Protocols and Performance Specs

7/28/2019 Protocols and Performance Specs

1/61

000.00

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR

CONTINUOUS MONITORING OF GASEOUS EMISSIONS

FROM THERMAL POWER GENERATION

Environmental Protection Series

REPORT EPS 1/PG/7 (REVISED)

December 2005


2/61

ENVIRONMENTAL PROTECTION SERIES

Sample Number:

EPS 3 / HA / 1

Report number with the qualifier EPS 3/HASubject Area CodeReport CategoryEnvironmental Protection Series

Categories

1 Regulations/Guidelines/Codes of Practice2 Problem Assessment and Control Options3 Research and Technology4 Literature Reviews5 Surveys6 Social, Economic and Environmental Impact

Assessments7 Surveillance8 Policy Proposals and Statements9 Manuals

Subjects

AG AgricultureAN Anaerobic TechnologyAP Airborne PollutantsAT Aquatic ToxicityCC Commercial ChemicalsCE Consumers and the EnvironmentCI Chemical IndustriesFA Federal ActivitiesFP Food ProcessingHA Hazardous WastesIC Inorganic ChemicalsMA Marine PollutantsMM Mining and Ore ProcessingNR Northern and Rural RegionsPF Paper and FibresPG Power GenerationPN Petroleum and Natural GasRA Refrigeration and Air ConditioningRM Reference MethodsSF Surface FinishingSP Oil and Chemical SpillsSRM Standard Reference MethodsTS Transportation SystemsTX TextilesUP Urban PollutionWP Wood Protection/Preservation

New subjects and codes are introduced as they become necessary. A list of EPS reports may be obtained from EnvironmentalProtection Publications, Environmental Protection Service, Environment Canada, Ottawa, Ontario, K1A 0H3, Canada.

Her Majesty the Queen in Right of Canada (Environment Canada) 2005.


3/61

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FORCONTINUOUS MONITORING OF GASEOUS EMISSIONS FROMTHERMALPOWERGENERATION

Report EPS 1/PG/7 (revised)

December2005


4/61

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATIONii

Library and Archives Canada Cataloguing in Publication

Protocols and Performance Specifications for Continuous Monitoring ofGaseous Emissions from Thermal Power Generation / Emissions Research andMeasurement Division, Environmental Technology Centre, Environmental

Protection Service, Environment Canada.

(Report; EPS 1/PG/7 (rev.))Issued also in French under title: Protocoles et spcifications de rendement pour la surveillance continuedes missions gazeuses des centrales thermiques.Includes bibliographical references: p. 42ISBN 0-662-41349-0Cat. no.: En83-2/1-7E

1. Steam power plantsEnvironmental aspectsCanada.2. AirPollutionMeasurement.3. Sulphur dioxideEnvironmental aspects.

4. Nitrogen oxidesEnvironmental aspects.5. Environmental monitoringCanada.I. Canada. Environment Canada.II. Environmental Technology Centre (Canada). Emissions Research and Measurement Division.III. Series: Report (Canada. Environment Canada) EPS 1/PG/7 (rev.).

TD888 S7 P68 2005 363.738'7 C2005-980226-X


5/61

This document has been reviewed by officials of the Emissions Research and Measurement Divison ofthe Environmental Technology Centre at Environment Canada, and approved for publication. Mentionof trade names or commercial products does not constitute recommendation or endorsement for use.

This document does not purport to address all of the safety aspects associated with its use. Anyoneusing this document has the responsibility to consult the appropriate authorities and to establish healthand safety practices in conjunction with any regulatory requirements prior to its use.

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATION iii

REVIEWNOTICE


6/61

Comments on or inquiries about the content of this report should be addressed to:

Emissions Research and Measurement DivisionEnvironmental Technology CentreEnvironmental Protection ServiceEnvironment CanadaOttawa, OntarioK1A 0H3

Additional copies of this report are available from:

Environment Canada PublicationsEnvironment CanadaOttawa, Ontario

K1A 0H3

Toll-free telephone: 1 800 734-3232Telephone: (819) 953-5750Facsimile: (819) 994-5629E-mail: [email protected]: http://www.ec.gc.ca/publications/index.cfm

This report is also available online at:http://www.ec.gc.ca/cleanair-airpur/caol/electricity_Generation/protocols_performance/toc_e.cfm

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATIONiv

READERS COMMENTS


7/61

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATION v

This report outlines specifications for the design, installation, certification, and operation of automatedcontinuous emission monitoring (CEM) systems used to measure gaseous releases of sulphur dioxideand nitrogen oxides from thermal power generation. The procedures used during certification testing ofeach installed CEM system are also presented. This report also describes quality assurance and qualitycontrol (QA/QC) procedures, including the contents of a site-specific QA/QC manual, which must bedeveloped by the system operator for each installed CEM system.

ABSTRACT


8/61

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATIONvi

ACKNOWLEDGEMENTS

This report was developed in consultation with the following persons:

Anthony Bielecki New Brunswick Power

Tracey Canney Cement Association of Canada

Christine Cinnamon Canadian Gas Association

Lorie Cummings Environment Canada

David Duthie Canadian Boiler Society

David Ebsary Ontario Power Generation

Ryan Hunting ATCO Electric Ltd.

John Hutchison Ontario Ministry of the Environment

David Law Air Testing Services

Denis Maftei Ontario Ministry of the EnvironmentGeorge Marson Environment Canada

Debra McLellan Nova Scotia Power

Stephanie Millson Ontario Ministry of the Environment

Lisa Minotti Ontario Ministry of the Environment

Jim Mulvale Association of Independent Power Producers of Ontario

Geoff Ross Environment Canada

Adolfo Silva Canadian Petroleum Products Institute

George Venta Cement Association of Canada

The assistance of Dr. James Jahnke of Source Technology Associates and Brian Williams during thedevelopment of this report is also gratefully acknowledged.


9/61

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATION vii

TABLE OF CONTENTS

ABSTRACT..................................................................................................................................................v

ACKNOWLEDGEMENTS ........................................................................................................................vi

SECTION 1.0 INTRODUCTION..........................................................................................................1

SECTION 2.0 SUMMARY OF SPECIFICATIONS AND PROTOCOLS ............................................2

SECTION 3.0 DESIGN SPECIFICATIONS AND TEST PROCEDURES ........................................43.1 Sample Interface/Conditioning Subsystem Specifications..................4

3.1.1 Location of the Calibration Gas Injection Port ..................................4

3.2 Gas Analyzer Subsystem Specifications ....................................................43.2.1 Operating Range ..................................................................................43.2.2 Interference............................................................................................43.2.3 Temperature-Response Drifts ..............................................................4

3.3 Flow Monitor Subsystem Specifications...................................................53.3.1 Operating Range ..................................................................................6

3.4 Data Acquisition Subsystem Specifications ............................................................................ 63.4.1 Averaging Time ....................................................................................63.4.2 Reporting Basis ......................................................................................73.4.3 Backfilling of Missing Data ..................................................................7

3.5 Overall System Specifications......................................................................73.5.1 Cycle Time Time-Shared Systems....................................................7

3.6 Test Procedures for Verification of Design Specifications...................73.6.1 Analyzer Interference ............................................................................73.6.2 Analyzer Temperature-Response Zero and Span Drifts ....................83.6.3 System Cycle Time ................................................................................83.6.4 Manufacturers Certificate of Conformance ......................................8

SECTION 4.0 INSTALLATION SPECIFICATIONS ............................................................................94.1 Location of Sampling Site ............................................................................94.2 Representativeness...........................................................................................9

4.2.1 Stratification Test Procedure ................................................................9

SECTION

5.0 CERTIFICATION

PERFORMANCE

SPECIFICATIONS AND

TEST

PROCEDURES .....

