EPA-340/1 -86-010
Recommended Quality Assurance
        Procedures for Opacity
         Continuous Emission
          Monitoring Systems
                     Prepared by

                   James W. Peeler
                 CEM/Engineering Division
               Entropy Environmentalists, Inc.
           Research Triangle Park, North Carolina 27709
               Under Contract No. 68-02-3962
              Work Assignments 2-52 and 3-101
                   With JACA Corp.
                Fort Washington, PA 19039
                     Prepared for

               EPA Project Officer: John Busik
        EPA Work Assignment Managers: Anthony Wayne
                            and Mary Cunningham
           U.S. ENVIRONMENTAL PROTECTION AGENCY
             Stationary Source Compliance Division
            Office of Air Quality Planning and Standards
                 Washington, D.C. 20460

                    February 1986

-------
                                   DISCLAIMER

    This material has been funded wholly or in part by the United states
Environmental Protection Agency under Contract 68-02-3962 to Entropy
Environmentalists, Inc.  It has been subject to the Agency's peer and
administrative review, and it has been approved for publication as an EPA
document.  Mention of trade names or commercial products does not constitute
endorsement or recommendation for use."

-------
                           TABLE OF CONTENTS

                                                                       Page

1.0  Introduction	«	»	....     1

2.0  Elements of An Opacity GEMS QA Program.	     2
     2.1  Daily QA Checks	,	     2
     2.2  Periodic QA Checks/Preventive Maintenance	.'	     4
     2.3  Corrective Action Procedures	     8
     2.4  Accuracy Checks	     9

3.0  Organization of QA Activities	    11

4.0  QA Plan Implementation and Revision	    13
     4.1  Establish Initial QA Objectives	    13
     4.2  Develop Draft QA Plan.	    15
     4.3  Initial Trial Implementation and Debugging.	    15
     4.4  Finalize Initial QA Plan.	    16
     4.5  Implement Current QA Plan.	    16
     4.6  Review of QA Records and Information.	    16
     4.7  Revision of the QA Plan	    16
     4.8  Notify/Train Involved Personnel	    17

5.0  Organization of QA Responsibilities	    18


     Appendix A.  Example Quality Assurance Plans

     Appendix B.  Example Quality Assurance Data Summaries

-------

-------
                               1 . 0;  INTRODUCTION
    This report presents recommended quality assurance (QA) procedures for
opacity continuous emission monitoring systems (CEMS's)  installed at electric
utility steam generating stations.  The recommended procedures are based on
experience and information gained from the Opacity GEMS Pilot Project that was
conducted in Missouri.  As part of this project, QA procedures for opacity
CEMS's were developed, implemented, and evaluated during approximately one-year
of field study.

    The field study included opacity CEMS's installed on six coal-fired
electric utility generating units located at four generating stations, each of
which is owned by a different utility company.  The sources included in this
study were selected to be representative of a wide range of monitoring
applications and conditions, and were equipped with contemporary opacity
monitoring instrumentation provided by the three major CEMS manufacturers.
Detailed descriptions of the sources, opacity CEMS's, and QA procedures that
were evaluated during the project are included in "Evaluation of Opacity CEMS
Reliability and QA Procedures, " CEMS Pilot Project Report, March 1986.

    The recommended QA procedures contained in this report are intended to
provide a simple, cost-effective approach to the development and implementation
of opacity CEMS QA plans.  The recommended approach allows for much flexibility
in the selection of monitor- and spurce-specific procedures, the establishment
of QA control limits, and the organization of the quality assurance elements.
Because of the flexibility retained in the recommended procedures, source
personnel may use them to develop a QA plan that most effectively meets their
specific needs.  Thus, no attempt has been made to determine or define the
minimally acceptable QA procedures or requirements.

    Section 2.0 of this report outlines the major elements of a QA plan:
(1) daily QA checks, (2) periodic QA checks and preventive maintenance,
(3) corrective action procedures, and (4) accuracy checks.  Procedures that may
be included are identified and described for each of the four elements.
Recommendations regarding the appropriate procedures for an initial QA plan are
also provided.  Section 3.0 describes briefly the organization and interaction
of the elements of the QA plan.  Section 4.0 presents an approach that may be
used for the implementation and revision of the QA plan to optimize its
usefulness and cost-effectiveness.  Section 5.0 provides comments regarding
considerations affecting the assignment of the various QA responsibilities at a
particular plant.  Example QA procedures developed during the opacity CEMS
Pilot Project are included in Appendix A and example QA data summaries are
included in Appendix B.

-------

-------
                    2.0   ELEMENTS  OF  AN  OPACITY GEMS QA  PLAN


    This section  describes each  of the major  elements  of an  opacity  CEMS QA
Plan.  These elements include:  (1) daily QA checks;  (2)  periodic QA  checks and
preventive maintenance;  (3)  corrective action procedures; and  (4)  accuracy
checks.  The following discussions describe the nature and types of  activities
that may be included in  each of  these elements.  The interaction and
organization of the above elements are described in Section  3.0 of this report.


2.1  DAILY QA CHECKS

    Daily QA check procedures for  opacity CEMS's should  be developed and
implemented to allow for  the identification of  monitoring problems and, thus,
for the timely initiation of corrective  action.  Adequate daily QA check
procedures can be performed  from the monitor  control unit/data recording
location in virtually all cases.   The daily check procedures should require
only a few minutes per day to complete for each monitoring system, and can be
performed by personnel who are not thoroughly familiar with the technical
details of the opacity monitoring  instrumentation.

    The daily QA checks may include the  following:


    1'   Zero and Span Checks - Opacity monitoring regulations require that
        checks of the zero (or low level) and span drift be performed at least
        once daily.  (Adjustments to the monitoring system are required when
        the zero or span  drift exceed specified limits.)   Daily zero and span
        checks should be  included in all opacity CEMS QA plans, since these
        checks are required by the applicable regulations, and because they
        provide an overall indication of the performance (i.e., accuracy and
        precision) of virtually the entire monitoring system.

        The daily check procedures should contain a written procedure for
        performing the zero and span checks.   This procedure should indicate
        that the determination of zero and span drift be based on the
        measurements displayed by the permanent data recording system that is
        used as the basis for preparation of  quarterly reports.  The zero and
        span drift check procedure should also specify the criteria that are
        used to determine when adjustments to the CEMS will be performed.


    2.   Fault Indicators - Virtually all contemporary opacity CEMS's
        incorporate fault indicators in the monitor control unit.   When
        activated, the fault indicators warn  of monitor malfunctions  and/or
        operating conditions that may affect  the quality  of the monitoring
        data.   The number and type of fault indicators are monitor-specific and
        vary both between monitor manufacturers and,  in some cases, among the
        different models provided by the same manufacturer.

        For opacity CEMS's equipped with computerized data recording systems,
        the software may be designed to provide flags and/or error messages in
        the computer printout to warn of problems with the monitoring system.

-------
      These messages should be considered equivalent to the fault indicators
      incorporated in the monitor control unit.

      It is strongly recommended that a check of all available fault
      indicators  and/or error messages in computer printouts be included in
      the  daily QA check procedures.


   3.  Auxiliary Monitor Performance Parameters - Some  opacity CEMS's provide
      easy access to important monitor performance parameters, such as
      reference  current values or zero compensation levels, at the monitor
      control unit.  Inclusion of checks of these parameters in the daily QA
      check procedures allows data to be compiled that can be used to assess
      both (a) the  stability of  these parameters and  (b)  whether the fault
      indicators for the  same parameters are operating reliably.  It is
      recommended that checks of these monitor performance parameters be
      included in the initial daily  check procedures  and that the frequency
      of checking these parameters be  subsequently  adjusted based on the
      observed monitor performance.


   4.  Auxiliary Data Recording/Data Display Devices - Many opacity  CEMS's use
       a computerized data recording system or data logger as  the primary data
       recording system.   Most of these CEMS's are also equipped with  strip
       chart recorders that serve either as back-up data recorders  or  as
       analog  data display devices for use by the boiler operators.   In
       addition, the opacity GEMS control unit usually includes  either an
       analog  or a digital panel meter.  At some sources, measurements that
       are used to assess monitor performance are obtained from the panel
       meter.  Therefore, at these sources the accuracy of the panel meter can
       affect  the accuracy of the opacity GEMS data.

       Checks  of  auxiliary recorders and/or  data display  devices are important
       in those cases where  (1) back-up recording devices are used during
       malfunctions of the primary data recording system, or (2) the accuracy
       of panel meter readings can affect either the determination of the need
       for adjustment of the GEMS calibration or the accuracy of the GEMS
       data.   Checks of the  auxiliary data  recording/data display devices can
       be  accomplished by either (1) expanding the daily  zero and span check
       procedures to include evaluation of  the responses  of the auxiliary
       devices in addition to the responses of the primary data recorder, or
        (2) comparison of  simultaneous measurements provided by the permanent
       data recorder and  the auxiliary device.  It  is  recommended that  such
       checks be included in the daily QA check procedures until sufficient
       data are  obtained  to  justify  less frequent checks. Alternatively,
       these checks could be performed only on those  occasions when data from
        auxiliary data  recording/data display devices  are actually used.


    Decisions regarding which of the above  parameters should be  included in
daily QA checks depend on many monitor- and source-specific factors.   Among  the
most important factors to consider are:  (1)  the design and operational features
of the specific opacity monitor, (2)  the data recording/data  display devices
used in the opacity CEMS's,  (3)  historical monitor performance and previously
encountered problems, and (4) zero and span drift adjustment procedures.

-------
Although an attempt  should be made to  choose  the  best parameters  for  the  daily
checks at the outset, periodic  review  and  revisions  of  the QA plan  (see Section
4.0) will allow for  the inclusion of additional parameters and/or for the
deletion of unnecessary checks  relative to the initially  established  daily QA
check procedures.

