of the
probe can be replaced quickly and easily, thereby making field service possible. Figure 6-2 provides a more
detailed view of the probe tip.
6-2
-------
FIGURE fr-2 2000A PROBE WfTH ZRCONIUM OXIDE SENSOR
On LSI's Dynatron™401 System, the zirconium oxide cell is located at the probe tip and is encased in a
ceramic diffuser. A paniculate matter deflector protects the diffuser from erosion and creates a tangential flow
across the sides of the diffuser. Two airfoils, positioned at 120° angles to the deflector, divert heavy paniculate
matter, they also aid in channeling the gas sample to the sheltered downstream side of the diffuser. This gas flow
creates a "self-cleaning" effect which helps minimize paniculate matter build-up on the sides of the diffuser. LSI
can also provide a cell protector when the probe is to be used in especially dirty environments. These special
features enable this in-situ system to operate successfully in harsh conditions.
The fuel cell sensor on the Dynatron™401 System has porous, platinum electrodes. When the cell is heated
above 593°C, it is permeable to oxygen ions and becomes an oxygen ion-conducting solid electrolyte. Reference
gas (instrument air) on one side of the cell and sample gas on the other side of the cell create a pressure differential
which causes oxygen ions to travel from one side of the cell to the other. The resulting electronic imbalance
provides a voltage potential between the electrodes. The voltage is a function of the cell temperature and the partial
pressures (concentrations) of oxygen on each side of the cell.
The output voltage from the cell is related to the concentration of oxygen in the sample gas by the following
equation called the Nemst equation:
(Eg. 6-1)
where
EMF
R
T
F
c.
c, =
EMF = KR)
-------
for temperature and an analog meter can also be added. The analog meter provides an indication of percent oxygen
on a log-scale. The meter is available with or without adjustable high and low setpoints.
FIGURE 6-3 CONTROL UNIT
The 401 System can be equipped with a manual stack-mounted reference and calibration gas package. The
manual package facilitates interconnection to a source of reference and calibration gases. Manual zero and span
checks can be initiated at the stack.
Summary
LSI has designed and marketed me Dynatron1M401 Oxygen Monitoring System foruse in a variety of sources.
This system uses the principle of electrocatalysis and an in-situ design for monitoring O2 concentration.
The fuel cell-t>pe sensor in the tip of the probe has porous, platinum electrodes. When the cell is heated above
593°C, it is permeable to oxygen ions and becomes an oxygen ion-conducting solid electrolyte.
Reference gas (instrument air) on one side of the cell and sample gas on the other side of the cell create a
pressure differential which causes oxygen ions to travel from one side of the cell to the other. The resulting
electronic imbalance provides a voltage potential between the electrodes. The voltage is a function of the cell
temperature and the partial pressures (concentrations) of oxygen on each side of the cell. The output voltage from
the cell is converted to the concentration of oxygen in the sample gas by using a mathematical relationship called
the Nemst equation.
The primary features of the 401 System are:
1. the zirconium oxide sensor encased in a ceramic diffuser,
2. a paniculate deflector which protects the diffuser from erosion, and
3. two airfoils which aid in channeling the gas sample to the sheltered downstream side of the diffuser.
6-4
-------
REVIEW QUESTIONS
The LSI Dynatron 401 uses the principle of
basis for the measurement of O2 concentration.
a. opacity
b. absorbance
c. electrocatalysis
d. electrolysis
as the
2. The 401 system uses an,
a. extractive
b. in-situ
. design.
1. c
3. True or false. Sample conditioning systems are re-
quired with the 401 System.
2. b
4. True or false. The zirconium oxide cell is only utilized
in low temperature environments.
3. False
5. True or false. The zirconium oxide cell is encased in a
ceramic diffuser
4. False
6. The voltage in the fuel cell sensor is a functionof cell
and the of oxygen on each side of the cell.
a. pressure... absorbance
b. temperature... partial pressures
c. temperature... absorbance
d. shape... partial pressures
5. True
7. The fuel cell sensor is permeable to.
a. oxygen ions, 120°C
b. oxygen ions, 593°C
c. oxygen, 120°C
d. oxygen, 593°C
above
6. b
6-5
-------
8. The pressure differential in the fuel cell sensor
results in a .
a. high resistance
b. low temperature
c. zero setting
d. voltage potential
7. b
9. The voltage is related to oxygen concentration by the.
tion.
a. Reynolds
b. Nemst
c. Menton
d. Faraday
.equa-
8. d
10. The.
. protects the diffuser from erosion.
9. b
a. compact design
b. ceramic case
c. paniculate deflector
d. platinum electrode
10. c
6-6
-------
REFERENCES
1. Kilkelly Environmental Associates. 1988. Continuous Emission Monitoring Guidelines: Update.
Electric Power Research Institute, Inc. Palo Alto, CA. Report CS-5998.
2. Lear SieglerMeasurement Controls Corporation. 1989. Dynatron 401 In-situOxygenMonitoring
System (Advertising Brochure). DYNA401/PL/3/89. LSI, 74 Inverness Drive East, Englewood,
CO80112-5189.
3. Skoog, D. A. and West, D. M. 1982. Fundamentals of Analytical Chemistry. Saunders College
Publishing, Philadelphia, PA.
4. U. S. Environmental Protection Agency. 1986. Quality Assurance Handbook for Air Pollution
Measurement Systems, Volume III. Stationary Source Specific Methods. EPA-600/4-77-027b.
6-7
-------
LESSON 7
Operation of General Purpose Analyzers
Lesson Goal and Objectives
Goal
To explain the operation of Horiba General Purpose Gas Analyzers.
Objectives
At the end of this lesson, you should be able to —
1. identify the analytical techniques of and the gases monitored by these systems,
2. explain the two primary modes of analysis, and
3. describe the special features of these instruments.
Introduction
The priorlessons have presented material on the operation of several types of commercially available CEMSs.
Likewise, this lesson will discuss the operation of the CMA/CFA/MPA Series of general purpose gas analyzers
manufactured by Horiba. Each instrument in this line of equipment uses a different number and combination of
non-dispersive, infra-red spectroscopy (NDIR) analyzers and magnetopneumatic analyzers to monitor exhaust
gases.
Modes of Analysis
As Lesson 4 explained, CEMSs which use spectrophotometers rely on the interaction of light energy with
gaseous molecules in order to determine the gas concentration. Since light energy and wavelength are related,
analyzers can employ light from different parts of the spectrum formonitoring. When agas absorbs some energy
from a light beam, the intensity of the light will decrease. This decrease in intensity can then be converted to the
concentration of the gas. Because the monitors described in this lesson use a combination of analytical techniques
including spectroscopy, this topic will be briefly reviewed.
7-1
-------
CEMSs which use paramagnetic analyzers operate on the principle that molecules behave differently when
placed in magnetic fields. The behavior is classified as either diamagnetic or paramagnetic. Most molecules are
diamagnetic and are repelled when placed in a magnetic field. However, a few molecules possess paramagnetic
properties and will be attracted by and interact with the magnetic field; such is the case with oxygen.
General Purpose Gas Analyzers
Horiba Instruments, Inc. has developed a line of general purpose gas analyzers, the CM A/CFA/MPA Series,
for the measurement of NOS, SO2, CO, CO,, and O2 in exhaust gas streams. Each model requires the extraction
of sample gas from the stack and deb'very to the analysis unit CFA-311/321 analyzers use NDDR cross-flow
modulation to measure up to two of the above components, with the exception of Or The MPA-311 uses the
magnetopneumatic method of measuring oxygen content and is applicable only to O,. The CMA-321/331
combine one or two NDIR analyzers and a magnetopneumatic analyzer in a single unit for measurement of oxygen
and up to two other components.
The CMA/CFA/MPA Series of analyzers provides the opportunity to monitor several components of flue gas
with one extractive system. As the number of gases being monitored increases, the complexity of the analyzer
also increases. The CMA-331, which monitors two gases in addition to O2, is depicted in Figure 7-1.
D-pUy
FIGURE 7-1 CMA-331 MONITORING SYSTEM
7-2
-------
Spectrophotometry
NDIR cross-flow modulation analysis requires the sample gas and the reference gas to be introduced
alternately into the measurement cell by a rotary valve. When the cell contains the reference gas, the IR beam
reaches the detector without any energy loss. When the cell contains the sample gas, however, the energy received
by the detector is less due to some IR absorption by the target gas.
Contained within the detector is a movable membrane which senses pressure differential. Because of the
energy loss due to IR absorption, the pressure which corresponds to sample gas analysis will be less than the
pressure which corresponds to the reference gas. The displacement of the membrane is converted to an electrical
signal and is amplified for output This process is depicted in Figure 7-2. Horiba's cross-flow modulation method
is used for monitoring such gases as NOX, S02, CO, and C02.
Light Source
Amplifier
(Comp)
Sample Gas
FIGURE 7-2 NDIR CROSS-FLOW MODULATION PROCESS
Magnetopneumatics
Magnetopneumatic oxygen analysis requires O2 to be introduced to an uneven magnetic field, where the O2
is drawn to the stronger side of the field and causes a pressure rise. A pressure rise is also produced outside the
field using a non-magnetic gas (e.g.. nitrogen); the differential pressure (between the two gases) is measured using
a condenser microphone. The pressure difference is related to oxygen concentration according to the following
equation:
(Eq. 7-7)
where
AP =
H =
X =
C =
AP
(X)(O
difference in pressure
strength of magnetic field
magnetic susceptibility of 02 (known), and
concentration of 02
7-3
-------
A stable signal is produced and transmitted from the detector by exciting the magnet intermittently and
processing an alternating current (signal). There will be no signal if there is no 02 in the sample gas, and therefore,
no zero drift. Precise oxygen measurement is possible because the magnetic susceptibility of the gas is accurately
known. Figure 7-3 shows a basic set-up for a magnetopneumatic analyzer.
Outlet
t Electromagnet
Magnetopneumatic
Cell
Sample Gas
FIGURE 7-3 MAGNETOPNEUMATIC ANALYZER
Horiba CMA/GFA/MPA Series models are designed so that all routine adjustments, operations, and mainte-
nance work can be conducted from the front of the instrument which reduces time and labor requirements. Use of
a small sampling volume also reduces maintenance frequency.
Summary
Horiba Instruments, Inc. has developed a line of general purpose gas analyzers, the CMA/CFA/MPA Series,
for the measurement of NO,, SO2, CO, CO2, and 02 in exhaust gas streams. Each model requires the extraction of
sample gas from the stack and delivery to the analysis unit These models differ according to the number and types
of gases that can be simultaneously monitored.
NDIR spectroscopy cross-flow modulation analysis requires the sample gas and the reference gas to be
introduced alternately into the measurement cell by a rotary valve. When the cell contains the reference gas, the
TR beam reaches the detector without any energy loss. When the cell contains the sample gas, however, the energy
received by the detector is less due to some IR absorption by the target gas. The electrical signal from the detector
is amplified and sent to the readout as concentration.
Magnetopneumatic oxygen analysis requires 02to be introduced to an uneven magnetic field. As the O2 enters
the magnetic field, it will be drawn to the stronger side of the field and cause a pressure rise. A non-magnetic gas
produces a pressure rise outside the field, and the pressure differential from the two gases is measured using a
condenser microphone. It can then be related to 02 concentration which is presented on the readout
7-4
-------
UNITS
SYSTEM DESIGN
-------
REFERENCES
1. Horiba Instruments, Inc. 1989. General Purpose Gas Analyzers (Advertising Brochure).
Bulletin: HRE-2833B. Horiba Instruments, Inc., 1021 Duryea Ave., Irvine, CA 92714-5583.
2. Kilkelly Environmental Associates. 1988. Continuous Emission Monitoring Guidelines: Update.
Electric Power Research Institute, Inc. Palo Alto, CA. Report CS-5998.
3. Skoog, D. A. and West, DM. 1982. Fundamentals of Analytical Chemistry. Saunders College
Publishing, Philadelphia, PA.
4. U. S. Environmental Protection Agency. 1986. Quality Assurance Handbook for Air Pollution
Measurement Systems, Volume III. Stationary Source Specific Methods. EPA-600/4-77-027b.
7-8
-------
9. In the NDIR cross-flow modulation process, the detector contains a
moveable membrane which senses .
a. porosity
b. temperature changes
c. ion concentration
d. pressure diffential
8. False
10. In the Horiba magnetopneumatic oxygen analyzer, a is used
to measure a pressure differential which is related to oxygen con-
centration.
a. flowmeter
b. magnet
c. condenser microphone
d. zirconium oxide cell
9. d
10. c
7-7
-------
REVIEW EXERCISES
1. CEMSs which use spectrophotometers rely on the interaction
of with gaseous molecules to determine gas concentratioa
a. magnetic fields
b. light energy
c. particulates
d. filters
2. CEMSs which use operate on the principle that molecules
behave differently when placed in magnetic fields.
a. UV light
b. infrared light
c. spectrophotometers
d. paramagnetic analyzers
1. b
3. The Horiba general purpose gas analyzers use an.
a. extractive
b. in-situ
.design.
2. d
4. The MPA-311 uses the magnetopneumatic method to measure
a. CO,
b. Oa
c. CO
d. SO2
3. a
5. True or false. The magnetopneumatic analyzers are unaffected
by temperature fluctuations.
4. b
6. True or false. The use of a large sampling volume in the Horiba
Series reduces maintenance frequency.
5. True
7. True or false. The CFA-311/321 units are capable of measuring
more than one type of gas at a time.
6. False
8. True or false. NDIR cross flow modulation analysis requires
the sample gas and the reference gas to be introduced simul-
taneously into the measurement cell
7. True
7-6
-------
This line of monitors from Horiba offers several important features. The Horiba magnetopneumatic analyzers
provide linear output and respond only to oxygen. They are said to be free from zero drift and unaffected by
temperature fluctuations. The Horiba cross-flow modulation method, used in their NDIR analyzers formonitoring
NOX, S02, CO, and CO2, provides stable, long-term operation. Because some of the monitors in this series
combine both types of analyzers in a single CEMS, it is possible for the user to monitor several gases of interest
7-5
-------
LESSON 8
Extractive Systems Design
Lesson Goal and Objectives
Goal
To describe extractive continuous emissions monitoring systems (CEMSs) design in terms of monitoring
objectives as well as systems' components and their functions.
Objectives
At the end of this lesson, you should be able to —
1. define the objectives of an extractive CEMS as related to system requirements and user needs,
2. list the criteria to be evaluated prior to selection of the monitoring system, and
3. describe the primary subsystems of an extractive CEMS.
Introduction
As noted in Lesson 1, CEMSs can be divided into two general classifications based on the means by which
sample gas is delivered to the analyzer. Those two categories are extractive systems and in-situ systems. Extractive
systems withdraw flue gas from the stack, condition it, and transport the gas to the analysis section which is usually
located within the plant. In-situ CEMSs have at least some pan of their analysis subsystem mounted in the stack.
Because they perform analyses at the stack, in-situ systems neither condition the sample gas nor transport it to a
remote analytical unit
Extractive sampling permits the analyzer to be located at some convenient site within the plant The separation
of the probe and the analyzer does, however, impose a time lapse between extraction of the sample gas and its
analysis. For this reason, the output reading will always be somewhat behind the process.
Understanding all aspects of a system's design, including its objectives, components, and functions, will assist
in guiding the user through the selection and installation period. This knowledge will also be invaluable during
routine calibrations and maintenance.
8-1
-------
Objectives
Obtaining accurate and on-going information about stack emissions and other parameters is a primary goal
of continuous emissions monitoring projects and is dependent upon many variables. Therefore, it is imperative
to begin such a project correctly. Proper planning and investigation are essential for a successful program.
Knowing the requirements of the system and the needs of the potential users should be the first step in designing
a OEMS.
System Requirements
The requirements of a continuous emissions monitoring system can be considered two-fold. The two main
questions regarding the requirements are:
1. What are the intended functions of the system, or equivalently, what is the GEMS expected to do?
(e.g., process control, compliance demonstation, general research) and
2. What criteria must be met in order for the CEMS to execute these functions properly?
In response to the first question, extractive systems, in general, should be able to:
1. remove a representative gas sample from the source on a continuous basis,
2. maintain sample integrity during its transportation,
3. condition the sample so that it win be compatible with the selected analytical method,
4. accurately measure the concentration of the subject gas, and
5. allow for the accurate calibration and QA/QC of the system.
In response to the second question, of meeting the system's requirements, it is suggested that the user.
1. study the applicable regulations with respect to the gases and other parameters to be monitored;
2. review the specifications and operating characteristics of several analyzers capable of monitoring
the selected gases;
3. determine the gas stream parameters and the most feasible and representative sampling sites;
4. select the best site;
5. select the analyzer that is most compatible with the site and the gas parameters; and
6. ensure access to electricity, high pressure air, steam, water or other items required for proper
operation of the system's components.
8-2
-------
In an effort to meet the requirements of the system, representative sampling sites should be located as
described in the appropriate regulations. Among the parameters needed to chose the appropriate location are stack
gas temperature and velocity profile, as well as the distance from the proposed location to the nearest upstream
and downstream flow disturbances (including process and control equipment). It will also be necessary to
determine the amount of moisture in the stack gas in order to calculate the water removal and drainage needs of
the system.
Representative sampling also requires that existence of stratification within a stack be assessed. A more
detailed discussion of stratification is given Lesson 12.
Another physical factor to keep in mind is the system's operating environment The location of the equipment
should not limit or restrict its utility due to temperature extremes, support facilities, vibration, or other prohibitive
ambient conditions such as corrosion or participate buildup. A protective housing is a recommended way of
protecting the entire system.
User Needs
Satisfying the requirements of the hardware accomplishes only one of the main design objectives. Meeting
the needs of the user is also of great importance. There are several physical factors which should be evaluated.
One of these factors includes accessibility to the monitor and the interface for day-to-day usage, routine
maintenance, and QA/QC. Being able to reach controls and inspect equipment with ease is a must since these
activities will be ongoing. The potential loss of time due to inappropriate equipment orientation should be avoided.
Other user needs which should be addressed during the design phase of a continuous emissions monitoring
project include the overall system design and its response time. The overall system design refers to the way in
which the individual components connect and interact If an existing system is being expanded or if equipment
is being purchased from several vendors, it will be necessary to consider the separate functions of each piece,
whether it will be possible to join them in such a manner which ensures the delivery of a representative sample
to the analyzer, and whether results suitable for the application will be obtained reliably.
Consideration of the system's response time is also suggested. This parameter is especially important when
dealing with extractive CEMSs. If the operator were able to observe a discrete portion of sample gas as it moved
through the entire system, it would become quite evident that there is, in fact, a time delay between sample
extraction and analysis. This lag is attributed to the time required for the sample gas to travel through the tubing
as it is being conditioned, diluted, and transported to the analysis unit As the length of sample line increases, so
does the response time. This issue is most important when CEMSs are being used as part of a process control
scheme. Also, some regulations specify a maximum allowable response time.
From the few preceding paragraphs, the importance of carefully considering the system's and the user's needs
should be apparent Once these objectives are understood, developing a successful extractive system design may
proceed to the next step.
8-3
-------
Components and Functions
After clearly defining and understanding the objectives of the CEM project, the next step of the design phase
may begin. There are generally two approaches taken in extractive system design. One approach is to condition
the gas near the analyzer, the other is to condition the gas as near to the stack as possible.
In the first approach, a probe is inserted into the stack where it takes in flue gas which passes through a coarse
paniculate filter and into a heated sample line. This sample line, which is sometimes rather long (possibly over 100
feet), would enter some kind of control room or environmental enclosure where the gas would then be conditioned.
The conditioning system cools the gas and reduces the amount of water vapor. Reduction of the amount of water
vapor is usually accomplished by some type of refrigeration, dilution, or permeation device. A fine filter is placed
just before the entrance of the analyzer to prevent any small particulates from entering. Pumps or aspirators are used
to transport the sample from the probe to the analyzer.
In the second approach, sample gas is extracted through the probe and conditioned at the stack. Filters, chillers,
and/or dilution systems are located at the sampling site. Of these three devices, only the chillers are designed to
remove moisture from the sample gas. In the case of dilution systems, the concentration of all constituents in the
sample, including moisture, is greatly reduced by combining the sample stream with a stream of non-interfering,
dry gas. Each of these methods enables a low moisture sample to be delivered to the analyzer and avoids long
sections of heat-traced or insulated sample lines.
Regardless of which approach is used to condition the flue gas removed from the stack, all extractive CEMSs
have five common subsystems. Those subsystems are:
1. the effluent/monitor interface subsystem,
2. the sample transport subsystem,
3. the sample conditioning subsystem,
4. the analysis subsystem, and
5. the data acquisition subsystem.
The first subsystem (the effluent/monitor interface subsystem) consists primarily of the sample probe. The
second subsystem (the sample transport subsystem) is comprised of tubing and a pump. Moisture reduction devices
and more filters are found in the third subsystem (the sample conditioning subsystem). Calibration equipment,
analyzers, detectors, and convenors are pan of the fourth or analysis subsystem. The data recording and processing
equipment is in the fifth subsystem.
As indicated above, some extractive designs combine the sample transport and conditioning subsystems into
one overall sample handling subsystem. A schematic of a typical extractive CEMS is given in Figure 8-1.
The following list of equipment identifies the most common components found in extractive monitoring
systems. They are usually constructed from items such as valves, filters, tubing, tube fittings, and solenoids with
which plant maintenance personnel are familiar. Though individual manufacturers may deviate somewhat in their
systems' designs, most systems usually include:
8-4
-------
1. coarse (in-stack) filter,
2. sampling probe,
3. gas transport tubing,
4. sampling pump,
5. moisture removal system,
6. fine filter,
7. analyzer,
8. calibration system, and
9. data processor/recorder.
Heated
Transport
Line
Conditioning
System Analyzer
and Detector
' Stack Gas
FIGURE 8-1 TYPICAL EXTRACTIVE CEMS CONFIGURATION
Design considerations for these components should be made in light of their expected, predetermined
functions. Some suggested parameters for evaluation of the components within each subsystem are given in the
following sections.
Effluent/Monitor Interface Subsystem
The basic purpose of the interface is to provide continuous access to a representative portion of stack gas which
will be extracted and analyzed. It is therefore necessary to position the probe in a proper location and to make sure
that it is not adversely affected by the conditions of its operating environment That is to say that the probe should
be:
1. located to obtain a representative gas sample;
2. protected from a build-up of paniculate matter,
3. unaffected by temperature, moisture, and vibration;
' 4. capable of accepting calibration gases; and
8-5
-------
5. constructed of some corrosion resistant material such as Hastelloy", 316 stainless steel, glass, quartz,
or ceramic material in order to minimize deterioration and maximize its operating life.
Sample Transport Subsystem
The sample transport subsystem begins at the junction of the probe and the sample line and travels from the
stack to the analysis section of the monitoring system. The first primary component of this subsystem, the tubing,
has several aspects which should be considered. Some of the more important factors to evaluate when selecting
the type of tubing for a particular system include corrosion resistance, heat tolerance, and chemical resistance/
stability. These parameters deserve special attention due to the harsh environments typically encountered in
combustion and other sources and the nature of the gases being sampled.
The other main component of the sample transport subsystem is the pump. Though there are numerous pumps
marketed, it is generally recommended that diaphragm or bellows pumps be used in transport subsystems for the
reasons which follow. Those types of pumps:
1. provide adequate suction and discharge for stack gas analysis applications,
2. do not require a shaft seal (which means that they would not be prone to seal failure and subsequent
air in-leakage),
3. do not require internal lubrication (which could lead to sample contamination), and
4. are relatively inexpensive.
5. may also need to be heated and corrosion resistant
In some continuous emissions monitoring applications, the sample gas will leave the transport subsystem and
enter the conditioning system. Various aspects of that section will be discussed in the upcoming paragraphs.
Sample Conditioning Subsystem
The purpose of this subsystem is to provide a clean, dry gas sample to the analytical section. Sample
conditioning equipment often includes primary and secondary paniculate filters as well as some type of water
removal apparatus.
The primary filter is usually located at the probe inlet Since this filter is the first one to be contacted by stack
gas, it performs the initial screening of particulates and removes large particulates (greater than 50 microns in
diameter) from the extracted sample gas stream. The secondary paniculate filter, located downstream of the
primary filter, remove nearly all of the remaining particulates that are greater than one micron.
8-6
-------
Secondary filters (or fine filters) can generally be divided into two broad categories which are surface filters
and depth filters. Surface filters, as depicted in Figure 8-2, have a porous matrix which allows anything smaller
than 1 micron to pass while collecting the larger particulates on the "front" side of the filter.
FIGURE 8-2 SURFACE RLTER
As shown in Figure 8-3, depth filters are usually made of some type of loosely packed material, such as fibers
or glass wool, and have the ability to entrain the undesirable particulates while permitting the passage of gaseous
molecules.
FIGURE 8-3 DEPTH FILTER
Moisture removal is usually accomplished by condensation, permeation, or dilution. Condensation of the
moisture is accomplished by rapidly cooling the sample and trapping the condensed water in a collection vessel
which is periodically emptied. To avoid absorption of the target gases by the condensate through prolonged
contact, design precautions can be takea The first and most common design involves the standard compressor-
type refrigeration unit More recent approaches are the thermoelectric plate cooler, which is a solid-state unit with
no moving parts, and the vortex chiller which uses the chilling effect due to rapid expansion of a compressed gas.
Permeation dryers can be used with or in place of refrigerated condensers. The operation of permeation dryers
is based on the selective permeability of water through a membrane. As illustrated in Figure 8-4. permeation
occurs continuously as moist stack gas flows along one side of the membrane and dry purge air flows counter-
currently along the other side of the membrane. Since the water vapor will tend to migrate from a point of high
concentration to low concentration, the stack gas will become dryer as the purge air acquires some of that moisture
through the membrane.
The third conditioning method employed by some units is the addition of dilution air (usually cleaned and
dried ambient air) to the sample stream as a means of reducing the moisture concentration of the stream to a level
which is acceptable forthe particular analyzerthat is being used. This step may take place in the stack, at the sample
interface, or at a remote location. Dilution requires very accurate metering and monitoring of the air so that the
diluted sample concentration measured by the analyzer can be corrected to the actual stack concentration.
8-7
-------
In any case, the gases of interest, the detectionprinciple, and parameter of the individual analyzers will dictate
the complexity of the conditioning subsystem used to supply the clean, dry sample.
Dry
Purge
Air
H2O H2O
Permeable Membrane
FIGURE 8-4 PERMEATION DRYING PROCESS
Analysis Subsystem
The overall purpose of the analysis subsystem is to determine and to provide the accurate concentration value
of the target gas to the user. This goal is accomplished by the integration of the functions of several components
belonging to this subsystem.
The first aspect of the analysis subsystem to be examined is the calibration procedure. Calibration gases
are used to validate the performance of each component of the analysis subsystem and to correct the pollutant
concentration values as required. The calibration gases should be injected as close to the probe as possible, and
the calibration check should be performed at the normal operating conditions (i.e., temperature, pressure, flow
rate, etc.) encountered during the analysis of sample gases.
The other components of the analysis subsystem include analyzers, detectors/convenors, readouts, and
recorders. A variety of analyzers have been developed to monitor certain pollutants. The principles employed
by these analyzers include spectroscopic absorption, luminescence, electroanalysis, and paramagnetism. As these
analyzers measure selected parameters, a signal will be received by the detector, converted to an electrical current,
and sent to a readout instrument as a concentration value. Often, there will be some type of data recorder included
in this subsystem for the purpose of maintaining written documentation about stack conditions, concentration
readings, and other parameters of interest Discussions of these analytical techniques and devices have been
presented separately in prior lessons, so please refer to those sections for more details.
Summary
Lesson 8 defined the objectives of an extractive CEMS as related to system requirements and user needs.
Furthermore, it identified the four primary extractive CEMSs' subsystems as well as their components and
functions.
8-8
-------
When establishing a continuous emissions monitoring program, it is essential for the user to define clearly the
system requirements by asking the following questions:
1. What are the intended functions of the system, and
2. What criteria must be met in order for the OEMS to execute these functions properly?
The CEMS should also meet the various needs of the user in terms of:
1. accessibility tothemonitorandtheinterfaceforday-to-dayusage, routine maintenance, and QA/QC;
2. the overall system design which should allow for some future modifications or adjustments as
necessary; and
3. a reasonable response time.