115.1 Certification Performance Specifications...............................................115.1.1 Operational Test Period ......................................................................115.1.2 Calibration Drift ..................................................................................115.1.3 Electronic Drift ....................................................................................145.1.4 System Response Time ......................................................................14


10/61

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATIONviii

5.1.5 Relative Accuracy ................................................................................145.1.6 Bias........................................................................................................145.1.7 Orientation Sensitivity ........................................................................14

5.2 Calibration Gases..........................................................................................155.3 Certification Test Procedures ....................................................................15

5.3.1 Operational Test Period ......................................................................155.3.2 Calibration Drift Test Protocols..........................................................15

5.3.3 System Response Time Test Protocols ..............................................165.3.4 Relative Accuracy Test Protocols........................................................165.3.5 Bias Test Calculations ..........................................................................205.3.6 Orientation Sensitivity Test Protocols ..............................................20

SECTION 6.0 Q UALITYASSURANCE AND QUALITYCONTROL............................................216.1 Quality Assurance/Quality Control Manual ........................................21

6.1.1 Quality Assurance Activities ..............................................................216.1.2 Quality Control Activities ..................................................................21

6.2 Daily Performance Evaluations ................................................................246.2.1 Calibration Drift ..................................................................................246.2.2 Electronic Drift ....................................................................................29

6.3 Quarterly Performance Evaluations ........................................................296.3.1 Cylinder Gas Test ................................................................................296.3.2 Stack Gas Flow Test ............................................................................316.3.3 F-Factor System Test ..........................................................................33

6.4 Semiannual Performance Evaluations.....................................................346.4.1 Relative Accuracy and Bias Tests ......................................................346.4.2 Exemptions from Semiannual Evaluations........................................35

6.5 Annual Performance Evaluations .............................................................356.5.1 Availability ............................................................................................356.5.2 Independent Inspection ....................................................................35

6.6 Criteria for Acceptable Quality Assurance/Quality ControlProcedures .......................................................................................................35

6.7 Quality Assurance Reporting Requirements.........................................35

GLOSSARY................................................................................................................................................37

UNITS, ABBREVIATIONS, ANDACRONYMS ........................................................................................39

REFERENCES............................................................................................................................................41

BIBLIOGRAPHY........................................................................................................................................42

APPENDIXA CALCULATION OF EMISSIONS BYF-FACTORS........................................................43A.1 Introduction ...................................................................................................43A.2 Oxygen-Based Dry Measurement Systems ............................................43A.3 Oxygen-Based Wet Measurement Systems ............................................44A.4 Carbon Dioxide-Based Measurement Systems.....................................44A.5 Mixed-Basis Measurement Systems .........................................................45A.6 Combined Combustion of Fuels ..............................................................46A.7 Calculation of Customized F-Factors .....................................................46


11/61

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATION ix

APPENDIXB DETERMINATION OF MASS EMISSION RATES ......................................................47B.1 Introduction ...................................................................................................47B.2 Method A: Energy Input Method Metering of Fuel Flows ........47

B.2.1 Determination of Heat Input Rate for Liquid Fuels ........................47B.2.2 Determination of Heat Input Rate for Gaseous Fuels......................48B.2.3 Determination of Heat Input Rate for Solid Fuels ..........................48

B.3 Method B: Determination Using Real-Time Stack Gas FlowMonitors ..........................................................................................................48

B.4 Method C: Energy Balance Method........................................................49

LIST OF FIGURES

FIGURE 1 Summary of Specifications and Protocols for ContinuousEmission Monitoring Systems ...............................................................................3

FIGURE 2 Quality Assurance/Quality Control Schematic ....................................................21

LIST OF TABLES

TABLE 1 Design Specifications for Continuous Emission Monitoring Systems ....................5

TABLE 2 Location of System Calibration Gas Injection Ports for Specific ContinuousEmission Monitoring Systems ...............................................................................6

TABLE 3 Certification Performance Specifications .............................................................12

TABLE 4 t Values................................................................................................................19

TABLE 5 Table of Contents for Quality Assurance/Quality Control Manual .....................22

TABLE 6

Daily, Quarterly, Semiannual, and Annual Performance Evaluations Summary ......25TABLE A-1 F-Factors for Selected Canadian Fuels .................................................................43


12/61

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATION 1

This report provides specifications for thedesign, installation, and operation of automated

continuous emission monitoring (CEM) systemsused to measure releases of sulphur dioxide (SO2)and nitrogen oxides (NOX) from thermal powergenerating facilities. The procedures used todetermine the various CEM system parametersduring initial certification testing and subsequentlong-term operation of the monitoring system arepresented.

No specific monitoring system has beendesignated in this report. Any system that meetsinitial certification criteria, specified parametersand quality assurance/quality control (QA/QC)requirements is acceptable. In situ or extractiveCEM systems based on dynamic dilutiontechnology or direct measurement of the targetspecies may be used. Time-shared CEM systemsusing a single set of analyzers to determineemission rates for several sources are acceptable.

Guidance has been provided to assist the operatorin developing a site-specific QA/QC plan, inconjunction with the appropriate regulatoryagency. The resulting plan forms an integral partof the overall requirements for the operation of

each CEM system.

This report provides guidance for the acquisitionof technically valid CEM data, which may be usedfor multiple purposes, including emission budgetprograms. It does not, however, address issuesspecific to any emission trading program, suchas reporting formats, seasonal averaging, dataretention requirements, etc., which should becompatible with the policies of the program anddefined by the corresponding regulating authority.

While SO2 and NOX are the pollutants mostoften associated with the flue gases released

from thermal power generating facilities, someor all of the concepts and procedures describedherein could also be used, as appropriate, for themeasurement and monitoring of SO2 and NOXin other streams or for the measurement ofother species, regardless of their origin. In suchcases, the appropriate regulatory authority thatmandates the monitoring conditions may adjust,expand, or reduce the requirements detailedin this document in order to reflect the specificconcerns and/or constraints related to the needto measure and monitor the particular species inquestion.

The personnel performing the initial certificationand subsequent audits must be trained andexperienced.

The application of this method will entailhealth and safety hazards. Individuals performingthe certification and audits are responsiblefor obtaining the required training to meetoccupational health and safety standards.

SECTION 1.0 INTRODUCTION


13/61

The specifications that must be met and theprocedures that must be followed for the

installation, certification, and continued operationof a CEM system are summarized in Figure 1.This summary is intended to help the operator toplan and carry out the numerous tasks involvedin installing and operating the monitoring systemand to continue to generate accurate long-termemissions data.

Section 3 outlines the specifications for theoverall CEM system and subsystems, along withassociated procedures for measuring theseparameters. This section will assist the operatorduring the initial design and/or purchase stages.Specific requirements are provided for the dataacquisition system, including emission reportingrequirements.

Specifications for installing the CEM system aregiven in Section 4. These are used to ensure thata test location meets some minimal requirementswith respect to the representativeness of the gasflow and the accessibility of the equipment formaintenance.

After installation, the CEM system is testedfollowing the protocols provided in Section 5.

The emissions data are compared with thosefrom manual reference methods or acceptableautomated measurement techniques to ensurethat the specifications have been met. When aninstalled system has met or surpassed all thesespecifications, it is deemed to be certified andcapable of generating quality-assured emissionsdata.

A QA/QC plan must be developed for each CEMsystem by the operator. Section 6 provides thebasis for the development of this plan. The QA/QCplan must encompass a diverse range of topics,including calibration procedures, maintenance,performance evaluations, and corrective actions.Each CEM system will require a QA/QC plan;however, if a number of identical CEM systemsare operated, a single QA/QC plan is acceptable,provided that appropriate records are maintainedfor each CEM system.

PROTOCOLS AND PERFORMANCE SPECIFICATIONS FOR CONTINUOUS MONITORING OF GASEOUS EMISSIONS FROM THERMAL POWER GENERATION2

SECTION 2.0 SUMMARY OF SPECIFICATIONS AND PROTOCOLS


14/61


Figure 1 Summary of Specifications and Protocols for Continuous EmissionMonitoring Systems

CEM System Specifications and Protocols

Sections 3 and 4Design and

Installation Information

Section 3Design

Specifications andTest Procedures

Section 4Installation

Specifications

Section 5Certification of the

System

Section 6Quality Assurance and

Quality Control

3.1 Interface/Conditioning

Location ofCalibration Port

3.2 Gas Analyzer Operating Range Interference Temperature-

Response Drift

3.3 Flow Monitor Operating Range

3.4 Data Acquisition Averaging Time Reporting Basis Backfilling of Missing

Data

3.5 Overall System Cycle Time

3.6 Test Procedures Interference Rejection Temperature-

Response Drift System Cycle Time Manufacturers

Certificate ofConformance

4.1 Site Location

4.2 Representativeness Stratification Test

5.1 Performance Specifications Operational Test Period Calibration Drift Electronic Drift System Response Time Relative Accuracy Bias Orientation Sensitivity

5.2 Calibration Gases

5.3 Test Procedures Operational Test Period Calibration Drift System Response Time Relative Accuracy Bias Orientation Sensitivity

6.1 QA/QC Manual QA Activities QC Activities

6.2 Daily Performance Evaluations Calibration Drift Electronic Drift

6.3 Quarterly PerformanceEvaluations

Cylinder Gas Test Stack Gas Flow Test F-Factor System Test

6.4 Semiannual PerformanceEvaluations Relative Accuracy and Bias Exemptions

6.5 Annual PerformanceEvaluations Availability Independent Inspection

6.6 Acceptance Criteria

6.7 Reporting Requirements


15/61

A CEM system consists of the following threesubsystems: sample interface/conditioning;

gas analyzers; and

data acquisition.