    At most electric utility generating stations, a  number of people  will be
involved in the performance of  the daily QA checks.  Therefore, it  is extremely
important that clear, concise written  procedures  be  developed for the daily QA
checks to ensure that all involved personnel  will perform them in a consistent
manner.  It is recommended that a one-page daily  log be developed for recording
of the necessary data.  The daily log  can  be  formatted  in a manner  that (1)
minimizes the recording burden, (2 ) encourages the consistent recording of
data, and (3) clearly delineates the acceptance criteria  for the  daily checks
and whether corrective action is necessary.   It is strongly recommended that
the daily log require the person performing the daily check to sign or initial
the log sheet.  Examples of detailed instructions for daily checks  and daily
log sheets for several opacity  monitoring  systems are included in Appendix A of
this report.

    If only a small  number of people (i.e., one or two) perform all of the
daily QA checks, or  after all personnel involved  in  the daily QA  checks nave
become completely familiar with the data recording procedures and criteria for
initiating corrective action, it may be possible  to  eliminate the daily log and
instead substitute a control chart or  condensed log.  Such a record keeping
procedure would allow for summarizing  all  data from  the daily checks  during a
one-month period on  a single log sheet.  (Examples of such summaries  are
included in Appendix B of this  report.)  It is emphasized that the  use of a
summary log or control chart will reduce the  volume  of  paper records  and  will
facilitate periodic  review of the data; however,  in  most  cases such a log is
not as effective as  a daily log sheet  in requiring consistent recording of data
or ensuring the initiation of corrective action when applicable control limits
are exceeded.
2.2  PERIODIC QA CHECKS/PREVENTIVE MAINTENANCE

    The periodic QA procedures provide for checks of monitoring system
components and operational status that are unfeasible, impractical, or
unnecessary on a daily basis.  These procedures include activities that are
only necessary on a semi-frequent basis (e.g., weekly, monthly, or quarterly,
etc.), and include checks of the instrumentation at both the actual monitoring
location and the control unit/data recording location.  The periodic QA checks
should be designed to identify developing or existing problems with the opacity
GEMS that cannot be detected in the daily checks.  The periodic QA procedures
should also incorporate preventive or routine maintenance procedures for the
opacity GEMS.  The nature of the periodic QA procedures requires that personnel
who are familiar with the operation, maintenance, and calibration of the
opacity CEMS perform these activities.  The time required to perform a periodic
QA check typically ranges from 2 to 8 hours.  The time required may be
significantly affected by the difficulty of the access to the monitoring
location, the availability of communications between the monitoring location
and the control unit location, and the complexity of the monitoring system.

-------
The periodic QA procedures may include the following:

1.    Transmissometer Optical Alignment - The optical alignment of the
      transceiver and reflector components is a critical factor in
      obtaining accurate effluent opacity measurements.  Virtually all
      contemporary opacity CEMS's provide an optical alignment sight for
      determining whether the transmissometer components are properly
      aligned.

      It is recommended that a check of the optical alignment and the
      recording of the alignment status be included in the periodic QA
      check procedures.


2.    Dust Accumulation on Optical Surfaces - The periodic QA procedures
      should include some means for estimating the amount of particulate
      (in terms of apparent opacity) that has accumulated on the exposed
      optical surfaces of the transmissometer, and should also provide for
      cleaning of the optical surfaces.  A simple method of accomplishing
      both of these objectives is to record the effluent opacity
      measurements, clean the transceiver window, and again record the
      effluent opacity measurements.  The procedure is then repeated for
      the reflector window.  The difference between the effluent opacity
      measurements before and after cleaning of each window provides a
      rough estimate of the quantity of dust deposited on each window.
      However, the validity of such determinations may be affected by
      fluctuations in the actual effluent opacity while the "before" and
      "after" measurements are being made.  Experience has shown that the
      suggested procedure provides adequate results at sources where
      significant levels of dust accumulation are encountered  (i.e.,
      > 4% opacity) and where the effluent opacity is relatively constant
      for at least short periods of time.  At sources where the levels of
      dust accumulation are small (i.e., < 2% opacity) and where effluent
      opacity fluctuations are relatively large, use of the suggested
      procedure often produces erroneous results.  In these cases, the
      optical surfaces should be cleaned; however, the quantification of
      dust accumulation can either be  omitted or attempted using an
      alternate approach.  For example, reference to the changes in the
      zero compensation level or the zero check response before and after
      cleaning will provide a rough estimate of the  quantity of dust
      deposited on the transceiver optics for some opacity CEMS's.


 3.    Auxiliary Monitor Performance Parameters - For most opacity CEMS's,
      periodic  checks  of  a few critical operating parameters can be very
      helpful ±n  the  identification of developing or existing  monitor
      problems.   The  nature and type of such  checks  is  inherently dependent
      on the design and operational features  of the  specific opacity
      monitor.  Examples  of  critical monitor performance parameters include
       (a)  operational status  of the automatic  gain  control  circuit  for LSI
      RM41 monitors,  (b)  lamp reference voltage for  Dynatron Model  1100
      opacity monitors,  and  (c) response  to  the simulated  zero opacity
      condition provided  by  the calibration  test kit (i.e., audit  device)
      for Contraves Model 400 opacity  monitors.

-------
 It is recommended that checks of critical monitor operating param-
 eters such as those mentioned above and those identified by the
 monitor manufacturer be included in the periodic QA check proce-
 dures.  It is also recommended that checks of important monitor
 performance parameters such as those described in Section 2.1, Item 3
 above (i.e., reference current values and zero compensation levels)
 be included in the periodic QA checks if they are not included in the
 daily QA checks.


 Auxiliary Data Recording/Data Display Devices - In addition to the
 auxiliary data recording devices and control unit panel meters
 described in Section 2.1,  Item 4 above, many opacity monitors are
 equipped with electronic test meters at the transceiver location
 which may be used to assess the need for adjustments or as a
 reference when adjustments are made.  It is recognized that at many
 power plants, instrument technicians ignore these test meters, and
 instead use digital voltmeters to obtain more accurate and precise
 electronic measurements than are provided by the installed meters.
 Checks of installed test meters may be performed by either (a) com-
 paring the meter responses for zero and span checks of the opacity
 GEMS to the permanent data recorder responses, 6r (b)  comparing the
 meter response to the responses of an external measurement device
 (i.e., calibrated digital  voltmeter).

 It is recommended that the periodic QA procedures include checks of
 all electronic data display devices (i.e., control unit panel meters
 and/or test meters at the  transceiver location)  which are used by
 station personnel to assess monitor performance.   It is also
 recommended that checks of all auxiliary data recording devices
 (e.g., strip chart recorders for opacity CEMS's equipped with
 computerized data acquisition systems  or data loggers)'be included in
 the periodic QA procedures.


 Purge Air System/Shutters  - All opacity CEMS's are equipped with
 purge air systems that are designed to provide a flow of highly
 filtered ambient air across the exposed optical surfaces of the
 transmissometer to minimize contamination of these optical surfaces
 by particulate in the effluent stream.   Some opacity CEMS's are also
 equipped with shutter systems which close  in the event of failure of
 the purge air system.   When operating properly,  the shutters  do not
 obstruct the transmissometer light beam while the purge air system is
 operating,  and they close  when the purge air system provides
 insufficient air flow to protect the optical surfaces.   Typical
 failures of the purge air  system and shutter devices include:  (a)
 plugging of air filters, (b)  leaks  in  the  purge air system supply
.hoses, (c)  failure of the  air flow sensor  to detect insufficient
 purge air flow,  (d)  partial  or complete closure  of the  shutters when
 the purge air system is working properly,  and (e)  failure of  the
 shutters to close when the purge air system  is inoperative.   Failure
 of the purge air system or shutter devices may result  in costly
 damage to the transmissometer components,  loss of data,  or a  high
 bias in all effluent opacity measurements  due to  dust  accumulation on
 the exposed optical surfaces.

-------
          It is recommended that the periodic QA check procedures  include a
          visual inspection of the purge air system filters and supply hoses,
          and checks that the shutters do not obstruct the transmissometer
          light beam during normal operation and that they close when the purge
          air flow rate is inadequate.


    6.     Preventive Maintenance - Preventive or routine maintenance procedures
          are typically specified by the monitor manufacturer in the operator's
          manual for the monitor.  These procedures usually include replacement
          of measurement lamps, fault lamp bulbs, purge-air filters, and
          desiccant cartridges.  Monitor manufacturers typically specify either
          a frequency for replacing expendable materials or criteria for
          determining when such components must be replaced.  Operational
          experience with the opacity monitor may indicate other maintenance
          activities that should be performed periodically, such as the removal
          of accumulated material from the transmissometer flanges or sampling
          ports.  Other preventive or routine maintenance activities are   .
          necessary to maintain data recording equipment in proper working  ; .
          order.  In addition to the maintenance practices specified by the    ,,
          data recorder vendor, common sense requires (a) reasonable cleaning, 7;
          of data recording devices, (b) checks and/or replacement of chart
          recorder inking supplies or printer ribbons, and (c) replacement of
          paper supplies as necessary.

          It is recommended that preventive or routine maintenance activities
          for the opacity monitor either specified by the monitor  manufacturer
          or developed based on operational experience be incorporated into the
          periodic QA procedures.  This approach will ensure that  these
          activities are performed at the same time other checks of monitor
          operational status are performed, and will therefore minimize
          additional trips to the monitoring location.  To the extent that
          routine maintenance activities of the data recording equipment are
          not included in the daily QA checks, it is recommended that these
          activities also be included in the periodic QA procedures in order  to
          avoid unnecessary GEMS downtime.


    Decisions regarding the exact procedures that should be included in the
periodic QA procedures depend on monitor- and source-specific factors.  In
addition, the appropriate frequency for performing the periodic QA procedures
is source-specific.  Choosing an appropriate frequency requires balancing
concerns regarding data quality against the amount of time and effort required
to perform the checks.  Important factors to consider in this decision are
(1) the reliability of previous monitor operation and  (2) the difficulty
associated with access to the monitoring location.  For example, at sources
where chronic monitor operational problems have been encountered and where
access to the monitoring location is  easy, it may be appropriate to conduct
periodic QA checks  every two weeks, at  least initially.  In contrast, at
sources where the opacity GEMS has operated reliably and where access to the
monitoring  location is difficult, it  may be sufficient to conduct periodic QA
checks only once every three or  four  months.