After clearly defining and understanding the objectives of the CEM project, the next step of the design phase
may begin. There are generally two approaches taken in extractive system design. One approach is to condition
the gas near the analyzer, the other is to condition the gas as near to the stack as possible. Regardless of which
approach is used to condition the flue gas removed from the stack, all extractive CEMSs have five common
subsystems. Those subsystems are:
1. the effluent/monitor interface subsystem,
2. the sample transport subsystem,
3. the sample conditioning subsystem,
4. the analysis subsystem, and
5. the data acquisition subsystem.
The first subsystem consists primarily of the sample probe. The second subsystem is comprised of tubing
and a pump. Moisture reduction devices and more filters are found in the third subsystem. Calibration equipment,
analyzers, detectors, and convenors as well as readouts and recorders are part of the fourth subsystem. Data
recording and processing occur in the fifth subsystem.
Understanding these subjects will assist in guiding the user through the selection and installation period. This
knowledge will also be invaluable during routine calibrations and maintenance.
8-9
-------
REVIEW EXERCISES
1. Which of the following functions is not performed by an extractive
OEMS?
a. In-stack analysis.
b. Sample transportation.
c. Sample conditioning.
d. Paniculate filtering.
2. The separation of the probe and the analyzer
a. decreases analysis time.
b. increases sampling volume.
c. causes a time lapse between extraction and analysis.
d. decreases paniculate buildup.
1. a
3. Extractive systems should not
a. condition the sample prior to analysis.
b. operate continuously.
c. permit paniculate matter to enter the analyzer.
d. deliver a representative portion of gas to the analyzer.
2. c
4. It is necessary to determine the moisture content of the stack
gas in order to
a. calculate the water removal needs of the system.
b. assess the effect of stratification.
c. calibi e the system.
d. estims 3 the amount of water recycled per hour.
3. c
5. Moisture reduction is accomplished by all the following
except
a. permeatioa
b. dilution.
c. refrigeration.
d. heating.
4. a
6. The sample transport subsystem contains
a. an in-stack analyzer.
b. paniculate filters.
c. moisture detectors.
d. a diaphragm pump.
5. d
7. Hastelloy* or quartz might be used in the construction
of a sample probe in order to
a. filter particulates.
b. reduce moisture condensation.
c. calibrate the system.
d. resist corrosion.
6. d
8-10
-------
8. Surface and depth filters are used
a. to remove water.
b. to collect participates between 1 and 50 microns.
c. protect the probe from paniculate buildup.
d. clean the purge air.
7. d
9. True or false. System calibration should be performed under the
same conditions encountered during sample gas analysis.
8. b
10. Heat tolerance, chemical stability, and corrosion resistance should
be considered when selecting
a. which gas to monitor.
b. the analyzer to be used.
c. sample transport tubing.
d. a signal processor.
9. True
10. c
8-11
-------
REFERENCES
1. Elliot, T.C 1986. "Monitoring Pollution." Power. 130:S.1-S.9.
2. Jahnke.LA. and Aldina,G. J. 1979. Continuous Air Pollution Source Monitoring Systems. EPA 625/6-
79-005.
3. Kilkelly Environmental Associates. 1988. Continuous EmissionMonitoringGuidelines: Update. Electric
Power Research Institute, Inc. Palo Alto, CA. Report CS-5998.
4. U. S.EnvironraentalProtectionAgency. 1986. QualityAssuranceHandbookforAirPollution Measurement
Systems, Volume III. Stationary Source Specific Methods. EPA-600/4-77-027b.
8-12
-------
LESSON 9
In-situ Systems Design
Lesson Goal and Objectives
Goal
To describe in-situ continuous emissions monitoring systems (CEMSs) design in terms of monitoring
objectives as well as systems' components and their functions.
Objectives
At the end of this lesson, you should be able to —
1. define the objectives of an in-situ CEMS as related to system requirements and user needs,
2. list the criteria to be evaluated prior to selection of the monitoring system,
3. describe the two design approaches of in-situ systems, and
4. describe the two analytical techniques used by in-situ CEMS in terms of their components and
functions.
Introduction
As noted in Lesson 1, continuous emissions monitoring systems (CEMSs) can be divided into two general
classifications based on the means by which sample gas is delivered to the analyzer. Those two categories are
extractive systems and in-situ systems. Extractive systems withdraw flue gas from the stack, condition it, and
transport the gas to the analysis unit which is located within the plant In-situ monitoring systems, as the name
implies, are designed to carry out most of their operations in the stack.
Iri-situ CEMSs have at least some part of their analysis subsystem mounted in the stack in direct contact with
the flue gas. Because they perform analyses at the stack, in-situ systems neither condition the sample gas nor
transport it to a remote analytical unit. The lack of conditioning and transport subsystems indicates mat in-situ
CEMSs designs require substantially less equipment than do extractive CEMSs designs.
9-1
-------
Understanding all aspects of a system *s design, including its objectives, components, and functions, will assist
in guiding the user through the selection and installation period. This knowledge will also be invaluable during
routine calibrations and maintenance.
Objectives
As stated in Lesson 8, obtaining accurate and repeatable information about stack emissions is a primary goal
of continuous emissions monitoring projects and is dependent upon many variables. Therefore, it is imperative
to begin such a project correctly. Proper planning and investigation are essential for a successful program.
Knowing the requirements of the system and the needs of the potential users should be the first step in designing
aCEMS.
System Requirements
The requirements of a continuous emissions monitoring system can be considered two-fold. The two main
questions regarding the requirements are:
1. What are the intended functions of the system, and
2. What criteria must be met in order for the CEMS to execute these functions properly?
In response to the first question, in-situ systems, in general, should be able to:
1. operate at the required source conditions (e.g., temperature, pressure, etc.);
2. analyze a representative portion of flue gas on a continuous basis;
3. accurately measure source-level concentrations of the subject gas;
4. minimize possible interferences from particulates. water vapor, or "non-subject" gases; and
5. allow for the accurate calibration of the system.
In response to the second question, of meeting the system's requirements, it is suggested that the user
1. study the applicable regulations with respect to the gases to be monitored;
2. review the specifications and operating characteristics of several analyzers capable of
monitoring the selected gases;
3. determine the gas stream parameters and the most feasible and representative sampling sites;
4. select the best site;
5. select the analyzer that is most compatible with the site and the gas parameters; and
6. ensure access to electricity, high pressure air, steam, water or other items required for proper
operation of the system's components.
9-2
-------
In an effort to meet the requirements of the system, reperesentative sampling sites should be located as
described in the appropriate regulations. Among the parameters needed to chose the appropriate location are stack
gas temperature and velocity profile, as well as the distance from the proposed location to the nearest upstream
and downstream flow disturbances (including process and control equipment). Stratified zones, as discussed in
Lesson 8, should be assessed.
Another physical factor to keep in mind is the system's operating environment The potential interferences
of the environment, including temperature extremes, support facilities, corrosion, or paniculate buildup, should
not restrict the utility of the analyzer. The use of a measurement cavity which is enclosed by a filtration device
(e.g., a ceramic thimble) will allow gases to permeate while preventing interference from particulates. A high
pressure blowback system can be used to remove accumulated particulates from the surface of the ceramic thimble
or other filtration device.
User Needs
Satisfying the requirements of the monitoring system accomplishes only one of the main design objectives.
Meeting the needs of the user, which includes having access to the monitor and the interface for day-to-day usage
and routine maintenance, is also of great importance. Being able to reach controls and inspect equipment with
ease is a must since these activities will be ongoing. The potential loss of time due to inappropriate equipment
orientation or location should be avoided.
From the proceeding paragraph, the importance of carefully considering the system's and the user's needs
should be apparent Once these objectives are understood, developing a successful in-situ system design may
proceed to the next step.
Components and Functions
After clearly defining and understanding the objectives of the CEM project, the next step of the design phase
may begin. There are generally two approaches taken in in-situ system design. One approach is to monitor a
certain path of stack gas; the other is to monitor at a single point within the stack. EPA distinguishes between path
and point monitors by the percentage of the stack diameter (or equivalent diameter for non-circular ducts)
represented by the measurement path. An instrument that measures gas concentrations along a path which is
greater than 10 percent of the diameter is said to be a path analyzer. If the measurement path is less than or equal
to 10 percent of the diameter, the instrument is considered a point analyzer.
Regardless of the type of in-situ design that is to be implemented, calibration procedures are required.
Calibration gases are used to validate the performance of each component of the analysis subsystem and to correct
the pollutant concentration values as required. The calibration gases should be injected as close to the probe as
possible, and the calibration check should be performed at the normal operating conditions (i.e., temperature,
pressure, flow rate, etc.) encountered during the analysis of sample gases. When it is not feasible to inject
calibration gases, it is possible to use gas filled cells for calibration checks. These techniques are discussed further
in Lesson 10.
9-3
-------
In-situ, Path Monitors
In-situ, path monitors are usually made up of two units which are aligned opposite of each other in the stack.
This system analyzes flue gas that passes through the specific 'line of sight" between the units. This line of sight,
or path, typically ranges from a few feet to the entire width of the stack. For accurate readings, it is critical to ensure
proper alignment of these units. In-situ, path monitors use spectroscopic absorption techniques in which either
infrared (IR) or ultraviolet (UV) light is directed through the flue gas. As discussed in Lessons 2 and 3, the percent
absorption of light energy at a specific wavelength quantifies the amount of target gas present in the stack.
Path monitors may be designed to direct the light through the gas once or twice. "Once-through" systems are
named single-pass monitors, whereas "twice-through" systems are identifed as double-pass monitors.
Insingle-passmonitors,atransmitterandareceiver comprise the stackunits as shownmF/gure9-l. The transmitter
contains an IR or UV light source which beams the light directly across the stack to the receiving unit The receiver
detects the amount of light energy present in the beam and converts that signal to a concentration reading. The
transmitter and receiver are protected by windows, over which is blown a constant flow of purge air (generally
filtered ambient air).
Signal Processor/
Readout
Transmitter
RGURE 9-1 SINGLE-PASS, IN-SfTU, PATH MONfTOR
The purge air.
1. prevents the build-up of paniculate matter on the windows,
2. aids in cooling the units mounted on the stack, and
3. assists in preventing the condensation of water or corrosive materials on the cooler instrument
windows.
Double-pass monitors consist of a transceiver and a retroreflector. Light is sent out of the transceiver and
through the flue gas. Upon reaching the retroreflector unit on the opposite end of the path, the light is returned
through the flue gas and enters the transceiver for analysis. Since the light travels through the flue gas twice, the
9-4
-------
beam is considerably weaker than the one in single-pass systems. Therefore, most commercial path monitors are
designed as single-pass units. The double-pass systems are protected by windows and employ purge air for the
same reasons as given for single-pass units. Figure 9-2 is an illustration of a double-pass monitor.
Retro-
reflector
Signal Processor/
Readout
Stack
RGURE 9-2 DOUBLE-PASS, IN-SfTU, PATH MONITOR
Path monitors require a number of accessories to protect them from the hostile environments often
encountered at their installation sites. Transmitters, retroreflectors, and receivers need hoods or covers to protect
them from meteorological phenomena such as wind, rain, hail, and lightning. Anti-vibration equipment may be
needed to prevent optical and electrical components from shaking loose. A dedicated, constant voltage
transformer can be useful in eliminating voltage transients which could upset the sensitive electronics of many
instruments.
Iri-situ, Point Monitors
Like path monitors, in-situ, point monitors can use absorption spectroscopy for flue gas analysis. Point
monitors are constructed in much the same way as path monitors. Such point monitors are made up of two units
which are aligned opposite of each other in the stack The light is emitted from the first unit and sent through the
flue gas. Upon reaching the receiving unit, the amount of light energy is measured and the gas concentration is
related to absorbance. A ceramic thimble encases the probe and allows gaseous molecules to enter its shell while
retaining particulates in order to minimize interference with analysis. Though the ceramic thimbles used in point
monitors do not require blowers for cleansing purposes, they should be checked and replaced periodically to ensure
proper operation.
Another type of point monitor based on absorption spectroscopy utilizes an open probe containing a
retroreflector as seen in Figure 9-3. In this device, the transceiver is placed behind a window and the retroreflector
9-5
-------
is supported on the other end of the monitoring path. Since there is no ceramic thimble or other filtration device
included on the probe, blowers supply purge air which:
1. prevents the build-up of paniculate matter on the windows,
2. aids in cooling the units mounted on the stack, and
3. assists in preventing the condensation of water or corrosive materials on the cooler instrument
windows.
Readout
Transceiver
Stack
FIGURE 9-3 IN-Smj, POINT MONITOR
One widely marketed point monitor uses electroanalysis for the measurement of O2 and SO2. Such point
monitors have an electrocatalytic cell which is mounted on the probe and is covered by a ceramic thimble. As in
other applications, the ceramic thimble allows gaseous molecules to enter the shell, but it retains particulates so
that a clean sample may be introduced to the electrocatalytic cell. A thin film, applied to the solid electrolyte's
surface, catalyzes a reaction which allows gaseous molecules to migrate through the solid and generate a
measurable current More detailed discussions about this type of analyzer have already been presented in Lessons
2 and 3.
Point monitors require protective accessories similar to those used with path monitors. Transceivers and
retroreflectors should be protected from the elements. Anti-vibration equipment may be needed to prevent optical
and electrical components from shaking loose or becoming detached. A dedicated, constant voltage transformer
can be useful in eliminating voltage transients which could upset the sensitive electronics of many instruments.
9-6
-------
Summary
Lesson 9 defined the objectives of an in-situ OEMS as related to system requirements and user needs. It
described the two design approaches of in-situ systems, and finally, it described the two analytical techniques used
by in-situ GEMS in terms of their components and functions.
When establishing a continuous emissions monitoring program, it is essential for the user to define clearly the
system requirements by asking the following questions:
1. What are the intended functions of the system, and
2. What criteria must be met in order for the GEMS to execute these functions properly?
Additionally, the CEMS should meet the needs of the user in terms of accessibility to the monitor and the
interface for day-to-day usage and routine maintenance. Their location and orientation should be established in
a manner which does not restrict personnel from accomplishing these and other tasks.
After clearly defining and understanding the objectives of the CEM project, the next step of the design phase
may begin. There are generally two approaches taken in in-situ system design. One approach is to monitor a
certain path of stack gas; the other is to monitor at a single point within the stack.
In-situ, path monitors are usually made up of two units which are aligned opposite of each other in the stack.
This system analyzes flue gas that passes through the specific "line of sight" between the units. This line of sight,
or path, typically ranges from a few feet to the entire width of the stack. In-situ, path monitors use spectroscopic
absorption techniques in which either infrared (TR) or ultraviolet (UV) light is directed through the flue gas. Path
monitors may be designed to direct the light through the gas once or twice. "Once-through" systems are named
single-pass monitors, whereas "twice-through" systems are identifed as double-pass monitors.
Like path monitors, in-situ, point monitors can also use absorption spectroscopy for flue gas analysis. These
point monitors are constructed in much the same way as path monitors. They can be made up of two units which
are aligned opposite of each other in the stack, yet are closer than the units used by path monitors. Another type
of in-situ analyzer uses an electrocatalytic cell which is mounted on the probe and covered by a ceramic thimble.
Understanding all the design aspects just reviewed, including the objectives, components, and functions, will
assist in guiding the user through the selection and installation period. This knowledge will also be invaluable
during routine calibrations and maintenance.
9-7
-------
REVIEW EXERCISES
1. True or false. In-situ CEMSs should always be found at some
conveniently located site within the laboratory.
2. In-situ CEMSs require equipment for a single sample point
compared to extractive CEMSs.
a. more
b. less
c. the same amount of
1. False
3. In-situ systems do all of the following tasks except
a. minimize interference from particulates.
b. operate in high temperature combustion sources.
c. analyze a representative portion of stack gas.
d. remove moisture from stack gas.
2.b
4. A allows gases to permeate and prevents particulates from
entering the in-situ analyzer.
a. ceramic thimble
b. retrorefiector
c. high pressure blowback system
d. stratified zone
3.d
5. "Line of sight" refers to the path between the.
a. probe... stack wall
b. transceiver... retroreflector
c. window... mirror
and
4. a
6. A constant flow of purge air prevents and.
a. paniculate buildup... sample loss
b. vibration... water condensation
c. sample loss... corrosion
d. water condensation... paniculate buildup
5.b
7. Anti-vibration equipment is used to help maintain
a. optical alignment of the transmitter and receiver.
b. sample integrity.
c.' a constant flow rate of stack gas.
d. a constant pressure
6.d
9-8
-------
8. Thetwomostcommonmethodsofanalysisusedbyin-situCEMSsare 7. a
a. luminescence and paramagnetism.
b. spectroscopic absorption and luminescence.
c. electroanalysis and spectroscopic absorption.
d. magnetodynamics and mass spectroscopy
9. True or false. In-situ monitors which measure across a distance less
than 10% of the stack diameter are classified as path monitors.
8. c
10. True or false. Calibration gases should be injected into the sample
transport tubing
9. False
10. False
9-9
-------
REFERENCES
1. Elliot, T.C. 1986. "Monitoring Pollution." Power. 130:5.1-5.9.
2. Jahnke, J.A. and Aldina, G. J. 1979. Continuous Air Pollution Source Monitoring Systems. EPA 625/6-
79-005.
3. Kilkelly Environmental Associates. 1988. Continuous EmissionMonitoring Guidelines: Update. Electric
Power Research Institute, Inc. Palo Alto, CA. Report CS-5998.
4. U. S. Environmental Protection Agency. 1986. Quality Assurance Handbook for Air Pollution
9-10
-------
LESSON 10
Applications of Systems
Lesson Goal and Objectives
Goal
To discuss the applications of continuous emissions monitoring systems (CEMSs)
in terms of their uses, limitations, and other considerations.
Objectives
At the end of this lesson, you should be able to —
1. name several reasons for the application of a continuous emissions monitoring program
at a facility,
2. discuss the use of a data acquisition system (DAS),
3. list some considerations which should be made before the installation of either extractive or in-situ
CEMSs,
4. identify the capabilities and limitations of both kinds of designs, and
5. identify some parameters which characterize a successful monitoring program.
Introduction
Lessons 8 and 9 presented the objectives and requirements for the design of both extractive and in-situ CEMSs;
however, applications of these systems were not discussed in any detail. Lesson 10 will describe the application
of CEMSs in terms of theiruses, limitations, and other considerations. The evaluation and comprehension of those
topics should lead to the implementation of a successful continuous emissions monitoring program.
10-1
-------
Benefits of Monitoring
For the Plant
Though continuous emissions monitoring systems are mandatory for certain facilities subject to various
federal regulations (e.g., New Source Performance Standards (NSPS)), their installation can prove to be beneficial
to other facilities as well In addition to meeting regulatory requirements, properly installed and operating CEMSs
will provide large amounts of source emissions data. The data include far more than just the quantity of a pollutant
which is being emitted to the atmosphere; the data also include parameters which can be used to:
1. assess source operating efficiency;
2. correlate emissions with process variables (e.g., fuel consumption, amount of excess air for
combustion, raw material loss, etc.); and
3. evaluate the maintenance needs of control equipment
Having the ability to examine various aspects of the process can provide the user the opportunity to make
critical adjustments for process optimization and cost reductioa Consequently, it can be seen that facilities can
benefit from the installation of a CEMS.
For the Regulatory Agency
The advantages of operating a CEMS are not limited to a plant however. Regulatory agencies are also
benefited by the implementation of these programs. The data provided by a CEMS can be used to:
1. record potential non-compliance situations,
2. relate the effects of source emissions to ambient air quality,
3. assess regulatory needs, and
4. identify source operating and maintenance (O&M) inadequacies.
Having access to continuous emissions monitoring data provides the agency with a better picture of actual
emissions and enhances their ability to develop strategies for future programs directed at emissions reductioa
Accessing the Data
The data acquisition subsystem (DAS) is an integral part of every monitoring system. The manner by which
the pollution concentration data are output may vary widely from one system to another, depending upon
applicable regulations and intended use of that information. Traditionally, data from CEMSs were kept on strip
10-2
-------
chart recorders and/or operator log sheets. These methods required extensive manual data reduction to prepare
the emission reports required by the agencies. With the availability of computers, these activities have
consequently become more automated.
In general, a DAS receives an input from the detector that is proportional to the target gas concentration. It
then executes some combination of the following functions:
1. performs zero or span drift corrections,
2. calculates the emission rate for the target gas in units of the applicable standard,
3. performs averaging based on the applicable emission standard,
4. activates alarms when emissions exceed a predetermined level or when a
system failure occurs,
5. enables data editing,
6. provides long-term data storage, and
7. prepares .emissions data for reports in a spreadsheet or other format
As the number of functions and level of complexity increase, the ability of the DAS to perform adequately
could tend to decrease. Therefore, selection of hardware and software for the DAS should be done carefully, and
allowances for backup and future expansion or demands should be considered.
Use of a modem, computerized DAS can effectively reduce the amount of time a CEMS user would have to
dedicate for manual data collection and report preparation. Additionally, the sophistication of these systems
usually permits data acquisition and transfer for use in other investigations (e.g., dispersion modeling, risk
assessment, etc.). A well-designed DAS effectively complements the other components of the overall monitoring
system.
Considerations for System Implementation
In General
As discussed in Lessons 8 and 9, there are numerous factors which should be evaluated before the installation
of a new CEMS. The selection of a representative sampling site is crucial and should be confirmed before the
installation of the system. Consideration of temperature, paniculate build-up, corrosion, vibration, electrical and
other needs must be thoroughly examined. Control equipment layout and internal ductwork need to be reviewed
closely forpossible impacts on stratification. If the site should be deemed unacceptable after operation has begun,
the process of moving support platforms, test ports, sample lines, signal cable, and other equipment would be
enormously inconvenient to say the least
10-3
-------
Cost
Comparing the capabilities and limitations of both extractive and in-situ CEMSs is another consideration to be
made in conjunction with system implementation. Due to the inherent differences of both types of systems, each
will have its relative strengths and weaknesses. While certain strengths can be advantageous, there can also be
significant economic differences between the two generic types. When only one system is required at a facility, the
initial cost of an in-situ system may be less than a single point extractive system. However, extractive systems allow
for expansion via multiplexing (time-sharing among several sampling locations) which can lead to cost reduction
if monitoring requirements increase in the future.
Calibration Techniques
Cost is by no means the only basis of comparison for the two types of CEMSs. Calibration methods can also
describe important differences among the various types of systems. Extractive systems can be calibrated directly
with gases of a known concentration (cylinder gases), thereby providing a full check of the entire analytical
subsystem. In-situ, point monitors can also be calibrated directly with cylinder gases without much difficulty.
However, the difficulty in calibration generally occurs with in-situ, path monitors. Because the measuring path of
this monitor occupies a significant portion of the stack, introduction of a calibration gas to the path is not feasible
without modifications to the system. Such a modification might be the use of a movable pipe as an enclosure which
could temporarily isolate the measuring path from stack gas and allow that chamber to be flooded with calibration
gas.
With regard to any monitoring systems that use gas-filled cells for calibration, it should be noted that significant
changes in concentration values over time have been observed. Changes in the pressure of the cells during
transportation have also resulted in apparent changes in concentration. Use of this calibration technique should be
done with care. These two problems could possibly be avoided by the use of flow-through cells.
Stack Conditions
Since in-situ CEMSs analyze the subject gas at stack conditions, compensating for high levels of moisture or
paniculate matter can possibly pose problems. Futhermore, when exhaust temperatures are above 260°C, changes
can occur in the spectra of the flue gas, and interferences are likely to increase. Because extractive CEMSs condition
the sample gas prior to analysis, flue gas conditions are less critical. However, extractive systems require additional
routine maintenance of the conditioning subsystem in order to provide this advantage.
Sampling Location
Where sampling locations can be subject to stratification, in-situ, path monitors may possess an advantage when
compared to in-situ, point monitors and extractive systems since the latter two sample from a single point within
the stack. Even though in-situ, path monitors can also be affected by stratification, the error may not be as great
because they linearly average the concentration of the subject gas across the measuring path. Additionally, locating
a representative path may be easier than the location of a representative point
10-4
-------
If the desired sampling location is easily accessible, in-situ CEMSs can offer the ability of having all
components at that one location. However, the installation of all equipment at the stack can impose the use of a
protective housing which can slow the normal maintenance activities. Furthermore, the hostile environments at
some sampling locations can create more difficulties for maintenance personnel
The advantages and disadvantages of each of these general categories of CEMSs should be carefully considered
since they are important for a successful program. A summary of this comparison is given in Table 10-1.
TABLE 10-1
Advantages and Disadvantages
of
Both Extractive and In-situ CEMS
KX'IHACTTVE CEMS
Hal more cgmpjwnu and a '
' flioiT complicated
^mechanical design
Performs multiple point
sampling via time-sharing
Expands easily for addiiontl
sampling points
Has higher initial eon (for
: single point sampling)
Has time-lag during analysis
Analysis is performed m a
protected environment
Is relatively «m»««iM» for
service •n^ mantfmanoe
IN-STTU PATH CEMS
Has fewer components
Analyzes only one gas per
unit .
Does not allow for
-'f lyngflsifln
Offers lower initial cost
(for single point sampling)
Provides rapid response
during analysis
Operates ID a potentially
Ainti co viroomen!
MaybereUtively
•od p*M*ti*ifti>n«»>
IN-STTU POINT CEMS
Has fewer components
Analyzes only one gas per ,tmit
Does not allow for expansion
Offers lower initial cost (for
single point sampling)
Provides rapid response duiug
analysis
Operates in • potentially harsh
:CBVaTQODlCBl
•Maybe tehtiycly inaooenible
:Joc vcfvicc tt4 ^^^"""P*^
In general, there are a number of parameters that characterize a successful continuous emissions monitoring
program. Such programs have these similarities:
1. management backing, which should provide adequate funds as well as enough personnel to support
the project;
2. planning and site-specific engineering by the vendor and/or consultants, to avoid design pitfalls
and ensure its overall reliability;
3. good plant/agency relations, to promote communication and an understanding of the needs of both
parties; and
4. singular, designated responsibility for all aspects of the CEMS.
10-5
-------
Summary
Lesson 10 described the application of CEMSs in terms of their uses, limitations, and other considerations.
The evaluation and comprehension of these topics should lead to the implementation of a successful continuous
emissions monitoring program.
In addition to meeting regulatory requirements, properly installed and operating CEMSs will provide large
amounts of source emissions data to the facility. Having the ability to examine various aspects of the source can
provide the user the opportunity to make critical adjustments for process optimization and cost reduction.
The advantages of operating a CEMS are not limited to a plant however. Regulatory agencies are also
benefited by the implementation of these programs. Having access to continuous emissions monitoring data
provides the agency with a better picture of actual emissions and enhances their ability to develop strategies for
future programs directed at emissions reduction.
The data acquisition subsystem (DAS) is an integral pan of every monitoring system. Use of a modem,
computerized DAS can effectively reduce the amount of time a CEMS user would have to dedicate for manual
data collection and report preparation. Additionally, the sophistication of these systems often permits data transfer
for use in other investigations and applications. A well-designed DAS effectively complements the other
components of the overall monitoring system.
Other factors related to systems' applications, such as cost, calibration techniques, stack conditions, and
sampling locations need to be addressed.
Finally, successful continuous emissions monitoring programs often have the following similar characteris-
tics:
1. management backing,
2. planning and site-specific engineering by the vendor and/or consultants,
3. good plant/agency relations, and
4. singular, designated responsibility for all aspects of the CEMS.
10-6
-------
REVIEW EXERCISES
1. True or false. Monitoring systems are very limited in the
services provided since they can only record non-
compliance situations.
2. CEMSs data has generally been collected on
a. tape recorders and strip chart recorders.
b. logarithmic paper.
c. computers and strip chart recorders.
d. graphs.
1. False
3. A data acquisition system (DAS) can do all of the following
except
a. record data.
b. analyze the target gas.
c. edit data.
d. correct for span drift.
2. c
4. True or false. A useful application of CEMS data is to identify
O&M inadequacies.
3. b
5. The initial cost of an extractive CEMS is the initial cost
of an in-situ CEMS when monitoring at a single location.
a. about the same as
b. less than
c. greater man
4. True
6. Time-sharing can be accomplished in.
a. extractive
b. in-situ
c. both a and b.
d. neither a nor b.
. designs.
5. c
7. True or false. Gas filled, calibration cells are completely
reliable for performing routine calibrations.
6. a
10-7
-------
8. Flue gas conditions are more easily compensated for by the
application of a design.
7. False
a.
b.
extractive
in-situ
9.