A flow monitor may also be part of the CEMsystem. When utilizing a flow monitor, if thepollutant concentration is measured on a drybasis whereas the flow rate is measured on awet basis, a continuous moisture monitoringsystem for measuring and recording the moisture

of the stack gases may be required. Further detailson the use of stack gas moisture monitoringsystems are presented in Appendix B.

Specifications for these subsystems are givenin Sections 3.1 to 3.5, while Section 3.6 outlinesthe procedures for determining the value of theparameters, where applicable. The parameters andspecifications for these subsystems are shown inTable 1.

This protocol does not specify measurementtechniques. Components that meet the criteriaspecified in Sections 3.1 to 3.5 and that allow theoverall CEM system to achieve the certificationspecifications in Section 5 and the performanceevaluations in Section 6 are acceptable.

3.1 Sample Interface/ConditioningSubsystem Specifications

3.1.1 Location of the Calibration GasInjection Port

The location of the system calibration gas injectionport is the sole criterion for the sample

interface/conditioning subsystem, with thelocation of this port being specific to the type ofCEM system. The location of the ports for thevarious types of CEM systems is given in Table 2.

3.2 Gas Analyzer SubsystemSpecifications

3.2.1 Operating RangeThe chosen range of the analyzer must encompassall anticipated concentrations for the gasflow being monitored. The average monthlyconcentration for each analyzed gas should fallbetween 40% and 75% of the chosen range.If the average monthly concentration of anypollutant or diluent gas for each specific chosenrange falls outside these limits, the analyzer shouldbe adjusted such that the average is brought back

within these limits.

Note that numerous performance specificationsare defined with reference to the full-scale (FS)setting of the CEM analyzers (see Table 1, Table 3in Section 5, and Table 6 in Section 6). The gasanalyzer of a CEM system may be able to measurelevels higher than the defined FS level; however,this higher level cannot be applied to demonstrateconformance to the performance specifications,which are tailored to the characteristics of theemission source.

If concentrations vary widely, the use ofmultirange analyzers is strongly recommended.The highest range should include the maximumpotential concentration anticipated for theprocess. Note that data that fall outside therange(s) of an analyzer are considered as missingand must be backfilled using the criteria given inSection 3.4.3.

3.2.2 InterferenceEach analyzer must exhibit a response of lessthan 4.0% of FS for the sum of all interferences

due to other gas constituents, as measured bythe procedures given in Section 3.6.1.

3.2.3 Temperature-Response DriftsEach pollutant or diluent gas analyzer used inthe system must exhibit a zero drift less than 2.0%of the FS setting for any 10C change over thetemperature range of 535C. Additionally, eachanalyzer must exhibit a span drift of less than


SECTION 3.0 DESIGN SPECIFICATIONS AND TESTPROCEDURES


16/61

4.0% of the FS setting for any 10C change intemperature from 5 to 35C. Both the zero andspan drift tests are to be carried out within theacceptable temperature operating range of theanalyzer, as specified by the manufacturer. Theprocedures outlined in Section 3.6.2 must befollowed to determine the temperature-response

drift.

Analyzers installed and operated in a temperature-controlled environment are exempt from thisspecification.

3.3 Flow Monitor SubsystemSpecifications

The gas flow monitor should have the capabilityof carrying out daily checks at low and highflow rates as part of the daily system calibrationprocedures. Electronic simulation of low and high

flow may be adequate in some systems, providingthat daily zero and span drift can be calculated.The sensor must cover the full range of gasvelocities anticipated in the flue or duct. Any flowsbeyond the range of the sensor are deemed to bemissing and must be backfilled, as described inSection 3.4.3 of this report.


Table 1 Design Specifications for Continuous Emission Monitoring Systems

Text references

TestSubsystem Parameter Specification Specification procedures

See Table 2 3.1.1

Average monthly concentration between 3.2.1 40% and 75% of full scale (FS)


17/61

3.3.1 Operating RangeThe FS setting should be approximately 100% ofthe maximum potential flow rate.

Note that various performance specifications aredefined with reference to the FS setting of theCEM flow monitor (see Table 1 above, Table 3 inSection 5, and Table 6 in Section 6). The flowmonitor of a CEM system may be able to measurelevels higher than the defined FS level; however,this higher level cannot be applied to demonstrateconformance to the performance specifications,which are tailored to the characteristics of the

emission source.

If flow varies widely, the use of multirange flowmonitors may be advisable for a stack servingseveral units. The highest range should includethe maximum potential flow anticipated for theprocess. Note that data that fall outside therange(s) of a flow monitor are considered asmissing and must be backfilled using the criteriagiven in Section 3.4.3.

3.4 Data Acquisition Subsystem

SpecificationsThe CEM system must include a microprocessor-based data acquisition subsystem that accepts theoutputs of the pollutant and diluent gas analyzersand other associated equipment and convertsthese to emission rates of the pollutant gases inunits of the standard. The system must maintaina record of all parameters in a format and timeframe acceptable to the appropriate regulatory

authority. The system must also record andcompute daily zero and calibration drifts, providefor backfilling of missing data, and record anyother relevant data that the operator may wishto include.

3.4.1 Averaging TimeData must be reduced to 1-hour averages for thepollutant and diluent gases and other measuredparameters. The 1-hour averages must be used tocompute the SO2 and NOX as nitrogen dioxide(NO2) emissions, expressed in units of thestandard. A variety of methods for calculating

emissions are provided in Appendix B.

For time-shared systems, 1-hour averages mustbe computed from four or more values, equallyspaced over each 1-hour period, with theexception of periods during which calibrations,QA activities, maintenance, or repairs are beingcarried out. During these specific activities, avalid hour must consist of a minimum of two datapoints for a time-shared system or 30 minutes ofdata for a CEM system using dedicated analyzers.The calibrations should be conducted in a mannerthat avoids the loss of a valid hour of emissions

data every time that a daily calibration isconducted. This may be achieved by usingmixtures of several pollutant/diluent gases; byscheduling the calibration periods so that theemissions data loss is shared by two consecutivehours; or by scheduling the calibration of differentanalyzers at different hours of the day.


Table 2 Location of System Calibration Gas Injection Ports for Specific ContinuousEmission Monitoring Systems

SystemType Subsystem Specification for location of system calibration gas injection port

Calibration gas must be introduced no further than the probe exit.

Calibration gas must be introduced prior to dilution.

Calibration gas must flood the measurement cavity of the analyzer.

Calibration gas must provide a check on the internal optics and allelectronic circuitry. System may also include an internal calibration devicefor simulating a zero and an upscale calibration value.

Directmeasurement ofgas concentrations

Dilution (in-stackand external)

Point

Path

Extractive

In situ


18/61


3.4.2 Reporting BasisData must be prepared on a quarterly basis andmust be expressed as 720-hour rolling averages askg/MWh net energy output or in any units andaveraging periods required by the appropriateregulatory authority. The data must be available

in both digital and analog form, with the analogform presented as a trend plot in units of thestandard versus time over the reporting period.

3.4.3 Backfilling of Missing DataEmissions data that are missing due to amalfunction of the CEM system (eg. gas analyzer,flow monitor) may be substituted for a period ofup to 168 hours for any single episode using dataderived from emissions versus load data and fuelsulphur content correlations that have previouslybeen determined by the certified, quality-assuredCEM system. When a CEM system is installed

to monitor emissions at the discharge of fluegas control equipment, missing data must besubstituted by deriving data from emissions versusoperating parameter correlations (e.g., averagesulphur content of the fuel, the power load ofthe unit, or other appropriate parameters). Thebackfilling technique must be fully describedin the QA/QC manual developed for eachCEM system and approved by the appropriateregulatory authority.

When a CEM system (eg. gas analyzer, flowmonitor) malfunction extends beyond 168 hours

for any single episode, data must be generatedby another certified CEM system or validreference method. Other CEM systemsused for this purpose must meet all design andperformance specifications given in this report.When using another system, the stack gas samplemust be extracted from the sample port(s) usedfor the reference method during certification ofthe CEM system.

Data that are backfilled using a procedure otherthan a certified alternate CEM system or referencemethod cannot be credited towards meeting theCEM system availability criteria specified in Table 3(in Section 5).