-------
    If no previous monitor operational experience is available upon which to
base a decision, it is recommended that the periodic QA checks be performed 'on
a monthly basis during the initial period of the QA plan implementation.
Periodic review of the QA data and corresponding revisions to the QA plan (as
described in Section 4.0) will allow for appropriate revision of the frequency
of QA checks, as well as for the inclusion of additional checks and deletion of
unnecessary procedures based on actual monitor performance.

    As with the daily QA checks, it is recommended that concise, clearly
written procedures should be developed for performing the periodic QA checks to
ensure that all involved personnel perform the checks in the same manner, and
to ensure that corrective action is initiated when control limits are
exceeded.  In addition, it is strongly recommended that a periodic QA log be
developed and completed for each periodic QA check to provide a record for
assessing monitor performance.  Periodic QA check logs can be developed that
(a) minimize the data recording burden, (b) encourage consistent data recording
practices, and (c) clearly indicate when control limits are exceeded, and thus
when corrective action is necessary.  Example periodic QA check procedures and
log sheets for several opacity CEMS's are included in Appendix A of this
report.
 2.3   CORRECTIVE  ACTION  PROCEDURES

    Control limits must be established for  virtually  all  of the parameters
 included in either the  daily QA check  procedures  or periodic QA check
 procedures.  (Control limits are not necessary for parameters  or measurements
 recorded for informational purposes only.)   Whenever  the  control limits  are
 exceeded,  corrective action activities should be  initiated in  order  to resolve
 the problem with the opacity GEMS.  Corrective action may also be  initiated as
 a result of accuracy checks performed  by source personnel or performance audits
 conducted by the control agency.  In general,  corrective  action procedures and
 corresponding records should provide documentation of what was wrong with the
 CEMS, what was done to  correct the problem,  and performance check  results that
 demonstrate that the problem was actually resolved.

    Clearly, corrective action procedures are even more monitor- and source-
 specific than are the daily and/or periodic QA check  procedures.   To a large
 extent,  the corrective  action procedures are contained in the  monitor
 operator's manual within applicable sections that describe procedures for
 instrument (a) start-up, (b) calibration, (c)  trouble-shooting, and  (d)
 repairs/maintenance.  (These procedures can be incorporated into the QA  plan
 simply by reference to  the operator's  manual.) The  corrective action
 procedures should stress "before"  and  "after" measurements of  appropriate
 parameters in order to  provide a basis for  assessment of  the impact  of repairs
 and/or adjustments on fault indicators, zero and  span check responses, and
'. monitor calibration. In addition, the corrective action  procedures  should
 indicate the need for only the measurements, adjustments, and  repairs for those
 components or subsystems of the opacity CEMS necessary to return the entire
 opacity CEMS to operation within the  applicable control limits.  For example,
 if a zero offset develops between  the  control unit and strip chart recorder  (as
 indicated by the daily  QA check),  all  that  would  be  necessary  is  (a) recording
 of the pre-adjustment zero and span reponses, (b) adjustment of the  chart
 recorder calibration, (c) a final  zero and span drift check that  showed  the
 proper responses for the chart recorder and agreement between  the  control

-------
unit and chart recorder responses.  In contrast, failure of an important
electronic component in the transceiver might require a complete recalibration
of the opacity monitor, including an off-stack,"clear path check (i.e., zero
alignment).

    It is recommended that a corrective action log be developed and that it be
appropriately completed on each occasion that repairs or adjustments are made.
A corrective action log satisfies the regulatory requirement for maintaining
records of all GEMS adjustments and/or repairs, and also provides a very useful
basis for evaluating the effectiveness of repairs and adjustments and for
identifying chronic monitor problems.  Notes recorded by the person performing
the repairs are perhaps most useful, provided that they can be understood by
the other personnel who make repairs and adjustments to the monitoring system.
At a minimum, it is recommended* that the corrective action log provide for
recording of the same parameters that are included in the daily and periodic QA
checks both before and after adjustments are completed.  Examples of corrective
action logs for several opacity CEMS's are included in Appendix A of this
report.


2.4  ACCURACY. CHECKS

    As used here, accuracy checks refer to either  (a) an on-stack performance
audit, or  (b) an off-stack (or clear path) zero alignment and calibration
check.  These accuracy checks  are briefly described below.


2.4.1  Performance Audits

    Performance audits of an installed opacity CEMS may be conducted by  either
source personnel or the  control  agency to assess the operational status  of the
opacity CEMS and the  accuracy and precision  of the opacity measurements
relative  to the simulated zero condition provided  by the zero check device.
Performance audits  do not quantify  the absolute accuracy of  the opacity
measurements.

    Performance audit procedures for many contemporary opacity CEMS's  are
 contained in "Performance Audit  Procedures for Opacity Monitors"  (EPA  No.
 340/1-83/010).  Additional  information and suggestions for modifications to  the
 above referenced procedures  are  contained in "Opacity CEMS Audit Procedure
 Guidelines,"  Opacity CEMS Pilot Project Draft Report, March  1985.

     In essence, virtually  all of the checks  included in  the  referenced audit
 procedure documents should be included in  either  the daily or periodic QA
 checks described in Sections 2.1 and 2.2 of  this  report,  except  for the
 calibration error check.  In general,  the  calibration error  check  procedure
 uses an audit device and a set of three calibrated neutral  density filters to
 assess the precision, linearity, and accuracy of  the opacity CEMS  (relative  to
 the zero check response) at three test points over the measurement range of  the
 monitor.   Each of the three test filters  is  inserted into the  light path five
 times, and the calibration error check result for each filter  is calculated in
 the same manner as the calibration error test results described in Performance
 Specification 1, Appendix B, 40 CFR 60.   (Alternate procedures are used for
 some monitors.)

-------
                                                                                10
    Audit  devices are commercially available for most opacity CEMS's.  Neutral
density  filters  can be obtained from a number of suppliers and calibrated
neutral  density  filters  (traceable to the National Bureau of Standards) can
also be  obtained.

    It is  recommended that the calibration error check portion of the
performance audit procedures be conducted at least once per year for each
installed  opacity GEMS.  The calibration error check should also be performed
when there are questions regarding the accuracy or linearity of the opacity
data.
2.4.2  Clear Path Calibration - Zero Alignment

    For the daily and periodic QA checks and for the performance audit checks,
the accuracy of the opacity measurements is determined relative to the
monitor's response to the simulated zero check devices.  The procedure for
establishing the equivalency of the simulated zero check device and the actual
clear-path (zero opacity) condition is referred to as "zero alignment."
Performance Specification 1 (Appendix B, 40 CFR 60) requires that this
procedure be performed either under clear stack conditions or prior to
installing the opacity monitor at the measurement location.  Since some amount
of residual opacity is usually present at the monitoring location even when the
source is not operating, the zero alignment procedure is almost always
performed prior to the initial installation of the monitor.

    Prior to the March 30, 1983 revisions to Performance Specification 1, the
regulations required that the clear-path calibration/zero alignment be
performed .at least once per year.  The requirement for the annual check was
deleted, since the purpose of the Performance Specification is to ensure the
initial capability of the opacity GEMS to provide valid data, rather than to
provide QA guidelines.

    The frequency for performing a clear-path calibration or zero alignment
should be based on the typical rate of change between the simulated zero check
response and the true zero check response.  However, there is almost no avail-
able public information regarding this factor.  (One plant that participated in
the Opacity GEMS Pilot Project indicated that the zero alignment procedure was
performed once per year for each of the four transmissometers installed at that
plant.)

    It is recommended that a clear-path calibration check be performed once per
year or at some other frequency for which there are data supporting the choice
of the interval between clear-path calibration/zero alignment checks.

-------

-------
                                                                               11
                       3.0  ORGANIZATION OF QA ACTIVITIES


    The four elements of the QA plan for opacity GEMS's are (1) daily QA check
procedures, (2) periodic QA check procedures and preventive maintenance, (3)
corrective action procedures, and (4) accuracy checks.   These four elements
include all of the activities necessary for the acquisition of quality assured
opacity monitoring data after the monitor has been properly installed and after
it has been successfully demonstrated to comply with the requirements of
Performance Specification 1 (Appendix B, 40 CFR 60).

   . One possible arrangement of the four elements of the QA Plan is shown in
Figure 1.  Other organizational schemes are possible and may be advantageous in
certain circumstances.  The organization shown in Figure 1 is recommended
because it is extremely simple and flexible.  The four elements are
interconnected only by (1) the schedules for the daily QA checks, periodic QA
checks, and accuracy checks, and (2) the criteria for initiating (and
completing) corrective action.  Changes in the schedules for the periodic QA
checks and/or accuracy checks, and changes in the acceptance criteria (control
limits) for the three quality assessment activities would not require
reorganization of the QA plan.

-------
                                                                 12
        CONDUCT DAILY
           CHECKS
                               WAIT FOR NEXT
                                DAILY CHECK
                           TIME
                           FOR
                         ACCURACY
                           CHECK
 DAILY
CHECKS
  O.K.
                  ACCURACY CHECKS
                                  PERIODIC OA CHECKS
                                    AND PREVENTIVE
                                     MAINTENANCE
       1
CORRECTIVE ACTION
       RECTIV
      ACTION
       PLETE
            YES
           FIGURE I.  ORGANIZATION OF QA ACTIVITIES

-------
                    4.0  QA PLAN IMPLEMENTATION AND REVISION


    The development and implementation of a QA plan for an opacity GEMS is to
some extent a trial and error procedure.  In most cases, the QA plan that is
initially developed, should be'reviewed from time to time to determine if
additional or more frequent activities are necessary to ensure data quality
and/or -to determine if some of the QA procedures can be relaxed in order to
reduce resource expenditures in'those cases where the quality of the data is
more than adequate.  A majdr QA management goal should be to achieve the
desired level of data quality at-the lowest real cost.  Since opacity GEMS QA
plans are inherently monitor- and source-specific, achievement of this goal
requires that the QA plan be considered a dynamic rather than a static
document.

    The major steps in the implementation and revision of a QA plan are shown
in Figure 2.  These steps are briefly described below.