. monitors linearly average pollutant concentration
within the stack.
a. No
b. In-situ, path
c. In-situ, point
d. Extractive
8. a
10. True or false. An increasingly common application of
CEMS data is to assess source operating efficiency.
9. b
10. True
10-8
-------
REFERENCES
I. Elliot, T. C 1986. "MonitoringPollution." Power. 130:S.1-S.9.
2. Jahnke,J.A.andAldina,G.J. 1979. ContinuousAirPollutionSourceMonitoringSystems.EPA625/6-79-
005.
3. Kilkelly Environmental Associates. 1988. Continuous Emission Monitoring Guidelines: Update. Electric
Power Research Institute, Inc. Palo Alto, CA. Report CS-5998.
4. Turner J., and IrwinB. 1986. "Factors in the Selection of an Extractive Continuous Emissions Monitor for
a Paper Mill Boiler." Tappi Journal 92-97.
5. U. S.EnvironmentalProtectionAgency. 1986. QualityAssuranceHandbookforAirPollutionMeasurement
Systems, Volume ffl. Stationary Source Specific Methods. EPA-600/4-77-027b.
10-9
-------
UNIT 4
REGULATIONS
-------
LESSON 11
Regulatory Requirements
for
Gas Emission Monitoring Systems
Lesson Goal and Objectives
Goal
To provide an overview of the various regulatory requirements involving the use of continuous gas
emission monitoring systems.
Objectives
At the end of this lesson you should be able to -
1. list at least five source categories required to install continuous gas emission monitoring
systems (for both new and existing sources),
2. distinguish the difference between "direct Emission" compliance monitoring data and "indicator
of compliance" (excess emissions) monitoring data,
3. list the regulatory procedures used by state and local agencies to ensure compliance with
air quality standards,
4. describe how to use the Federal Register index and the LSA index in keeping track of newly
proposed regulations or changes in existing regulations and
5. describe the use of a performance specification in the continuous gas emission monitoring
program.
11-1
-------
Introduction
So far, we have discussed the technical aspects of how continuous gas emission monitoring systems are
designed and how they work. It is also important for the CEM user to understand how these systems relate to
local, state and federal regulations. These regulatory requirements stipulate who is required to install
continuous emission monitors, the specific type of monitor which can be used for a specific source category,
and the performance test criteria and quality assurance procedures necessary to ensure that the monitoring
system is reliable and accurate. These next three lessons deal with these important issues.
The Reasons for Monitoring Gas Emissions Continuously
The Clean Air Act as amended in 1977 (42 U.S.C. 1857 et seq.) requires that several types of regulations
be established by the U.S. Environmental Protection Agency (EPA) [e.g.. New Source Performance Standards
(NSPS) and National Emission Standards for Hazardous Air Pollutants (NESHAPS)]. In addition, the Act
requires that the National Ambient Air Quality Standards (NAAQS) be established and that states adopt
provisions for seeing that these ambient air quality standards are met In order to comply with state and federal
regulations promulgated pursuant to the Clean Air Act. many industries have installed air pollutant reduction
devices such as wet scrubbers, baghouses, electrostatic precipitators and carbon absorbers. Such equipment is
expensive to buy and install and must undergo a rigorous maintenance schedule in order to ensure proper
continuous operation. State air pollution control agencies must also ensure that sources are in compliance with
applicable air regulations. Both maintenance of control equipment and compliance with applicable regulations
can be enhanced through the monitoring of air pollutant emissions.
Monitoring air pollutant emissions can be done either continuously or by the use of EPA manual methods
which are performed on a periodic basis. Using the EPA manual methods creates problems in that they are done
infrequently as part of inspection programs and usually while the source in question is being operated under
optimum conditions. Normally, these results do not reveal the true day-to-day concentrations of gaseous
pollutants released from a given source. Continuous monitoring on the other hand, provides a continuous record
of gaseous emissions. Such records can be used by both the state and the source in the following ways:
1. as an indicator that the source is using good operating and maintenance practices on its process and
its control equipment in order to minimize gaseous air emissions.
2. as a continuous record of the source's ability to comply with gas emission standards,
3. as sufficient data to issue a Notice of Violation (NO V) in case the source is not complying with agency
regulations or standards, and
4. as data on upset conditions or trend data indicating degradation of control equipment performance.
For regulatory purposes, there are two types of emission monitoring data. They are direct emissions
compliance monitoring data and indicators of compliance (excess emission) monitoring data. Direct emissions
compliance monitoring data (e.g., when continuous emission monitors (CEMS) are the state-approved
11 -2
-------
compliance method) are used directly to determine an emissions limit violation requiring appropriate
enforcement action which could include issuing penalties to a given source offender. The excess emission
monitoring data are used whenever a continuous emission monitor is not the state or EPA region's approved
emissions compliance method. In this case, the data should be used to monitor the compliance status of a
source and to initiate follow-up action which includes on-site inspections, requesting further information,
and issuing a Notice of Violation. The use of continuous emission monitor (CEM) data for determining
compliance and enforcement is further discussed in the EPA document, "Guidance: Enforcement Applications
of Continuous Emission Monitoring Data", available through the EPA Office of Air Quality Planning and
Standards (O AQPS) and OECM (Office of Enforcement and Compliance Monitoring) dated April 22,1986.
In each instance where CEMS have been promulgated or approved by the regulatory agency as an
official method to determine source compliance with the applicable emission limitations, the regulatory
agency canrelyupon CEM data when making compliance determinations. Continuous Emission Monitoring
Systems (CEMS) have been specifically prescribed as the method to establish emission violations for one
or more pollutants in the following instances:
1. for electric utility steam generating units, regulated by 40 CFR 60 Subparts Da and Db (New Source
Performance Standards (NSPS)),
2. for primary nonferrous smelters regulated by 40 CFR Part 60 Subparts P, Q, and R (NSPS)
3. for NSPS stationary gas turbines regulated by 40 CFR 60 Subpart GG,
4. for various sources which are regulated by permits, orders or consent decrees in which CEM has
been specifically designated as the test method,
5. for various types of sources which are regulated by the State Implementation Plan and where the state
has specified in that plan, the use of CEM as the compliance test method, and
6. federal regulations promulgated as part of State Implementation Plans for non-ferrous smelters.
The reader should note that each subpart to 40 CFR Pan 60 contains CEMS requirements pertaining
to availability (e.g., the percentage of time which a monitoring unit must record source emissions). For
example, 40 CFR60.47b(c) (i.e., subpart Db which pertains to Industiral Commercial Insitutional Steam
Generating Uits, emission monitoring requirements for sulfur dioxide), states:
"(c) The owner or operator of an affectedfacility shall obtain emission data
for at least 75 percent of the operating hours in at least 22 out of 30
successive boiler operating days. If this minimum data requirement is not
met with a single monitoring system, the owner or operator of the affected
facility shall supplement the emission data with data collected with other
monitoring systems as approved by the Administrator or the reference
methods and procedures as described in paragraph (b) of this section."
The use of continuous gas emission monitors for ensuring regulatory compliance can be summarized
by .the following policy stated in a 1988 EPA Office of Air Quality Planning and Standards (OAQPS)
memorandum:
"QAQPS is committed to promoting, encouraging and utilizing CEMS data
as a compliance assessment measure. Our Office is also committed to the use
of CEMS in direct enforcement where CEMS is the compliance test method
11-3
-------
and for supporting enforcement where OEMS is not the compliance test
method OA&Sencouragesttieuseof CEMS databyStatesincompIiancemonitoring
and in supplementing or supporting enforcement actions. If it is technically
feasible, CEMS requirements should be incorporated into NSR preconstruction
reviews, operating permits and resolutions of enforcement actions including
consent decrees and administrative orders."
"OEMS should be used to assure continuous compliance of sources in both
attainment and non attainment areas. Resources should be allocated to monitor
continuous compliance of sources in areas where the greatest environmental
benefit is likely to occur. Therefore, priority should be given to NESHAPS
sources subject to continuous monitoring requirements, (currently 40 CFR 61,
subparts F, N, O and V) and to SIP (including major and minor NSR sources)
.and NSPS sources in non-attainment areas (for the pollutant for which the area
is non-attainment). Next, CEMS should be used to monitor the continuous
compliance of NSPS and PSD sources in attainment areas. Sources with
excessive emission limit excursions identified by CEMS data should be targeted
for follow-up action (on-site inspection or 114 letter). Where CEMS is the
compliance test method, CEMS data should be used to identify significant
violators. These sources will then be tracked in accordance with the Timety and
Appropriate Enforcement Response Guidance,' issued by f the Office of Air and
Radiation] OAR on April 11, 1986."
Additionally, except where prohibited by state law, both state and local agencies may impose more
stringent standards on sources in their jurisdiction. The regulatory procedures used to bring about these
additional standards can be in the form of operating permits, variances, prders. agreements or promulgated
regulations. All of these options can require continuous emission monitoring.
In meeting the NAAQS, states issue operating permits to both new and existing sources. Under the New
Source Review (NSR) program, State Implementation Plans must establish provisions for the review of new
or modified sources to ensure that the NAAQS will not be exceeded. Another program, called Prevention of
Significant Deterioration (PSD), is designed to ensure that the construction or modification of a facility wfll not
cause a significant degradation of air quality. The issuance of an operating permit may require the source to
install a continuous emission monitor so mat any significant emissions increase potentially resulting in a
NAAQS violation would be detected.
Variances are sometimes issued on a temporary basis to allow a source to pollute above the standard
provided that it meets the standard by a specified period of time. The granting of a variance may involve the
implementation of certain continuous emission monitoring requirements.
are usually issued as -a result of public hearings and are an authorized=method of assuring
compliance as stipulated in Section 1 13 of the Clean Air Act It takes a certain type of legal authority to issue
an order, there can be administrative orders, delayed compliance orders and court orders. An order may require
that a source submit a plan for installing, calibrating, maintaining and operating a CEM
• Sometimes the regulatory agency may end up negotiating an agreement with a source in order that it meet
the regulatory requirements within a certain period of time. There are several types of agreements, namely,
consent decrees, stipulation agreements, and court settlements. Again, the use of a CEM may be a pan of an
agreement with the agency.
11-4
-------
Finally, the promulgation of regulations is mandated through the federal requirements for New Source
Performance Standards (NSPS) and the State Implementation Plans (SIPs). SIPs are submitted by the states to the
EPA, explaining how the State plans to attain and maintain the NAAQS. Regulations promulgated as a result of
a SIP often require continuous emission monitoring.
Sources Required to Install Continuous Gas Emission Monitors
As mentioned above, the Clean Air Act mandates that EPA develop NSPS and NESHAPS. In addition, states
must adopt legally enforceable emission standards to ensure that NAAQS are met Each of these, NSPS,
NESHAPS, and state rules developed to meet NAAQS, contain testing and monitoring requirements. The NSPS
are found in the Code of Federal Regulations Title 40, Part 60 (40 CFR 60).
The NESHAPS rules are found in 40 CFR 61. Minimum emission monitoring requirements for State
Implementation Plans are found in 40 CFR 51, Appendix P. In the event that a state fails to adopt rules adequate
for attaining the NAAQS, the EPA is required to promulgate substitute regulations. Federal regulations
promulgated as pan of a SIP are found in 40 CFR 52. Table 11-1 is a listing of the pertinent subparts and sections
of the aforementioned code of federal regulation parts which specify the continuous monitoring requirements for
each source category as well as the gases required to be monitored.
Table 11-1. Continuous Monitoring Requirements for Each Source Category Emitting
Pollutant Gases
(40 CFR 51)'
Source Category 40 CFR 60 Appendix P 40 CFR 61 Gases Required to be
Subpart Section Subpart Monitored
FonilFudFired Steam Gencnun
Electric Utflay Qcncntinf Vtiu
'fffw4nctft«l rVwj|fflM^iBl_T|p^j^|lf^|}f)
Stem Genen&v Unto
Nitric Acid Hnti;
Suliuric Acid PJmtf
ftBoleam Aefoeo*
Prim try Copper Sntehcn
PrtDJoyZmcSmelttK
Primny Lad Smdten
KnftPu^Milk
Suiion«y Gu Turbine* ;
BhyleneDichloridePurificttianSyaaiu/
PUnts. Vinyl Chloride Pints, TolyvinyJ
CUoridePlatr
: Araenk Trioidde «nd Metdlie Aneoie •
PnxiuctiaoF*cUJuer
D
D»
Db
O
H
J
f
0
It
flB
OG
2J
32
23
:
V
T
SOrNO..(*0,oray
SOrNO^C^OO,
SO, NO.
KO.
SOj.^.&CO.inptrtH)
CO, SO,
so,
so
«>1
IRS {Toul Reduced SnffaX %O,
SO, NO,
Vinyl Chloride
:Afflbiem Air Mohiiadnt
of borinicAneak
' That r^uirai*nlsafoapptyto40CFR52,iodwaucoiiiuu&us gas aruMnmoiwriwrtq Ike Suit IntpUmtiuatiai Plan.
11-5
-------
Federal and state monitoring requirements are subject to changes, and it is important to stay abreast of these
changes in order to make sure that new requirements are met New regulations, revisions to existing regulations,
and notices of revisions to SIPs are all published in the Federal Register (FR). The FR is published daily and
codified annually in the Code of Federal Regulations. The FR is a useful tool in keeping track of daily updates
in the CFR. There are two indices which can be used in tracking the changes in the regulations. They are the
Federal Register Index which is published monthly and is accumulated on a twelve month basis, and the LSA
(List of CFR Sections Affected) Index which is published on a monthly basis. The steps employed in using these
references are contained in the discussions which follow.
Federal Register Index
The first step in learning to use the Feeral Register Index is to obtain a copy of it Most libraries subscribe
to this publication, a copy of which is shown in Figure 11-1.
ANNUAL
Index
RQUREH-1 FEDERAL REGISTER MOEX TITLE PAGE
After obtaining a copy, find the listing of the agency or topic which is affected by the regulation of interest
[in this case, the EPA, (see Figure 11-2)]. These topic or agency headings will be subdivided into Rules,
Proposed Rules and Notices. Under the subheading. Rules, look for the specific area of interest (say airpoUution;
standards of performance for new stationary sources). Under this area look for the specific topic under the
specific area of interest Figure 11-3 shows how to find anything regarding continuous emission rate monitoring
systems; specification and test procedures.
11-6
-------
oxi n^.. »t. rm M|4 4101. no. . 0~~ — « i •—
•»» toot. utu. m» jrw.yi» . ••>..» MC4mi
. CA.ITUI
KM. Mil
> liar. ON. IM4I
VA. 11170
Tnm l». TX TTfT
CA,II»I
FIGURE 11-2 FEDERAL REGISTER INDEX - FEDERAL AGENCY OR TOPIC
AND SUBTOPIS LISTINGS
You will notice that after each topic of interest, there is a page number. Remember that number as you turn
to the back of the index and find the section "Federal Register Pages and Dates" (Figure 11-4). The first column
gives the range of pages and the second column gives the date of the Federal Register notice that deals with the
specific topic/area that you are interested in. The third column gives the Federal Register volume number. Using
11-7
-------
the first column, find the page range which your issue falls into. The corresponding second and third columns tell
you the date and volume of the Federal Register Notice where the required information can be found. Once the
ER volume number has been found, find the page number [located in the upper left comer of each page, (set Figure
11 -5)] corresponding to the number referenced in the FR index. The FR notice which you were looking for will
be found on that page.
FIGURE 11-3 FEDERAL REGISTER INDEX LISTINGS OF SPECIFIC AREAS
OF INTEREST
-------
'i_r_nAL MEOvm PAOCI AMD tTfiTti nimunT n*rnatn
13441-11714
I37U.I40M
1MII-IIU4
II2U.II17I
-
1UM-IT009
1TOU-111M
ITirt. 11444
1T441.ITW1
ITUI-ITflO
17111. imo
1U1-IM4
:441-140.
IMI-tTU
173.-IMI
I
1I2U.IIM4
1I5U-IUI4
-
SSfcSJS
MKT.II404
U4U-I111I
4U1-4IU
4MMIU
3I4»-33_I
12M-UM
3U1
1W1-174J
37
>M4-4n<
4m-iu:
4IU-4741
3141-140IO
:40I|.|4M
14I4T-144M
II4I-M20
•431-MIO
It 1 14144
1743-uu
IU1-XU
•423-MM
•WI-47U
lilt-Mil
•U1.IWM
o>40
1034
10317-10111
I031I-IOMI
:UM-IIUO
11021-11:21
UITI.2UM
I9U1-2IIM
11417.1112.1
IIIU.1III4
I1I1I-I1MO
UMI.IJUI
i:m-irrro
SSU3&
ESSEX
-
30421OOCM
JM11-JMN
:nn- 1U44
:1941>13I43
:U41-IM43
FIGURE 11-4 FEDERAL REGISTER INDEX LISTING OF PAGES AND DATES
OF FEDERAL REGISTER NOTICES
ISA Index
The second reference that can be used to stay abreast of changes in regulations is the LSA (see Figure 11 -
6). This index is arranged according to the CFR title and part. In this example, we will look for any amendments,
revisions or proposals to Title 40 Part 60. First (refer to Figure 11-7). turn to the section in the index under Title
40 (the title number can be found at the top left comer of each page in bold print). Next, locate the section number
of interest Lets say that we want to research Section 60.8. Go down the column until you find 60.8 and look over
11-9
-------
jgrrge gmiT^ZZ.^.
Mr. WUU.. O»*T orMr. UfV T. '^rTTMt l^fM»Wfc--T-r
ol«OO«r>RBio>MNi
-------
LSA
List of CFR Sections Affected
February 1989
TttlM 1-16 (Annual)
Changes January 3,1989
through February 26,1989
Title* 17-27
Changes April 1,1988
through February 28,1989
Title* 28-41
Changes Juty 1,1988
through February 28,1989
Title* 42-50
Changes October 3,1988
through February 28,1689
Parade! Table of
Authorities and Rules
FIGURE 11-6 LSA - LIST OF C FR SECTIONS AFFECTED INDEX TITLE PAGE
Acceptable Monitor Criteria
When continuous gas emission monitoring systems are required in an NSPS, NESHAPS, or state rule, it is
usually to ensure that control equipment performance is maintained in such a way that a source remains in
compliance with the standard. In regard to NSPS, 40 CFR 60 states that an continuous emission monitoring
systems which require gas pollutant monitoring are subject to the performance specifications under Appendix B
11-11
-------
(Part 60) and, are subject to Appendix F (Part 60 Quality Assurances Procedures) if they are used to demonstrate
compliance with emission limits on a continuous basis. We will discuss the requirements of Appendices B and
F (Part 60) in lessons 12 and 13, respectively.
or cm (tenons unem
CHAMOIS JUIT i. ira TWOUOH miUAir a. iff*
• TTTU 40 dimwr t—CM.
(ci
FIGURE 11-7 LSA LIST OF CFR SECTIONS AFFECTED INDEX PAGE
CONTAINING THE CODE OF FEDERAL REGULATIONS TITLE PART
NUMBER OF INTEREST
40 CFR 51, Appendix P requires that each State Implementation Plan establish requirements for the use of
continuous emission monitors. This appendix includes requirements for source categories that must be monitored;
incorporates performance specification tests required under Appendix B of Part 60; and lists recordkeeping and
reporting requirements. The performance specifications mentioned under Appendix B of Part 60 are used to
11-12
-------
evaluate continuous emission monitors for acceptance at the time of installation (or soon thereafter) and whenever
the regulations specify that they be reevaluated. Essentially, the performance specification test is used to particular
location may not be sufficient for use at another location due to differences in heat, vibration, etc. between these
locations. Thus, each source and its continuous emission monitoring system must be evaluated to see that the
monitoring system is suitable for that source and that it is operated properly.
TABU Of KDBLAL IKHSTBI OSUt PAOH AND DATES
49287-49544
49545-49648
49648-49842
49843-49968
49969-50200
S0201-50372
50373-50306
50507-50910
50811-51088
51089-51216
51217-31534
51935-51734
51725-52110
52111-52396
52397-52622
36431-36556
36557-36774
36775-36948..
36949-37280
37281-37538
37539-37726
37721-37982
37983-38280
38281-3M86
38687-38938
38939-39072
39073-39224
39225-39432
39433-39582
39583-39738
39739-40012
40013-40200
40201-40394
40395-40714
40715-40864
52623-52970
52971-53376
271-386
387-594
595-786
787-960
1143-1324
1325-1674
1675-1922
1923-2080
2081-2984
2985-3404
3405-3576
3577-3768
3769-3978
3979-4248
4249-4748
4749-5070
5071-5206
5207-5404
5405-5582
5583-5920
5921-6114
6115-6262
62634380
6381-6502
6503-6640
6641-6860
6861-7028
7029-7170
7171-7390
7391-7520
7521-7750
7751-7924
7925-4180
6181-8266
8267-8518
40865-41148
41149-41304
41)05-41550
41551-42930
42931-43184
43185-43412
43413-43672
43673-43842
43843-43998
43999-44166
44167-44372
44373-44584
44585-44852
44853-45058
4SOS9-4S248
45249-45442
45443-45750
45751-45880
45881-46078
46079-46426
46427-46600
46601-46842
46843-47178
47179-47490
47491-47656
47657-47798
47799-47926
47927-48242
48243-48504
48505-48628
48629-48894
48895-49110
49111-49286
FIGURE 11-8 LSA - LIST OF CFR SECTIONS AFFECTED INDEX TABLE OF
FEDERAL REGISTER ISSUE PAGES AND DATES
40 CFR 61 lists the continuous emission monitoring requirements for sources regulated under the National
Emission Standards for Hazardous Air Pollutants (NESHAPS). Specifically, these requirements are delineated
in Subparts F and P.
11-13
-------
VoL «. No. a TTutfon. Ftbfwv II. UN / RulM ua *****
FIGURE 11-9 FEDERAL REGISTER NOTICE PAGE CONTAINING THE
NOTICE REGARDING THE AMENDED CODE OF FEDERAL REGULATIONS
TITLE PART
Summary
.The U.S. Clean Air Act as amended in 1977, mandated EPA to promulgate New Source Performance
Standards (NSPS) and National Emission Standards for Hazardous Air Pollutants (NESHAPS) and to establish
National Ambient Air Quality Standards (NAAQS). States must see that sources are adequately controlled
through a state-established permit review system in order to meet the NAAQS.
11-14
-------
Continuous gas emissions monitoring requirements have been established in order to demonstrate whether
a source is or is not in compliance with the applicable air quality standards. The continuous gas emission
monitoring requirements for sources are contained in the applicable Code of Federal Regulation Parts. As we will
learn in Lesson 12, performance specification tests (contained in 40 CFR 60, Appendix B) are required to be
performed on CEMS in order to assure that such equipment is reliable, accurate and durable.
11-15
-------
REVIEW EXERCISES
1. True or false. Excess emission monitoring data are used whenever
the use of continuous emission monitoring data is the state's
approved compliance method.
2. List three regulatory procedures which state or local agencies use
in order to assure compliance with air quality standards.
1. False
3. Continuous Emission Monitoring Emissions Standards for Haz-
ardous Air Pollutants (NESHAPS) are contained in 40 CFR Pan
2. Any three of the fol-
lowing variances,
operatingpermits,
orders, agreements,
or promulgated
regulations.
4. Title 40 Part of the Code of Federal Regulations deals with
continuous emissionmonitoringunderStatelmplementationPlans.
3. 61
5. 40 CFR Part 51 (Appendix P) Section 2.1 pertains tocontinuous
emission monitoring requirements for
a. Electric Utility Generating Units
b. Kraft Pulp Mills
c. Primary Zinc Smelters
d. Fossil Fuel Steam Generators
4. 51
6. List the two federally published indices which can be used to keep
track of changes in the Code of Federal Regulations.
5. d
7. Performance Specification Tests Index and the LSA Index.
for continuous emission monitors are found in Appendix .
6. Federal Register
11-16
-------
8. A performance specification test will determine if a continuous
emission monitoring system is:
a. accurate
b. durable
c. reliable
d. all of the above
7. 40 CFR 60, B
9. True or false. Continuous emission monitoring requirements
for sulfuric acid plants are described in both 40 CFR 60 subpart H
and in 40 CFR 51 Appendix P section 2.3.
8. d
10. True or false. Continuous emission monitoring requirements
for nitric acid plants are found only in 40 CFR 51 Appendix P,
section 2.2.
9. True
10. False
11-17
-------
REFERENCES
1. Protection Agency (EPA). 1975. Requirements for Submitted oflmplementatio Plans-Standards for New
Stationary Sources-Emission Monitoring. 40 FR 46240 (October 6,1975).
2. EnvironmentalProtectionAgency(EPA). l9B2.Memorandum:GuidarKe&ncerningEPA'sUseofContinuous
EmissionMonitoringData:KM.BennetttoRegionalOfficeDirectors.EPAAirProgramsPolicyandGuidance
Notebook-Office of Air Quality Planning and Standards. Research Triangle Park, N.C. Policy Notebook
Number PN113-82-08-12-014. (Available for public inspection and copying at EPA Regional Offices).
3. Environmental Protection Agency (EPA). 1984. APT1 Course SI:476A TransmissometerSystems-Operation
andMaintenance,AnAdvancedCourse-Se^InstructidnalHandbookJEPA450/2-M^^
1986).
4. Environmental Protection Agency (EPA). 1988. Memorandum: Transmittal of ReissuedOAQPS CEMs
Policy: Gerard A. Emison, Director, OAQPS to Regional Office Directors. EPA Air Programs Policy and
Guidance Notebook-Office of Air Qaulity Planning and Standards. Research Triangle Park, N.C. Policy
Notebook Number PN113-88-03-31 -048. (Available for public inspection and copying at EPA Regional
Offices).
11-18
-------
LESSON 12
Performance Specification Test Procedures
Lesson Goal and Objectives
Goal
To explain the performance specification test procedures for continuous emission monitoring systems as
required by Appendix B of Part 60 Title 40 of the Code of Federal Regulations (40 CFR 60) and to provide a basic
outline of the requirements for the pollutant gases addressed in that appendix.
Objectives
At the end of this lesson, you should be able to -
1. describe the parameters which are measured in a performance specification test;
2. define span value, relative accuracy, calibration drift, EPA reference method, and stratification;
3. describe the specifications used for properly installing the continuous emission monitor (OEM) and how
locations for taking performance specification test measurements are determined; and
4. define the variables used in calculating the relative accuracy and calibration drift.
Introduction
As we learned in Lesson 11, federal regulations pertaining to the installation and use of continuous gas
emission monitoring equipment are found in 40 CFR Parts 51,52 and 60. The GEM requirements mentioned in
these Parts reference Appendix B to 40 CFR 60. This appendix delineates certain performance specification test
procedure requirements which continuous emission monitors (CEMs) must pass before they can be used for
regulatory purposes. These tests are performed on a one time basis at the time of or soon after (within 30 days)
the initial installation of the monitoring system. Additionally, these tests may be required at other times by the
EPA pursuant to Section 114 of the dean Air Act The performance specification tests are required in order to
ensure that the monitoring equipment meets certain design and specification requirements with the intent being
12-1
-------
that the equipment will be reliable once it has been installed for use. These performance specification tests were
not designed to evaluate the installed CEM performance over an extended period of time nor do they identify
specific calibration techniques and other auxiliary procedures to assess the CEM performance. The source owner,
or operator, however, is responsible for proper calibration, maintenance, and operations of the CEM as stated in
40 CFR Part 60, Appendix B and 40 CFR Part 51, Appendix P.
The use of the phrases "performance specification test", and "performance tests" have been used interchange-
ably, but there are a couple of distinctions in the actual definitions of these terms. First of all, a "performance test"
(as defined in 40 CFR Part 60) is an emission compliance test which is not related to testing a CEM system. This
performance test is used to ensure that sources subject to continuous emission monitoring regulatory requirements
(e.g., NSPS) are in compliance with the applicable emission standards. Secondly, a "performance specification
test" is a set of procedures required by 40 CFR Part 60.13 to evaluate the accuracy and reliability of systems
(GEMS).
CEM performance specification tests were initially promulgated in 1975 for opacity monitors (PS1), SO, and
NOX (PS2), and C02 and O2 (PS3). Since that time, these performance specification tests have undergone various
revisions including reorganization of each of these tests, improved descriptions of CEM location requirements for
PS 1, clarification of calibration drift (CD) and relative accuracy (RA) test requirements in PS2 and PS3, and an
increased reliance on the RA test results for assessing CEMS performance in PS2. Finally, since 1975 three
additional performance specification tests have been promulgated. They are CO (PS4) which was published on
August 5,1985 total reduced sulfur (TRS) (PS5) which was published on July 20.1983, and Continuous Emission
Rate Monitoring Systems (GERMS) (PS6) which was published on March 9,1988. The reader should note that
PS6 involves Performance Specifications Tests for equipment required for the determination and recording of a
pollutant mass emission rate (in terms of mass per unit time). CERMS are not the subject of this course and PS6
will not be discussed in this lesson.