All emissions data should be quality auditedto identify suspected data using proceduresdescribed in the QA/QC plan (Section 6.1).The procedures may include automatic flaggingof a) out-of-range concentrations and flows,b) abnormal system calibration response time,

c) abnormal heat rate levels (for systems fittedwith fuel flow monitors), d) abnormal flow-to-input or flow-to-output levels (for systems fittedwith stack gas flow monitors), and e) abnormalconcentrations during periods when thegenerating unit did not burn fuel.

The QA-flagged data must be investigated andeither accepted or backfilled. The QA-flaggeddata should be identified in the quarterly report,along with a summary of reasons for acceptanceor backfilling.

3.5 Overall System Specifications

3.5.1 Cycle Time Time-Shared SystemsThe specification for cycle time applies to time-shared systems measuring emissions from anumber of sources using a single set of pollutantand diluent gas analyzers. One completemeasurement cycle of all streams must becompleted in 15 minutes or less, generating aminimum of four sets of concentration andemissions data for each hour of operation. For aCEM system measuring the emissions from nstacks, the maximum time available for eachsource being monitored would be 15/n minutes,including switching, stabilization, and analyzeroutput integration times.

3.6 Test Procedures for Verification ofDesign Specifications

3.6.1 Analyzer InterferenceThis test may be carried out after the analyzershave been installed in the CEM system or in alaboratory or other suitable location before theanalyzers are installed. Sufficient time must beallowed for the analyzer under test to warm up,and then the analyzer must be calibrated byintroducing appropriate low- and high-level gasesdirectly to the analyzer sample inlet. After theinitial calibration, test gases must be introduced,each consisting of a single interfering gas at aconcentration representative of that species in the

gas flow to be monitored. The magnitude of theinterference of each potential interfering specieson the target gas must then be determined.

The analyzer is acceptable if the combinedresponse of all interfering gases is less than4.0% of the FS setting.


19/61


3.6.2 Analyzer Temperature-Response Zeroand Span Drifts

The analyzer must be placed in a climate-controlled chamber in which the temperaturecan be varied from 5 to 35C. Sufficient timemust be allowed for the analyzer to warm up, and

then the analyzer must be calibrated at 25C usingappropriate zero and span gases. The temperatureof the chamber must be adjusted to 35, 15and 5C. It should be ensured that the analyzertemperature has stabilized. The power to theanalyzer must not be turned off over the durationof this test.

When the analyzer has stabilized at each climatechamber temperature, each of the calibrationgases must be introduced at the same flow orpressure conditions, and the response of theanalyzer must be noted.

The temperature-response zero drift is calculatedfrom the difference between the indicated zeroreading and the reading at the next higher orlower temperature. The analyzer is acceptable ifthe difference between all adjacent (i.e., 5/15,15/25, and 25/35C) zero responses is less than2.0% of the FS setting.

The temperature-response span drift is calculatedfrom the differences between adjacent spanresponses. The analyzer is acceptable if thedifference between all adjacent span responsesis less than 4.0% of the FS setting.

3.6.3 System Cycle TimeThe system cycle time is set by the manufacturerduring design and must meet the specificationgiven in Section 3.5.1.

3.6.4 Manufacturers Certificate ofConformance

It may be considered that specifications for bothinterference and temperature-response drifts havebeen met if the analyzer manufacturer certifiesthat an identical, randomly selected analyzer,manufactured in the same quarter as the deliveredunit, was tested according to the procedures given

in Sections 3.6.1 and 3.6.2, and the parameterswere found to meet the specifications.


20/61


This section contains guidance for selecting asuitable sampling site on the flue or duct and

determining the representativeness of the desiredlocation with respect to the homogeneity of thegas flow.

4.1 Location of Sampling Site

The probe or in situ analyzer must be installedin a location that is accessible at all times andduring any weather conditions, so that routinemaintenance can be performed on schedule, asoutlined in the QA/QC manual. Sufficient sheltershould be provided on outdoor installations sothat maintenance can be safely performed during

any weather conditions without detriment toeither the CEM system or service personnel. Thedegree of exposure, seasonal weather conditions,servicing and maintenance, susceptibility tolightning strikes, and vibration of the duct and/orplatform are some of the considerations whensiting a probe or in situ analyzer.

Before a flow rate sensor is permanently installed,it should be ensured that cyclonic flow is notpresent at the desired sampling location. Thepresence of a cyclonic flow pattern will addconsiderable complexity to both certification

and operation of the installed sensor. It isrecommended that an alternate location befound if cyclonic flow patterns are verified at aproposed site. The protocols given in this reportrelate only to sources for which the gas flowpattern has been demonstrated to be non-cyclonic.

4.2 Representativeness

The probe or in situ analyzer must be installed in alocation where the flue gases are well mixed. Thedegree of turbulence and mixing time are major

factors that influence the extent of stratificationof the flue gases.

The extent of stratification of the flue gases at anylocation must be determined using the applicabletest methods. It is therefore highly recommendedthat the procedures outlined in Section 4.2.1 becarried out at a proposed analyzer installation siteto determine the extent of stratification before

installing the CEM system. If significant gasstratification of any of the measured species is

present at the proposed location, then seriousconsideration should be given to relocating thesystem to another location where the flow hasbeen determined to be non-stratified.

If stack flow monitoring is a component ofthe CEM system, then it is highly recommendedthat the adequacy of the sampling site be assessedwith respect to the selected flow monitoringsystem as well as to the reference method to beused for the initial certification and for the annualor semiannual evaluations.

Before the flow monitor is installed, a numberof velocity traverses must be carried out at theproposed sensor installation location over a rangeof loads using the equipment and proceduresfound in Method B of Reference Method EPS1/RM/8 (Reference Methods for Source Testing:Measurement of Releases of Particulate fromStationary Sources, Environment Canada,December 1993, as amended). The degree ofcyclonic flow is determined using the proceduresfound in Method A of Reference Method EPS1/RM/8. These measurements will provide a

basis for the location of the sensor and will alsodemonstrate the absence of cyclonic flow (averagerotational angle 15 degrees). The location ofsampling ports must be selected so as to avoidinterference between the flow monitor, theconcentration measurement point(s) or path,and the RM probes.

If a single-point velocity sensor is being installed,the sensing tip must be located at a point yieldingvelocity measurements within the specificationsover the full range of loads. The velocity profiledata must be used to select the optimum

measurement point.

4.2.1 Stratification Test ProcedureA minimum of nine sampling points must be usedin the stack or duct, applying the procedures forselecting sampling points found in ReferenceMethod EPS 1/RM/8. Using two automatedsystems with similar response characteristics, theconcentration of a target gas must be measured

SECTION 4.0 INSTALLATION SPECIFICATIONS


21/61

at each of the sampling points in the matrix withone system (traversing system), whilesimultaneously measuring the target gasconcentration at a fixed or reference location,usually at the centre of the flue or duct.

Note that a stratification test must be carried outfor each gaseous species measured by the installedCEM system, including the diluent gas(es).

The concentration of the gas measured at thefixed location (stability reference measurement) inthe flue/duct is used as an indicator of the stabilityof the gas flow. If this concentration varies bymore than 10% of the average concentration forlonger than 1 minute during this test, thestratification test must be carried out when morestable conditions prevail.

Note that the installed analyzer in the CEMsystem, which withdraws a sample from a fixedpoint, may be used as the stability referencemeasurement for the stratification test. Theresponse characteristics of the reference and thetraversing analyzers should be similar.

The concentration of a target gas must bemeasured at each of the sampling points in thematrix. At the conclusion of the traverses, themeasurement of the concentration must berepeated at the initial measurement point. If theconcentrations differ by more than 10% for the

pre- and post-test values at this point, stratificationmust be retested when more stable conditionsprevail.

The degree of stratification for each species iscalculated at each traverse point within the gasflow using Equation 1.

where:STi = stratification (%)Ci = concentration of the measured species at

point iCavg = average of all measured concentrations

The flow in the stack or duct is considered to bestratified if any calculated value using Equation 1exceeds 10%.

STi = Equation 1(CiCavg)

Cavg

x100



22/61

To achieve certification, an installed CEM systemmust meet all of the performance specificationsoutlined in Table 3. The specifications are relevantto each pollutant and diluent gas measured,as well as the stack gas flow measurement (ifapplicable) and the overall CEM system. A systemmay be partially certified for example, for SO2or NOX and then fully certified at a later datewhen deficiencies in specific portions of thesystem have been corrected.

The specifications are described in Section 5.1.The gases used during certification testing aredescribed in Section 5.2, while the applicable testprocedures are outlined in Section 5.3.