4.1  ESTABLISH INITIAL QA OBJECTIVES /„..-•
               . >v           '  v • •*  "?*v..  ••- "'*  •• •
    Management of tne QA.plan implementation to achieve the necessary level of
data quality at the lowest cost requires that the QA objectives be clearly
defined.  In the most general sense., ,the quality of the data must be compatible
with the intended use of the data... .. Two potential,'uses of opacity GEMS data at
electric utility plants are: (1) the plant personnel may use the data to
identify process and control system1 problems in order to initiate actions that
will maximize efficiency and minimize equipment deteriorization, and (2) the
control agency will rely on the data to .identify periods of excess emissions in
order to prioritize source inspections and/or other follow-up activities.  The
data quality required by the source will be subject to the conditions imposed
by particular generating units and other internal considerations of the utility
company.  The data quality requirements imposed by a plant's own internal use
of the data are considered to be.no less rigorous, than those imposed by the
agency's use of the data.

    The use of the opacity GEMS data to identify periods of excess emissions is
somewhat dependent on the proximity of the effluent opacity for the particular
generating unit under normal operating conditions to the opacity level used to
identify periods of excess emissions.  Coal-fired electric utility generating
units are typically subject to opacity limits in the range of 20% to 40%
opacity.  For most generating units, the opacity under normal operating
conditions is generally well below 20% opacity (regardless of the applicable
standard); however, when process or control system problems are encountered,
the effluent opacity is often much greater than the standard (e.g., greater
than 150 percent of the standard).  In these situations, if the overall
accuracy of the opacity monitoring data is within +_ 5% opacity, there is little
chance that periods of actual excess emission will be overlooked or that
periods of operation below the opacity standard will be inadvertently reported
as periods of excess emissions due ^to monitor inaccuracy.

-------
                                                            14
      ESTABLISH INITIAL QA OBJECTIVES.
                  I
          DEVELOP DRAFT OA PLAN
                  I
INITIAL TRIAL IMPLEMENTATION AND DEBUGGING
                  I
         FINALIZE INITIAL QA PLAN
                  I
       IMPLEMENT CURRENT OA PLAN
                  I
         REVIEW QA RECORDS AND
              INFORMATION
      I.
      2. PwMIc QA checks. |«f«. «rf
      3. C«rr«cUv»
      4. Acoracy chMk
      5. CMMMMU ft-ma
          REVISE QA PROCEDURES
    NOTIFY/TRAIN INVOLVED PERSONNEL
 FIGURE 2.  QA PLAN IMPLEMENTATION AND REVISION

-------
    Using an accuracy of +_ 5% opacity as the overall objective, appropriate
individual QA objectives (i.e., control limits) can be developed.  For example,
a particular generating unit might choose the following control limits:
                Parameter	Control Limit

             Zero Alignment               +_ 2% opacity
             Calibration Error            +_ 3% opacity
             Dust Accumulation            +_ 2% opacity
             Calibration Drift  (zero      _+ 2% opacity
               or span drift)
Assuming that  an  opacity CEMS  is  operating  at  the maximum  for  each of  the
control limits simultaneously  (which  is  grossly pessimistic  from the source's
perspective),  and assuming  that the associated biases  are  randomly distributed
(which is  optimistic  from the  source's perspective,  since  dust, accumulation
results only in a positive  bias),  the overall  probable error can be calcu-
lated.  For  the above control  limits  and assumptions,  the  probable error would
be  on the  order of +^4.6% opacity.  If the  calibration drift control limit were
increased  from +^  2%~~opacity to +_  4% opacity, then the  probable error would be
increased  from +_  4.6% opacity  to  approximately 4^ 5.7%  opacity.

     The above  example analysis is somewhat  simplistic  from a statistical point of
view.  Nevertheless,  the  example  shows  (1)  that  overall QA objectives  can be es-
tablished  based on considerations of  how "good"  the data needs to be and  (2) how
control limits that allow attainment  of  the overall objectives can be  selected.
 4.2   DEVELOP DRAFT QA PLAN

     The draft QA plan may be developed in accordance with the recommendations
 provided in Section 2.0.   Again, monitor- and source-specific factors and
 previous operational experience with the opacity CEMS are among the most
 important factors to consider.  The example QA procedures evaluated at four
 generating stations for Lear Siegler, Dynatron, and Contraves Goerz opacity
 CEMS's equipped with a variety of data recording systems provide much back-
 ground information for the development of a draft QA plan.  Extensive dis-
 cussions of these QA plans are contained in "Evaluation of Opacity CEMS
 Reliability and QA Procedures," Opacity CEMS Pilot Project, Draft Report, April
 1985.
 4.3  INITIAL TRIAL IMPLEMENTATION AND DEBUGGING

     It is recommended that the draft QA procedures be implemented for a short
 time period in order to identify ambiguous instructions, problems with data logs,
 and procedures that are simply overly cumbersome.  Some of these problems are
 bound to occur in even the most carefully prepared QA plan.  The trial
 implementation/debugging period should last for several weeks or at least until
 most of the personnel who will perform the daily QA checks have worked through
 the instructions and completed the daily log, and at least until one periodic QA
 check has been performed.

-------
                                                                               16
4.4  FINALIZE INITIAL QA PLAN

    The problems identified in the initial trial implementation/debugging
period must be resolved.  The comments of the personnel performing the checks
are probably the most useful source of information for revising the
instructions and log sheets for each element of the QA plan.  It may also be
necessary to revise the QA objectives and/or individual control limits at this
point in the development of the QA Plan.


4.5  IMPLEMENT CURRENT QA PLAN

    After all initial revisions to the QA plan have been completed, the plan
should be implemented for at least several months.  Implementation of the QA
plan for six months will provide sufficient data to allow appropriate
adjustments to each of the elements of the QA plan to be made within the same
time frame.
4.6  REVIEW OF QA RECORDS AND INFORMATION

    At the end of the first implementation period of the initial QA plan, the
available QA documentation should be reviewed.  This review should include
daily QA check logs, periodic QA check results, corrective action records,
accuracy check results, and comments from the personnel who have carried out
each of these procedures.  It is recommended that data from the daily QA checks
be consolidated into monthly summaries or control charts to facilitate analysis
of data for trends.  The objectives of the review and analysis of data are (1)
to identify problem areas and/or recurring monitor problems that require
additional attention and (2) to identify those parameters that are documented
to be very stable and thus, no longer warrant checking at the initially
scheduled frequency.  Examples of this type of analysis are provided in
"Evaluation of Opacity GEMS Reliability and QA Procedures," Opacity GEMS Pilot
Project Draft Report, April 1985.

    If no changes to the QA plan appear to be necessary, either the review and
analysis are too superficial or the QA plan has already reached a maximum cost-
effectiveness level.  In the latter case, the QA plan should continue to be
implemented without changes.
4.7  REVISION OF  THE QA PLAN

    The QA plan should be revised in accordance with the results of the review
and analysis of the QA data from the implementation period.  It is important
that changes in both the instructions and the QA data logs be made, even though
most of the personnel involved with the implementation of the QA procedures
will no longer refer to the instructions.   (The use of out-dated instructions
with revised data sheets can  create many problems when personnel changes
occur.)

-------
                                                                               17
4.8  NOTIFY/TRAIN INVOLVED PERSONNEL

    Personnel responsible for the implementation of specific elements of the QA
plan must be properly notified of changes in the specific QA procedures.  One
result of the initial period of QA implementation may be the identification of
the need for additional training of specific individuals.  If this situation
occurs, either the necessary training should be provided or the division of QA
responsibilities should be altered to relieve the need for training.

    After all involved personnel have been notified of changes in the specific
QA procedures and after all training issues have been resolved, implementation
of the revised QA plan should begin again at Step 5, "Implement Current QA
Plan."

    The entire process should be repeated indefinitely; however, the periods
between review of the QA documentation may be extended during subsequent
implementation periods.  Eventually, the most cost-effective QA plan will be
developed through iteration.

-------

-------
                                                                                18
                    5.0  ORGANIZATION OP QA RESPONSIBILITIES


    The implementation of opacity CEMS QA plans usually  involves multiple
personnel from several different departments or groups at  electric utility
generating stations.  Therefore, the clear delineation and assignment of the
responsibilities for various aspects of the QA plan to different groups and/or
individuals, as well as coordination among the different groups and/or
individuals, is usually necessary.                               '.

    The management and organizational structure used for power plant personnel
vary among electric utility companies, and may also vary among different
generating stations operated by the same company.  Union labor agreements
differ greatly for different companies.  The number of people who may be
involved with an opacity monitoring program can vary dramatically among
companies and generating stations.  Finally, different approaches are often
taken by different companies regarding the interaction of  corporate environ-
mental representatives and station personnel.  All of these factors must be
considered in the development and implementation of an opacity CEMS QA plan.

    To be successful, an opacity CEMS QA plan must be compatible with the
existing management and organizational structure at the  generating station.
The QA plan should be designed to rely on the established  lines of communi-
cation and decision making authority among the various groups and/or
individuals involved with the opacity monitoring program.

    The opacity CEMS QA plan delineated in the preceding sections involves
various activities included in the four elements of the  QA plan: (1) daily QA
checks, (2) periodic QA checks and preventive maintenance,  (3) corrective
action, and (4) accuracy checks.  Decisions regarding the  assignment of QA
responsibilities should take into account the differences  in the technical
skills and knowledge required by the specific activities associated with each
of the QA plan elements.  In essence, the daily QA checks  do not require a
detailed under-standing of how the opacity CEMS works; however, the other three
elements do.  Another major difference between these two groups of elements is
the frequency with which activities are performed.  The  daily QA checks are
obviously performed on a daily basis; the other activities  are performed
relatively infrequently.  Because of these differences,  some power plants may
assign responsibilities for performing daily QA checks to  one group of people
and assign the responsibilities for the other activities to a second group.
Three of the four generating stations participating in the  Opacity CEMS Pilot
Project chose this option.  (The fourth station assigned responsibility for all
of the QA activities to the instrument department personnel.)  Of the three
stations that divided the QA responsibilities, two stations chose to have the
boiler operators perform the daily QA checks; the third  station had an
individual from the "Results Department" perform the daily  QA checks.  In all
cases, instrument technicians performed the other activities included in the QA
plan.                                                          f,

    In some cases, a large number of people (10-15 individuals) may be involved
in performing the daily QA checks.  In this situation, the  importance of clear,
step-by-step instructions, completion of the daily QA logs, and keeping track
of who performed the checks is increased, since verbal communication among all
the different individuals is not likely to provide much  continuity of operation.