In this lesson, we will explore the technical requirements of Appendix B. In particular, we will discuss the
performance specification tests for continuous gas emission monitors, define some of the important terms used
in this appendix, describe the CEM parameters (calibration drift and relative accuracy) which are required to be
tested and the acceptable specification test values, discuss the specifications for locating and installing a CEM,
and discuss the methods used for determining the calibration drift and relative accuracy.
Overview of Appendix B
Table 12-1, lists the topics covered in performance specification tests, numbers 2-5, for CEM equipment used
tomonitorSO,andNOx (PS2), C02 and 02(PS3), CO (PS4),andTRS (PS5). Itshould be noted, that specification
tests 3-5 use almost the same procedures outlined in specification test 2. Because of this, the following discussions
focus on specificiation test number 2. We wfll first address Section 2.0, Definitions.
«
Performance specifications tests 2-5 have several key terms in common and these terms are important to learn
in order to understand each test procedure. The terms we will address are: 1) Span Value, 2) Relative Accuracy
(RA), 3) Calibration Drift (CD), and 4) Reference Methods (KM).
Span Value refers to the full range of the monitoring system and is defined as "The upper limit of a gas
concentration measurement range specified for affected source categories in the applicable subpart of the
regulations".
12-2
-------
PART60-APPENDKB
PERFORMANCE SPECIFICATIONS
1.0
2.0
3.0
4.0
5.0
(.0
7.0
to
9.0
10.0
11.0
1
PERFORMANCE
SPECIFICATION TEST 2
MONfTORS OF SOj AND NOX
- APPLICATION AND PFfNOPLE
-j DEFINmONS
_ INSTALLATION AND
MEASUREMENT SPECIFICATIONS
_ PERFORMANCE AND EQUIPMENT
SPECIFICATIONS
4TgB!88SSFBe"™
iffigw*"™
• S&gg&JSg""**™
- CALCULATIONS. OATAANALYW
- REPORTING
-1 ALTERNATIVE PROCEDURES
j BIBLIOGRAPHY
1.0
2.0
3.0
-
•
-
1
PERFORMANCE
SPECFCATONTEST3
MONfTORS OF CO2 AND Of
APPLICATION AND PRINCIPLE
PERPORMANCE AND EQUIPMENT
SPECIFICATIONS
RELATIVE ACCURACY TEST
PROCEDURES
NOTE: SECTIONS 2,3.5.6 «8 AND
10 OF SPECIFICATlONtEST 2
APPLYTO SPECIFICATION TEST
3 EXCEPT AS SPECIFIED IN
SECTION 3 OF THIS
SPECIFICATIONTEST
1.0
2.0
3.0
I
PERFORMANCE
SPECIFICATION TEST 4
MONfTORS OF CO
APPLICATION AND PRINCIPLE
PERFORMANCE AND EQUIPMENT
SPECIFICATION
RELATIVE ACCURACYTEST
PROCEDURES
NOTE: SECTIONS 2.3.S.6.8AND*
OF SPECIFICATION TEST 2
APPLY TO SPECIFICATION TEST
4 EXCEPT AS SPECIFIED W
SECTION} OP THIS
SPECIFICATION TEST
1.0
2.0
3.0
4.0
•
I
PERFORMANCE
SPECIFICATION TEST 5
TOTAL REDUCED SULFUR
APPLICATION AND PRINCIPLE
PERFORMANCE AND EQUIPMENT
SPECIFICATION
RELATIVE ACCURACY TEST
PROCEDURES
BIBLIOGRAPHY
NOTE: SECTIONS 2. »ASA« AND
• OF SPECIFICATION TEST 2
APPLYTO SPECIFICATION TEST
5 EXCEPT AS SPECIFIED W
SECTION2OFTHIS
SPECIFICATIONTEST
TABLE 12-1 CODE OF FEDERAL REGULATIONS, TITLE 40, PART 60, APPENDIX B
PERFORMANCE SPECIFICATION TEST PROCEDURE SECTIONS (SPECIFICATION
TESTS #2, #3, #4, AND #5
The Relative Accuracy (RA) of a gas CEM is the percent difference in the continuous monitor's measurement
of the pollutant versus the value of the measurement as a result of using an EPA Reference Method. Because the
accuracy of the monitoring system is determined by the use of a second, independent measurement and not by a
known or actual value, the term Relative Accuracy and not. Accuracy, is used.
The Calibration Drift (CD) refers to the comparison of the monitor's response to a known concentration of
gas. If the continuous emission monitoring system has a separate pollutant and diluent monitor, then the CD test
must be performed separately for each monitor. We will discuss the difference between a pollutant and a diluent
monitor later.
The Reference Method is a way to sample and analyze for an air pollutant These methods are found in
Appendix A of Pan 60 and are required to be used in the performance specification test procedures of Appendix
B. For the most part, these reference methods rely on equipment specifications, but a few of the methods do rely
on performance criteria. That is to say, some reference methods have established performance criteria by which
any system design can be used as long as it meets the criteria.
It is important to note, that whenever source owners are required to perform specification tests on newly
installed CEM equipment, the regulatory agency will appoint a representative to observe the tests in order to ensure
12-3
-------
that they are performed in accordance with the specification requirements. A detailed explanation of the required
specification test(s) as well as specific instructions and recommendations for the test report which the observer must
file are contained in an EPA document entitled, 'Performance Specification Tests for Pollutant and Diluent Gas
Monitors: Reporting Requirements, Report Format, and Review Procedures," EPA-340-1/83-013. These
regulatory personnel can be from the state control agency or the EPA regional office and must be specifically trained
to work with GEMS.
Tables 12-2 through 12-5 provide a basic overview of the accepted specification test values for zero
drift,calibration drift and relative accuracy, as well as the period of time required for performing a calibration drift
test Each table is specific to a different specification test The applicable Part 60 Appendix A Reference Method(s)
required to be used for each of the gases monitored in either specification test #2, #3, #4, or #5 are displayed in Table
12-6. The reader should be aware that some Subparts to 40 CFR 60 may contain alternate Reference Methods to
those contained in Table 12-6 which may be used in conducting a performance specification test
Since the Reference Methods procedures are designed to analyze the gas pollutants on a dry basis, a correction
factor to account for moisture is necessary if the CEM system being tested is not corrected for this factor. Reference
Table 12-2 Performance Specification for SOt 1 NOt System
Specification Test $2
PARAMETER
ACCURACY*
ZERO DRIFT
CALIBRATION DRIFT
CALIBRATION DRIFT TEST PERIOD
* Hutficmd as not of absolute tnf an value Tins
** Span values an at spedfied In die applicable!
burned.
*** PorSOjnnittianitandanb between 030 and
standard; below 0.20 ZMMMBtaweTO* of*
SPECIFICATION
< 20% of the mean value of the reference
method test data or 10% of the applicable
standard, whichever is greater. **
23% of Span Value***
23% of Span Value***
16? hours, minimum (at 24 hour interval)!
for 7 consecutive days)
>5* confidence interval of asenes of teas.
egulations and area function of Ihe type oTfod
030 a/MMBm. nse 1 5* of the applicable
e fiHJuiun standard.
Method #4 is used to provide this moisture correction and is used concurrently with the regular gas pollutant
measurement methods.
CEM Installation and Measurement Location
Section 3.0 performance specification test #2 of Appendix B describes m specifications for installing the
monitoring system and selecting locations where representative measurements can be taken. It is the source owners'
responsibility to assure that the CEM is located in such a position which will yield a representative concentration
12-4
-------
Table 12*3 Performance Specification ford and CO^
Specification Test §3
PARAMETER
ACCURACY*
CALIBRATION DRIFT
CALIBRATION DRIFT TEST PERIOD
SPECIFICATION
< 20% of the mean reference method test
data or 1 .0% Q or CQ of the applicable
standard, whichever is greater
<0.5% Q or CQ from reference value of
gas, gas cell, or optical filter
168 hours, minimum (at 24 hour intervals
for 7 consecutive days)
• Expressedas«iniofabsoluJeme«nvalueplus95%CQnfidenceiiuerva]of«««iesofleo«.
Table 12-4 Performance Specification Jbr CO
Specification Test #4
PARAMETER
ACCURACY*
CALIBRATION DRIFT
CALIBRATION DRIFT TEST PERIOD
SPECIFICATION
<10% of the mean reference method test
data value or 5% of applicable standard,
whichever is greater
-------
Table 12-5 Performance Specification for TRS
SpecificationTesttiS
PARAMETER
ACCURACY*
CALIBRATION DRIFT
CALIBRATION DRIFT TEST PERIOD .
SPECIFICATION
< 20% of the mean value of reference method test
data or 10% of applicable standard, whichever is
greater
<5%(1.5PPM) of established span value of
30ppm for 6 out of 7 test days *»
168 hours, minimum (at 24 hour intervals for?
consecutive days)
* ExpresMdufnmofabiolutetneanvalueplus95%cor^denceimervalc/aseriesoftesU.
*• Span valuw ate as specified in the applic^lereguUUons and are a funoicB of the type of fuel
or can be corrected to yield a representative concentration of the total emissions from that facility. As indicated
earlier, the specifications listed in Section 3.0 (Le., performance specification test 2) are also used in the other CEM
system performance specifications.
In general, the suggested CEM location (i.e., where the sample taken will most likely provide data which meets
the RA requirements) should be at least two equivalent diameters (i.e., at a distance two times the diameter of the
stack or duct) downstream from the nearest control device, the point of pollutant generation or other point at which
a change in the pollutant concentration or emission rate occurs, and at least 05 equivalent diameters upstream from
the effluent exhaust or control device. It is worthy to note that some individual subparts of Part 60 may contain
. additional requirements. For example, steam generating facilities must locate the CEM downstream of the air
preheater.
Section 32 of Appendix B addresses the sample locations for the Reference Method (RM) measurement
locations and traverse points during the relati veaccuracy tests. Basically, this section states that in performing the
relative accuracy test, the RM sample is taken by traversing the measurement line at three points. If the
measurement line is 2.4 meters long and stratification is not a problem (discussed below), the sample points may
be located on this line at 0.4.12 and 2.0 meters from the stack or duct walL If the measurement line is longer man
2.4 meters, then the sample points can be located at 16.7.50 and 83.3 percent of the length of the measurement
line.
It should be noted that the CEM measurement location and the Reference Method traverse location need not
be the same for determining the relative accuracy. For example, the CEM may be located in the exhaust duct of
a control device because it is an accessible location, but the area may not contain ports for taking Reference Method
samples. The Reference Method sample could then be taken further downstream where ports would be present
and which would meet Die requirements of the performance specifications. The CEM results are still compared
to results obtained by using the Reference Methods in determining the relative accuracy.
Personnel who perform the performance specification test, need to be aware of the problem of stratification.
Stratification refers to the amount of variation in pollutant emission rates at the Reference Method measurement
locations. Stratification results from the mixing of two dissimilar gases immediately upstream of the measurement
12-6
-------
location. A specific example of where stratification can occur is when a CEM is located in duct work which
provides better accessibility for maintenance and repair purposes. In this situation, by-passed gas can be mixed
with cleaned or scrubbed gas for reheating purposes. Another example of where stratification occurs is due to poor
scrubber operation especially where plugging of gas passageways and liquid-slurry spray nozzles has occurred.
Stratification involving poor scrubber operation is very difficult to measure and is often temporally variable.
TABLE 1M APPLICABLE REFERENCE METHODS FROM 40 CFR £ i
PART «0, APPENDIX A M\^;:.^< '. '' 'ff?^
PERFORMANCE APPLICABLE
SPECIFICATION REFERENCE TITLE
1EST METHOD®
2
3
4
5
Method 3
Method 3B
Method4
Method 6
l||Me^fii::|
Methods
Method 3B
Methodic
Method 1QA
Method 10B
Method 16
Method 16A
Gas analysis for the determination of dry
molecular weight
Gas analysis for the determination of
emission rale correction factor or excess
air
Determination of moisture content in
stack gases
Determination of sulfur oxide emissions
from stationary sources
Determination of nitrogen oxide
emissions from stationary sources
Gas analysis for the determination of dry
molecular weight
Gas analysis for the determination of
emission rate correction factor or excess
air
Determination of carbon monoxide
emissions from stationary sources (Note:
Section 10.1 of method 10 specifies the
use of an alternative interference trap
when using nondispersive infrared
CEMS)
Determination of CO emissions in :
certifying continuous emission
monitoring systems at petroleum
refineries
Determination of CO emissions from
stastionary sources
Sernicontirraouff • ftfrtCTmination of sulfur **
•emissions from stationary sources
Determination of total reduced sulfur
emissions from stationary sources
(Impinger Technique)
(Nau: St**S&paiiU>40CFIlfmltOma)coi>uiinll*natnfmiic*mttke4i wkick maj kt iai4iuuari cf
•Under conditions where stratification is evident, the Reference Method sampling points should be taken along
a line perpendicular to the line of such gas mixing. There are methods for locating the sample points whenever
stratification is suspected for both rectangular and circular ducts. These procedures are outlined in an EPA
12-7
-------
document entitled "Gaseous Continuous Emission Monitoring Systems - Performance Specification Guidelines
for S02, NOIf C02, (\ and TRS", EPA-450/3-82-026. Interestingly enough, it has been shown that taking the RM
samples at three points along the traversed measurement line will yield good results even if stratification is a
problem.
Calibration Drift, Relative Accuracy, Test Methods and Procedures
Section 4.0 of performance specification test #2 inAppendixB, addresses relative accuracy and the calibration
drift (CD) requirements for CEMs. The actual procedures for performing the calibration drift test and the relative
accuracy test are contained in sections 6 and 7, respectivley. As stated earlier, if a CEM has both a pollutant and
a diluent monitor, then the calibration drift test must be performed separately for each monitor. According to
Appendix B, a pollutant monitor is one which measures the concentration of the pollutant gas and will display an
output which is proportional to the concentration of the gas being sampled. A diluent monitor is used to measure
the concentration of the diluent gas which is either 02 or C0r These gases are measured concurently with SO,
or NO, gases in order that the pollutant gas concentrations can be converted from parts per million (ppm) to pounds
of pollutant per million Btu heat input (Ib/MMBtu). This type of conversion is usually required for electric utility
steam generators and may be required for additional sources such as sulfuric acid plants. A diluent monitor is
required to comply with performance specification test #3. Section 5.0 of performance specification test #2 of
Appendix B gives the required time period for performing both calibration drift and relative accuracy tests. The
minimum required operation time for performing these tests is 168 hours. These tests are performed for 7
consecutive days at 24 hour intervals.
Sections 7.5 and 8.0 of performance specification test #2 describe the necessary calculations and the
equations which are used in reporting the relative accuracy results. In addition to the equation for relative accuracy,
Section 8.0 gives the equations for computing the standard deviation and the confidence coefficient
Calibration Drift
The calibration drift test is done by using cylinder gases, gas cells or optical filters. This test is performed
over a period of seven consecutive days (per Section 5.0) and involves the introduction of a zero and a span gas
of known concentrations into the system. The system output readings are then compared to the known values
(certified by the National Institute of Standards and Technology (NIST) of the calibration gases. During this test,
the CEM recorder is usually preset at 5% or 10% offset to detect negative drift. A 24-hour drift test involves 8
sets of zero and span measurements correlated with 7 days of zero and calibration drift measurements. The zero
drift values are determined by taking the difference between the zero value recorded after each 24 hour period
and substracting the correct zero value. The calibration drift value is the span value measured at the end of each
24 hour period minus the correct span value. Following each zero and span measurement, a corresponding zero
and span value adjustment is required. If the zero value adjustment is made before the span reading is made, men
the effects of zero drift are removed and the span measurement adjustment is not necessary. Mathematically, the
calibration drift is expressed in terms of the monitor span value for each individual day in the following manner
(Eq.12-1) Calibration Drift (%) « (Sr-S.) x 100
(Sv)
where: Sr = the monitor span reading (ppm). and
S,s the monitor span value (ppm).
12-8
-------
The calibration drift data should be recorded in a form similar to that as displayed in Figure 12-1. This
format is the type as suggested in Appendix B.
FIGURE 12-1 DATA SHEET USED FOR RECORDING
CALIBRATION DRIFT MEASUREMENTS
L
O
W
L
E
V
E
L
H
1
C
M
L
E
V
E
L
DAY
DATE*
TIME
CALIBRATION
VALUE
MONITOR
VALUE
DIFFERENCE
PERCENT
OF SPAN VALUE
1
MOTE: LOW-IEVO. AND MOMLCVE L BEFEII TO THS tEMCENTAOE O* CCU DATA MECOMED FUU. KALE
WHERE THt CAUMATION DRIFT UEASMCMENTS ARE MADE. AN ACCEPTAtLE TUT WOULD IE TO DCTEH-
•METMECAUIKATnN MFT tETWEEN TIN AND TWCNn PERCENT OF TME NUHUVEL VALUE KM THE
LOW-LEVEL TEST AND AT A VALUE IETWEEN FTT» AND ONI -MUNCHED PERCENT V THE MBMEVEL VALUE
K>K THE HKStiUVB. TEST. tECTKJH 4.1 <*EMOI(HANCE AND COUPUENT tPECmCATIONS) APTENDn 1. W
CfK PAHT ». Of t&un* THE HETHOO WMCM THE tOUKX OWNE* OH O»€IUTO« HAY UtE M CHOCMMQ
THE OOIUD MOH4JVEL VALUE.
Relative Accuracy
The relative accuracy test involves the use of the specified EPA Reference Methods which are delineated
in Appendix A to 40 CFR Pan 60. This test is essentially a comparison between a CEM's measurement of
a pollutant and the value of the pollutant's concentration as determined by an EPA Reference Method.
Because a second (the reference method) independent measurement is used to assess the accuracy of the
12-9
-------
monitor, this test is referred to as a relative accuracy test. This test requires a series of nine sets of measurements.
However, the source owner may choose to perform more than nine sets of measurements. If this option is chosen
up to three sets of data comparisons can be rejected as long as a minimum of nine sets of measurements are used
in calculating the relative accuracy. All data, including the data sets which were rejected, are reported. The
measurements are taken at three points along the cross-section of the duct or stack. There are three terms which
are used in calculating the relative accuracy (expressed as a percent). They are: 1) the algebraic mean difference
between the recorded concentrations of the monitor and those as determined by the Reference Method procedure,
2) the 95% confidence interval associated with the determination of the mean difference, and 3) the mean
concentrations as determined by the Reference Method used. We can mathematically express this relationship
in equation 12-2 as follows:
(Eg. 12-2) Relative Accuracy (%) = ldl + CC^ x 100
where:
Idl = the absolute mean difference between the monitor value and reference method value.
corrected for O2 or C0r
C.C.M% = the 2.5 percent confidence coefficient, (the 95% confidence interval associated with
the observered differences) and
^] = the average result of the Reference Method used for a series of test, corrected for 0,
Section 8 presents the equation used for determing the value of d, its associated standard deviation, and the
confidence coefficient, CC. Equation 12-3 is the equation used for determining the mean difference between the
Reference Method data and t he GEMS data and is expressed as follows:
(Eg. 12-3) d = n 2, (Xi - Yi)= n 2- d,
M M
"d" = arithmetic mean,
n . = number of data points.
X, = concentration from reference method, ng/J,
Y, = concentration from the CEMS, ng/J
d, s signed difference between individual pairs, X, and Y,, ng/J, and
Id, - algebraic sum of the individual differences, d,, ng/J.
In order to calculate the 2.5 percent confidence coefficient, we must first calculate the standard deviation S4
using Equation 12-4 as follows:
(Eg. 12-4)
12-10
-------
The 2.5 percent (error) confidence coefficient, CC, is also presented in Section 8 and is calculated using
Equation 12-5 as follows:
(E0.12-5) CC= t
OUTS
where l^ = A statistic which represents a 95% probability that the sample value is correct
within a particular sample population. A standard table of t-values is contained in Section 8 ofAppendixB. This
table is presented in Table 12-7 below.
TABLE 12-7 VALUE OF T FOR 95 PERCENT
PROBABILITY
an
lli*l|:if
||::|3:;v::::;:;:::;::
|i:f4|fj:|
ii$3il
J|!*IP
'0.975
1Z706
4303
3.182
1776
2J71
•
n
lliflt
lllffti
li^;i!l
iSlil
l;:;||j|;:S;P
'0.975
2.447
2365
2306
97^0
2228
•
n
12
»
14
IS
16
1 0.975
2201
2.179
2.160
2.145
2.131
: : : The vahtei in thi* table are «lre*dy eonected for n-1 deflect of freedom. U«e it eqotl
'.-•xtothe niinibBi'-xtf inrfJVitfmTvtlngi;'1
The reader should note that the values obtained from the monitor and the Reference Methods need to be
corrected for O2 or CO2 (either correction is applicable) in determining the relative accuracy in PS2 and PS4 for
combustion sources, and that an O correction is necessary in performing PS5. Performance Specification Test
#3 does not require this correction.
Basically, the smaller the RA value, the more accurate the monitor. Performance Specification Test #2
requires that the RA value be less than 20% in order for the CEMs to be acceptable for use. This required value
differs slightly for Performance Specification Tests #3 and #4 (see Tables 12-2 -12-5). The reader should note
12-11
-------
differs slightly for Performance Specification Tests #3 and #4 (see Tables 12-2 -12-5). The reader should note
that the relative accuracy results can be affected by errors in both the monitor data and the Reference Method data
due to the variations in the ability of the individual(s) performing these tests. These types of errors cannot be
quantified and the cause of an excessive RA value (e.g., poor testing practices or poor CEM performance) is
difficult to determine. Therefore, an acceptable RA value may demonstrate the adequacy of a CEMS, but an
unacceptable value does not necessarily mean that the CEM is providing inaccurate data.
Now that we have presented the methodology for determining the RA, we will perform an example
calculation. The S02 emission standard for Electric Utility Steam Generating Units (listed in 40 CFR Pan 60
Subpart Da) which bum solid solvent refined coal (SRC-I) is 520 ng/I. Let's say that during a Performance
Specification Test at one of these facilities using PST, the nine sets of data (corrected for O2 or C02) yielded the
following results contained in Table 12-8.
TABLE 12-8 Example Performance Specification Test Results
' TEST
1
2
3
4
5
6
7
8
9
MONITOR DATA
(ng/J)
420
440
450
479
485
490
435
550
460
IRMA
REFERENCE METHOD DATA
(ng/J)
425
436
444
484
490
482
445
565
480 _
rf] * 4723 Idl
2d,
**?
SIGNED DIFFERENCE
(ng/J)
45
-4
- 6
45
45
- 8
, 4-10
415
±22
e 8.67
- 78
« 916
As we can see from Table 12-8, we now have two values (e.g., Idl and [RM ] which will be used in the
equation for determining the RA. However, we still need to determine the value of the 25 percent confidence
coefficient, C.C^J. As we just learned, the confidence coefficient is determined as follows:
1. first of all, determine the standard deviation using equation 12- 4:
S
d
\
1
£7 [916 - 1/9 C782)] = 5.48ngfl
12-12
-------
2. next, determine the value for the 2.5% confidence coefficient as follows:
C.C.,,, = 2.306
5.48
~
= 1.83 ng/J
Now that we have determined the value for C.C.95%, we can determine the RA which is shown below:
RA = 8.67+1.83 x 100% = 2.2%
472.3
Since the value for RA is well below the allowable level (<20%) as specified in Appendix!*, the CEM passes
the Performance Specification Test
In situations where the value for (RM^p is very small, a higher RA value will result which may fall outside
of the acceptable range for RA. This could happen in a situation where the desired values for the monitor and
Reference Method (i.e., in ppm) are very low. To obtain the desired results, an alternative mathematical
expression can be used. This mathematical relationship is expressed in Equation 12-6 as follows:
(Eq.U-€) . IXI + €.1.^x100
Relative Accuracy (%) = Applicable Standard
The Applicable Standard would be in units applicable to a specific source category. The reader should note
that this alternate method for expressing the RA is only reserved for sources whose actual emissions are < 50%
of the emissions standard. Additionally, the RA is required to be < 10% using this mathematical expression.
In this situation, the concern is not to determine whether or not the emissions standard has been exceeded but
rather to ensure that the CEM is producing accurate and reliable data.
12-13
-------
Figure 12-2 Show the suggested Appendix B data sheet which is used for recording the relative accuracy
values from each test run.
FIG 12-2 DATA SHEET USED FOR RECORDING RELATIVE
ACCURACY MEASUREMENTS
RUN
NO.
1
2
3
4
B
6
DATE*
TIME
AVERAGE
scfe
RM
M
DIFF
PPMC
CONFIDENCE INTERVAL
ACCURACY
NO,"
RM
M
DIFF
PPMC
%% \^" Vk''fe' ^*. W
^^.^X^
$$5$
v v?WJ* •x^cSC'w.s
**
RM
M
DIFF
MASS/GCV
^«>
%5^
\ *s
''$$&
fe
'N^»
Ntf
RM
M
DIFF
MASS/GCV
^*S^»
HV
S&8S8
•
*FOR STEAM GENERATORS: AVERAGE OF THREE SAMPLES; °MAKE SURE RM ft M DATA ARE ON
CONSISTENT BASIS. EITHER WET OR DRY.
RM - EPA REFERENCE METHOD DATA
M- GEM MONITOR DATA
GCV • GROSS CALORIFIC VALUE
|N- - N^i I'sVJjf ^ ^A11 .sjO'sV- «-»ijx->--^*^.^^'5»-5«^>S.s4?^ ^v>/"*\.*iwji>^SSC**<"*»^«i'»''&0£>i. ^X•.•.^ «*%\\ NSj^JSSs" 5 -\ ^ \X$S " X*
11
I
1
i
Reporting
Section 9.0 of performance specification test #2 of Appendix B, gives the minimum requirements for re-
porting the test results for both relative accuracy and calibration drift Appendix B requires that the results of
both the calibration drift test and the relative accuracy test must be summarized in tabular form and all data
sheets used, calculations, monitor charts showing the CEM's response, cylinder gas concentration certifications
and calibration cell response certifications (if applicable) must be reported to ensure the adequacy of the
performance specification test
-------
Alternative Procedures
Sources whose emissions are well below the applicable standards may petition the EPA Administrator for
permission to use an alternative relative accuracy procedure. This alternative procedure is explained in Section
10 of performance specification test #2 of Appendix B, and applies to S02, NO,, 02, and C02 GEMS. In this al-
ternative test, RA is the difference between the monitor response and the calibration gas or cell value. Table 12-
9 delineates the accepted values for the relative accuracy, zero drift, calibration drift and operational period
obtained by using the alternative performance specification test procedures specified in Section 10.0 for S02and
NOx. The reader should note that diluent monitors, which (as we discussed earlier) are operated concurrently with
SO2 and NO,, monitors, are subject to the requirements of performance specification test 3. The alternative
procedures are applicable to performance specification test 2 and 3. In this alternative procedure, the response of
the GEM is checked at two measurement points within the ranges specified in Table 12-10 on both the pollutant
and diluent monitors. The RA difference is expressed as a percent difference for SO2 and/or NOX and as an absolute
difference for C02 or 02.
Table 12-9
ALTERNATIVE PERFORMANCE SPECIFICATIONS FOR
SO2/Nq> SYSTEMS
ACCURACY*
DILUENT MONITOR
ACCURACY
ZERO DRIFT
CALIBRATION DRIFT
OPERATIONAL PERIOD
< 15% of span gas or cell
<0.7% O2orCO2
2.5% of span value **
2.5 % of span value **
168 hours, minimum
* Expressed as sum of absolute difference faetween monitor
response and cylinder gas or calibration value.
•** Span values are as specified in the applicable regulations and are
• function of the type of fuel bumed.
Preparing for the Performance Specification Test
Now that we have discussed the various sections to AppendixB, let us say a few words regarding the necessary
steps involved in preparing for the performance specification test Several steps should be considered. First of
all, the performance specification test requirements should be carefully reviewed and if necessary, the control
agency should be contacted if there are any problems encountered in the proper interpretation of these require-
12-15
-------
used in the test. Lastly, all parties involved in the performance specification test should meet prior to the actual
test in order to discuss such issues as test scheduling, agency participation, alternative procedures to be used,
and the performance specification test procedures to be employed. Participants who would be involved in such
a meeting would include representatives from the company (source), control agency, the testing contractor (if
such a company is used), and the GEM vendor.