5.1 Certification PerformanceSpecifications

It is recommended that after the CEM systemhas been installed according to the manufacturerswritten instructions, the entire CEM systemshould be operated for a conditioning period ofnot less than 168 hours, prior to the operationaltest period (OTP), during which the emission

source should be operating. During theconditioning period, the entire CEM systemshould operate normally that is, analyzing theconcentration of the pollutant and diluent gases with the exception of periods during whichcalibration procedures are being carried out aswell as other procedures indicated in the QA/QCmanual.

5.1.1 Operational Test PeriodThe OTP is a 168-hour cumulative timeperiod during which most of the performancespecification tests are carried out. The 168-hour

period may be contiguous or fragmented inperiods of no less than 24 hours. No unscheduledmaintenance, repairs, or adjustments to theCEM system are allowed during the OTP. Theprocedures in the QA/QC manual must befollowed as if the CEM system were generatingemissions data.

CEM systems installed at peaking stations areexempted from the OTP and calibration drift tests.

5.1.2 Calibration DriftThe calibration drift specification is applicableat the three concentration ranges indicated inTable 3 and is applicable to each pollutant anddiluent gas analyzer. Table 3 also includes flowmonitoring calibration drift specifications.

For the gas analyzers, this procedure tests bothlinearity and calibration drift.

At 24-hour intervals over the 168-hour OTP, theCEM system response to the pollutant or diluentcalibration gases, as indicated by the dataacquisition system, must not deviate from thecertified value of the appropriate gas by anamount exceeding the greater of:

Pollutant gas analyzerLow level: 2.0% of the FS setting or 2.5 ppm,

absolute differenceMid level: 2.0% of the FS setting or 2.5 ppm,

absolute difference

High level: 2.5% of the FS setting or 2.5 ppm,absolute difference

Diluent gas analyzerLow level: 0.5% O2 (or CO2)Mid level: 0.5% O2 (or CO2)High level: 0.5% O2 (or CO2)

At 24-hour intervals over the 168-hour OTP, theCEM system response to the stack gas flow (andstack gas moisture, if applicable), as indicated bythe data acquisition system, must not deviate froma concurrent RM measurement by an amount

exceeding the greater of:

Flow monitorLow level: 3.0% of the FS setting or 0.6 m/s,

absolute differenceMid level: 3.0% of the FS setting or 0.6 m/s,

absolute difference


SECTION 5.0 CERTIFICATION PERFORMANCESPECIFICATIONS AND TEST PROCEDURES


23/61

PROTOCOLSANDPERFORMANCESPECIFICATIONSFORCONTINUOUSMONITORINGOFGASEOUSEMISSIONSFROMTH

ERMALPOWERGENERATION

12

Table 3 Certification Performance Specifications

Parameter Component Level Specification

24-hourcalibrationdrift

Electronicdrift

Systemresponsetime

Relativeaccuracy(RA)

SO2/NOX gas analyzer

O2/CO2 gasanalyzer

Flow monitor

Stack gasmoisture monitor*

Dedicated analyzer

Time-shared system


O2/CO2 gas analyzer

Flow monitor

Stack gasmoisture monitor*

Mass emission

Low level (020% FS)Mid level (4060% FS)High level (80100% FS)




Minimum safe and stable loadMid load (4060%)Full load (90100%)

Minimum safe and stable loadMid load (4060%)Full load (90100%)

the greater of2.0% FS or 2.5 ppm absolute difference2.0% FS or 2.5 ppm absolute difference2.5% FS or 2.5 ppm absolute difference

0.5% O2 (or CO2)0.5% O2 (or CO2)0.5% O2 (or CO2)

the greater of3.0% FS or 0.6 m/s absolute difference3.0% FS or 0.6 m/s absolute difference3.0% FS or 0.6 m/s absolute difference

2.0% of (100 %Bws)2.0% of (100 %Bws)2.0% of (100 %Bws)

3.0% FS in 24 hours

200 seconds for 90% change

15 minutes for 90% change

the greater of10.0% RA or 8 ppm avg. absolute difference

the greater of 10.0% RA or 0.5% O2(or CO2) avg. absolute difference

the greater of10.0% RA or 0.6 m/s avg. absolute difference10.0% RA or 0.6 m/s avg. absolute difference10.0% RA or 0.6 m/s avg. absolute difference

10.0% RA of (100 %Bws)10.0% RA of (100 %Bws)10.0% RA of (100 %Bws)

the greater of 10.0% RA or 7.3 g/GJ inputavg. absolute difference


24/61

PROTOCOLSANDPERFORMANCESPECIFICATIONSFORCONTINUOUSMONITORINGOFGASEOUSEMISSIONSFROMTHERMAL

POWERGENERATION

13

Table 3 Certification Performance Specifications (contd.)

Parameter Component Level Specification

* The performance specifications for stack gas moisture monitoring systems based on wet O2 and dry O2 measurement (calibration drift, response time, relative asummarized in Table 3, Section 5, and in Table 6, Section 6. Other moisture monitoring systems may be proposed for use with Equation B-3 in Appendix B, if tthat the system meets the required specifications. The specific QA activities related to the moisture monitoring system must then be described in the QA/QC m

* Further details on the use of stack gas moisture monitoring systems are presented in Appendix B. Note that %Bws is the moisture content of the stack gas (% v/

Bias

Orientationsensitivity


O2/CO2 gas analyzer

Flow monitor

Stack gas moisture monitor*

the greater of 5.0% of FS value or 5 ppmavg. absolute difference

the greater of 5.0% of FS value or0.5% O2 (or CO2) avg. absolute difference

the greater of 5.0% of FS value or0.6 m/s avg. absolute difference

5.0% of mean FS (100 %Bws ) value

4.0% of value measured at zero orientation


25/61

High level: 3.0% of the FS setting or 0.6 m/s,absolute difference

Stack gas moisture monitorLow level: 2.0% of (100 %Bws)Mid level: 2.0% of (100 %Bws)

High level: 2.0% of (100 %Bws)

Calibration drift must be tested according toprocedures in Section 5.3.2.

Further details on the use of stack gas moisturemonitoring systems are presented in Appendix B.

5.1.3 Electronic DriftThe electronic drift of stack gas flow monitorsthat do not perform daily flow system calibrationchecks must not deviate from the value of theelectric input signal by more than 3.0% FS.

5.1.4 System Response TimeFor CEM systems using dedicated analyzers,the system response time is acceptable if theaverage of three increasing and three decreasingvalues is no greater than 200 seconds, for eachanalyzer, for a 90% response to a step change inconcentration of gas at the probe exit. Note thatthis includes the lag time.

For time-shared systems, the system responsetime is acceptable if the average of threeincreasing and three decreasing values is no

greater than 15 minutes, for each analyzer oneach stream, for a 90% response to a step changein concentration of gas at the probe exit. Notethat this includes the lag time.

System response time must be tested according toprocedures in Section 5.3.3.

5.1.5 Relative AccuracyThe relative accuracy for an SO2 and NOX gasanalyzer must not exceed 10.0% or 8 ppmaverage absolute difference (|d|), whichever isgreater.

The relative accuracy for an O2 (or CO2) gasanalyzer must not exceed 10.0% or 0.5% O2(or CO2) average absolute difference (|d|),whichever is greater.

The relative accuracy for a stack gas flow monitormust not exceed 10.0% or 0.6 m/s averageabsolute difference (|d|), whichever is greater.

The relative accuracy for a stack gas moisturemonitor must not exceed 10.0% of (100 %Bws).

The relative accuracy for SO2 and NOX massemissions must not exceed 10.0% or 7.3 g/GJheat input average absolute difference (|d|),

whichever is greater.

Meeting the relative accuracy for SO2, NOX, O2,and CO2 concentrations and stack gas flow doesnot guarantee meeting the relative accuracy forSO2 and NOX mass emissions.

Relative accuracy must be tested according toprocedures in Section 5.3.4.


5.1.6 BiasThe bias for an SO2 and NOX gas analyzer mustnot exceed 5.0% of the FS value or 5 ppmaverage absolute difference, whichever is greater.

The bias for an O2 (or CO2) gas analyzer must notexceed 5.0% of the FS value or 0.5% O2 (or CO2)average absolute difference, whichever is greater.

The bias for a stack gas flow monitor must notexceed 5.0% of the FS value or 0.6 m/s averageabsolute difference, whichever is greater.

The bias for a stack gas moisture monitor mustnot exceed 5.0% of the FS value for (100 %Bws).

Bias must be tested according to calculations inSection 5.3.5.


Should there be any bias as defined inSection 5.3.5, either positive or negative, in anymeasurements made by the CEM system, thedata that are subsequently generated must becorrected for the bias before any subsequentuse is made of the data.