-------
                                                                               19
    The periodic QA procedures, corrective action activities, and accuracy
checks all require that the personnel performing these activities have a
working knowledge of how the opacity GEMS works.  Usually, these activities
will be performed by a relatively small group of individuals whose jobs
routinely require effective exchange of information.  Thus, the instructions
associated with these activities may be less detailed; however, the acceptance
criteria should be very clear.  In addition, since these elements of the QA
plan are performed relatively infrequently and/ or in response to the
unpredictable development of problems, communications with other departments or
individuals and keeping track of schedules are important.

    In addition to the responsibilities associated with the four elements of
the QA plan, someone must be assigned the responsibility for overseeing the
on-going QA program.  The responsibilities associated with this task include
making sure that:  (1) required activities are performed, (2) data logs are
properly completed, and (3) proper decisions are made when exceedances of the
established QA control limits occur.  This person or group of people may also
be responsible for the periodic review of QA data to determine whether an
increase or reduction in the frequency for specific QA procedures is
appropriate.  Ideally, the person responsible for making these decisions would
not be involved in performing the specific activities required by the QA plan;
however, this is not absolutely necessary.

    The person or group responsible for overseeing the implementation of the QA
plan should be in contact with those persons responsible for the preparation of
the quarterly reports that are submitted to the control agency.  This is
necessary:  (1) to ensure that the correct information regarding GEM performance
is included in the report, and  (2) to close the feedback loop regarding any
response of the control agency to reported GEMS performance data.  Finally, the
person overseeing the implementation of the QA plan should be in contact with
other station personnel and/or corporate personnel who use the GEMS data to
ensure that the level of data quality is sufficient for the internal use of the
monitoring information.

-------
                                  APPENDIX A.

                      EXAMPLE QUALITY ASSURANCE PROCEDURES

    This Appendix contains quality assurance procedures that were used by the
four sources participating in the CEMS Pilot Project.   Four separate sets of
quality assurance procedures are included which address the monitoring
instrumentation provided by three major CEMS manufacturers and the source-
specific constraints encountered at each of the four participating sources.
(All company and plant names and other identifiers have been deleted.)
Additional information regarding these QA procedures may be found in
"Evaluation of Opacity CEMS Reliability and QA Procedures," CEMS Pilot Project
Report, March 1986.

-------

-------
                                 EXAMPLE A

                                       STATION
                 OPACITY MONITOR QUALITY ASSURANCE .PROCEDURES.

                     •""''••'        ?             "STATION

                DYNATRON MODEL 1100 OPACITY MONITORING SYSTEMS  -.



                           OVERVIEW OF QA PROCEDURES

     A proposed quality assurance program has been developed for the
                      Station opacity monitoring  systems.  The specific QA
procedures have been developed to be compatible with the             Station
(1) opacity monitoring instrumentation and monitoring system configuration,
(2) data recording devices, (3) effluent handling system, and (4) management
and organizational structure.  The proposed QA program will be field-tested
during implementation, reviewed and evaluated periodically, and revised as
necessary over a one-year period.  Through this process, it is expected that
QA procedures will be developed and demonstrated which are both adequate for
maintaining high levels of data quality and cost  effective in terms of
necessary time and material resource expenditures.

     The following elements are included in the proposed             Station
opacity monitor quality assurance program.

     (1)   Daily Log, Daily Check Instructions - The Daily Log is to be
           completed by the personnel responsible for checking the monitoring
           system on a daily basis.  Step-by-step Daily Check Instructions are
           provided for completing the Daily Log.  The Daily Check and Daily
           Log do not require extensive time to complete, nor do they require
           that the person performing the procedure be intimately familiar
           with the opacity monitoring instrumentation.  The Daily Check and
           Daily Log provide for identification of monitoring problems and
           initiation of corrective action.

     (2)   Corrective Action Log and Instructions - The Corrective Action
           Instructions and Log are used when adjustment, repairs, and/or
           other non-routine corrective action is necessary as indicated by
           the Daily Checks.  The Corrective Action procedures are to be
           utilized by technical personnel experienced in resolving problems
           with the monitoring systems.  The specific corrective action
           procedures are relatively extensive, but will provide adequate
           documentation for future refinement of QA procedures and
           demonstration of their effectiveness.  The corrective action
           procedures are utilized only on an "as necessary" basis.

     (3)   Periodic QA Check and Instructions - The Periodic QA check is
           intended to be performed in conjunction with the opacity monitor
           routine maintenance program performed monthly at the
           Station.  The Periodic QA Check procedures provide for checks of
           monitoring system components and operating status which are
           unfeasible, impractical, and unnecessary on a daily basis.

-------
(4) The documentation of the QA program will be reviewed periodically to
    determine if modification to the proposed procedures are
    appropriate.  Such modifications may be made as additional experience
    and data are obtained.

-------
                                                   STATION

                           DAILY   CHECK INSTRUCTIONS

                DYNATRON  MODEL  1100  OPACITY MONITORING  SYSTEMS

     The  following  are  step-by-step  instructions  for  conducting  the Daily
Check  of  the              Station Unit #  and Unit  #   opacity monitoring
systems.  Criteria  are  provided for  determining when  corrective  action should
be  initiated.  The  person conducting the Daily Check  of each monitoring  system
should complete all blanks on the  Daily Log as indicated below.  (Separate
forms  are used for  the  Unit #4 and #5 monitoring  systems.)

  I. GENERAL INFORMATION

    o Enter name of  person performing check, the date  (month, day, year), and
     the  time of day (24-hour clock) that the check is begun.

    o Hours Boiler Down  -  Enter the number of hours the boiler was not in
     operation (i.e., not combusting fuel) during  the preceding 24-hour
     period.

    0 Hours Monitor Down - Enter the  number of hours the monitor did not
     provide a record of  effluent  opacity during the  preceding 24-hour
     period.  Describe  monitor downtime in "Comments" section (i.e., cause of
     outage, time began,  time ended, and any corrective action taken to  return
     monitor to service).

 II. FAULT LAMPS

    o Examine the fault  lamps on the  Control Unit.  Check the appropriate box
     on the Daily Log to  indicate  the status of each  fault lamp (check "YES"
     if lamp is illuminated).

   o If any fault lamps are illuminated, corrective action should be initiated
     as soon as possible.

III. ZERO/SPAN CHECK DATA

   o Examine the strip  chart to determine the zero and span responses of the
     most recent monitor calibration.  Record the zero and span calibration
     values on the Daily Log in units of %  opacity.

   o Examine the data logger tape to determine the zero and span response of
     the most recent monitor calibration, and record these values on the Daily
     Log.

   o Check the appropriate boxes on the Daily Log to indicate whether the zero
     and span values are outside acceptable limits.

   o Check the data logger tape to determine if  the proper time of day is
     displayed.   Reset the timer of the data logger if necessary.

-------
     Note:   IF YES ANSWERS ARE INDICATED FOR ANY OF THE QUESTIONS ON THE
             DAILY LOG, CORRECTIVE ACTION SHOULD BE INITIATED AS SOON AS
             POSSIBLE.

IV. COMMENTS

   o Describe any problems observed during the performance of the Daily Check
     and/or any other  apparent problems which may affect monitor performance.

   O ENTER THE TIME OF DAY THAT THE DAILY CHECK IS COMPLETER (PART I).

-------
                           Opacity Monitoring System

                                   DAILY LOG

                                          STATION
                              UNIT:
  I.  GENERAL INFORMATION

      Name:
Date:
Time Start:
Time Complete:
      Hours Boiler Down:

 II.  FAULT LAMPS
             Hours Monitor Down:
FAULT LAMPS ON?
Lamp
Window
Air Flow
NO



YES



III.  ZERO/SPAN CHECK DATA

      Chart Recorder, Zero Value:
      Data Logger, Zero Value:
                     Span Value:_
                     Span Value:

Does Zero Value exceed acceptable limits of H
Does Span Value exceed acceptable limits of -i
- 2% opacity?
- 27, opacity?
NO


YES



| Data Logger
Timer
Correct

Incorrect

      IF YES ANSWERS ARE INDICATED FOR ANY OF THE ABOVE QUESTIONS,  CORRECTIVE
      ACTION SHOULD BE INITIATED AS SOON AS POSSIBLE.
 IV.  COMMENTS:

-------

-------
                                     .,            , STATION


                  PERIODIC QUALITY ASSURANCE CHECK INSTRUCTIONS
                       DYNATRON OPACITY MONITORING SYSTEM


      The following are step-by-step instructions for performing the Periodic
 QA Checks of the             Station opacity monitoring systems.  Initially,
 the Periodic QA Checks are to be performed Once per month.  The person
 performing the monthly checks should complete all blanks on the Periodic QA
 Check Log.  (A separate data form should be used for each monitor.)   The
 manufacturer's instructions for cleaning optical surfaces, performing
 electronic adjustments, and inspecting/servicing the purge-air system should
 be followed throughout the Periodic QA Check.

   I.  GENERAL INFORMATION

       o   Enter the Unit number for the monitor on the QA Log.

       o   Enter the name of the person performing the check, the date (month
          day,  year),  and the time of day (24-hour clock)  that  the  check is
          begun.

  II.  CALIBRATION CHECK DATA

      (1)  Place the "CYCLE  TIME HOURS"  knob on  the "MANUAL  POSITION."   Rotate
          the "METER DISPLAY"  knob to the "OPACITY" position.   Depress
          "ZERO/SPAN"  switch and  record the low range calibration check
          responses of the  panel  meter  chart recorder,  and  data logger
          (%  opacity).

      (2)  Record  the high range calibration check responses  of  the  panel meter,
          chart recorder, and  data logger (% opacity).

      (3)  Corrective action  should be initiated  if all  "zero" and "span"
          responses  are  not  within +_ 2%  opacity  of the  proper values.   A
          Corrective Action  Log  should  be completed if  such  action  is
          necessary.

      (4)  Using an  accurate  volt meter, measure  the Lamp Reference Voltage  from
          available  test points on the back  of the control unit.  Record the
         .voltage on the QA log.   If  the Lamp Reference voltage is outside  the
          range of 6.5 + 1.0 volts, adjust  the monitor according  to the
         manufacturer's instructions and record  the post-adjustment Lamp
         Voltage on the data sheet.