Table 12-10
MEASUREMENT RANGES FOR RELATIVE ACCURACY DETERMINATIONS
FOR
Dilute Monitor
Measurement
Point
1
2
Pollutant Monitors
20 to 30% of span value
50 to 60% of span value
co,
5to8%byvolune
10 to 14% by volume
o,
4 to 6% by volume
8 to 12% by volume
Summary
The requirements for performance specification tests for continuous gas emission monitors are contained
in Appendix B to 40 CFR Pan 60. These tests are performed on a one time basis (or as required by the EPA
pursuant to Section 114 of the U.S. Clean Air Act) in order to ensure that the GEMS are acceptable for
determining regulatory compliance by the source. There are performance specification tests for SO2 and NOs
(PS2), CO2 and O, (PS3), CO (PS4) and TRS (PS5).
Most of the test procedures for PS2 are the same procedures which are used for PS3, PS4 and PS5. These
procedures require the measurement of two parameters: 1) calibration drift (CD), and 2) relative accuracy (RA).
The calibration drift is determined by comparing the monitor's response to a known concentration of gas. The
relative accuracy is a percent difference between the measurement of the stack gas concentration obtained by
the monitor versus the same concentration obtained by using an EPA Reference Method. These Reference
Method procedures are contained in Appendix A to Pan 60. In some cases, an alternative relative accuracy
procedure may be approved by the EPA for sources which emit below the applicable standards. This procedure
applies to S02, NO,, CO2 and O2 OEMS.
12-16
-------
Performance specification test #2 describes the methods used in installing the GEM and when/how
Reference Method measurements are taken. Both the GEM measurement location and the Reference Method
measurement location need not be the same. The Reference Method measurements are taken by traversing the
stack or duct cross-section at three points. Stratification involves the mixing of two dissimilar gases and can cause
significant differences in the amount of pollutant gas measured at the Reference Method measurement locations.
Preparing for the performance specification test at a source facility requires the development of a written work
plan and cooperation between the company, control agency, the test contractor (if applicable) and the OEM vendor.
Finally, the reader is advised to obtain copies of both Appendix A and Appendix B to 40 CFR Pan 60 in order
to become familiar with the methods and requirements contained in both of these Appendices.
12-17
-------
REVIEW EXERCISES
1. True or false. Appendix Bis contained in 40 CFR Part 60 and provides
alternate reference method procedures which can be used during a
performance specification test
2. True or false. Performancespecification tests are used
to evaluate a OEM's performance over an extended
period of time.
1. False
3. True or false. Unless otherwise stated, the procedures delineated in the
sections to performance specification test #2 are applicable to perfor-
mance specifications test #3-5.
2. False
4. The term Span Value refers to:
a. The upper limit of a gas concentration measurement range
specified for affected source categories in the applicable subpait
of the regulations.
b. The lower limit of a gas concentration measurement range
specified for affected source categories.
c. The Upper and Lower Limit of a gas concentration measurement
range specified for affected source categories in the applicable
subpait of the regulations.
d. Can be all of the above.
3. False
5. True or false. An alternative mathematical expression for determin-
ing the relative accuracy may be used whenever the actual emissions
from a source are < 50% of the emissions standard.
4. a
6. What section of performance specification test #2 of Appendix B,
describes the specifications for installling a CEM system.
5. True
12-18
-------
7.
"Stratification":
a. results from the mixing of two dissimilar gases immediately
upstream of the measurement location.
is the non-uniform distribution of pollutant gas along a given
length of a stack or duct
refers to the amount of variation in pollutant emission rates at
the Reference Method measurement locations.
Both a and c.
b.
c.
d.
6. Section 3-D
8. The minimum period of time required in testing a continuous
emission monitor for calibration drift is:
a. 8 hours
b. 24 hours
c. 14 hours
d. 168 hours
7. d
9. True or false. The equation used for determining relative accuracy
can be found in Section 8.0 of performance specification Test #2
Appendix B.
8. d
10. The terms used to calculate the relative accuracy of a
continuous monitor are:
a. the algebraic mean difference between the recorded concentra-
tions of the monitor and those as determined by the Reference
Methods.
b. the 95% confidence interval associated with the
determination of the mean difference.
c. the mean concentation as
determined by the Reference Methods.
d. all of the above.
9. True
10. d
12-19
-------
REFERENCES
1. Code of Federal Regulations, Title 40 Pan 60 Appendix B. PerfbnnanceSperifications.
2. Environmental Protection Agency, 1986. The Quality Control Handbook for Air Pollution Measurement
Sterns Volume HI. Stationary Source SpecificMethods. Section 3.O.9., Continuous EmissionMonitoring
(CEM) Systems Good Operating Practices. EPA-600/4-77-02b.
3. Environmental Protection Agency. 1983. Guidelines for the Observations of 'Performance Specification
Test of Continuous Emission Monitors. EPA 625/6-79-005.
4. EnvironmentalProtectionAgency. 1982. Gaseous ContinuousEmissionMonitoring Systems' Performance
Specification Guidelines for S02,NOx, CO2,02,andTRS. EPA-450/3-82-Q26.
5. Electric PowerResearchlnstitute. 1988. Continuous Emissions Monitoring Guidelines: Update.EPRl CS-
5998.
6. Paley, Lit 1980. Agency use of Continuous Emission Monitoring Data to Ensure Stationary Source
Achievement of Emission Reductions. Air Poll. Control Assoc. Meeting Paper. Montreal 80-70.1.
7. Peeler, J.W. 1983 A Compilation ofS02 and NOf Continuous Emission Monitor Reliability Information.
EPA-340/1-83-012.
8. Winbeny, William T. Jr. 1985. Technical Assistance Document for Monitoring TotalRediiced Sulfur (TRS)
from Kraft Pulp Mills. EPA-340/1-85-0139.
12-20
-------
LESSON 13
The Requirements of Appendix F
Quality Assurance Requirements
for
Continuous Gas Emission Monitors
Lesson Goal and Objectives
Goal
To describe the requirements of the Code of Federal Regulations, Title 40 Pan 60 Appendix F, Procedure 1
• Quality Assurance Requirements for Continuous Gas Emission Monitoring Systems.
Objectives
At the end of this lesson you should be able to -
1. define the two functions of the quality control (QQ procedures required in Procedure 1 of Appendix F,
2. describe the minimum QC requirements which source owners must implement in their continuous
emission monitoring program,
3. define the criteria for determining excessive drift in a continuous emission monitoring system
(GEMS), and
4. describe the three acceptable methods for performing Data Accuracy Assessments.
Introduction
In the last lesson, we learned about the regulatory requirements of Appendix's (Specification Test Procedures)
of Part 60 of Title 40 of the Code of Federal Regulations. In this lesson, we will explore the requirements in
Appendix? of Part 60. AppendixFis entitled "AppendixF- Quality Assurance Procedures, Procedure 1 - Quality
Assurance Requirements for Gas Continuous Emission Monitoring Systems for Compliance Determination" and
was promulgated at 52 FR 21007 on June 4,1987. These procedures apply to steam generating units which are
13- 1
-------
subject to the requirements of 40 CFR 60 Subparts Daand Db. Appendix F requires that the source implement
written quality assurance procedures, states the acceptable values in performing calibration drift assessments
and data accuracy assessments, and gives the requirements for reporting the data to the regulatory agency.
The requirements of Appendix F are summarized in Figure 13-1.
FIGURE 13-1 AN OVERVIEW OF THE REQUIREMENTS OF APPENDIX F
QAPKOGRAM
1. MUSTDJCLUEE WRITTEN PROCEDURES
CALIBRATION DRIFT
1. CHECK. RECORD. AND QUANHFYDAILY
2. MUST REVISE QC PROCEDURES TO
ENSURE ANY CEM DEHOENOES THAT
CAUSE EXCESSIVE INACCURACIES ARE
CORRECTED
2- TAKE CORRECTIVE ACTION AND REPEAT
CD CHECKS IF THE CEM FAILS TO MEET
CALIBRATION REQUIREMENTS (DATA
TAKEN DURING PERIODS THATTHE CEM
FAILS CALIBRATION REQUIREMENTS
CANNOT BE USED FOR COMPLIANCE
PURPOSES.)
APPENDIX F
(40 CFR 60)
CEM AUDIT
1. ONCE EACH CALENDAR QUARTER
2. RATA ONCE EVERY 4 QUARTERS
RETENTION OFDATA
1. RETAIN ALLMEASUREMENTDATA ON FILE
FOR 2 YEARS
3. RAA OR CGA MAY BE PERFORMED 3 OUT
OF4 CALENDAR QUARTERS
4. IF AUDrr CRITERIA ARE NOTMET. TAKE
CORRECTIVE ACTION AND REPEAT
DATA ASSESSMENT SHORTS
1. REPORT ACCURACY AND DRBTINFORMATION
2. SUMMARlZEANYCORRECTIVEACnONTAKEN
In this lesson, we will take a look at each of the following sections of Appendix F:
• Applicability and Principle (Section I),
• Definitions (Section 2),
• Quality Control Requirements (Section 3),
13- 2
-------
• Calibration Drift (CD) Assessment (Section 4),
• Data Accuracy Assessment (Section 5),
• Calculations for CEMS Data Accuracy (Section 6), and
• Reporting Requirements (Section 7).
Applicability and Principle - Section 1
Section 1.0 , Procedure 1 of Appendix F, sets forth procedures for evaluating the effectiveness of quality
assurance/quality control procedures as well as evaluating the quality of data produced by CEMS that are used
for determining compliance with emission standards.
This section also states that the QC procedures required by Procedure 1 have two distinct but equally important
functions. The first function is assessment of CEMS data quality by estimating the accuracy of the monitor. The
second function is that of control and improvement of data from a continuous emission monitor (CENT) through
the implementation of QC policies and corrective actions. A control loop is formed through the interaction of these
two functions as follows; whenever the assessment function demonstrates that the data quality is inadequate, the
control effort has to be increased until the quality of the data is acceptable.
Section 1 also specifies the methods for assessing response drift and accuracy of the CEM and these methods
are based on methods found in Appendix B, Part 60. Procedure 1 allows source owners a great deal of latitude in
the development of QC procedures in order for these to be as effective and efficient for source specific situations.
In some circumstances, source owners may specify certain sections of the CEMS instrument manual (i.e.. the
manual which is published by the manufacturer) as their written procedures.
Definitions - Section 2
Section"! pro\adestriedefiriitiorisofseveraltennsardccntmuouseinissionmonitor(CEM)pararneters. They
are:
1. continuous emission monitoring system;
2. diluent gas;
3. span value;
4. zero, low-level and high-level values;
5. calibration drift (CD); and
6. relative accuracy (RA).
13- 3
-------
The definitions for GEMS, calibration drift, relative accuracy and span value are the same as those contained
in Appendix B of Pan 60. These terms were discussed in the last lesson. A "diluent gas" is defined as a major gaseous
constituent in a gaseous pollutant mixture. C02 and 02 are the major gaseous constituents of interest for combustion
sources. "Zero, low-level and high-level values" refer to the CEMS response value as it relates to the source-specific
span value. Procedures for determining these values are presented in the appropriate performance specification tests
in Appendix B.
Quality Control Requirements - Section 3
Section 3 requires source owners to develop and implement a quality control program which, as a minimum,
must include detailed procedures for the following activities:
1. calibration of the CEMS;
2. calibration drift determination and adjustment of the CEMS;
3. preventive maintenance of the CEMS (which includes maintaining a spare pans inventory);
4. data recording, calculations, and reporting for emissions and QA data;
5. accuracy audit procedures which includes sampling and analysis methods; and
6. the program of corrective action for the malfunctioning CEMS.
The reader should note that Section 3 does not provide detailed discussion of these activities. The following
sections to Appendix F contain requirements for these activities:
• Section 4 • Calibration drift determination and adjustment of the CEMS, corrective action for the
malfunctioning CEMS, and data recording;
• Section 5 - Accuracy audit procedures and corrective action for the malfunctioning CEMS;
• Section 6 - Data recording and calculations; and
• Section 7 - Reporting requirements for emissions and QA data.
Since none of the sections in Appendix F specifically address calibration of a CEMS, we will briefly mention
this term. The calibration of a CEM refers to the adjustment of the instrument relative to specified gas standards
or independent effluent measurements. Written procedures for calibrating a CEM are required and each source may
develop its own procedures. As stated earlierinSecrio/i 1, applicable sections of an instrument's operating manual
may suffice for these written procedures. The reader should take note mat mere are no current regulations regarding
specific calibration frequencies or specific criteria for initiating calibration procedures. Therefore, sources should
implement their own procedures for determining the calibration frequency and criteria for calibrating a CEM based
on operating experience or manufacturer's recommendations.
13- 4
-------
Preventive maintenance is also not addressed in detail in Appendix?. Preventive maintenance consists of such
activities as routine maintenance and repairs and instrument adjustments performed on an as-needed basis. We
will talk more about maintenance for continuous gas emission monitors in Lesson 15.
Finally, Section 3 states that if excessive inaccuracies in data occur for two consecutive quarters, then the
source owner must either revise the current written procedures or modify or replace the CEMS in order to correct
this deficiency. Written QC procedures must be kept on file and must be available for inspection by the regulatory
agency. In the next lesson, we will talk about quality control in more detail.
Calibration Drift (CD) Assessment - Section 4
This section discusses the requirements for performing the calibration drift (CD) assessment tests. As we
learned in Lesson 12, "calibration drift" refers to the difference between the CEMS output reading and a reference
value after a period of operation during which no unscheduled maintenance, repair, or adjustment took place. Part
60.13 of Title 40 of the Code of Federal Regulations, requires that daily zero (or low-level) and span drift checks
be performed. These same two checks are also used to ftjlfiU me CD assessment r^ Written
procedures are required for specifying how the zero and span calibration drift determinations will be performed.
Also, these procedures must be consistent with the methods prescribed by the CEM vendor.
Table 13-1 is a compilation of the performance specification criteria and the required response by the source
owner or operator if the zero and span calibration drift exceeds the allowed limit It should be noted that a source
may develop response requirements more stringent than those as stated in this table. The written procedures for
the performance of calibration drift, should incorporate the criteria for adjusting the CEMS should the calibration
drift test show that the instrument is out of line. Making corrections for excessive drift involves any type of
Table 13-1
CEMS CALIBRATION DRIFT CRITERIA
PARAMETER
CRITERION*
ACTION REQUIRED
Zero (or low)
level calibration
drift
CD>2x(Spec)"
CD> 2 x (Spec) for 5
Bonifcntive 24-hour •
periods
CD>4x(Spec)
Adjust CEMS for calibration drift
CEMS out-of control period begins at end of 54 diy the CD
exceeds 2 x (Spec); perform corrective action and repeat CD check
CEMS ovMii'CCuiiul peiiud hryni at ihe 0306 tmcipondmg to
die completion of die Ian acceptable CD check preceding the CD
check which exceed* 4 x (Spec); perform corrective action and
repeat the CD check
drift
CD>2x(Spec)**
CD>2x(Spec)for5
consecutive 24-hour
periods
CD>4x(Spec)
Adjua CEMS for calibration drift
CEMS onuof-control period begins at end of 5th day die CD
exceeds 2 x (Spec); perform corrective action and repeat CD check
CEMS out-of-control penod Pf£in* at the time 3Us!fidiB»y bepre«BB«d by
13- 5
-------
adjustments which the CEM operator deems necessary in order to correct the observed drift. Activities such as
routine checks and adjustments of calibration and sample gas flow rates, pressures, filter elements and moisture
removal systems; verification of the status of monitor-specific auxiliary monitoring parameters (i.e., sample cell
temperature and pressure, gas flow capillaries, detection optics and electronics, temperature compensation,
linearization signal output circuitry); and adjustments of zero and/or span potentiometers are included in the
procedure for compensating for excessive drift. The source owner should have written procedures for performing
drift adjustments and these procedures should include the criteria for demonstrating that the adjustments mat were
made are adequate.
If the zero flow-level) or the span (high-level) values exceed four times the applicable drift specification as
stated in Appendix B, the CEM is considered to be out-of-control. Whenever this happens, Appendix F requires
that corrective action be taken. The type of action taken will depend on the nature of the problem. Written
procedures are required to be available for addressing instrument start-up and trouble shooting. CEM instrument
operation and/or repair manual sections applicable to out-of-control problems will suffice for these written
procedures. It is also suggested that additional quality assessment procedures be written for the purpose of
verifying that the CEMS is in control following any repair or adjustment Alternative methods used for monitoring
emissions during the out-of-control period should be listed in these quality assessment procedures. These
procedures should also list the office or individual responsible for handling an out-of-control event Such a listing
would include the personnel in charge of approving the corrective action to be taken, and procedures for
determining when alternative methods for emission monitoring should be performed. The criteria for determining
that a CEMS is out of control must include the procedure from Appendix F for determining excessive drift and
excessive inaccuracy.
Section 4 of Appendix F also states requirements for CEMS mat have automatic adjustments for resetting the
CD. Such systems must be programmed to record the unadjusted concentration measured in the CD prior to
resetting the calibration if performed, or record the amount of adjustment (e.g., a microprocessor control would
do this).
Lastly, Section 4 states that all measurements from the CEMS are required to be retained on file for at least
two years as required in 40 CFR Part 60.7(d). However, emission data recorded on each day that the CEMS is
out-of-control need not be included in the daily data requirement of the applicable subpart in Pan 60.
Data Accuracy Assessment • Section 5
Appendix F became effective on December 4,1987 and the first required relative accuracy test was to be
performed by March 4, 1988 or by the date a GEMS initial performance test was required by the applicable
regulation, whichever date was later.
This section lists the three acceptable techniques which can be used for performing an audit on the CEM.
Audits are required to be performed at least once each quarter. The three acceptable techniques are:
1. Relative Accuracy Test Audits (RATA),
2. Cylinder Gas Audits (CCA), and
3. Relative Accuracy Audits (RAA).
13- 6
-------
In addition to these procedures, AppendixP states that the EPA may approve alternative procedures to be used
during three of the four calendar quarters. However, one RATA is required at least once every four calendar
quarters. The reader should note that currently, there have been no alternative procedures approved by the EPA
and only one procedure for an RAA test for in-situ systems is currently under review.
A CEMS which does not demonstrate adequate accuracy during any quarterly audit test is considered to be
out-of-control with corrective action necessary. As we learned in Section 3, if the CEMS does not demonstrate
an acceptable accuracy for two consecutive quarters, then either the QA program must be revised or the CEMS
must be replaced or modified. Table 13-2 delineates the specific requirements and performance criteria for each
of the three acceptable audit techniques. Appendix F requires that the QC program must have written sampling
and analysis procedures to be used during the required quarterly accuracy audits.
Table 13-2.
Requirements and Criteria for Appendix Ff Procedure 1 Audit Techniques
TECHNIQUE
Relative Accuracy
Ten Audit
Relative Accuracy
Audit
Cylinder Gu
Audit
REQUIREMENTS
Conduct as per applicable
performance ipecification (PS) io 40
CFR Pan 60 Appendix B,
2fcrSO and NO )
Analyze appropriate performance
audit ample* from EPA
Conduct as per applicable PS in
Appendix B except only 3 runs are
ventured •
Use relative difference between the
mtn nfritm't mtlhod vafura and
the mean of the CEMS nipcusM to
assets the accuracy of the CEMS data
Challenge both pollutant and dflutant
channel! (if applicable) of CEMS
three times at me two points cpecified
in Procedure 1
Use gases that have been certified by
companion to NBS SRM"f or
NBS/EPA approved |u
manufacturer's CRMt
Operate analyzer in normal sample
node
actual gat value and tJUBuailiatien
indicated by CEMS to access
accuracy
PERFORMANCE CRITERIA
Relative accuracy mutt not exceed 20%
or 10% of applicable standard.
whichever it greater
For SO, standards from 020 to 030to /
million Bui, relative accuracy must not
exceed 15% of the Handard
For SO, standards below 0.20 Ib/10'
Bui, relative accuracy mutt not exceed
20% of me standard
Inaccuracy must not exceed ± 15% or
7 .3% of the applicable standard.
whichever U greater
Inaccuracy must not exceed * 15%
WHEN
REQUIRED
Once every four
calendar quaaters
May be performed
in three of the four
calendar quarters
May be performed
in three of me four
calendar quarters
13- 7
-------
As stated earlier, Appendix F requires that these accuracy audits be performed once each calendar quarter.
Additionally, successive audits must occur no closer than two months apart and must be conducted as follows:
1. The RATA test must be performed once every four calendar quarters. This test is performed the same way that
the performance specification test #2 (for S02 and NO,) as outlined in Appendix B is done. However, Appendix
F also requires that EPA performance audit samples be taken sequentially (i.e., shortly before or after) with
the RATA samples. Both the EPA audit sample and the RATA sample must be analyzed by the same person.
The logic behind conducting these two tests sequentially is that the RATA will approximate the "truth" by the
Reference Method test results which are then checked for analytical accuracy by the EPA audit method.
The EPA audit samples are performed the same way that the applicable Reference Methods (e.g., Methods 6 for
S02 and 7 forNOJ as delineated inPart 60 Appendix A, areperformed. The EPA audit samples and the data samples
are concurrently analyzed in the same manner in order to determine that the technique of both the sample analyst
and the audit sample preparation are effective. The results of the EPA audit sample must agree within five percent
of the audit concentration on each of two S02 audit samples or within ten percent of each other on each of two NOx
samples.
2. Whereapph'(^le,aCX}Amaybeperfomiedmthreeofthefourcalendarquarters. Anexampleofasituationwhere
a CGA could not be performed would be in the case of apathin-situ monitor which could not undergo this test unless
a flow-through cell is used.
This audit is performed by checking the CEMS (both pollutant and diluent monitors, if applicable) with an audit
gas following EPA Traceability Protocol No. 1 (see citation in the Reference section). An audit gas must be
certifiable through comparison with a National Institute of Standards and Technology -Standard Reference
Materials (NIST-SRM) gaseous reference orNIST/EPA-approved gas manufacturers CRM Certified Reference
Material (CRM). This audit gas of a known concentration is used to check the CEM at two points within the
following ranges:
Audit range
Audit
1
2
Pollutant monitors
20 to 30% of span value
50 to 60% of span value
Dilute monitors for - •
CO,
5 to 8% by volune
10 to 14% by volume
02
4 to 6% by volume
8 to 12% by volume
It should be noted that a separate audit gas cylinder must be used for each point checked. This cylinder gas cannot
be diluted. Check the CEM at eachpointthreetimes and use the average response valueforcomputing the accuracy.
Also, each point should be sampled long enough to assure absorption-desorption of the CEMS sample transport
surfaces has stabilized. Audit each monitor in its normal sampling mode (i.e., allow the audit gas to pass through
all filters, scrubbers, conditioners as well as the other components used during normal sampling and as much of
13- 8
-------
the sampling probe as possible). The audit gas should be injected at the connection between the probe and the
sample line. The accuracy is the difference between the concentration of the audit gas and the concentration as
indicated by the monitor.
3. The RAA test may be performed three of the four calendar quarters. This test is performed the same way in which
the relative accuracy test of Appendix B is performed with two exceptions; 1) only three sets of measurements are
required, and 2) EPA performance audit samples are required to be analyzed concurrently with the RAA samples.
The EPA audit sample and the RAA test sample must be taken by the same person. The CEMS accuracy is
determined by the relative difference between the mean of the CEMS values Cm terms of the standard) and the mean
of the Reference Method results. That is to say, the audit sample verifies the accuracy of the Appendix B Reference
Method test used, then the Reference Method test is used to "audit" the GEM.
The data accuracy assessment is an important check of the CEMS ability to produce accurate emissions data
over time. One should not expect that accuracy will remain constant over each quarter because of changes in
calibration gases, analysts and environment
Calculations For CEMS Data Accuracy - Section 6
This section gives the types of calculations necessary to assess the accuracy of the CEM. For the RATA
relative accuracy calculation, the equations used are contained in Section 8 of Appendix B, Performance Speci-
fication Test 2. We talked about these equations in the last lesson. Before proceeding any further in this section,
the reader may wish to go back to Lesson 12 and review these equations. While the RATA and RAA must be
calculated in units of the applicable standard (e.g., IbTmillion BTU), the CGA must be calculated in units of the
appropriate concentration (i.e., ppm SO2or percent 02).
Equation 13-1 can be used by combustion sources to convert a pollutant concentration to units of the
applicable standard as follows:
(Eg. 13-1) E = CF
20.9
20.9 - percent O2
where
E = pollutant emission, ng/J (Ib/mfllion Btu),
C = pollutant concentration, ng/dsm3 (lb/dscf)*
F = factor representing a ratio of the volume of dry fuel gas generated to the caloric value of
the fuel, dsmVJ (dscfymillion Btu),
20.9 = a constant which refers to the amount (percent) of 02 in the ambient air.
Percent O2 = oxygen content by volume (expressed as percent), dry basis in the sample stream.
* dsm = Dry Standard Cubicmeter
dscf = Dry Standard Cubic Feet
13- 9
-------
Data from CEMSs arc sometimes collected on a wet basis. When comparing the CEM data with the
Reference Method data it is necessary to convert the wet basis data to a dry basis. This conversion can be done
as shown in equation 13-2:
XPP».«I
(Eg. 13-2) :
where:
X = Concentration from the CEMS
Bw = Moisture fraction of the CEMS gas sampled. This value is determined by using
Part 60 Appendix A, Method 4.
Part 60, Appendix A, Method 19 presents a complete explanation of equations 13-1 and 13-2.
As we stated earlier, the RATA relative accuracy is determined by using the equation contained in Section
8 of Appendix B, Performance Specification Test 2. The RAA test accuracy and the accuracy of the CEM per
the CGA method are calculated by using the same equatioa Equation 13-3 is the mathematical expression
used for determining both the RAA and the CGA and is presented as follows:
(Eq. 13-3) C. - C.
A = x 100
where:
A = The accuracy of the CEMS (percent),
Cm = average CEMS response during auditin units of applicable standard, (for RAA).
or
C. = average audit value (CGA certified value or three - run average for RAA) in
units of the applicable standard or appropriate concentration.
Reporting Requirements - Section 7
The results of each CEMS accuracy audit must be reported in the form of a Data Accuracy Report (DAR).
After each quarterly audit, one copy of the DAR must be submitted to the state regulatory agency or in some
cases the EPA Regional Office along with a report of emissions required under the applicable regulation (i.e.,
the requirements of subpans Da and Db in Part 60). The minimum information required to be submitted with
a DAR is as follows:
1. Source owner or operator name and address.
2. Identification and location of monitors in the CEMS.
•3. Manufacturer and model number of each monitor in the CEMS.
4. Assessment of CEMS data accuracy and date of assessment as determined by a RATA, RAA or CGA,
including the relative accuracy for the RATA, the accuracy for the RAA or CGA, the Reference
Method results, certified values for the cylinder gases, the CEMS responses, and the CEMS accuracy
13- 10
-------
calculation results. If the accuracy audit results show the CEMS to be out-of-control, the CEMS
operator shall report both the audit results showing the CEMS to be out-of-control and the results
of the audit following corrective action showing the CEMS to be operating within specifications.
5. Calibration drift (CD) assessment results.
6. Results from the EPA performance audit samples.
7. Summary of all corrective actions taken when the monitor was determined out-of-control.
Figure 13-2 is an example of what a DAR form looks like.
Figure 13-2 Example format for data assessment report (DAR).
Period ending date . Year
Company name
Plant name Source emit no..
CEMS manufacturer ' Model no.
CEMS serial no. CEMS type (fcg, la-titu)
CEMS sampling location (fcg., control device outlet)
CEMS span values as per the applicable regulation, SO, ppm
O, percent, NO, ppm, CO, percent
L Accuracy asseismentreiultt (Complete A. B, or C belowfor eack CEMS or for each pollutant and diluent analyzer, as applicable.)/fine
q uarterly audit results show the CEMS to b* out -if- control, report the netnils of both the q uanerty audit and the audit followinf the eorrtaivt action
showing the CEMS to be operating property.