5.1.7 Orientation SensitivitySome stack gas flow monitors may be sensitiveto the probe orientation in the gas stream. Forthese monitors, the indicated gas flow rate of thesensor at orientations other than that at the zero



26/61

degree measurement must not differ from thezero orientation by more than 4.0%.

Orientation sensitivity must be tested according toprocedures in Section 5.3.6.

5.2 Calibration GasesThe gases used by both the CEM system and thereference method during the relative accuracy testmust be U.S. Environmental Protection Agency(EPA) protocol grade.

Gases used during the calibration drift andresponse time tests must be certified to anaccuracy of 2.0% by the supplier, but protocolgases may be used if desired.

The QA/QC manual should specify a methodof cross-referencing successive gas cylinders toidentify out-of-specification cylinders before thenew cylinders are used to calibrate the CEMsystem.

5.3 Certification Test Procedures

5.3.1 Operational Test PeriodDuring the OTP, the CEM system must continueto analyze flue gases without interruption andproduce a record of the emissions data usingthe data acquisition system. This record must bekept for the duration required by the appropriateregulatory authority. Sampling may be interruptedduring this test period only to carry out systeminstrument calibration checks and specifiedprocedures contained in the QA/QC manual.

During this period, no unscheduled maintenance,repairs, or adjustments to the CEM system maybe carried out; otherwise, the OTP must berestarted. Calibration adjustments may beperformed at 24-hour intervals or more frequentlyif specified by the manufacturer and stated in theQA/QC manual. Automatic zero and calibrationadjustments made without operator intervention

may be carried out at any time, but theseadjustments must be documented by the dataacquisition system.

If the test period is fragmented due to processshutdown, the times and dates of this periodshould be recorded and the test continuedwhen the source resumes operation. If the testis interrupted due to CEM system failure, the

entire test period must be restarted after theproblem has been rectified.

The performance specification tests outlinedin Sections 5.3.2 to 5.3.6 must be carried outduring the OTP, with the exception of the relative

accuracy test (Section 5.3.4), which may beconducted during the OTP or during the 168-hourperiod immediately following the OTP. It isrecommended that the calibration drift tests becompleted before attempting the relative accuracytests, to minimize the risk associated withrepeating the latter.

5.3.2 Calibration Drift Test ProtocolsThe calibration drift test period may befragmented in subperiods that are not lessthan 24 hours each.

The calibration drift must be determined for eachpollutant gas analyzer, diluent gas analyzer, andstack gas flow monitor at approximately 24-hourintervals over the 168-hour test period.

The following procedures are used during this test.

5.3.2.1 Calibration Adjustments. Automaticor manual calibration adjustments may be carriedout each day. The calibration drift test must beconducted immediately before these adjustmentsor in such a manner that the magnitude of thedrifts can be determined. Since the test is carried

out before adjustments, the magnitude of anydrift occurring in the system or analyzer over the24-hour period is incorporated into the reportedresult.

5.3.2.2 Test Procedures. On the first day of theperformance test period, the calibration of thesystem must be checked by injecting the threecalibration gases (Section 5.2) at the primary CEMsystem calibration port, as indicated in Section3.1.1 of this report.

The system must be challenged three times dailyat approximately 24-hour intervals with each ofthe low-, mid-, and high-level calibration gasesfor the pollutant and diluent species, for a totalof nine tests each. The three ranges for each gasmust not be introduced in succession or in thesame sequence, but must be alternated with otherreference gases. The response of the system, asindicated by the data acquisition system, must berecorded, and the average system response of the



27/61

three calibration checks for that day must becalculated.

If the CEM system is fitted with a stack gas flowmonitor, then RM flow measurements mustbe performed three times daily at approximately

24-hour intervals. The concurrent flow from thedata acquisition system must be recorded, andthe average flow error for that day must becalculated. If the RM flow runs of each day areperformed consecutively, then a single moisturedetermination encompassing the three flowruns may be used.

5.3.2.3 Gas Concentration Calculations. Thegas calibration drift for the responses to the low-,mid, and high-level test gases is calculated usingEquation 2.

where:Dc = concentration calibration drift (%)A = average of the three system responses to the

low-, mid-, or high-level calibration gas (% orppm)

R = certified concentration of thelow-, mid-, or high-level test gas(% or ppm)

FS = full-scale setting of the analyzer

(% or ppm)

5.3.2.4 Gas Flow Calculations. The flowmeasurement calibration drift is calculated usingEquation 3.

where:Df = flow calibration drift (%)Af = average gas velocity or flow rate,

as measured by the CEM system(m/s or m3/s)

RM = average gas velocity or flow rate,as measured by the reference method(m/s or m3/s)

FS = full-scale setting of the flow monitor(m/s or m3/s)

Note: The daily drift results must not be averagedwhen reporting the calibration drift measuredduring the system certification.

5.3.2.5 Acceptance Criteria for Certification.The performance specifications presented in

Section 5.1.2 must be met.

5.3.3 System Response Time Test ProtocolsThis test is performed for each pollutant anddiluent gas analyzed, with the results expressedin concentration units, and on the overall CEMsystem, with the results expressed in terms ofthe standard. The test is carried out with theCEM system fully operational. Sample flow rates,pressures, and other parameters must be at thenominal values specified in the manufacturerswritten instructions and outlined in the QA/QCmanual.

5.3.3.1 Test Procedures. Low- and high-levelcalibration gases must be introduced alternately atthe system calibration gas injection port specifiedin Section 3.1.1. Sufficient time must be allowedfor the system to stabilize, and then the dataacquisition system responses to these gases mustbe recorded. This sequence must be carried outthree times, thus generating a total of threeincreasing and three decreasing concentrationchanges. When a time-shared system is beingtested, the injection of the calibration gases mustbe timed to produce the longest possible response

time for the system.

5.3.3.2 Calculations. Using the output of thedata acquisition system, the time required forthe system to achieve a 90% response to theconcentration difference between the low-and high-level gases for both increasingand decreasing gas concentrations must bedetermined. The lag time of extractive systems(i.e. the time necessary to convey the gas samplethrough the sampling line) must be includedwhen determining the time that the systemtakes to reach 90% response change.

5.3.3.3 Acceptance Criteria for Certification.The performance specifications presented inSection 5.1.4 must be met.

5.3.4 Relative Accuracy Test ProtocolsThis test is a comparative evaluation of CEMsystem performance using an independentreference method, which may be either a manual

Df= Equation 3(AfRM)

FSx100

DC= Equation 2(AR)

FSx100



28/61

or automated procedure, as specified by theappropriate regulatory authority. The test iscarried out on each pollutant and diluent gasanalyzer as well as on the stack gas flow monitorand pollutant mass emissions. F-factor-basedmethods may be used for the certification and

for the semiannual performance evaluations forCEM systems that calculate emissions by F-factors(Appendix A). Data from this test are also usedto calculate a system bias.

The emission source must be operating atnormal capacity or at greater than 50% maximumheat input (the latter for new generating unitsor units that did not operate in the previous twoquarters) while combusting the primary fuelnormal for that unit. The CEM system must beoperated in a routine manner during this test,and no adjustments, repairs, or modifications to

any portion of the system may be carried outother than those actions outlined in the QA/QCmanual. As the system includes the hardware andsoftware associated with data acquisition, datamanipulation, and system control, parametersin this subsystem may not be modified duringthe test.

5.3.4.1 Reference Method Sampling Pointfor Non-stratified Flow. Where it has beendemonstrated, using the procedures outlined inSection 4.2.1, that the flue gases are not stratified,the RM testing may be carried out at a single test

point in the flue or duct, with the gas extractionpoint being no closer than 7.5 cm from any wall.

When certifying extractive or in situ point systems,the RM probe tip must be located no closerthan 30 cm from the CEM probe. For in situ pathsystems, the RM probe must be no closer than30 cm from the inner 50% of the measurementpath. The RM probe must be positioned so that itwill not interfere with the operation of the CEMsystem under test.

5.3.4.2 Location of Reference Method SamplingPoints in Presence of Stratified Flow. If the gasflow has been found to be stratified using theprocedures outlined in Section 4.2.1 or if thestratification test has not been performed, theRM sample must be collected at several pointsin the gas flow.

A measurement line that passes through thecentroids of the flue or duct must be established.

This line should be located within 30 cm ofthe CEM sampling system cross-section. Threesampling points must be located at 16.7, 50,and 83.3% along the length of the measurementline. Other sampling points may be selected ifit can be demonstrated that they will provide a

representative sample of the bulk gas flow overthe period of the test.

5.3.4.3 Test Methods. Either integratingmanual or automated methods specified bythe appropriate regulatory authority may beused as the reference methods for this test.