      (5) Record the minimum 6-minute  average opacity indicated by the chart
          recorder for the hour preceding the QA check.

III.  TRANSMISSOMETER CHECK/SERVICE

     This segment of the Periodic QA Check requires that effluent opacity
measurements "before and after" both alignment adjustments and cleaning of

-------
optical windows be obtained in order to assess the impact of these
activities.  These measurements may be obtained by two methods: (1) have an
assistant record real-time data if two-way communication between the control
unit and monitoring location is available, or (2) sychronize watch and chart
recorder, and note the exact time each action is performed to facilitate
recovery of opacity measurements from the chart recorder after the various
activities are completed.

     (1) Alignment Check - Determine the optical alignment status of the
         transceiver and reflector components by looking through the alignment
         sight on the transceiver and observing whether the image is within
         the circular target (acceptable), or outside the circular target
         (unacceptable).  Indicate the position of the light beam on the
         diagram, and check the appropriate box for alignment status on the QA
         Log.

         If the optical alignment is unacceptable, note the effluent opacity,
         and then realign the opacity monitoring system in accordance with the
         manufacturer's instructions.  Record the effluent opacity after
         realignment is completed.  If a shift in the baseline opacity occurs
         after realignment, note the magnitude of the change which was
         observed in "Part V COMMENTS."

     (2) Reflector Window Check - Record the exact time (or effluent opacity)
         prior to cleaning the reflector optics.  Remove the protective window
         (slide) from the reflector and clean the glass according to the
         manufacturer's instructions.  Replace the reflector protective
         window.  Record the exact time, and wait at least two full
         integration periods before performing the next step, (or record
         directly the effluent opacity).

     (3) Transceiver Window Cleaning/Check - Record the exact time (or
         effluent opacity) prior to cleaning the transceiver optics.  Remove
         the protective window (slide) from the transceiver and clean the
         glass according to the manufacturer's instructions.  Replace the
         transceiver protective window and record the exact time, (or record
         the effluent opacity).

     (4) Purge Air Service - Inspect and service, as necessary, the purge-air
         blowers, air filters, and shutter mechanism as per the manufacturer's
         instructions.  Note any corrective action taken on the QA Log.

 IV. FINAL MEASUREMENTS

     (1) If necessary, examine the data record to obtain the average opacity
         values corresponding to the times recorded on the data sheet (i.e.,
         before and after alignment adjustments, before and after  cleaning  of
         the reflector window, and before and after cleaning of the
         transceiver window).  Record all data on the QA Log.

     (2) Record the minimum 6-minute average effluent opacity value during  the
         hour period following completion of all adjustments, repairs, and
         service for the monitor.

-------
   (3) Note periods of alignment adjustment  and window cleaning on the
       permanent data record to ensure that  these periods will not be later
       mistaken for excess emissions.  Reset alarms if activated during QA
       checks.  Note that the periodic QA check was performed in the "Part V
       COMMENTS" section of the Daily Log.

V. COMMENTS

   All observations regarding monitor performance should be detailed and
   explained.

   ENTER TIME OF DAY PERIODIC QA CHECK IS COMPLETED (PART 1).

-------
                              Opacity Monitoring  System

                                 PERIODIC QA CHECK

                                             STATION
                                                       UNIT:
  I.  GENERAL INFORMATION

     Name:	
                              Date:
   Time  Start:	
   Time  Complete:_
 II. CALIBRATION CHECK DATA

m
r?>
m
(4)
Low Range Response
High Range Response
Adjusted Low Range Response
Adjusted High Range Response
% Opacity
Panel Meter




Chart Recorder




Data Logger




  Lamp Reference Voltage:_
                                 Adjusted Lamp  Reference Voltage:
  Minimum 6-min. opacity value (hour preceding QA check):	

III. TRANSMISSOMETER CHECK/SERVICE

     1. Alignment Status (Note position of light beam on diagram)
       [Alignment Acceptable?
                                  YES
                                        NO
See instructions if
alignment is not acceptable
     2. Before reflector cleaning
     3.
        After reflector cleaning
Before transceiver cleaning
        After tranceiver cleaning
                                            TIME
                                                   EFFLUENT OPACITY
 IV. FINAL MEASUREMENTS

     Minimum 6-min. opacity value  (hour following QA Check):

  V. COMMENTS:

-------
                                     ,             STATION

                         CORRECTIVE ACTION INSTRUCTIONS
                       DYNATRON OPACITY MONITORING SYSTEM

     The following are step-by-step instructions for completing the Corrective
Action Log when adjustments and/or repairs of either the Unit #  or Unit //
opacity monitoring systems are necessary.  The person performing the corrective
action should complete all blanks on the Corrective Action Log as indicated
below.

  I. GENERAL INFORMATION

   o Enter the Unit number corresponding to the monitor for which repairs or
     adjustments are performed.

   o Enter the name of person performing repairs or adjustments, the date
     (month, day, year), and the time of day (24-hour clock) that the corrective
     action is initiated.

 II. FAULT LAMPS

   o Enter "ON" or "OFF" for each fault lamp on the Log.

   o Monitor calibration and completion of Part III of the Log are not required
     if an Air Flow problem initiates the Corrective Action.  However, if the
     "WINDOW," or "LAMP" fault lamps are illuminated, data listed below must be
     obtained prior to adjusting the monitor.

   o If "LAMP" is illuminated, (1) measure and record the lamp reference
     voltage, (2) adjust reference voltage to manufacturer's specifications, and
     (3) record the post-adjustment reference voltage.

   o If "WINDOW" is illuminated, (1) record the effluent opacity indicated by
     the data recorder, (2) clean transceiver and reflector windows, and (3)
     record the effluent opacity indicated by the data recorder.

   o Record the time when the fault is corrected.  Describe all corrective
     action taken in the "Part IV COMMENTS."

III. CALIBRATION DATA/ADJUSTMENTS

     (1) Place the "CYCLE TIME HOURS" knob on the "MANUAL POSITION."  Rotate the
         "METER DISPLAY" knob to the "OPACITY" position.  Depress "ZERO/SPAN"
         switch and record the low range calibration check responses of the
         panel meter, chart recorder, and data logger (% opacity).

     (2) Record the high range calibration check responses of the panel meter,
         chart recorder, and data logger (% opacity).

     (3) Adjustment of  the monitoring system is necessary when the low range
         check responses of the panel meter, chart recorder or data logger
         exceed +_ 2% opacity from the correct low range check value.  (Correct
         value of the low range check should be labeled on the front cover of
         the  control unit.)

-------
        If no adjustment is made, enter "NA" on line 3 of the Corrective
        Action Log, and proceed to step (4).

        Describe all adjustments and/or corrective action in "Part IV
        COMMENTS."  After all adjustments are completed, repeat step (1)
        above, and record the post-adjustment low range responses on line 3
        of Log.

    (4) Adjustment of the monitor is necessary if the high range responses of
        either the panel meter, chart recorder, or data logger exceed +_ 2%
        opacity from the correct value.  (Correct value of the high range
        check should be labeled on front of control unit.)

        If no adjustment is made, enter "NA" on line 4 of the Log.  Describe
        all adjustments and/or corrective action in "Part IV COMMENTS."
        After all adjustments are completed, repeat steps (1) and (2) above,
        and record post-adjustment span response on line 4 of Log.

IV. COMMENTS

        All repairs and/or adjustments performed as a result of the above
        procedures should be described.  Sufficient explanation should be
        provided to determine what was done and what effect it had on monitor
        performance.

        In addition, any preventive or non-routine maintenance performed on
        the monitor should be detailed in this section.

        ENTER TIME OF DAY ALL CORRECTIVE ACTION IS COMPLETED. (Part I)

-------
                               Opacity Monitoring System

                                 CORRECTIVE ACTION LOG

                                         STATION        UNIT:
  I.  GENERAL INFORMATION

      Name:
Date:
Time Start:
Time Complete:
 II.  SYSTEM/MONITOR FAULTS
FAULT LAMPS
LAMP
WINDOW
AIR FLOW
ON/OFF



DATE/ TIME CORRECTED



      "LAMP" ON    Lamp Voltage - before adjustment:
      "WINDOW" ON  Recorded Opacity - before cleaning:

III. CALIBRATION DATA/ADJUSTMENT
                            after adjustment:
                              after cleaning:

(1)
(2)
(3)
(4)
Low Range Response
High Range Response
Adjusted Low Range Response
Adjusted High Range Response
% Opacity
Panel Meter




Chart Recorder




Data Logger




 IV.  COMMENTS

-------

-------
                              EXAMPLE B
                          	 STATION
                                            STATION

                       DAILY CHECK INSTRUCTIONS

                     LSI OPACITY MONITORING SYSTEM

          * RM41 Transmissometers                       •
          * 622 Emission Monitor Combiner and Chart Recorder
          * DP-30 Data Logger and Printer


    The following are step-by-step instructions for conducting  the  .
Daily Check of the       Station Unit #1 and Unit #2 opacity monitor-
ing systems.  Criteria are provided for determining when corrective
action should be initiated.  The person conducting the  Daily Check of
each monitoring system should complete all  blanks on the Daily  Log as
indicated below.  (Separate forms are used  for the Unit #1 and  #2
monitoring systems.)

  I.  GENERAL INFORMATION

    * Enter name of person performing check, the date (month, day,
      year), the time of day (24-hour clock) that the check is  begun,
      and the total boiler 0.0.S time for the previous  24 hours.

 II.  COMBINER DATA

    * Examine the fault lamps on the Combiner.  Check the appropriate
      box on the Daily Log to indicate the  status of each fault lamp
      (check "YES" if lamp is illuminated).

    * Rotate the Measurement switch to the  "COMP" position.  Rotate
      the Analyzer switch to position (1) and record the 2ero compen-
      sation level for Monitor A.   The panel meter readout is in units
      of optical density (OD).  Rotate the  Analyzer switch to position
      (2) and record the zero compensation  level for monitor B.
      Return the ANALYZER switch to the "EXIT" position and return the
      Measurement switch to the "30% OPACITY" position.