A. Relative accuracy tea audit (RATA) far _______________________^^_^^_ (e.g,.SO1inng/J).
1. Date of Aodii .
2. Reference methods (RM's) used * (e-g.. Methods 3 and 6).
3. Avenge KM value (e.g., ng/}, mg/don9, or percent volume).
4. Avenge CEMS value .
5. Absolute value of the mean difference Hi_
6. Confidence coefficient ICO
7. Percent relative accnncy (RA) percent
8. EPA performance audit results:
a. Audit lot number (I) (2)
b. Audit sample number (1) C2)
c. Results (mg/dsm>) (1) (2)
d.Acmal value (rng/dsm1)* (1) (2)
e. Relative error* (1) (2)
B. Cylinder gas audit (CCA) far (tg.. SO, in ppm).
1. Date of audit .
Audit Audit
point 1 point 2
2. Cylinder ID number
3. Date of certification ^—— -^——^^.^
4. Type of certification (e-g. .EPA Protocol 1 or CRM).
13- 11
-------
Figure 13-2 (Continued)
5. Certified audit vtliie
6. CEMS response vilne
7. Accuracy
C Relalnt accuracy audit (RAA) for
1. Dale of radii
2. Reference methods (RM'i) used
3. Avenge RM value
4. Avenge CEMS value
5. Accuracy percent.
6. EPA performance audit results:
a. Audit lot number 0 )
b- Audit ufnple number (1)
c. Results (mg/dsm1)* (1 )
d. Actnal value (mg/dsm')* (1)
e. Relative error* (1)
D. Corrective action for excessive inaccuracy.
1. Out -of -control period*.
^ D«u(.}
h. Nmpho- of &*y*
2. GorPB£*WC >o>cin tflken
(e.g.. ppm).
fee., pom).
perocoii
(e.g..SO,inng/J).
(e.g.. Methods 3 and 6).
fej-na/H.
m
m
C)
C)
(2)
3. Results of audit following corrective action. (Use format of A, B, or C above, as applicable.)
U. Calibration drift assessment.
A. Om-of -control periods.
l.D»te(i) . _ _
2. Number of days
B. CoiTEflive maawt ukep
To be eampUttd try Out Agency
Summary
The requirements of Pan 60 of Title AOAppendixF, Procedure 1 apply to steam generating units listed in Pan
60 Subparts Da and Db. Appendix F requires that these sources implement written quality control procedures for
evaluating the quality data produced by continuous emission monitoring systems. In particular, the source is
required to perform calibration drift (CD) tests (i.e., daily zero and scan value checks.) Both the zero and span
values cannot exceed the applicable drift specifications as stated in Appendix B to Pan 60 and the measurement
data must be retained on file for two years except in cases where the GEM is deemed out-of-controL
The source is also required to perform quarterly audits of the GEM. The three acceptable audits which can
be used are the relative accuracy test audit (RATA), which must be performed at least once ever four calendar
quarters, the relative accuracy audit (RAA), and the cylinder gas audit (CGA). Both the RAA and CGA may be
13- 12
-------
performed in three of the four calendar quarters. The RATA and RAA are performed in the same manner which
the relative accuracy test as required by Part 60 Appendix B are performed. The RAA test only requires that
three sets of data be used as opposed to taking nine sets of data which Appendix B requires. Additionally, EPA
audit samples must be analyzed sequentially with the RATA and/or RAA test samples. The CGA test is
performed by checking the CEM at two point within specific ranges with an audit gas.
The results of CEMS accuracy tests must be recorded on a Data Accuracy Report (DAR) form each quarter
and a copy of this form along with a report of source emissions required under Subparts Da and Db must be
submitted to the regulatory agency.
13- 13
-------
REVIEW EXERCISES
1. True or false. The two distinct but equally important functions of the QC
procedures required by Appendix F are to 1) assess the quality of the
CEMS data by estimating the monitors accuracy, and 2) to control and
improve the quality of the CEMS data through the implementation of
QC policies and corrective actions.
2. Trueorfalse. The inclusion of a spare parts inventory is one of the several
QC requirements which source owners must implement in their CEMS
program.
l.True
3. True or false. Appendix F states that if excessive inaccuracies in data
occur for two consecutive quarters, then the source owner shall replace
the CEMS.
2. True
4. In performing the daily calibration drift test for both zero and span drift,
the CEMS is considered to be out-of-control if:
a. The CD > 4x the allowable limit
b. The CD > 2x the allowable limit
c. The CD > Sx the allowable limit
d. The CD > lOx the allowable limit
3. False
5. True or false. The RAA Test requires that only 3 sets of measurements
betaken.
4. a
6. Trueorfalse. A CGA Test may be performed in any of four calendar
quarters.
5. True
7. The audit gas used in performing the CGA Test is used to check the CEM
at point(s) within the specified ranges.
6. False
8. The RAA Test may be performed during of the four calendar
quarters.
7. Two
-------
9. The performance criteria for the RATA Test requires that the RA must g. Three
not exceed or of the applicable standard, whichever is
greater.
a. 20% or 10%
b. 10% or 15%
c. 20% or 15%
d. 5% or 10%
10. True or false. The equation used to convert a pollutant concentration
to units of the applicable standard is as follows:
9. a
E=CF
T 20.
20.9 - pe
percent
The percent 02 in this equation is the oxygen content by volume
(expressed as percent), wet basis in the sample stream.
10. False
13- 15
-------
REFERENCES
1. Code of Federal Regulations, Title 40 Part 60, Appendix f- Quality Assurance Procedures.
2. Environmental Protection Agency. 1987. Appendix F - Quality Assurance Procedures. Procedure 1
— Quality Assurance Requirements for Gas Continuous Emission Monitoring Systems used for
Compliance Detenninatioa Summary of Comments and Responses. EPA/450/3-87/D09.
3. Environmental Protection Agency. 1985. Calculation and Interpretation of Accuracy for Continuous
Emission Monitoring System (GEMS). Section 3.0.7 of the Quality Assurance Handbook for Air
PoUutionMeasurementSystems, Volume fflStationary Source SpecificMethods.EPA-600/4-77-0276.
4. Environmental Protection Agency. 1986. Continuous Emission Monitoring (CEM) Systems Good
Operating Practices. Section 3.0.9 of the Quality Assurance Handbook for Air Pollution Measurement
Systems, Volume H Stationary Source Specific Methods. EPA -600/4-77-0276.
5. Environmental Protection Agency. 1985. Guideline for Developing Quality Control Procedures for
Gaseous Continuous Emission Monitoring Systems. Section 3.0.10 of the Quality Assurance Handbook
for Air Pollution Measurement Systems, Volume n Stationary Source Specific Methods. EPA -600/4-
77-0276.
6. Environmental Protection Agency. 1978. Traceability Protocol for Establishing True Concentrations of
Gases used for Calibration and audits of Continuous Source Emission Monitors (Protocol Number 1).
Section 3.04 of the Quality Assurance Handbook for Air Pollution Measurement Systems, Volume ID.
Stationary Source Specific Methods. EPA -600/4-77-0276.
7. Electric Power Research Institute. 1988. Continuous Emission Monitoring Guidelines: Update. EPRI
CS-5998
13- 16
-------
UNITS
CONTINUING OPERATIONS
-------
LESSON 14
Quality Assurance/Quality Control and Audit Programs
for
Continuous Gas Emission Monitors
Lesson Goal and Objectives
Goal
To gain an understanding of the basic components of a quality assurance/quality contol (QA/QQ program
and the importance of an audit program.
Objectives
At the end of this lesson you should be able to -
1. define the following terms:
• quality control • quality assurance
• audit • intralaboratory audit
• interlaboratory audit • schedule system inspection program
• level system inspection program • phase audit procedure
• level audit procedure
• 2. Describe quality assurance procedures associated with purchase and installation of a continuous emission
monitoring system (CEMS),
3. list objectives of a QA program for continuous emission monitoring,
4. define the three levels of a QA program and the three stages of a QC program for CEMSs, and
5. describe the types of audits used in a QA program.
Introduction
In the last three lessons, we have learned about some of the regulatory requirements involved with the use of
continuous gas emission monitors. One way to ensure that a continuous emission monitor (CEM) is maintained
in a manner which provides assurance to the source owner that the system is in regulatory compliance, is through
14-1
-------
the establishment and implementation of a good, sound quality assurance program. In the past, much confusion
has come about regarding the application of quality control methods to CEMSs. This was because there were three
groups involved with these types of procedures. These three groups were:
• the regulatory agency which required the air emissions monitoring,
• the source owner or operator responsible for purchasing and installing the CEM, and
• the CEM manufacturer.
Although each group wants the system to perform at a satisfactory level, there is some uncertainty about each
participant's role in ensuring the success of the CEM.
It is always important to remember that unless the source has entered into a service contract with a vendor
the CEM belongs to the source owner. However, once the unit has been purchased from the vendor, that party
has lost control of it Also, the regulatory agency has very little control over the unit even though the agency
required the source to install it Normally, the source owner will not allow the vendor or the agency to touch the
unit unless they have been contracted. It is therefore the responsibility of the source to design and implement a
satisfactory quality assurance/quality control program which will enable the source to operate the CEM to the
satisfaction of the regulatory agency, meeting the manufacturer's specifications. Quality control refers to ac-
tivities which will provide a quality product and quality assurance refers to activities which will provide assurance
that the quality control program is performing in a satisfactory manner.
There are three phases involving the development of a continuous emission monitoring program. They are:
1. purchasing an instrument that meets design specifications,
2. installing and testing the performance of the new instrument, and
3. ensuring continued operation of the monitoring system in accordance with the specifications and
regulatory requirements.
Figure 14-1 shows a basic overview of these three phases. For me first two phases, there are quality assurance
procedures and techniques which can be applied in order to ensure that these phases are performed adequately.
For the third phase, two types of quality assurance programs are involved. The first type of program is designed
by the source to ensure the continuous operation of the CEM as well as to audit the system. The second type of
program is designed by the regulatory agency in order to inspect the CEM. This lesson will provide an overview
of these quality assurance/quality control programs and will provide a discussion of audits which are a necessary
element in ensuring that the QA/QC program is performing up to standards and the CEM is displaying accurate
and reliable data.
Quality Assurance Involving the Purchase and Installation of Continuous Gas
Emission Monitors
Procuring a CEM must be done on a case-by-case basis. This is because a CEM which functions well at one
plant may not be suitable at another facility due to the differences in the design parameters at each of the two
14-2
-------
facilities. In purchasing a CEM, there are three areas which should be incorporated into a quality control program.
They are:
• prepurchase evaluation/selection,
• writing of the purchase contract specification, and
• recordkeeping.
A discussion of these three areas follows below.
Prepurchase Evaluation/Selection
CEMs which are available on the
market should be compared according
to instrument specifications, evalu-
ated according to their operational
design with respect to the installation
site, and compared by contacting
owners of systems which are already
installed and field tested if possible
(this is recommended for multiple
purchases). An instrument which does
not meet the required specifications is
probably not designed for regulatory
purposes.
FIG 14-1 TYPES OF QUALITY ASSURANCE PROGRAMS FOR
CONTINUOUS MONfTORING SYSTEMS
CEM OA PROGRAMS
PROCUREMENT
QUALITY CONTROL
OA-PERFORMANCE
BPEOnCATIONTESTI
OA-CONTMUDM
OPERATION
In making comparisons of opera-
tional designs, methods used in per-
forming zero and calibration checks,
vulnerability to vibration and tem-
perature extremes, and data process-
ing options should be considered in
selecting a system based on the in-
stallation site, the data requirements
and plant management systems.
Source owners who are currently
using CEMs should be consulted both
for their experiences and suggestions
about the equipment This informa-
tion is quite useful in evaluating
available systems even though differences in installation may result It is also advisable to field test a system if
the source wishes to make a multiple purchase. If a source were to purchase a dozen or so units at a reduced cost
(a SQ called package deal) and, after installing the first unit unresolvable problems were to result the source would
be stuck with a number of units, all which would be incompatible for use at that facility. At that point it would
be difficult to back out of an agreement with the vendor.
14-3
-------
Writing the Purchase Contract
There are several features which should be incorporated into a contract with the vendor
1. the performance specifications for the selected CEM should be guaranteed (i.e., the manufacturer's test data
which demonstrates that the system meets the required design and performance test specifications should
be included in the contract);
2. the payment or a substantial pan of the payment is not due until the CEM passes the performance test;
3. a minimum of a one-year warranty should be given which should not start until the CEM has passed an
acceptance test conducted by the purchaser,
4. the operating manual supplied with the CEM must be consistent with the equipment purchased;
5. operator training should be provided by the vendor,
6. the vendor of the CEM must subject the monitor to a bum-in procedure to detect immediate malfunctions
due to defective components or poor assembly, and
7. a two-year supply of spare parts and analyzer consumables (listed in the operating manual) should be
furnished with the purchase.
Such specifications as shipping costs, service fees, and manpower requirements can be included in the contract
but these specifications are usually verbally negotiated.
Recordkeeping
In the area of recordkeeping, a file system should be maintained which covers the entire lifetime of the CEM.
Such items as vendor literature, phone logs, meeting notes and financial records should be included in this file. Also,
the results of performance specification tests, copies of quarterly excess emission report, and instrument log books
should be kept on file so that if there is ever a change in personnel, or a failure of memory, the company will have
on file documentation of the experiences gained from the use of the system.
Quality Assurance for Performance Specification Tests
The design specifications for a newly purchased CEM are normally checked by the CEM vendor and a
contractor usually performs the Performance Specification Test Both of these parties should have an established
14-4
-------
quality assurance program for doing these tests. Unless the CEM is not functioning properly, the Performance
Specification Test (which gives specific data and reporting requirments should) yield good quality results.
Quality assurance techniques which are not specifically addressed in the specification tests but which may be
used as pan of the testing program are as follows:
1. strip chart recorders may be run at a faster rate during the test to aid the operator in reading the data;
2. after a series of zero and span readings have been taken, both plant and agency representatives can initial
the results in order to demonstrate agreement between the two parties;
3. a separate (portable) monitoring system can be run in parallel with the system undergoing the test to aid
in verifying the data produced by this system; and
4. audit devices can be used before and after the performance tests to uncover any inconsistences in the
system.
Techniques such as these can be used by the instrument operator, the independent contractor (if one is
employed) or the CEM vendor if they are present during these tests.
Quality Assurance for Continuing Operation and Maintenance
Quality Assurance versus Quality Control
So far, we have used the terms quality assurance and quality control interchangeably. Both fall under the
overall quality assurance program for a source. Figure 14-2 shows an overview of such a program. Quality
control involves "Activities" which will provide a quality product (i.e., adequate CEM data) while quality
assurance involves "actvities" which will ensure an adequate quality control program.
Objectives of a Source Quality Assurance Program
What then, is the primary objective of a quality assurance program? The primary objective of such a program
is to ensure that all data collected are precise, meaningful and accurate within stated acceptance criteria. A quality
assurance program should be rigid enough so that all established procedures are strongly adhered to and flexible
enough to allow a continual evaluation of the adequacy and effectiveness of the implemented procedures. The
objectives of a quality assurance program can be summarized as follows:
• to monitor routine performance of personnel and/or equipment;
• to provide for prompt detection and correction of conditions that contribute to the collection of poor
quality data, and
• to collect and supply information necessary to describe the quality of the data.
14-5
-------
A good sound quality assurance program must contain the following elements:
• routine training and/or evaluation of operators,
• routine monitoring of the variables and/or parameters which may have a significant effect on data quality,
• development of techniques to detect defects,
• development of methods and written procedures to qualify data, and
• action strategies to increase the level of precision in the reported data and/or to detect equipment defects
or degradation in equipment performance.
I
Figure 14-2
Source Specific Quality Assurance Program
Quality Assunnce Program
Quality Assurance
Quality Control
Daily Zero/
Span Check
Control Chatti
Development
Detenuiiiaticfi of
Monitor "Out of
Control"
Relative
Accuracy
Audit (RAA)
Cylinder Gas (CCA)
Quarterly
Audilf
Control Chart!
Development
Audit Samples
Generated by
G«s Cylinders
Cometh*
Action
Procedures
Calibration
ofCEMS
Calibration
Drift Deter-
Prevenuve
Maintenance
Data Recording
MunnctioD
•*=•
Overall, quality assurance program involves the incorporation of three general categories. These categories
are; 1) management, 2) measurement and documentation, and 3) data reduction, validation, analysis and statistics.
Let us take a moment and comment on each of these categories.
14-6
-------
Management
The first category pertains to the development of written policies concerning the use of quality assurance
procedures as a part of the management effort. Also, any administrative procedures relevant to all measurement
systems would be incorporated in this category. These procedures should comply with current EPA regulations.
QA reports and graphics which would demonstrate data quality of CEMs should be summarized on a routine basis
in order that the QA coordinator may distribute them for internal review. Cost-effectiveness considerations for
obtaining high quality data at the least cost should be incorporated here as well.
Measurement and Documentation
The second category to consider in developing a QA program, measurements and documentation includes a
QA manual which delineates step-by-step procedures for the following: CEM start-up, calibration, internal QA
checks on each CEM, performance audits, system audits, calibration standards certification, CEM certification,
data validation, maintenance (non-routine and scheduled) and intralaboratory quality assessment activities.
Procedures which incorporate the requirements of Appendix?, Procedure 1 should be a part of this document
A document control system should also be implemented. Such a system would consist of the identification
of and revisions to all QA documents associated with CEMS. A numbering system should be used in order to
identify all log books used. Types of log books used should include the following: monitor log books, procurement
log books, calibration standards log books, laboratory log books, excess emission log books, daily monitor log
books, and periodic audit log books. Also the document system should provide for maintaining records of
monitoring equipment (by manufacturer's serial number and monitor location), spare parts inventory, the status
of all CEM components (e.g., return to vendor, discarded, under repair, etc.) and all QA audit standards (National
Institute of Standards and Technology (NIST) traceable, recertified date). Finally, the source owner should
develop a document distribution system that provides the most current procedures to concerned personnel for
review and implementation.
Data Reduction, Validation, Analysis and Statistics
Finally, the third category incorporates the methods for data reduction and calculations so that the data can
be represented in units of the applicable standard. The use of statistics in a quality assurance program is essential
in that it provides assurance that the CEM is performing to the required specifications and that the data obtained
is valid. These statistics are also used to develop control charts which are then used to detect and assess long-term
trends. From these trends, such items as out-of-control criteria limits and confidence limits used to identify any
significant changes from past CEM performance are established.
Developing and Implementing a Quality Assurance Program
In developing the QA program, the following questions need to be answered:
• How does the CEM system work?
• What possible things can go wrong?
• How does one determine that something has gone wrong?
14-7
-------
The person developing the QA program has various sources to consult These sources include the operation
and maintenance manuals, the service department of the vendor who supplied the CEM, technical personnel who
work for the source and who are familiar with the system, and outside contractors who have experience with such
systems. Remember that a quality assurance program is source specific and that a program developed for one
source may not be applicable to another source *s needs. The developer of the Q A program will need to determine
which CEM parameters need to be tested and how often these tests should be performed. Several issues must be
taken into consideration when designing test programs:
• the probability of the CEM parameter changing,
• the impact such a change would have on the quality of the CEM data,
• the degree of difficulty that would be encountered in checking the CEM parameter, and
• any additional reasons for documenting specific CEM parameters.
Once these issues have been resolved, the next step is to determine at what frequency certain CEM parameters
should be checked. Certain parameters will require testing on a daily or more frequent basis while others may be
checked periodically. It is probably best to divide these frequencies up into weekly, monthly and quarterly checks
and then determine which frequency category each parameter test should belong to.
In determining the optimum frequency for performing various CEM activities, it is best to implement a trial
period whereby the CEM parameter would be checked more often. After a period of time, the source would have
a large enough data base for use in determining this optimum frequency.
It is also a good idea to get representatives of each department which would be involved in or affected by the
CEM program, active in the development of the QA program. The input and suggestions received by the various
departments will be most useful in developing and implementing a successful QA program.
Once the QA program has been drafted and is approved, it should be implemented on a trial basis. In doing
so, such problems as wasted motion, inadequate documentation and confusing instructions which were not
apparent during the development of the QA program, will be clearly recognized.
Finally, a QA program needs to undergo a periodic review either once a year or twice a year to determine any
necessary changes in the program which would reflect changes in the CEM's reliability due to instrument age or
other factors, changes in repair personnel, changes in manpower availability or source operating practices. Such
a review would include an inspection of the daily QA checks, periodic checks and corrective action logs. This
review would act to fine tune the frequency with which the various CEM parameters would need to be tested. For
example, if several months of data indicated that certain daily checks were shown to be quite stable, these checks
could then be performed on a less frequent basis.
14-8
-------
Overview of a Quality Assurance Document
Figure 14-3 gives an outline of a written quality
assurance program document for stationary source test-
ing. We will discuss those sections that are relevant to this
lesson. They are, Section 1.0, Introduction, Section 2.0,
Quality Assurance Overview, Section 3.0, Organization
and Individual Responsibility, Section 4.0 Quality As-
surance/Quality Control and Section 7.0 Quality Assur-
ance Audits. The basic content which should be found in
the first three of these sections, is described below. The
other two sections (i.e., Section 4.0 to 7.0) are discussed
in subsequent parts of this lesson.
In the introduction, a commitment by the source to
maintain the QA program in compliance with all existing
environmental rules and regulations should be stated.
Also, a commitment should be made to acquire GEM data
of the highest quality possible and to maintain documen-
tation of system performance.
In the QA overview section, the specific objectives of
the quality assurance program should be clearly stated.
We have already discussed these objectives. Such activi-
ties as designation of responsible individuals, assurance
of data integrity, documentation, training programs and
corrective action activities should be discussed.
The section on organization and individual responsi-
bilities should discuss the key personnel involved in the
CEM QA program and provide an organization chart of
personnel and activities associated with the major ele-
ments of the program. This section should include a
definition of all QA/QC maintenance, data reduction,
recordkeeping and communication activities and then
assign these responsibilities to the appropriate personnel
(by job category and name). Also, the mechanisms used
for initiating each activity should be defined and the
specific frequency for performing certain activities (i.e.,
preventive maintenance and QA audits) should be stated.
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
Figure 14-3
EXAMPLE OUTLINE
Source Quality Assurance Program
Written Document
Introduction
Quality Assurance Overview
2.1 General
22 Specific
Organization and Individual Responsibility
3.1 Personnel Assignments arid Responsibilities
3.1 .1 Organization
3.1.2 Program Management
3.1.3 Field Operations
3.1.4 Laboratory Operations
3.1.5 Data Management
3.2 QA Responsibilities
3.2.1 Quality Assurance Coordinator
322 Field Personnel
Quality Assurance/Quality Control
4.1 Quality Assurtance
4.1.1 Activities
4.12 Personnel Responsibilities
42 Quality Control
42.1 Recordkeeping
422 Activities/Corrective Action
Data Handling, Validation and Reporting
5.1 Data Logistics
52 Data Handling and Statistical Analysis
5.3 Control Charts
5.4 Data Validation Criteria
5.5 Data Reporting
5.6 Data Forms
Operation and Maintenance Program
6.1 General
62 Preventive Maintenance
6.3 Corrective Maintenance
6.4 Spare Parts Inventory
6.5 Maintenance Documentation
Quality Assurance Audits
7.1 Self Auditing
72 Corporate Auditing
7.3 Performance Audit
7.4 Data Quality Audits
Rocordkeeping and Reporting ReojmremontB
|,. ,.,:.•:,-,- :..,,:, ,,...,.•; -? .',,, . ,,:.:,,,,,,::
-
This section on organization and individual responsibilities should also define the department responsible for
initiating any malfunction activities. Such activities include instrument repair, QA audits following maintenance,
and operating alternative measurement methods. Additionally, this section should contain flow charts which
indicate the flow of information between parties involved in the QA program and a table which identifies all
responsible personnel involved in the CEM program and lists these individuals by name, title, address and
14-9
-------
telephone number as well as each participant's individual role. An example of a flow chart demonstrating the flow
of activities between the various departments involved in a CEM QA program is shown in Figure 14-4.
AUWT/OA REPORT!
ffEEOroHWARD) .
1
QA/ AUDIT RESULTS 1
i
i SOURCE MONfTORING
* DIRECTOR
1 QA COORDINATOR
1
OA ENGINEER
ENVIRONMENTAL
, ENGINEERING
. . MANAGER
_ QA./ AJfOQ.RE.BUUP (EftBPtt1'!
CPU lUTl I T
DATA SERVICES CEMO&M
ENGINEER QC ENGINE
1 CHECK CALCULATION!
COMPARE RESULTS
1 WITH CEM DATA
1
1
V AUKTSAMPL
1
1
1 S
AMBIENT 1
MONITORING s
DIRECTOR i
|
LABORATORY I
SERVICES 1
LABORATORY 1
ER SUPERVISOR 5
s
» 1 A
1 ' 1 |
SAMPLE ANALYSIS ' 1 1
1 PREUMMARV CALCULATIONS _f , £
EB
J 1
- 1
i
FIGURE 14-4 PERSONNEL FLOW CHART WITHIN SOURCE Q A i
PROGRAM 1
VXftXXf&ttXXiiVxSiff:
W^fi^W^imSSffS!!SfSSi!ffffffffffiSSSfiSiiffiflfS!t:
x&myAyf?xfs#$f-;f&w-8f&x-&?issi!sm!$s
The Three Levels of a Quality Assurance Program
QA programs contain at least three different levels, Level I, Level n, and Level in QA functions. Table 14-
1 is a summary of the key responsibilities of the three QA levels. The following discussion summarizes the role
of each level.
Level 1
The Level I function involves the implementation and support of the QA program. Such duties as
administering company QA policy, administrative support, local assistance, development of strategies, system
analysis, promulgation of permit requirements and interfacing between the company and the regulatory agency
are some of the functions of Level I QA.
Level II
The responsibility of implementing the QA program lies within Level JJ. Personnel involved with Level n QA
are responsible for developing and implementing the Q A/QC programs which includes statistical procedures and
techniques necessary to meet the regulatory requirements. The scheduling and performance of non-routine and
quarterly audits, the evaluation of new QA procedures for applicability, the evaluation of data quality and
maintenance of QC charts and records are all a pan of the Level n QA responsibilities. Also. Level D QA is
14-10
-------
responsible for responding to QC problems and coordinating such activities with Level IQA. The Level n QA
coordinator should summarize in monthly and quarterly reports the status of the reliability of the OEM system and
its components, the quality of generated data, any CEM problem areas, any corrective actions taken and their
results, and quarterly excess emission reports (EER). Such information is fedback to Level I QA for approval and
distribution.
TABLE 14-1
RESPONSIBILITIES OF LEVELS 1,11, AND 111 IN A
SOURCE QA/QC PROGRAM
LEVEL I AND LEVEL II
LEVEL II AND
LEVEL III
LEVEL III
SUPERVISOR &
COODINATOR
SUPERVISOR &
OPERATOR
OPERATOR
• Document
Control/Revisions;
• QA Policy & Objective;
• Organization;
• Quality Planning;
• Training;
• Quality Cost;
• I nterlaboratory Testing
• Audit Procedures;
• Data Validation; and
• Quality Reports to
Management.
' Corrective Action;
and
' Procurement Quality
Control. '
• Preventive
Maintenance;
• Data Reporting;
• Calibration; and
• Document Control.
Level III
Level in QA consists of ensuring that good quality data are obtained from instrument operation. Level in
QA activities are the responsibility of the CEM operator. Such responsibilities as the daily QA zero/span checks
and subsequent recordkeeping, preventive maintenance procedures, sample collection and data reporting are a pan
of this Q A level.
14-11
-------
Three Stages of Quality Control Involved in a CEMS Program
In developing a quality assurance program for a continuous gas emission monitoring system, there are three
basic stages of quality control which need to be addressed. These three stages are:
1. operation checks (daily checks, observations and adjustments),
2. routine (periodic preventive maintenance) and corrective maintenance, and
3. performance audits.
The operation checks involve daily or routine checks of the monitoring system to ensure that the equipment
is performing adequately. Perhaps the most common QA/QC check involves checking the zero and span drift on
a daily basis. Other operation checks include checking the reference signals from control panels, flow rates,
pressures, vacuum levels, other gauges, valves and temperature settings. These types of checks can compare
results to baseline values and control limits. If deficiencies are found, corrective action should be taken.
Routine maintenance is performed at regular intervals. This type of maintenance includes preventive
maintenance which involves such activities as the replacement of filters, lamps, motor bearings and other parts.