Manual grab sampling reference methods are notacceptable for CEM system certification.

5.3.4.4 Sampling Strategy. A minimum of ninecomparisons of the RM and CEM results must be

conducted to evaluate the performance of theCEM system being tested. When manual samplingreference methods are being used, the samplingmust be carried out at a fixed sampling rate; thatis, the sampling rate must not be adjusted overthe duration of the test, except to maintain theflow at the initial rate. Sampling must be carriedout for 30 minutes during each test, dividedequally over the three sampling points forstratified flow testing or at the single point fornon-stratified flow.

The operator may choose to carry out more

than nine sets of comparisons. Should this optionbe exercised, the results of a maximum of threetests may be rejected from the test data if anappropriate statistical test applied to the datademonstrates that these results are outliers.A minimum of nine RM tests must be availableafter statistical rejection of data. All data must bereported, including the outliers, along with allcalculations.

All appropriate diluent gas and moisturemeasurements must be conducted simultaneouslywith the RM pollutant concentration measurements.

If the CEM is fitted with a stack gas flow monitor,then RM concentration measurements must beconcurrent with RM flow measurements. Twoadditional sets of flow comparisons must becompleted either during the OTP or during the168-hour period immediately following the OTP,at different load levels.



29/61

5.3.4.5 Correlating Reference Method andContinuous Emission Monitoring SystemMeasurements. In order to correlate thedata from the CEM system and RM tests, it isimperative that the beginning and end of eachtest period be clearly marked on the CEM data

acquisition record and that the CEM time besynchronized with the RM test crew time.After each test is completed, compare theresults from the CEM system with the dataderived from the RM results over the exacttime period that the test was performed.

The CEM system and RM results must becorrelated on the same basis. Thus, correctionsmay need to be applied for moisture,temperature, pressure, etc. The auxiliarymeasurements of the RM testing (such as stackgas moisture or barometric pressure) are used

to make any adjustments to the RM results. Theauxiliary measurements of the CEM system areused to make adjustments to the CEM results.

5.3.4.6 Calculations. The relative accuracy of thesystem must be calculated for each pollutant anddiluent gas measured by the system in terms ofconcentration in ppm or percent (by volume), aswell as stack gas flow in terms of m/s or Sm3/h.Additionally, the relative accuracy for SO2 andNOX emissions in units of the standard must becalculated.

(i) Calculation of relative accuracyThe relative accuracy is calculated usingEquation 4.

where:RA = relative accuracy (%)d = mean difference between the CEM system

and RM resultscc = confidence coefficient

RM = average of the reference method results

When the pollutant gas concentrations are lessthan 250 ppm, the FS setting of the analyzermust be substituted for the value of RM whencalculating the relative accuracy using Equation 4.

(ii) Calculation of differencesThe absolute value of the difference between theCEM system and RM results is calculated usingEquation 5.

where:di = difference between an RM value and

a corresponding CEM system value(di = CEMi RMi) for the ith test run

n = number of data pairs

Note: The numeric signs for each data pair mustbe retained. The absolute value of the sum ofdifferences is used, not the sum of absolute values

of the differences.

(iii) Calculation of confidence coefficient andstandard deviationThe values of the confidence coefficient andstandard deviation are determined from Equations6 and 7, respectively.

where:cc = confidence coefficientt0.025 = t value from Table 4 for a one-tailed t-test

corresponding to the probability that ameasured value will be biased low at a95% level of confidence

Stdev = sample standard deviation ofthe differences of the data pairs from therelative accuracy test, calculated usingEquation 7


where parameters are as defined above.

(iv) Calculation of reference flow-to-outputand heat-to-output ratiosIf the CEM system includes a stack gas flowmonitor, the flow-to-output or heat-to-outputratio may be required as reference for the

Stdev= Equation 71d i2( )

nn 1

n

d i

2

( )i=1

n

i=1

cc = Equation 6t0.025 x Stdev

n

d = Equation 5

1

di

n

ni i=1

RA= Equation 4d + cc

RMx100



30/61

quarterly stack gas flow test in subsequentquarters.

Depending on the option chosen for the quarterlystack gas flow test, as described in Section 6.3.2,one or more of the following averages must becalculated, based on the process data and theRM measurements obtained during the relative

accuracy test of the initial CEM certification orsubsequent relative accuracy test audit (RATA): flow to electric output ratio;

heat input (if the CEM system is fitted toperform this calculation on the basis ofmeasured stack gas flow and measured diluentconcentration or measured fuel flow rate andfuel gross heating value); and

heat input to electric output ratio (only forCEM systems that can calculate heat input).

The flow to electric output ratio is calculated with

Equation 8, from the results of the RM runs.

where:Rref = reference value of flow to electric output

ratio on a wet basis, from the most recentRATA (WSm3/MWh)

Qwi = flow measured during each test run on awet basis (WSm3/h)

MWi = mean gross electric output during eachtest run (MW)


The heat input is the average of the grossheat input during each RM test run. It may becalculated on the basis of the measured stack gasflow and F-factors (as shown in Table A-1) or, inthe case of gaseous and liquid fuel, on the basis ofthe measured flow rate and gross heating value of

the fuel. The heat input corresponding to an RMtest run may be calculated with Equations 9 to 12.

where the following values are RM test runaverages:HI = heat input (MJ/h)QW = stack gas flow on a wet basis (WSm3/h)%BWS = moisture content of the stack gas (%)Fd = fuel-specific F-factor from Table A-1

or calculated as per Appendix A,Section A.7

Fc = fuel-specific F-factor from Table A-1or calculated as per Appendix A,Section A.7

%O2W = stack gas oxygen concentration (% wetbasis)

%CO2d = stack gas carbon dioxide concentration(% dry basis)%O2d = stack gas oxygen concentration (% dry

basis)

The reference heat to electric output ratio (orgross heat rate) is calculated with Equation 13,from the results of the RM runs.

HI = QW Equation120.209 (100 %BWS) %O2W

20.9Fd

HI = QW Equation11100 %BWS

100Fc

%CO2d

100

HI = QW Equation101Fc

%O2w100

HI = QW Equation 9100 %BWS

100Fd

20.9 %O2d

20.9

Rref = Equation 81

n MWi

Qwin

i=1


Table 4 t Values

n 1 5 6 7 8 9 10 11 12 13 14

t0.025 2.571 2.447 2.365 2.306 2.262 2.228 2.201 2.179 2.160 2.145

Note: These are t values for a one-tailed t-test at a 95% confidence level.


31/61

where the following values are RM test runaverages:GHRref = reference value of gross heat rate, from

the most recent RATA (MJ/MWh)HIi = gross heat input during each test

run (MJ/h)MWi = mean gross electric output during each

test run (MW)n = number of data pairs

5.3.4.7 Acceptance Criteria for Certification. Theperformance specifications presented in Section5.1.5 must be met.

5.3.5 Bias Test CalculationsA bias or systematic error is considered to be

present if, in the measurements of a pollutant gas,diluent gas, or stack gas flow:

As presented in Section 5.1.6, acceptable bias is

(|d| |cc|) 5.0% FS

or|d| 5.0 ppm for pollutant concentration

or|d| 0.5%O2 (or CO2) for diluent concentration

or|d| 0.6m/s for stack gas flow monitor

It is highly recommended that the sources of biasin the system be investigated and remedied.

If bias is present, as determined by Equation 14,and it is within the above levels, then thesubsequent measurement of the CEM system mustbe corrected by a bias adjustment factor (BAF),

using Equations 15 and 16.

where:CEMadjusted = data adjusted for biasCEMmonitor = data provided by the monitorBAF = bias adjustment factor, defined by

Equation 16

where:BAF = bias adjustment factor CEMRATA avg = average CEM results during RATARM = average of the reference

method results

The use of a BAF in any measurement must bestated in the QA/QC manual.

5.3.6 Orientation Sensitivity Test Protocols

This test is intended as a check for flow monitorsthat are sensitive to the orientation of the sensorin the gas flow, such as differential pressure flowsensors. This test is carried out at the same threeload levels previously defined in the relativeaccuracy test.

5.3.6.1 Test Procedures. During a period ofsteady flow conditions at each load, the sensor inthe gas flow must be rotated a total of 10 degreeson each side of the zero-degree position (directlyinto the gas flow with no cyclonic flow patterns)in increments of 5 degrees, noting the response of

the sensor at each angle. A total of five flows mustbe generated for each load condition, at -10, -5,0, +5, and +10 degrees relative to the zero-degreeposition.