    * Examine the strip chart to determine  the zero and span responses
      of the most recent monitor calibration.   Record the zero and
      span calibration values on the Daily  Log in units of % opacity.
      The zero and span calibration comes in every 4 hours.

III.  DP-30 HOURLY REPORT DATA

    * Record on the Daily Log the zero and  span values  displayed on
      the most recent Hourly Report.

    * Check the appropriate box on the Daily Log to indicate whether
      the zero and span values are outside acceptable limits.

    * Circle on the Daily Log the Reason Code and/or Flags which ap-
      pear on the most recent Hourly Report.

-------
                                 - 2 -
    * Check the appropriate box on the Daily Log to indicate whether
      the "M" or "R" Flags are present.

    * Check the appropriate box on the Daily Log to indicate whether
      the listed error messages are shown on the Hourly Report.

NOTE:  IF YES ANSWERS ARE INDICATED FOR ANY OF THE QUESTIONS ON
       THE DAILY LOG, A JOB REQUISITION MUST BE INITIATED.

 IV.  COMMENTS

    * If a J.R. is required, enter J.R. number in space provided.*

    * Describe any problems observed during the performance of the
      Daily  Check and/or  any other apparent problems which may affect
      monitor performance.

    * Enter  the TIME OF DAY that  the Daily Check is completed.
     (*JR means "job requisition")

-------
                        CO.
                   Opacity Monitoring System

                           DAILY LOG

                               STATION
  I.  GENERAL INFORMATION

     Name:
                      Date:
           83
                          UNIT:
Time Start:
     TOTAL BOILER O.O.S. TIME (PREVIOUS 24 HOURS):

 II.  COMBINER DATA

                              NO   i
FAULT LAMPS ON?
           Filter
           Shutter
      :	_Ref
           Windows
      	Over Range
YES
     Zero Compensation Level:  Monitor A COMP (OD):	

     Chart Recorder, Zero Value:	

III. DP-30 HOURLY REPORT DATA

     Zero Calibration: 	 Span Calibration:
                                                Monitor B COMP  (OD):_

                                               Span Value:	
                                                                       NO
     Does zero value exceed acceptable limits of + 2.5% opacity?
     Does span value exceed acceptable limits (67.5% to 72.5% opacity)?
     Reason Code/Flags (Circle those present on hourly report)
               ##
                                    M
                                                         NO
                          A

                          YES  I
         Is the "M" or "R" Flag present?	

         Error Messages on Hourly Report?                NO      YES

         1.  A/D REFERENCE FAULT
         2.  DIGITAL/ANALOG DIFFERENTIAL ALARM
         3.  A/D TIME OUT

IF YES ANSWERS ARE INDICATED FOR ANY OF THE ABOVE QUESTIONS, A JOB
REQUISITION SHOULD BE INITIATED
IV.  COMMENTS: J.R. No.
                          was initiated
                                                                     YES
     TIME COMPLETED:
Route:  Operating Engineer
        File A/13/26(d)
SP-192

-------
                                                           Page  1  of  8
                      JOB INSTRUCTION BREAK-DOWN (J.I.B.)
Job
Periodic  QA Check
              J.I.B. No.  1-13-5.1
Prepared by
                                  Date
      6-15-83
      1-10-79
Job Class.    Tech.
The following are step-by-step instructions for performing the
Periodic QA checks of the       Station opacity monitoring systems. .
The Periodic QA checks are to be performed at least once per month.
The person performing the monthly checks should complete the Periodic
QA Check Log.  (A separate data form should be used for each monitor.)

I.   General Information
     *   Enter the Unit number (1 or 2) and the monitor indentifier (A
         or B).

     *   Enter the name of the person performing the check, the date
         (month, day, year), and the time of day (24-hour clock) that
         the check is begun.

II.  Calibration Check Data

     (1) Enter PA/PD commands at the DP-30 printer to initiate display
         of instantaneous analog and digital computer values at one-
         minute intervals for the monitor being evaluated.

         Example:  Control is gained by pressing the CTL and I keys
         simultaneously.  The DP-30 system response and the operator
         response should be as follows:
         System Response

         DP-30 I/O REQ

         STACK ID?

              (or 2)

              (or 2)

              (or 2)
                        Enter



                        1 (or 2)

                        P A

                        P D

                        Return
System Response     Enter
  1 (or 2)          Return
   PA '              Return
   PD               Return
      (2) Rotate the Analyzer switch on the combiner to the ^position
         corresponding  to the monitor being evaluated.  Position 1
         corresponds to the A monitor and position 2 corresponds to
         the B monitor.  Rotate  the measurement switch to the "REF"
         position and record the reference current value (ma) on the
         Log. Return the Analyzer switch to the "EXIT" position and
         the Measurement switch  to the "30% Opacity" position.

-------
.B. 1-13-5.1                                            page 2 Qf 8


  (3) Switch the monitor (duct) not being evaluted out-of-service
      at the Combiner and initiate a manual zero calibration.
      Record the zero calibration responses of the panel meter and
      strip chart recorder.   Record the analog system zero calibra-
      tion response from the DP-30 printout (see sample DP-30 prin-
      tout sheet attached).

  (4) Rotate the Analyzer switch to the position corresponding to
      the monitor being evaluated.  Rotate the Measurement switch
      to the "COMP" position and record the zero compensation value
      indicated^by the panel meter.  Return the Analyzer switch to
      the "EXIT" position and the Measurement  switch to the "100%
      Opacity"  position.

  (5) Initiate  a span calibration check and record the  panel  meter
      and strip chart responses on the  Log.  Record the analog  sys-
      tem span  response  from the DP-30  printout.

  (6)  Record  the minimum 6-minute average  opacity  indicated by  the
      chart recorder  and  DP-30  printout  for the  hour  preceding  this
      check.                                    ~~	SL

  (7)  Switch  the out-of-service monitor  (duct)  back to  operate.

  (8) Repeat  steps  2  through 7  for  the monitor  (duct) which was
     previously taken 0.0.S.   Be  sure to  use a  separate form.

  (9) Return  the combiner unit  back to normal operation, that is...

     *    "Operate" light illuminated

     *   Measurement switch in the "30% Opacity" position

     *   Analyzer switch in the "Exit" position.

(10) Reissue the PA/PD command to stop the DP-30 printout.

     Example:  Control is gained by pressing CTL and I keys
     simultaneously.  The DP-30 system response and the operator
     response should be as follows:


     System Response        Enter      System  Response     Enter

     DP-30 REQ

     STACK ID?               1 (or 2)    1 (or 2)            Return

     ?/l (or 2)             P A        PA                   Return

     ?/1              *1>        TO                   Return

     ?./l (or 2) .            P D        PD                   Return

     ?/l (or  2)             Return

-------
J.l.B. 1-13-5.1   -    •                                    Page 3 of 8


III. Transmissometer Check/Service

     All the time spent performing this procedure on each instrument
     is counted as out-of-service hours.  It is very important that
     this job be done as quickly and as thoroughly as possible so the
     system remains as continuous and trouble free as possible.

     (1) In the Control Room:

     Before starting to work on any monitor, transceiver, and/or re-
     flector, log that particular monitor out-of-service with the "OS"
     commands.  It is suggested that both monitors be taken 0.0.S.
     before preceeding out to the field in order to save time.

     Example:  Control is gained by pressing the CTL and I keys
     simultaneously.  The DP-30 system response and the operator
     response should be as follows:
         System Response        Enter      System Response     Enter

         DP-30 REQ

         STACK ID?

         1/1 (or 2)

         INSTRUMENT?

         ?/l (or 2)

         INSTRUMENT?

             (or 2)
1 (or 2)
0 S
MONA
0 S
M 0 N B
Return
1 (or 2)
OS
MONA
OS
M 0 N B

Return
Return
Return
Return
Return

     NOTE:   Both field monitors cannot be taken O.O.S.  on the same in-
     struction line, that is, "MONA and MONB" is not a  valid entry and
     will result in the computer issuing the instruction "BAD ENTRY".

     (2) At the Reflector:

         A.  Remove, empty, and wipe clean with a rag,  the pre-cleaner
             collector .

         B.  Remove, empty, and wipe clean with a rag,  the pre-filter
             collector.

         C.  Remove and inspect the filter element and  replace it with
             a new or clean one.

             NOTE:  These filter elements can be cleaned and several
             times before a new one is needed by following the in-
             structions on the blower housing.

-------
J.I.B. 1-13-5.1   .    •                                     Page 4  of 8


         D.  After cleaning, inspect the filter by dropping a light
             down inside the element.  If any ruptures or pin holes
             are observed, discard and use a new element.

         E.  Replace the filter element and the pre-filter collector
             only.

         F.  Before replacing the pre-cleaner collector, place your
             hand over the intake to completely obstruct air flow.

             *   Check to see that the shutter falls.

             *   Check that the "shutter" and "filter" alarms occur,
                 on the combiner panel in the Control Room, for that
                 particular duct by positioning the analyzer switch.

         G.  Replace the pre-cleaning collector.

         H.  Restore the shutter to the open position by de-energizing
             the blower motor and then re-energizing it again.

             NOTE:  If shutter fails to open after the above proce-
             dure, the shutter solenoid or T.D. relay may be faulty.
             Replace the defective part and try the above procedure
             again.

         I.  Inspect all air hoses for mechanical integrity.  Test all
             hose  clamps for tightness.

         J.  Clean the reflector lens completely by removing the iris
             plate and wiping with the clean, dry, lint  free cloth
             stored inside the cylinder located in the reflector
             mounting flange.

         K.  If any moisture, small beads, are present within the re-
             flector (between glass cover and reflector) then the des-
             sicant cartridge should be changed.

             NOTE:  Be sure the reflector module is replaced in the
             exact 3-hole position found, when changing  dessicant.

         L.  Inspect the duct mounting flange for any accumulation of
             dust.  Rod out any accumulation with'wire brush rods.

         M.  Close up the reflector unit.

         N.  Perform an alignment check at the reflector by determin-
             ing  if the light beam is within the circular target of
             the  viewing port.  Indicate  the position of the image
             within the circular  target on the diagram and check the
             appropriate box for  alignment status on the QA Log.