Electronic and optical system checks may also take place. The purpose of this type of maintenance is to prevent
problems before they arise or find incipient problems and resolve them before they produce out-of-control
situations. Corrective maintenance is performed whenever there is a malfunction in the monitoring system and
there is a need to bring it back into proper operation. Corrective maintenance is also called non-routine
maintenance which refers to an unscheduled repair. We will talk more about these types of maintenance programs
in the next lessoa
Performance audits are performed in order to check the operation of the system, identify any operational
problems, identify the need to improve preventive maintenance schedules/procedures, or identify the need for
corrective maintenance. We will talk more about audits later.
A Quality Assurance Perspective for Regulatory Agency Inspections
In past lessons, we have learned that federal, state and local agencies require that a source install, calibrate and
maintain CEMS in order to demonstrate matitismamtaining compliance with the established airquality standards.
The regulatory agency has the right to conduct periodic inspections of a source in order to ensure that the
monitoring system being used at a plant is operating in a satisfactory manner. Control agencies which have good
inspection procedures for performing audits of a source's CEM program will have a higher percentage of CEMSs
which demonstrate good QA/QC practices within their jurisdiction because these CEMs will be maintained and
operated in a manner necessary to produce accurate and reliable air quality data. Agency inspection programs are
necessary in order to stimulate sources to establish and implement a sound CEM program as well as good quality
assurance procedures. The source will find it much easier to maintain a monitoring system man to constantly end
up in court due to a poor inspection record. With that in mind, there are two types of agency inspection programs.
They are:
• schedule system programs, and
• level system programs.
14-12
-------
The schedule system program involves the periodic inspection of a given source at various time intervals. This
type of inspection program involves more manpower and resources than does a level system inspection program.
Schedule system inspection programs may call for quarterly or yearly inspections and may be announced (i.e., the
actual date of the inspection is known to the source) or can be unannounced (i.e., the source may know that an
agency inspection will be conducted within a given year, but not know the specific date). During the actual course
of the inspection, the content may vary. The inspection may only consist of the inspector's use of a checklist to
determine the condition of the monitoring system with such items as strip chart records, meter readings and
computer outputs reviewed. Such checklists have been developed by the EPA Stationary Source Compliance
Division. Some inspections, however, may require the use of audit devices to check the performance of the CEM
in the presence of the inspector. Usually, the agency will hire an independent contractor to perform these audits.
The level system inspection program relies on the required quarterly excess emission report (HER) require-
ments contained in the applicable subparts of 40 CFR Part 60. For both new and existing sources, the information
required to be contained in this report includes details on exceedences of emissions standards; the nature and cause
of the excess emissions; and periods during which the CEM was inoperative and during which calibration checks,
adjustments and maintenance were performed. Because the specific format for reporting the excess emissions
varies from one agency to the next, it is a good idea to contact the agency prior to developing any software for
handling and reporting this type of data.
The excess emission report is reviewed by the agency once received, and a determination made as to the status
of the control equipment and monitoring system performance. Enforcement action may result if these reports
reveal extensive periods of excess emissions. Depending on the severity of the problems noted in these reports,
the agency will determine the next level of action to take. This action may be in the form of an agency call (or
a letter) to the plant asking for clarification as to why the monitoring system malfunctioned, or it may require an
agency inspection visit to the plant in order to review data and calibrations using checklists as previously
described. If this level of review does not determine the nature of the problem, a full scale audit or a repeat of parts
of the performance specification test (Part 60, Appendix B) may be required. Although a level system inspection
program requires less manpower, this type of program may encounter problems. Such problems include sources
going for years before undergoing an independent audit or agency inspection, and excess emission reports that may
mask actual problems in the monitoring program.
Audits
Overview
Earlier in this lesson, we discussed the three levels of quality control used in CEM programs. As you may
recall, one of those levels involved the use of performance audits. In general, audits are independent checks of
the monitoring system by someone other than the day-to-day instrument operator. These audits are usually
performed by an outside consultant, someone from a different facility or someone in the quality assurance group
of the company. The audit is performed in addition to the normal quality control checks of the monitoring system
using audit standards/devices different from those used in the normal day-to-day calibration checks.
The audit will provide an independent means for determining if the reported data (from day-to-day) are
precise and accurate, that the QA and maintenance programs are adequate and quality control is satisfactory. The
audit should be performed while the monitoring system is operating under normal conditions (i.e., without any
special preparation or adjustments) and both announced and unannounced audits should be scheduled.
14-13
-------
In the last lesson, we talked about the requirements of Part 60 Appendix?. This type of audit as you may recall,
is required on a quarterly basis for regulatory purposes. It should be noted that while a GEM calibration is the most
importantpart of quality control, the audit program is the most important part of quality assurance. Basically, there
are two types of laboratory audit programs:
• intralaboratory
• interlaboratory
Intralaboratorv program audits, involves the analysis of standard samples using the same procedures that are
used for analyzing the actual air samples. Each standard sample would contain the same concentration and
therefore, each participating laboratory should obtain the same results as long as each lab uses the same identical
procedure. The intralaboratory audit is performed by the same CEM technician who also performs the daily OEM
checks (Le. calibration drift test). The CGA test which we learned about in Lesson 13 (e.g., Pan 60 Appendix
F requirements) is an example of an intralaboratory audit because this test is usually performed by the same person
who performs all daily CEM checks. Figure 14-5 shows a general diagram which exemplifies the intralaboratory
audit method.
FIGURE 14-5 INTRALABORATORY AUDIT (STANDARD
SAMPLE)
STANDARD SAMPLE
ANALYSES
The Intel-laboratory Program audit includes the performance audit which provides a quantitative assessment
of the monitoring system in order to verify the adequacy of the quality assurance procedures used to determine
adequate CEM performance. In performance audit surveys, a single source will send samples to various
participating laboratories with instructions on bow to analyze the samples. Each laboratory will send their sample
analysis results to the source for evaluation. The Interlaboratory audit is performed by personnel not involved
in the CEM program (e.g.. by source QA personnel not involved with the CEM program or outside contractors).
An example of this type of audit is the RATA test requirement of Part GQAppendixP. As we will learn in the next
subsection, performance audits may comprise only a pan of a source's audit program. Figures 14-6 shows a
general diagram which demonstrates the performance audit method.
14-14
-------
FIGURE 14-6
INTERLABORATORY AUDIT (PERFORMANCE
AUDIT SURVEY)
B
C
I 0
* I
Types of Audits Incorporated In A Source Quality Assurance Program
From a QA standpoint, a source may incorporate four different types of audits in its CEM program. The four
types of audits are:
• Management Systems Audits (MSA),
• Technical Systems Audits (TSA),
• Performance Audits (PA), and
• Data Quality Audits (DQA).
The management system audit provides an evaluation of the role of management in the QA program. This
type of audit is conducted to determine if the company's management policy includes quality assurance in the
normal plant activities, if plant personnel involved in the QA program have job descriptions and performance
standards, if adequate resources have been appropriated to the QA program, if the QA program has an adequate
system for document control involved in audits and other QA projects, and if there is a satisfactory program for
ensuring mat QA training programs and communications are adequate and upgraded when necessary. It is the
source which determines the specific criteria for an MSA and incorporates this into its QA program.
A technical systems audit evaluates qualitatively, the total measurement system which includes selection and
use of the monitoring system, (both in the field and laboratory). This type of audit is intended to ensure that the
source has adequate sampling and data analysis techniques, that personnel have adequate training and are routinely
evaluated, that the equipment and facilities used in the CEM program are satisfactory and that written procedures,
recordkeeping, calibration and equipment maintenance are all adequate to produce data of acceptable quality and
minimize out-of-control conditions. The QA program should establish the necessary criteria for the TSA and
guidelines for developing this type of audit program may come from federal, state or local regulations.
As we have just learned, a performance audit is a quantitative assessment of the monitoring system for
demonstrating adequate QA procedures used in verifying adequate CEM performance. The performance audit
provides this quantitative assessment of the monitoring system by using reference samples ofknown concentrations
or replicate samples. We also stated that the RATA test requirement of Pan 60 Appendix F is an example of a
14-15
-------
perfbnnance audit In addition, the "Quality Assurace Handbook for Air Pollution Measurement Systems"
consisting of five volumes, contains the EPA's Quality Assurance program. The reader should note, that of these
five volumes, only volume 3 pertains to source CEM programs. Volume 3 gives quality assurance procedures
for specific source categories and provides audit procedures for some of the reference methods (RM) which we
mentioned in Lesson 12. Basically, a good performance audit will answer several questions. These are:
1. What is the accuracy of the CEM at the time of the audit?
2. Is the system operating within the established control limits?
3. Are QC data obtained in routine system operations correctly reflecting the quality of data generated from
the system?, and
4. How do these data compare to that generated from previous audits?
The PA results can sometimes be used to determine corrective action for out-of-control situations. However,
if this cannot be done, a technical systems audit may be needed in order to determine such corrective action needed
as a result of an out-of-control situation. In mis case, the TSA may determine that there is an insufficient
documentation of procedures and the next step would be to conduct a data quality audit
The data quality audit evaluates the documentation which is used to show that the data obtained is of a known
quality. Essentially, this audit ensures mat quantitative and qualitative indicators of data quality are available in
the Q A program. Several questions need to be answered in order to determine if these indicators are present:
• Are monitoring procedures documented in such a way that they could be used by personnel (with similar
technical qualifications) other than the original data collectors)?
• Did personnel involved in sampling sign the appropriate documents which demonstrates that the required
procedures were used?
• Does the data obtained contain enough information (i.e., such indicators as precision, bias, and
representativeness) so that the limitations of the data can be evaluated and if the intended use of the data
is appropriate?
A successful QA program should implement all four of these audit types at some point in time. Only the
performance audit provides a quantitative assessment of precision and bias in the measurement of CEM
parameters.
Audit Procedures Used By The Regulatory Control Agency
A regulatory agency may use an inspection audit program consisting of the administrative "phase" and
technical CEM evaluation "levels". Let us discuss each of these approaches.
The "phase" approach consists of the administrative activities involved with the initial CEM system
application, performance testing and final approval This will allow the agency to approve the CEM through
established certification procedures. The "phase" approach consists of these activities:
1. Phase I -me control agency grants the initial approval of the source CEM application as required through
the source permit.
14-16
-------
2. Phase n - the control agency observes the performance specification testing (PST) of the installed CEM,
and
3. Phase ID - the control agency reviews the PST report and gives approval or disapproval.
The "level" approach begins after the completion of the administrative phases. This approach extends from
a relatively simple to a more complex evaluation of the source's CEM program. The lowest level would consist
of source records review and the highest level would consist of stack testing compliance determinations. This
approach involves the inspection program which determines or confirms compliance and identifies causes of
excess emissions. The level approach consists of four levels and the activities at each level are as follows:
1. Level I - agency review of records including excess emission reports, previous inspection reports, source
"working" file and permits;
2. Level II - on-site inspections involving review of monitor recordkeeping (maintenance, monitor and
control equipment logs), monitor fault indicator, monitor internal zero/span check, strip chart review and
electronic checks;
3. Level in - evaluation of the source CEM through external audit evaluation involving the use of neutral
density filters for opacity monitors and gas cylinders/permeation tubes for gas monitors; and
4. Level IV - a comparative evaluation of the source CEM through performance testing utilizing federal
Reference Methods or portable CEMS (e.g., this level involves a recertification of the CEM).
The regulatory audit procedure using the "phase" and "level" approach was designed so that by completing
an activity, it can be known whether or not the CEM has achieved the necessary stage of compliance before starting
the next level or phase.
Additional information regarding specific audit procedures and techniques used by regulatory field inspectors
is contained in an EPA document entitled "Field Inspections Audit Techniques", EPA 340/1-89-003.
Summary
A successful Q A/QC program for a CEM involves considerations in the purchasing, installation and testing,
and the continual operation of the monitoring system. Quality control refers to methods used to assure a quality
product. Quality assurance is used to ensure the adequacy of the quality control program. The overall objective
of a quality assurance program is to ensure that CEM data are precise, meaningful and accurate within given
acceptance criteria. QA involves the implementation of four general categories, namely management, measure-
ment, documentation and statistics. There are three levels involved in a quality assurance program. The first level
involves the implementation and support of the QA program, the second level is program implementation and the
third level involves the duties of the CEM operator. The three levels of CEM quality control are operation checks,
routine and corrective maintenance and performance audits.
Regulatory agencies may use either a schedule system inspection program involving the periodic inspections
of sources or may use a level system program which relies on required quarterly excess emission reports from the
source. A good agency inspection program will yield a high number of sources which have an adequate QA/QC
program for their CEM systems.
14-17
-------
An audit provides a means (outside of the QA program) for the source to ensure that its monitoring system
is providing accurate and reliable data. Intralaboratory audits involve using the same personnel and procedures
for analyzing both standard and actual air samples while interlaboratory audits employee personnel outside of the
CEM program to analyze the air samples in order to compare the results of the independent laboratory to those
obtained by the source. Regulatory agencies may use a "phase" and "level" audit procedure approach in order
to ensure that a sources CEM program is in regulatory compliance.
14-18
-------
REVIEW EXERCISES
In conducting a prepurchase/ evaluation of a CEM, a source
owner needs to consider which of the following:
a. Having the CEM field tested (if purchasing multiple units.)
b. The operational design with respect to the installation site,
c. Contacting the owners who have already installed that particu-
lar CEMS model.
d. All of the above.
2. True or false. A minimum warranty of one year should be
included in a CEM purchase contract This warranty should not
begin until after the CEM passes an acceptance test conducted by
the purchaser.
1. d
3. True of false. Quality control refers to activities which will
provide a quality product and quality assurance refers to activities
which provide assurance that the quality control program is
satisfactory.
2. True
4. Which of the following is not a primary objective of a quality
assurance program:
a. Collect and supply information necessary to describe the
quality of the data.
b. Provide routine performance evaluation of personnel
and/or equipment.
c. Provide for prompt detection and correction of condi-
tions that contribute to the collection of poor quality data.
d. Provide an initial screenning of new personnel through
the use of a lie detector system.
3. True
5. True or false. A quality assurance program consists of the
incorporation of three general categories:
1. Management
2. Measurement and documentation, and
3. Data reduction, validation, analysis and statistics.
4. d
6. The person made responsible for developing a CEM Q A program
should consider which of the following questions:
a. How does one determine that the CEM is malfunctioning?
b. How does the CEM work?
c. What possible things can go wrong with the CEM?
d. All of the above.
5. True
14-19
-------
7. True or false. The person developing a CEM QA
program will need to consider testing CEM parameters
on both a daily and a periodic basis.
6. d
8. True or false. Once a CEM QA program has been
implemented and has undergone an acceptable trial re-
view, there is usually no need to review the program
adequacy on a periodic basis.
7. True
9. In a quality assurance program, the Level in function
involves which of the following activities:
a. the overall implementation of the Q A program.
b. the duties of the CEM operator.
c. Administering company QA policy.
d. procuring operating permits from the regulatory
agency.
8. False
10. The three stages of quality control in a CEMS program
are , and .
9. b
11. True or false. A level system inspection program used by
a regulatory agency involves the periodic inspection of a
given source at various time intervals.
10. Operation checks, rou-
tine and corrective
maintenance, perfor-
mance audits.
12. True or false. Audits are independent checks of the
monitoring system, are performed in addition to the
normal quality control checks, and are performed by
some person other than the normal day-to-day instru-
ment operator.
11. False
14-20
-------
13. A technical systems audit:
a. Provides a quantitative assessment of the monitoring system
using reference samples of known concentrations or replicate
samples.
b. Provides an evaluation of the managements role in the QA
program.
c. Evaluates the total measurement system which includes select-
ing the CEM, ensuring adequate written procedures, and ad-
equately trained personnel.
d. None of the above.
12. True
14. The
audit will provide a quantitative assessment of
precision and bias in the measurement of CEM parameters.
13. c
15. True or false. Inaregulatoryagency'sinspectionauditprogram.the
"phase" approach consists of administrative activities involved with
the initial CEM system application, performance testing and final
approval
14. Performance
15. True
14-21
-------
REFERENCES
1. Environmental Protection Agency 1986. Continuous Emission Monitoring (OEM) Systems Good
Operating Practices. Section 3.0.9 of the Quality Assurance Handbook for Air Pollution Measure-
ment Systems. Volume m. Stationary Source Specific Methods. EPA-600/4-77-0276.
2. Environmental Protection Agency. 1985. Technical Assistance Document for Monitoring Total
Reduced Sulfur (TRS)from Kraft Pulp Mills. EPA-340/1-85-013a.
3. Environmental Protection Agency 1984. APTI Course:476A Transmissiometer Systems - Operation
andMaintenance.AnAdvancedCourse-SelflnstructionalHandbook. EPA450/2-84-004. (Revised
March 24,1986).
4. Environmental Protection Agency. 1981. Air Pollution Control Orientation Course. Unit 4.
Sampling and Analysis of Air Pollutants. APTI Course SI:422 3rd Editioa EPA 45Q/2-81-017d.
5. Reynolds, W.E. 1985. Development and Evaluation ofSO2 CEM QA Procedures. APCA (TR-3).
Pittsburgh, PA.
6. Stanley, T.W. 1985. The Intent of a Performance Audit Program. APCA (TR-3). Pittsburgh, PA.
7. Winberry, W.T. 1984. Field Inspection Notebook for Monitoring Total Reduced SutfurfTRS) from
Kraft Pulp Mills. EPA 340/1-85-013B.
14-22
-------
LESSON 15
Maintenance Procedure - Problems and
Troubleshooting for
Continuous Gas Emission Monitoring Systems
Lesson Goal and Objectives
Goal
To describe the recommended maintenance activities for continuous gas emission monitors and to discuss the
potential problems with their use and suggested solutions to these problems.
Objectives
At the end of this lesson, you should be able to -
1. describe some of the activities involved in preventive maintenance, routine maintenance, and corrective
maintenance programs;
2. describe the three basic problem areas encountered in a continuous emission monitor (CEM) maintenance
program;
3. describe some of the activities specific to preventive maintenance, routine maintenance, and corrective
maintenance programs for extractive monitoring systems; and
4. describe some of the activities specific to preventive maintenance, routine maintenance, and corrective
maintenance programs for in-situ monitoring systems.
Introduction
In Lesson 13, we discussed the requirements of Appendix F to Pan 60 which involved the quality assurance
procedures that sources subject to subparts Da and Db of Pan 60 must incorporate into their monitoring programs.
As you may recall, one of the requirements called for written procedures for performing preventive maintenance
activities to the monitoring system. A successful maintenance program for continuous gas emission monitors can
mean the difference between "life and death" to the monitoring program. These instruments are complex, electro-
opticaj systems which are operated in a variety of adverse environments. It is no wonder why a successful
maintenance program must include procedures for maintenance to be performed at certain intervals in order to
ensure that the system does not fail for ( short or) extended periods of time.
15-1
-------
This lesson presents a basic overview of a complete maintenance program which includes the three types of
maintenance programs used. This lesson also describes the maintenance programs which are specific to both
extractive and in-situ type monitoring systems.
A Basic Overview of Maintenance Programs For Continuous Emission Monitors
There are three types of maintenance programs. They are preventive maintenance, routine maintenance, and
corrective maintenance. Preventive maintenance involves any activities which are designed to detect and prevent
the development of monitoring problems. This would include daily operational checks of the CEM. A preventive
maintenance program is important because it reduces the overall maintenance of the system and increases the
continuous emission monitors (OEMs) ability to obtain reliable data. In the development of a preventive
maintenance program, there are a number of factors to consider. They are:
1) providing the CEM with adequate protection against environmental effects such as heat, cold, dust, rain.
and flue gas;
2) locating vibration-free locations for installation in order to minimize damage to mechanical components
and electrical connections;
3) installing the CEM in an accessible location for convenience to maintenance personnel;
4) providing a stable power supply which is relatively free of electrical noise;
5) providing a system air supply which is adequate to prevent filter and sample line pluggage;
6) installing CEMs in areas which avoid extreme flue gas conditions such as high temperature flue gases,
in areas of high paniculate loading, and areas with a high moisture content (especially entrained droplets);
7) providing adequate mountings for on-stack CEM components to eliminate stack movement (expansion/
contraction) and vibration;
8) sdectmgtheappropriatern-stackcomponents(e.g.,sdectingro^
depending on the CEMS location;
9) maintaining maintenance records which are adequate to determine problem trends which could help in
identifying system weaknesses; and
10) selecting a CEM which is appropriate to the application (e.g., using path in-situ systems in gas streams
which contain entrained water droplets is not recommended).
All of these considerations will be important in determining the types of, and frequency for performing CEM
maintenance. Routine maintenance is actually a pan of a preventative maintenance program and involves the
periodic inspection and/or replacement of worn or damaged CEM components. Corrective maintenance involves
the correction of any immediate malfunctions in the monitoring system (i.e., on an as-needed basis). Trou-
bleshootingthe monitoring system whenever something goes wrong is an integral pan of a corrective maintenance
program. As you may recall, in the last lesson we defined the three levels of quality control for CEMs and we
briefly discussed each. Two of these levels were operation checks (routine maintenance) and preventive/
corrective maintenance.
15-2
-------
Maintenance Providers
The maintenance program can be carried out either under a maintenance contract with a service contractor or
CEM supplier, by plant personnel trained in CEM maintenance, or by a combination of contract maintenance and
plant personnel. If a full-service maintenance contract is used, it is a good idea to have several plant employees
trained in the maintenance of the system since it will allow these people to conduct basic checks of the CEM. Also,
they will be able to notify the maintenance contractor whenever non-scheduled maintenance is necessary.
On the other hand, if an in-house maintenance program is used, the required level of effort will depend on such
factors as type of system(s) used at the facility, the number of and locations of the monitoring systems, and local
conditions (e.g., the plant environment).
The training of in-house personnel forperforming CEM maintenance should be provided by the CEM supplier
or manufacturer. Individuals to be trained for an in-house maintenance team should ideally have experience with
CEMs, or should have a technical background and mechanical and electronic maintenance skills. The reader
should note that continuous emission montioring systems (CEMS) contain complex electronic components and
some systems incorporate mechanical components which are highly sophisticated in design. It will take an
individual with a strong technical background to learn and perform some of the more difficult maintenance tasks.
Problem Areas
Recent surveys have showed that there are three basic problem areas which can be found in a CEM
maintenance program. These areas are spare parts, emergency service, and data handling systems.
Obtaining spare parts has been shown to be one of the most persistent and recurring problems encountered
in a maintenance program. The most obvious way around this type of problem is to always maintain a large in-
house inventory especially if a facility installs several monitoring systems of the same type. By having spare parts
available, monitor down time is minimized.
Emergency service is also a problem since it is often difficult to obtain a quick response from a vendor, and
sometimes the person(s) who has been called to perform the maintenance may not be very proficient in diagnosing
the problem and performing the repair in a timely manner. A possible solution to this problem is to specify certain
plant personnel to be responsible for CEM maintenance and to provide these people with vendor training on
instrument diagnosis and repair. Another possible solution would be to initiate a service contract with someone,
other than the CEM vendor, who has the capability to provide qualified service technicians on an on-call basis.
The problems associated with automated data handling systems are 1) the lack of software support by vendors,
and 2) the use of inappropriate calculation parameters. The lack of software support by vendors makes it difficult
for the CEM user to make the necessary modifications to their software programs which are usually required after
the initial installation of the program. Many times, the vendor will lose the original program documentation
whenever there has been a change-over in programming personnel. One way to overcome this type of problem
is for the source owner to employ in-house information specialists or consultants to provide constant surveillance
over the computer programming system used at the facility. The second problem involving automated data
handling systems, pertains to the use of generic equations that contain process specific conversion factors. For
example, the F-factor (EPA Method 19) that converts the pollutant concentration in parts per million (ppm) to units
of the emission standard in pounds per million Btu for combustion sources, is different depending on the type of
fuel burned. When the data handling system is installed, the F-factor in the generic equation must be checked to
ensure that it corresponds with the fuel burned at the facility.
15-3
-------
Finally, the reader should note that these automated data handling systems require stable environments (i.e.,
clean, dry and temperature controlled). Additionally, the power supply used for these systems must be stable and
free of electrical interferences. A faraday cage is a good preventive maintenance device to use in situations where
the data system is subjected to strong radio frequency signals.
Now that we have discussed the basic types of CEM maintenance, considerations to be addressed in developing
a maintenance program and problem areas encountered in maintenance programs, the remainder of this lesson will
focus on maintenance programs which are specific to 1) extractive monitoring systems, and 2) in-situ monitoring
systems.
Recommended Maintenance For Extractive Monitoring Systems
Preventive Maintenance
Daily operation checks should be performed by a qualified and trained instrument operator who is capable of
recognizing problems from these daily checks. The use of automatic check systems, which are incorporated into
many types of extractive monitoring systems for performing daily zero and calibration drift checks as well as other
internal self-checks without the need of the operator, reduce the level of operator attention. However, this lack of
attention can cause small problems to develop into much larger ones. This type of problem can be alleviated to a
certain extent by using CEMS which have incorporated a more intelligent system which monitors key system
parameters and will report an out-of-control situation at a remote panel.
The first item which should be observed during a daily operation check is the strip chart record and/brother data
recording devices. All pertinent information for calibration purposes, the time/date of the readings, the name of the
person recording the information and chart recorder settings, should be written directly onto the chart Always check
the paper in the recorder or printer (if applicable) to ensure that there is an ample supply of paper for the next 24-
hour run. If the data obtained by the monitoring system are both checked and adjusted automatically (by the
appropriate calibration factors) by means of a microprocessor controller, the controller unit will have to be
preprogrammed to record the unadjusted values before the CEM is adjusted to the required values. If both a strip
chart recorder and a microprocessor are used together, the adjusted data should be recorded on both the strip chart
and the microprocessor printer output The problem involved in this case is that sometimes the microprocessor is
programmed to record the adjusted number only, without physically adjusting the CEM. When mis happens, the
CEMmeter readings and strip chart readings win differ from the microprocessor reading making data interpretation
very difficult
Once the strip chart recorder has been checked, me next thingto inspect is the indicator lights on the system or
monitor control panel. The indicator lights system will alert the operator whenever an out-of-control situation or
other problems occur. These problems need immediate attention since subsequent data may be in error. If reset
buttons designed to override the indicator lights are installed on the monitoring system, they should not be used until
the problem has been corrected. It is advisable to record the system status in ink in a hardbound logbook.
Additionally, all othermaintenance, including unscheduled repairs or system modifications should also be recorded
in this logbook. This book is useful for tracking the long-term performance of the CEM and will allow other system
maintenance personnel to become more familiar with the monitoring system. The status of other system indicators
should also be recorded at this time. These would include vacuum, or pressure gauges, sample flow rates, and lamp
and detector reference levels (if applicable).
15-4
-------
The daily monitoring system check should also include a visual inspection of the conditioning system
(moisture removal) purge cycle. The proper operation of all valves and verification that the condenser system has
been completely drained is the focus of this inspection.
A calibration check of the system should be performed next. We have already discussed the regulatory
requirements for performing calibration checks on continuous emission monitoring systems (CEMs) in Lessons
12 and 13. We also discussed the required performance specifications (i.e., AppendixB and Appendix? toPait
60). In performing a calibration check it is desirable to inject the calibration gas at a point in the monitoring system
where many of the conditioning system components will be checked. In some extractive systems, this can be
performed at the probe itself with the advantage of allowing the system to be tested for sample line losses. The
reader should note some of the problems involved with the injection of calibration gases. First of all, the use of
gas cylinders for injecting the gas has a tendency to pressurize the system. When this happens, the flow rate of
the gas into the analyzer sample cell may differ from the flow rate of the extracted stack gas. Secondly, cylinder
gas is a dry gas and does not contain any moisture. The calibration process can be affected if flow rates and
moisture content of the calibration gases entering the analyzer sample cell are not similar to the sample gas.
Thirdly, whenever the system is pressurized, system leaks will go undetected because ambient air is prevented
from entering the system as it might otherwise if there were a leak during normal monitoring procedures. Finally,
the reader should recall from Lessons 12 and 13 that gases which are used for performing these daily checks should
be either EPA certified gas concentrations or validated against certified calibration gases.
Figure 15-1 shows a suggested format (from the EPA's "Quality Assurance Notebook for Air Pollution
Measurement Systems, Volume HI, Stationary Source Specific Methods") for recording the daily checks for an
extractive monitoring system.
Routine Maintenance
For CEMs which have just been installed, it is recommended that routine maintenance be performed on each
system every 30 days. As more experience with each system is gained, the period for such maintenance can be
modified for either shorter or longer periods of time depending on the specific problems encountered with each
system type. These maintenance intervals may reflect maintenance frequencies for subsystems or individual
components.