5.3.6.2 Acceptance Criteria for Certification.The performance specifications presented inSection 5.1.7 must be met.

BAF = RMCEMRATA avg

Equation16

CEMadjusted = CEMmonitorx BAF Equation15

d cc Equation14

GHRref = Equation131

n

HIi

MWi

n

i=1



32/61

In cooperation with the regulatory agency, theoperator must develop a QA/QC manual for eachinstalled CEM system.

The QA policies (high level) and QC procedures(standard operating procedures, working level) areoutlined in the written QA/QC manual. Themanual must be followed to ensure and documentthe quality of the environmental data beingcollected and reported. Establishing the QA/QCmanual ensures that the environmentalmonitoring and reporting procedures are verifiedand documented so that uncertainties in thereported data can be controlled and quantified.Figure 2 shows a schematic of a QA/QC manual.

A QA program is defined as a managementprogram to ensure that the necessary QC activitiesare being adequately performed, while QCactivities are those that detail the day-to-dayoperation of the system.

6.1 Quality Assurance/Quality ControlManual

The QA/QC manual must describe a completeprogram of activities to be implemented to ensurethat the data generated by the CEM system will

be complete, accurate, and precise. As aminimum, the manual must include the QA/QCprocedures specified in this report. Therecommended Table of Contents for the QA/QCmanual is shown in Table 5.

6.1.1 Quality Assurance ActivitiesThis section of the manual should describe howthe QA program is managed, provide personnelqualifications, and describe the QA reportingsubsystem. It must describe the CEM system, howit operates, and the procedures for calibration andinspection. It must also include preventativemaintenance and performance evaluationprocedures.

6.1.2 Quality Control ActivitiesThis section should provide detailed descriptionsof the step-by-step procedures required tooperate and evaluate the system, including detailsabout daily, quarterly, semiannual, and annualperformance evaluations. Procedures andminimum criteria for a selection of these activitiesare provided in Sections 6.2 to 6.5. A summaryof these performance evaluations is outlined in

Table 6.


SECTION 6.0 Q UALITYASSURANCE AND QUALITYCONTROL

Figure 2 Quality Assurance/Quality Control Schematic

QA/QCDocuments the QA/QCpolicies and procedures

Quality AssuranceSets out the policies toensure the collection of

high-quality data

Quality ControlSets out the procedures to

ensure the collection ofhigh-quality data


33/61


Table 5 Table of Contents for Quality Assurance/Quality Control Manual

Subsection Contents

Quality assurance policies and system descriptions

1 Quality Assurance Goals andObjectives

2 CEM System Description and DesignConsiderations

3 Exceptions/Clarifications/AlternateMethods

4 Organization and Responsibilities

5 Calibration and QualityControl Checks

6 Data Acquisition and Analysis

7 Preventative Maintenance Policy

8 Corrective Action Program

9 Performance Evaluations/Audits

10 Document Control System

11 Reports and Records

12 Modifications and Upgrades

Specific system goals relating to precision, accuracy, andcompleteness. Specific objectives as laid out in the regulations and

guidelines. Emission standards and emission reporting requirements.Detailed system description, including principles of operation,sample location layout, flow and temperature measurement, sampleconditioning system, analyzer layout, CEM shelter, and data handlingsystem. Design considerations and engineering evaluation of CEMsystem options, including sample location, extractive vs. in situ, flowmonitoring, and supplier. Should also include a detailed list of CEMsystem component serial and model numbers.

Any exceptions/clarifications or alternate methods relating to 1/PG/7,reference test methods, or regulations.

Description of the organization of personnel involved with the CEMsystem and its quality system. Defines the roles and responsibilitiesof the personnel involved as related to CEM system operation and

maintenance, control of documents/records, and control of data.Description of the calibrations and QC checks that are performed ona routine basis, generally daily, to determine whether the system isfunctioning properly. Includes daily zero and calibration checks andvisual checks of system operating indicators, such as vacuum andpressure gauges, rotameters, analyzer displays, LEDs, and so on.

Description of the data acquisition system and analysis program.Includes references to data completeness, validation, reporting,storage, and revision management. Roles and responsibilities of thepersonnel involved in the data handling should be included.

Description of the CEM system preventative maintenance program,including how preventative maintenance scheduling is determinedand maintained along with roles and responsibilities of the personnelinvolved.

Description of the policies for correcting any CEM system non-conformance. Parameters such as CEM system downtime/reliabilityshould be addressed. Roles and responsibilities of the personnelinvolved in the corrective action program should be included.

Description of the policies and specifications for performanceevaluations/audits (i.e., stack quarterly audits and RATAs). Describethe action necessary to ensure that the appropriate evaluations arecarried out on the appropriate schedule.

Description of the policies and systems used to control all thedocuments that form part of the CEM systems quality system.Lists how and where the related documents are located, howthey are reviewed and revised, and how they are approved foruse by authorized personnel prior to issue.

Description of all reports and records collected. Description shouldcontain method of collection, person responsible, data storagelocation, data security, data distribution, and length of data storage.

Description of the policies regarding modifications and upgradesto the CEM system. This section should include any regulatoryrequirements pertaining to modification or upgrade of the CEMsystem.


34/61


Table 5 Table of Contents for Quality Assurance/Quality Control Manual (contd.)

Subsection Contents

Quality assurance policies and system descriptions

13Training and Qualification Policy

14References

Quality control (standard operating) procedures

1 Startup and Operation

2 Daily CEM System Operationand Inspection

3 Daily and Manual Calibration

Procedures

4 Gas Bottle Check Procedures

5 Preventative Maintenance Procedures

6 Spare Parts List and Inventory

Procedures

7 Corrective Maintenance Procedures

8 Data Backfilling Procedures

9 Data Backup Procedures

10 CEM System Security

11 Data Approval and ReportingProcedures

12 Quarterly Audit Procedures

13 Semiannual Relative Accuracy TestAudit Procedures

Training and qualification policy for CEM system maintainers, CEMsystem coordinators, computer and programming technicians, data

validators, quarterly audit and RATA testers. Includes educational andexperience requirements, on-the-job training, job shadowing, andclassroom training requirements.

References for QA/QC plan.

Lists in detail complete, step-by-step procedures for the startup andoperation of the CEM system.

Detailed description of daily routine operation and inspection of theCEM system. Includes descriptions of equipment and data validationprocedures. Examples of daily equipment checks and/or logbookentries should be included.

Lists in detail complete, step-by-step procedures for daily and

manual calibrations. May make reference to specific OEMdocumentation/manuals. Includes schedule for manual (mid-point)calibration, if done.

Description of procedure to cross-reference cylinder gases. Gasescan be cross-referenced to previous gas bottles and quarterly bottles.Specifications for rejection of gas bottle to be stated.

Detailed description of the CEM system preventative maintenanceprocedures along with the preventative maintenance schedule.This could include a description of such things as a preventativemaintenance work order program for those facilities so equipped,along with reference to or examples of preventative maintenancework orders in use.

Detailed descriptions of the spare parts inventory available for

the CEM system, along with a description of the procedures forobtaining spare parts from inventory and ensuring that the spareparts inventory is maintained.

Detailed descriptions of the non-routine maintenance that isperformed when the system or part of the system fails. Maymake reference to specific OEM documentation/manuals.

Procedures for data backfilling when a CEM system is not available.Data backfilling algorithms to be based on process variables.

Procedures for regular backup of data in hard or soft copy.

Includes security actions for CEM equipment software and data.

Procedure for approval and reporting of CEM data. Includes anysystems for review, modifications, approval, summary, and release

of data.Detailed procedures on conducting quarterly audit procedures.Includes roles and responsibilities, gas bottle requirements,scheduling, and test methods.

Detailed pretest sampling plan for executing RATAs. Pretest plan toinclude organization plan, sampling points, scheduling, test methods,calibration requirements, reporting schedule, reporting format, andsite safety plan.


35/61

6.2 Daily Performance Evaluations

6.2.1 Calibration DriftCalibration of the CEM system is one of themost important aspects of the QA/QC program.The following summarizes the requirements forcalibration drift, all of which must appear in theQA/QC manual.

6.2.1.1 Frequency. The drift of each gasanalyzer and flow monitor must be determinedat least once daily, at 24-hour intervals. It is goodpractice to determine the drift of each analyzereven during periods when the generating unit isdown. The operator may, however, skip the dailycalibration during extended periods in whichthe generating unit does not burn fuel. Howeverthe CEM system should be successfully calibratedimmediately prior to or during the startup to avoidusing the backfilling option (Section 3.4.3).

6.2.1.2 Test Gases. Protocol gases or gases

certified to an accuracy of 2.0% may be usedfor the daily calibration of gas analyzers.

6.2.1.3 Calibration Gas Injection Port. Thelocation o