Conditioning System
It is the sample conditioning system rather than the analyzer system that is of most concern with extractive
monitoring systems. Before the analyzer receives the gas sample, the conditioning system filters remove
paniculate matter and water vapor. These filters must be periodically cleaned or replaced. Forcondensing-type
moisture removal systems, the condensed water must be drained.
CEM Plumbing
All plumbing associated with the CEM should be periodically checked for corrosion and leaks. This includes
an inspection of all fittings, valves and gas regulators. If solenoid valves are used to automate the system, these
valves should be checked to ensure that they move freely and on command as these types of valves are prone to
sudden failure. Motorized or air-activated rotary valves have been used on extractive systems in order to reduce
15-5
-------
FIGURE 15-1 EXAMPLE FORMAT FOR EXTRACTIVE GAS
MOMTORING SYSTEM
DAILY QC SHEET
PLANT DATE TIME
umrr UAUP
QASMONTTDRED „ °HONF
ANALYZER!. D . „ ., ,.., ... 7FROOFFS FT VALUE _
ROAM VALUE BATP f.fartetfn
CALIBRATION GAS VALUE DAD(
ZERO QAS VALUE (AIR, Nj. OTHEF
HOURS OPERATIMQ M PERIOD ,
k BTRI
•JK STATUS:
PCM ART
PRINTER
PART1 MOICATORS
MDCATDR LIQHTSOUJGES
SAMPLE PRESSURE/VACUUM
SAMPLE FLOW
LAMP
DETECTOR
STATUS
PROBLEM/ACTION TAKEN
PARTS CALIBRATION CHECK
UNADJUSTED READING
ZERO (LOW-LEVEL) GAS
CALIBRATION (HK3H1E VEL) GAS
STACK CONCENTRATION
TIME
METER
STRIP CHART DIGITAL PRINTER
PARTS ZERO A SPAN ADJUSTMENT (IF OUTSIDE OF CONTROL LOOTS)
CONTROL LIMfTt PPM
ADJUSTED READNG
ZERO (LOW-LEVEL) GAS
CALIBRATION (HIGH-LEVEL) GAS
STACK CONCENTRATION
TIME
METER
STRIP CHART DIGITAL PRINTER
OPERATOR SIGNATURE DATE SUPERVISOR SIGNATURE DATE
1
I
1
1
1
1
1
1
i
i
the frequency of valve failure. It is not a good idea to design a system which utilizes too many valves as me more
valves involved, the more valves there are to be checked. Always maintain a sufficient supply of valves in the
spare parts inventory.
15-6
-------
Electrical System
It is important to frequently inspect the CEMS electrical system which includes all electrical cables and heat-
traced lines. The electrical system is always prone to damage in the plant environment through either normal
operations or construction projects. The deterioration of electrical insulation is common in the ambient
atmosphere near such areas as where flue-gas leakage occurs or in stack down-wash areas. Also, both electrical
finings and plumbing may become corroded whenever acid gases are circulating near the stack. Finally, many
CEMS are equipped with electrical test points located on the back panel or on circuit boards. These test points
are used in conjunction with a voltmeter or an oscilloscope to check the electrical system for possible problems.
This should also be done on a periodic basis.
Pumps and Chiller Units
Since pumps and chiller units on extractive monitoring systems operate on a 24 hour basis, these must be
checked on a regular basis for excess wear in motor brushes, for pump diaphragm breakage, or other parts which
require oiling.
Overall Qeanliness
Since paniculate matter from flue-gas has a tendency to settle in unwanted locations, the overall cleanliness
of the system should be inspected. For systems which are located outside, near a stack, all sensitive components
should be installed inside a dust-free cabinet These cabinets should never be opened whenever fly-ash is in the
ambient air.
Lamps and Bulbs
There are certain components of extractive monitoring systems which have limited lifetimes. Such
components include lamps and bulbs. These components should be replaced before their lifetime expires because
a weakened bulb can produce erroneous results. Since monitoring systems are designed to operate within a given
range of lamp intensities, a bulb which is below this intensity range will not respond well to incoming signals. In
addition to replacing bulbs, detectors sometimes need replacement but this is not done very often.
Figure 15-2 shows an example of a check sheet format (from the EPA's "Quality Assurance Handbook for
Air Pollution Measurement Systems, Volume HI, Stationary Source Specific Methods") which could be used for
recording routine maintenance checks. Depending upon factors such as the adequacy of vendor instruction
manuals to provide routine maintenance information and experience with the system itself, it may take up to a year
to design such a check sheet The reader should keep in mind that the system's logbook is a very good reference
to use in developing these routine maintenance check sheets.
Corrective Maintenance
As we stated earlier, any individual involved with CEM maintenance must have a good technical background
and be well trained on the monitoring system. This is particularly true in the case where something goes wrong
and the technician is left with the task of locating the problem and solving it
15-7
-------
With extractive monitoring systems, maintenance problems generally occur in the gas transport and the gas
conditioning components. If routine maintenance is not performed on valves, finings, tubing and filters, then acid
gases, paniculate matter and continuous vibration will take their toll on the system. A good quality control/quality
assurance program can minimize the need for corrective maintenance. Table 15-1 provides a listing of some of
the common problems encountered with extractive monitoring systems. These problems range from physical
problems associated with the conditioning system to problems associated with the analyzer. These problems are
sometimes caused by a failure of the GEM manufacturer to understand problems regarding constraints imposed
by the plant environment and stack conditions. Additionally, the plant may have provided inadequate
specifications to the vendor at the time of purchase. Corrective maintenance problems are difficult to foresee and
often times the monitoring system must be redesigned so that the frequency of any corrective maintenance can
be minimized.
FIGURE 15-2 EXAMPLE FORMAT FOR EXTRACTIVE GAS
MONITORING SYSTEM
30-DAY ROUTINE MAINTENANCE CHECK SHEET
DLAMT DATE 71UP
UNIT NAME
RVKTCU i n "HONE
AMAIV7CDI n n*,f
AMAIV7PDI D AAC
AMAIV7CDI n n»C
REQUIRED MAINTENANCE CHECKS
EXTRACTIVE SYSTEM
PROBE FILTER
RNE FILTER
CONDENSATION SYSTEM DRAM
HEAT TRACE CONTINUITY
PUMP. BEARING NOISE
PUMPING LEAK CHECK - VACUUM
PRESSURE
CABLE INTEGRITY
CLEANLINESS
CORROSION LEVELS PROBE
SOLENOID PERFORMANCE
REGULATOR PRESSURES
AIR-OPERATED VALVES
AR PURGE/BLOWBACK
ZERO GAS
CALIBRATION GAS
ANALYZERS
LAMP
SENSOR
TEST POINTS
CHOPPER MOTOR
OPTICAL WINDOW STATUS
STATUS
ACTION
OPERATOR SIGNATURE DATE SUPERVISOR SIGNATURE DATE
I
|
|
| < * %\ -- - - - '* * -v - ",:
15-8
-------
TABLE 15-1
Extractive Monitoring System Problems and Possible Solutions;
COMMON PHYSICAL
PROBLEMS
A. Conditioning Systems
Probe plugging
Probe/filter corrosion
Probe breakage (due to vibration or
embrttt'lement from chloride)
Condensation in sample lines
Inadequate water removal
Dirt in sample lines, plugged values. '
plugged samples lines
Leaks in sample lines/f tilings
Pump failure
B. Analyzers
Internal corrosion/ damage
Poor response lime (False positive
zero values or poor calibration check
values)
Excessive drift
Component failures
Lamps, fan. chopper motors
Electronic problems
Loose circuit boards, poor contacts
Ground loops and noise
Large voltage drops when plant
equipment is started. Spikes in
strip chart record.
Static electrical charges
Burned out electronic circuits from
lightning strikes
No output from Instrument, no
calibration cycle, etc
Improper inslrument response faulty
cal Ibratlon. improper or no output
pa^JN-p— *-*«*«*,
POSSIBLE CORRECTIVE ACTION
Install blowback system, increase blowback frequency proble shield Relocate Change
probe design Change system design Enter probe at downward angle.
Relocate probe Obtain corrosion-resistant alloy for probe construction.
Support probe. Shorten Select resistant alloy.
Resize heaters. Don't let heat go off on heat trace. Use backup power. Avoid shorts -
don't loop lines. Install thermal conductivity sensor if continuing problem. Remove
water at stack probe. Filter at lower temperatirte (acid may be condensing) - increase
temperature or heat
Improve chillerdesign. Connect two chillers in series Back up chiller with Permapure
dryer (but heat front end of Permapure). Dilute the gas stream to lower moisture
content.
Decrease pore size of probe filter Increase sample flow rate to fine filler. Increase
diameter of line Use clear Teflon tubing lo detect areas of accumulation Redesign to
reduce number of valves Use redundant f i Iters.
Reduce number of fittings as much as possible. Detect leaks by pressurizing system and
using soap bubble Indicator. Check for leaks in gas regulators Don't wrench down on
compression fittings loo severely. Don't use glue, paint, glyptal, etc., lo cover leaks -
rebuild system if necessary.
Perform routine maintenance -check brushes periodically. Cheek diaphragms of
diaphragm pumps.
Check moisture removal system for failure. Build redundancy In system Add extra
chiller. Put thermal conductivity sensor In line lo slop pump when moisture breaks
through, when moisture breaks through, dismantle sample cell, clean, and dry. May have
to replace entire cell in some models Clean and dry all sample lines
Check sample line length. Shorten line or increase flow rale. Some analyzers have slow
response limes, increase lime for calibration gas flow during dally checks
Check foul ing of sample cell for dirt or waier. Electrical problems Passivation of cell
surfaces. Lamp weakening- I igrii levels loo low. Detector problems Electronic
problems. Erratic power supply.
Check component wear. Check and replace on regular schedule.
Check for vibration problems Install circull board clamps. Check for SO. corrosion in
exposed units. *
Trace and rewire.
•
Inslall transienl suppressor, dedicaled power transformer or constant
voltage/isolation transformer for monitoring system.
Connect probe case to dedicated earth ground.
Add phenolic gaskets between metal stack and probe. Add surge arresters at junction
box
Check fuses
Check electronics Check lo see that cards and components are secure. Use
trouble-shooting guide supplied by vendor to check electronic test points Replace
approriate components or replace cards Check software for errors In programming -
particularly in calibration adjustment routines.
15-9
-------
Such problems as loss of signal, inconsistent readings and poor calibration response are indications of
malfunctions in the CEM system. As we just stated, it is important that the CEM operator have sufficient skill
to diagnose these types of problems. The following general guidelines can be used for troubleshooting both
mechanical and electrical deficiencies of CEMs.
• Loss of signal or abnormally low values - check conditioning system for plugging, leaks, pump
failures.
• Noisy, erratic signals - check for electronic problems, electrical supply problems, weak lamps,
moisute condensation, paniculate matter in analyzer.
• Loss of linearity - check for sample cell contamination, leaking calibration manifold, incorrect gas
cylinder values.
• Slow response - check for leaks, water in lines, measuring cell failures.
A majority of the problems encountered with extractive monitoring systems are in the system's electronics.
Normally, if the electronics problem is severe, the instrument is compact enough such that it can be shipped back
to the vendor for repair. The instrument's maintenance manual is a good reference to use for troubleshooting
problems in circuit boards and components.
From a mechanical standpoint, extractive monitoring systems encounter their greatest problems when they
are located in severe environments orif the conditioning system fails. In Lesson 8, we mentioned that placing the
entire system inside a protective housing was one way to protect the CEM from the operating environment When
the conditioning system breaks down, acid gases and/or moisture may condense in the sample cell and paniculate
matter may cause blockage of the probe, sample lines, or the analyzer itself. Also, after unconditioned sample gas
has entered the analyzer, it can take months before the system will operate properly again. By far, a malfunctioning
conditioning system is one of the worst maintenance problems encountered with an extractive monitoring system.
Plugged paniculate filters are a prevalent problem with extractive monitoring systems. This can lead to
reduced sample flow to the analyzer which will result in component failure and erroneous emissions data.
Blockage of paniculate filters can be caused by an insufficient backpurge or by an inadequate schedule for filter
replacement. Filter blockage is evident whenever a steady drop in the sample flow rate occurs and may be
accompanied by a gradual increase in the system vacuum. Backflushing with purge air may slightly improve the
situation.
The use of valves which are not property designed for application to a particular monitoring system are another
cause of frequent failures in extractive monitoring systems. Valve failures are sometimes difficult to assess and
major maintenance efforts can result One example of an inappropriate use of a valve type in a system is the use
of a solenoid-type valve in a paniculate-laden gas stream. In this situation, the paniculate matter will eventually
wear out the valve seat causing the system to leak. Another problem with the improper use of valves is the use
of unsuitable material for frequent operation such as the use of polymeric materials which fail at accelerated
operation rates. Additionally, brass valves should not be used in corrosive gas streams. TeflonR and corrosion-
resistant steel alloys should be used in most situations.
One other problem encountered with extractive monitoring systems is the failure of heat-traced sample lines
which .transport moist flue gas to the conditioning system. These lines are heated to prevent condensation from
occurring. If a section of the heating element fails, condensation will occur and this can lead to sample gas loss
or contamination, and blockage of the sample line. In some instances, these heat-traced lines are not designed to
function in adverse environments such as those found in power plants. Providing additional insulation or using
a stronger sample line sheath are ways in which this problem can be minimized.
15-10
-------
Recommended Maintenance For In-situ Monitoring Systems
Preventive Maintenance
As we just learned, performing the daily operation checks on extractive monitoring systems requires a well
trained operator. The same holds true when performing daily checks on in-situ monitoring systems. Also, as with
extractive systems, using systems designed with automatic zero and calibration procedures can lead to a false sense
of security and additional system failures. However, these problems can be alleviated through the use of more
intelligent systems.
Strip chan recordings and computer printouts from the previous 24 hour operation should be reviewed with
any discrepancies recorded on both the daily check sheet and instrument logbook. As with extractive monitoring
systems, any problems identified by the system indicator lights should be resolved immediately.
Gas cells containing pollutant gas at a known concentration are normally used for calibrating in-situ gas
monitors. Optical in-situ analyzers are checked by placing the gas cell into the path of the light beam, and the signal
is attenuated by a specific amount The reader should note that this does not always perform a complete check
of the optical system over the full range of concentrations measured in the flue gas. Additionally, the gas
concentration of the pollutant gas inside the cells is not normally certified by independent laboratories. These cells
are also prone to leakage, and degradation by adsorption and internal reactions which can cause problems when
attempting to calibrate these systems.
In-situ point monitors can be calibrated with a cylinder gas which has been certified. This is accomplished
by flooding the volume within the ceramic thimble with calibration gas or with a zero gas. In either single-pass
or double-pass monitors, a "flow-through" gas cell can be used. This involves flowing gas of a certified
concentration through a fixed cell in the instrument Through this method, a calibration "traceable" to National
Institute of Standards and Technology (NIST) or other certified gases can be obtained. However, problems
involving the optical system may still be present
During any of the calibration sequences, the operator should note whether the temperature compensation
circuit had been disconnected so that problems in these circuits are not overlooked. This system is used to
compensate the temperature for the effects of elevated flue gas temperatures. For in-situ systems which use
temperature compensation circuitry, this circuitry must be used during instrument calibrations. If the calibration
is performed manually using calibration cells, mis circuitry can be bypassed.
In order to ensure proper operation of in-situ monitoring systems, it is important to check the alignment of the
optical system on a daily basis. It should be noted that if the system was not designed with an alignment sight a
more detailed electronic or optical check may need to be performed in order to optimize the system alignment
Figure 15-3 gives an example of a check sheet (from the EPA's "Quality Assurance Handbook for Air
Pollution Measurement Systems, Volume m, Stationary Source Specific Methods) to be used for recording daily
operation checks.
15-11
-------
FIGURE 15-3 EXAMPLE FORMAT FOR IN-SfTU MONITORING
SYSTEM DAILY QC CHECKS
Dl AWT HATE Till*
UMTT MAUC
QAS MQNrTABFn PHONF
6pA^ VAI VF P"T "F"^««"n
CALIPJMTWNnWLVAlllF . _ _ . weunop
ZERO GAS VALUE ..-"< \.^-*'*f
1
|
j
1
Routine Maintenance
Routine maintenance for in-situ monitors should be initially performed every 30 days. There are some
manufacturers of in-situ systems who will recommend a routine maintenance period of two or three months, but
until the system is "broken in," it is a good idea to perform this maintenance every 30 days.
15-12
-------
The first type of routine maintenance consists of ensuring that key system components (e.g., windows, filters,
and desiccants) are clean and operational. Whenever cleaning windows or optics, the technician should be very
careful as these components can be easily damaged. These components should be cleaned with lens tissue or soft
rags along with alcohol or water solutions. If the windows are exposed to flue gas, a mild detergent may have to
be used in order to clean these windows. One method which would ensure that optical surfaces are maintained
as clean as possible, is to replace purge air blower filters on a periodic basis. If these filters become clogged,
paniculate matter can scratch the optical surfaces. If stack gas enters the monitor cabinet, electronic components
will corrode and fail. The blower filters should be replaced concurrent with cleaning system optics and purge air
shutters should be checked at this time. Other optical components such as diffraction gratings are so sensitive
that they should never be touched or cleaned in the field. In this case special techniques should be used because
fingerprints or traces of cleaning material will severely affect the performance of these components.
As with extractive monitoring systems, the electronics involved with in-situ systems should also be routinely
checked for adequate performance. Digital voltmeters and oscilloscopes are used to check the electrical circuits
along the various test points. These tests should include checking lamp voltages, power supply voltages, and
detector outputs. A well-trained service technician or electronic technician will need to be employed for this
maintenance service.
In many in-situ monitoring systems, chopper motors are used in the transceiver assembly to modulate the
light beam. Routine maintenance on these motors involves checking the motor bearing for noise or excessive
vibration. Motors used for moving mirrors or gas cells into position should be checked by manual means to ensure
that the movements are smooth and complete.
A system logbook should be used for recording observations made during routine maintenance checks. This
logbook can be used to determine performance trends over a given period of time so that a determination can be
made as to the specific frequency for performing certain routine maintenance checks. Figure 15-4 gives an ex-
ample of a check sheet (from the EPA's "Quality Assurance Handbook for Air Pollution Measurement Systems,
Volume ffl, Stationary Source Specific Methods) which can be used for recording routine maintenance on in-situ
monitoring systems.
Corrective Maintenance
In-situ monitoring systems are complex in nature and require a person who has a high degree of troubleshooting
skill to perform corrective maintenance. A maintenance technician employed to perform maintenance tasks on in-
situ systems should be required to undergo formal training at the vendor's facility. These systems do not share
many of the problems associated with extractive monitoring systems but because they are directly mounted inside
stacks or ducts they are subjected to harsh environments. In general, such conditions as temperature cycling, acid
gases, and vibration can damage optical components, disrupt the optical alignment, and harm electrical
components located in the mounted system assemblies.
Recent surveys conducted with utility companies have identified three problem areas relative to locating in-
situ monitoring systems. These are:
• ambient dust and severe weather conditions;
• moisture condensation from the effluent stream; and
• sensitivity to vibration.
15-13
-------
FIGURE 15-4 EXAMPLE FORMAT FOR IN-SFTU MONITOR
30-DAY ROUTINE MAINTENANCE CHECK SHEET
BL4MT DATP T1UP
UNIT WAUP
MONITOR 1. D. punuc
PARTI REQUIRED MAINTENANCE ACTIONS
CLEAN OUTER SURFACES
HOOD
TRANSMrTOR/TRANSCEIVER ASSEMBLY
LAMP REFLECTOR ASSEMBLY
PURGE AIR SYSTEM
CLEAN INNER SURFACES
HOOD
TRANSMrTTER/TRANSCEIVER
ASSEMBLY (UNLATCHED)
LAMP RETROREFLECTOR
ASSEMBLY (UNLATCHED)
PRESEPARATOR • AIR PURQINQ SYSTEM
CLEAN TRANSMITTER/TRANSCEIVER WINDOWS
CLEAN LAMP/RETROREFLECTOR WINDOWS
REPLACE PURGE AIR FILTER
(OR CLEAN AND REPLACE)
TIGHTEN HOSE, CLAMPS. CABLES, MOUNTINGS
PART 2 REQUIRED MAINTENANCE CHECKS
CHECK ALIGNMENT
CHECK DESICCANTS
CHECK CABLES (CONTMUrTY,
PINCHING, CUTS, CORROSION)
CHECK HOSESJCONTINUmr.
PINCHING, CUTS, CORROSION)
CHECK SECURITY SEALS
BLOWER MOTOR (BEARING NOISE)
STATUS
STATUS
ACTION
ACTION
OPERATOR SIGNATURE DATE SUPERVISOR SIGNATURE DATE
I
|
1
|
1
|
j
jj
|
I ^-" % - v. v \- ^ *' -s v ->"5, f"o'Mi$"s5 * vJVvvA&s? ^*;,^>>w$w§*OsU^ i^>. ^..vx^ry^F
The most common problem identified was exposure to ambient dust and severe weather conditions.
Electronic drift which will cause erratic operation of the CEM can be caused by extreme fluctuations in ambient
temperature and/or cooling caused by wind. Exposure to ambient dust can cause problems with optical surfaces,
electronic components and purge air systems if purge air blower filters are not changed periodically. Some in-
situ systems are equipped with isolation shutters which protects the system from optical and electronic component
damage and should be checked frequently. The construction of an enclosure around the system large enough for
maintenance personnel to work on the system is a possible solution to this kind of problem.
15-14
-------
Moisture condensation from effluent streams can cause severe problems to exposed optics. The optics can
become clouded and this will cause a decrease in light transmittance which could possibly result in the system
recording higher pollutant concentration readings. Cleaning these optical surfaces offers only a temporary
solution. A longer term solution would be to use heated purge air, especially in colder climates. This air would
have to be heated above the dewpoint of the effluent gas and any exposed areas acting as heat sinks would need
to be insulated to prevent heat loss. Moisture condensation can also cause corrosion of metal parts (e.g., the
reaction of SO2 and NO, with water to form an acid solution). This corrosion can cause moving parts to freeze
and complete mechanical failure to occur. It is therefore important that fiberglass, teflonR or metal alloys which
are resistant to corrosion are used in in-situ systems. Problems involving moisture condensation are encountered
mostly whenever the system is mounted in areas downstream of a wet flue-gas desulfurization (FGD) system.
In-situ monitoring systems are very susceptible to vibration. A possible solution to this problem would be to
either locate the system in areas away from any stack or duct vibration or to mount the monitor on a floating
platform which is attached to a rigid structure (e.g., by placing shock absorbers between the monitor and the
platform).
In addition to these three common problems, there is one additional problem which is not as common as the
first three. This is the problem of alignment Whenever stack or duct walls expand or contract, the relative
orientation of the transceiver and the retroreflector is changed causing a misalignmento&he system to occur. By
mounting both ends of the system to fixed platforms and interfacing the system to the stack or duct with flexible
seals, this problem can be avoided. Across-stack, in-situ monitors are very sensitive to alignment problems.
Table 15-2 is a listing of some of the common problems associated with in-situ monitoring systems and
possible solutions to these problems.
Summary
The three types of maintenance programs are preventive maintenance, routine maintenance and corrective
maintenance programs. Preventive maintenance programs involve the daily operational checks of the system in
order to prevent long-term problems from occurring. Routine maintenance involves the periodic inspection/
replacement of CEM components and is an integral pan of a preventive maintenance program. Corrective
maintenance includes troubleshooting the monitoring system whenever a malfunction in the system occurs.
Troubleshooting a monitoring system can only be accomplished by a skilled technician who is knowledgeable in
the overall operations of the monitor.
There are a number of factors involving the system design as well as environmental factors to consider in the
development of a CEM maintenance program. The maintenance of spare parts, obtaining emergency service and
data handling systems are three basic problem areas in a CEM maintenance program.
A preventive maintenance program for extractive monitoring systems involves daily operation checks of the
system which include strip chart recorders/data handling devices, indicator lights, the conditioning system and a
daily instrument calibratioa For in-situ systems, daily checks include strip chart recordings and computer
applications, instrument calibrations using gas cells, and proper alignment of the optical system. Routine
maintenance should be performed every thirty days for both extractive and in-situ monitoring systems. For
extractive systems, such components as conditioning system filters, valves, plumbing (for corrosion and leaks),
15-15
-------
electrical components, pumps and chillers should be inspected and replaced if faulty operation of any of these
components is evident Routine maintenance for in-situ systems includes an inspection and/or replacement of the
optical system components, electrical parts and chopper motors. Corrective maintenance for both extractive and
in-situ systems involves a thorough knowledge of all CEM subsystems and a consideration of the severe
environmental conditions for which these monitoring systems are required to operate in.
TABLE 15-2 Irvshu Gas Analyzer Problems and Possible Solutions ^ vdlfp :|:;jlf '
PROBLEM
Excessive dirt Buildup on windows or
thimbles
Cyclic drift In signal unrelated to plant
performance - due to ambient
temperature changes. Signal becomes
erratic from high temperatures
Optics misalignment/electrical noise
due to stack or dust
Signal becomes erratic at high
opacities.
Misaligned system.
Probe/seal leaks (In-stack monitors).
Lightning strikes
Static charge bul Idup burning out
circuit boards
Lamp birnout/degradation
Gas eel Is unreliable
Spurious readings during plant
start-up, shutdown, etc.
Improper temperature compensation.
increased response time.
Deference- OualttvAssnnce Handbook ta
&A-60Q/4'77-
-------
REVIEW EXERCISES
1. True or false. A corrective maintenance program involves trouble-
shooting the monitoring system whenever it breaks down in order
to detect the cause of the problem.
2. True or false. Training in-house personnel for performing CEM
maintenance should consist of a thorough study of the CEM
owner's manual.
1. True
3. Which of the following has been shown by a recent survey as being
problem areas in CEM maintenance programs:
a. Data handling systems.
b. Spare parts.
c. Emergency service.
d. All of the above.
2. False
4. Daily CEM operation checks for an extractive monitoring system
include which of the following:
a. strip charts and/or other data recording devices.
b. system indicator lights.
c. conditioning system.
d. a, b, and c.
e. a, b, c, and a calibration check of the system.
3. d
5. The recommended period for performing routing maintenance
checks on extractive monitoring systems are every .
4. e
6. True or false. Inspecting the overall cleanliness of an extractive 5 39 days
monitoring system is really not necessary.
7. True or false. The majority of monitoring system problems encoun-
tered with extractive systems occur in the system electronics.
6. False
8. By far, extractive monitoring systems encounter their greatest
mechanical problems in
a. heat traced lines.
b. the indicator light system.
c. the conditioning system.
d. none of the above.
7. True
15-17
-------
9. True or false. If the system indicator lights on an in-situ monitoring
system give the indication of a potential problem, it should be
assumed that the light circuit has malfunctioned and the operator
may proceed with a daily calibration check of the system.
8. c
10. Gas cells which are used to calibrate in-situ monitors are:
a. not normally certified by independent laboratories.
b. are always EPA certified.
c. are not EPA certified but certified by OSHA.
d. certified by both EPA and OSHA.
9. False
11. True or false. If an in-situ monitor is calibrated manually using
calibration cells, the temperature compensation circuit can be by-
passed.
10. a
12. The three most common problem areas which have been associated
with in-situ monitoring systems are:
a. I) plugged filters, 2) moisture condensation. 3) improper CEM
design.
b. 1) improper CEM design 2) plugged filters, 3) improper stack
or duct design.
c. 1) ambient dust and severe weather, 2) sensitivity to vibration,
3) moisture condensatioa
d. In-situ systems are superior in design and never encounter any
operational problems.
11. True
12. c
15-18
-------
REFERENCES
1. Electrical Power Research Institute. 1955 Continuous Emission Monitoring Guidelines: Update.
EPRICS-5998.
2. EnvironmentalProtectionAgency. 1985. QualityAssuranceHandbookforAirPoUutionMeasurement
Systems, Volume HI, Stationary Source Specific Methods. Section 3.0.10 Guideline for Developing
Quality Control Procedures for Gaseous Continuous Emission Monitoring Systems. EPA 600/4-77-
0276.
3. EnvironmentalProtectionAgency. 1988. Quality Assurance Handbookfor Air PollutionMeasurement
Systems, Volume III, Stationary Source Specific Methods. Section 3.0.9 Continuous Emission
Monitoring (CEM) Systems Good Operating Practices. EPA 600/4-77-027b.
15-19
------- |