      (3) At the Transceiver:

         A.   Repeat Steps  2-A through 2-1.

-------
J.I.B. 1-13-5.1                                            Page 5 of 8
         B.  Clean the projection lens and the zero mirror face with
             the clean, dry, lint free lens cloth by positioning the
             mode switch from "Operate" to "Zero".

         C.  Return the mode switch back to "Operate" position.  Be
             sure the zero mirror returns to the rest (down) position.

             Do not change mode switch from "Operate" again until zero
             mirror cam comes to rest.

         D.  If any moisture, small bead , are present on the inside of
             the projection lens, the dessicant cartridge should be
             changed.

         £.  Inspect the duct mounting flange for any dust
             accumulation.  Rod out any dust with wire brush rods.

         F.  Close up the transceiver unit.

         G.  Check the AGC LED (Automatic Gain Control light emitting
             diode) on the right hand side of the transceiver, and
             record whether it is "ON" or "OFF".

             If the AGC is not on, repairs to the monitoring system
             must be completed before continuing.

         H.  Perform an alignment check by removing the plastic cover
             from the mode switch on  the transceiver and turn the
             switch to the "ALIGN" position.  Determine the optical
             alignment status of the  transceiver and retroreflector by
             looking through the "bull's eye" and observing whether
             the image is within the  circular target (acceptable), or
             outside the circular target (unacceptable).  Indicate the
             position of the light beam  on the diagram, and check the
             appropriate box for alignment status on the QA Log.

             Return the transceiver to the "OPER" position.

             *   If the optical alignment is unacceptable, realign the
                 opacity monitoring system in accordance with the
                 manufacturer's instructions.' If a shift in the base-
                 line opacity occurs  after realignment, note the mag-
                 nitude of the chance which was observed in "Part V
                 Comments".

          I.  Repeat the above instructions for the other field monitor
             starting at  step number  2.

      (4) In  the  Control Room:

          A.  At  the combiner unit make a manual  "zero"  calibration
              (1  minute) and  a manual  "span"  calibration (1 minute).

          B.   Issue  the  "IS"  command  to stop  the  DP-30  printout.

-------
1 (or 2)
I S
M 0 N A
I S
M 0 N B
See step
1 (or 2)
IS
M 0 N A
IS
M 0 N B
C
Return
Return
Return
Return
Return

J.l.B. 1-13-5.1                                            Page 6 of S


             Example:   Control is gained by pressing CTL and I keys
             simultaneously.  The DP-30 system response and the opera-
             tor response should be as follows:

         System Response        Enter      System Response     Enter

         DP-30 I/O REQ

         STACK ID?.

         1/1 (or 2)

         INSTRUMENT?

         1/1 (or 2)

         INSTRUMENT?

         1/1 (or 2)

         C.  Using the "AC" command type in the following comment for
             the particular stack:

                      "QA" MAINTENANCE FOLLOW-UP"

     Example:

         System Response        Enter      System Response     Enter

                                AC        AC                  Return

         "COMMENT  IS"           Type in    System will         Return
                                the above  print out
                                comment    comment
                                           entered

         ?/l (or 2)             Return

 IV.  Final Measurements

      (1) Upon completion of all the above steps , record the minimum 6-
         minute average effluent opacity value during the hour period
         following completion of all adjustments, repairs, and service
         for the monitor.

      (2) Record the final  zero compensation value for such monitor us-
         ing the procedure detailed in Part II  Step (4) above.

      (3) Write a J.R. for  any problems that cannot  be immediately
         corrected.  Write comments on the QA Log about the nature of
         the problems.

      (4) Write down the time on the QA Log when this procedure is
         completed.

-------
J.I.B. 1-13-5.1
                                 Page 7 of 8
  I.
     UNIT
     Name
                       Opacity Monitoring  System

                           PERIODIC  QA CHECK

                                 , * STATION
    MONITOR 	
Date:   -   - 83
Time Start:
 II. CALIBRATION CHECK DATA
     Reference  Current  (ma)
      Zero  Value (% opacity)
      Zero  compensation (OP)
      Span value (% opacity)
     Minimum 6-min.  average opacity
        value for preceding hour	
                                                Panel Meter
                                      Chart
                                     Recorder
                                     7777/7777""
                      DP-30
                     (Analog)
                                                  Ill/Hill
 III.  TRANSMISSOMETER. CHECK/SERVICE
1.

AGC

ON?
YES

NO

                                                   (See Instructions if AGC is not on.
      2.  Alignment Status (Note position of Light Beam on Diagram)

            Transceiver  -4- 4- 4-               Retroreflector   -f •»-
Alignment Acceptable?
Transceiver
Retroreflector
YES


NO


                                                   (See Instructions if alignment
                                                    is not acceptable.)
  IV. FINAL MEASUREMENTS

      1. Minimum 6-min. average effluent opacity during hour:
      2. Final Zero Compensation value (panel meter, OD): 	
   V. COMMENTS:
 Time Completed:
 Route:  Operating Engineer  •

-------
J.I.B. 1-13-5.1
                                                            Page  8  of  8
                     SAMPLE DP-30 PRINTOUT - PA/PD

                      COMMAND - QA PERIODIC CHECK
         STACK #  1
         ADRS.
         ADRS.
         ADRS.
         ADRS.
         ADRS.
         ADRS.
         STACK
         ADRS.
02/0
02/1
02/2
02/3
02/4
02/7
#  1
                                        Analog System Response Zero or
                                        Span (% Opacity)
3.88
3.80
   1.02 VOLTS
   0.95 VOLTS
   1.01 VOLTS
   1.00 VOLTS
0.36 VOLTS
0.97 VOLTS

-------
i   I                                                           Page 1 of 3
I   '»
;   «
•5'   ,                      JOB INSTRUCTION BREAK-DOWN
I
j   Job   CORRECTIVE ACTION LOG INSTRUCTIONS	 J.I.B. No. 1-13-5.3
1   {
         Lear-Siegler RM41 Opacity Monitoring System	
*
i
   Prepared by 	    . .,	 Date  	 Job Class.	
        The following are step-by-step  instructions  for completing the
   CORRECTIVE ACTION LOG when adjustments and/or repairs of either the
   Unit #1 or Unit #2 opacity monitoring system are  necessary.  The per-
   sons performing the corrective action should complete all blanks on
   the CORRECTIVE ACTION LOG as  indicated below.

     I. GENERAL INFORMATION

        *Enter the date, plant and unit number corresponding to the moni-
        toring system for which  repairs or adjustments are performed, and
        the number of the job requisition that initiated the corrective
        action.

        *Enter the name of the technician performing the repair or ad-
        justment and the time started.

    II. DESCRIBE SYSTEM PROBLEM   >;

        Describe the system problem  in  as much detail as necessary to
        clearly state the as found condition, to include meter readings,
        fault lamps, calibration data,  etc.  Use the reverse side of the
        CORRECTIVE ACTION if necessary. Refer to  the DAILY LOG which in-
        itiated corrective action.

   III. DESCRIBE ALL CORRECTIVE  ACTION  TAKEN

        Describe all the corrective  action  taken,  in as much detail as
        necessary  to clearly state  the  repairs or  adjustments or other
        actions taken.  Reference and attach all data sheets, DP-30 hand-
        copy, calculations, etc. which  resulted from this  corrective
        action.

    IV. RECORD TIME AND DATE CORRECTIVE ACTION WAS COMPLETED

        Record  the  time and date when the corrective action was com-
        pleted, i.e. when  the  system is again considered operational.

     V. THE  NEXT DP-30 HOURLY REPORT DATA

        Hand carry  the CORRECTIVE ACTION LOG to  the UCO responsible for
         the  affected generating  unit.  Notify  the  UCO that the opacity
        monitoring  system  is again  operational and that  it will be neces-
         sary for  the UCO  to obtain  certain  information  from  that unit's
        NEXT hourly  opacity report generated by  the DP-30  terminal.  Per-
         form the  following:

-------
J.I.B. 1-13-5.3                                            Page 2 of 3


         *0n the CORRECTIVE ACTION LOG, record  the zero and span cali-
         bration values printed on the next Hourly Report.

         *Check the appropriate box on  the CORRECTIVE ACTION LOG to
         indicate whether  the zero and  span values are outside accept-,
        • able  limits.

         *0n the CORRECTIVE ACTION LOG, circle  the Reason Code and/or
         Flags which appear on  the next Hourly  Report.

         *Check the appropriate box on  the CORRECTIVE ACTION LOG to
         indicate whether  the "M" or  "R"  Flags  are present.

         *Check the appropriate box on  the CORRECTIVE ACTION LOG to
         indicate whether  the listed  error messages  are  shown on the
         Hourly Report.                               ,

 NOTE:   IF  YES  ANSWERS  ARE  INCIATED FOR  ANY OF THE QUESTIONS,
        A JOB REQUISITION MUST BE  INITIATED.

 VI.   COMMENTS

      *If a J.R. is  required,  enter  the  J.R.  number  in space provided.

      *UCO  to  route  completed  CORRECTIVE ACTION  LOG  as indicated  at the
      bottom of the  LOG.       '...-.-

-------
U.JL.O.
        . — J.O— _/._>
                       Opacity Monitoring System

                         CORRECTIVE ACTION LOG
I.   GENERAL INFORMATION
     Date	
     Plant 	
     Unit No. 	
     J.R. No. 	
     Technician 	
     Time Corrective Action Started
II.  DESCRIBE SYSTEM PROBLEM
III. DESCRIBE ALL CORRECTIVE ACTION TAKEN
                                       Hrs,
IV.
RECORD TIME AND DATE CORRECTIVE ACTION WAS COMPLETED

Time	HRS.
Date
 V.
THE NEXT DP-30 HOURLY REPORT DATA

Zero Calibration:
                                         Span Calibration:
                                                                       NO  YES
     Does zero value exceed acceptable limits of +2.5% opacity?
     Does span value exceed acceptable limits (67.5% to 72.5% opacity)

     Reason Code/Flags (Circle those present on hourly report)
               i
          ##     *     D     M     R     A

Fls the
"M"
or
"R"
Flag
present?
NO

YES

Error Messages on Hourly Report?
1. A/D REFERENCE FAULT
2. DIGITAL /ANALOG DIFFERENTIAL ALARM
3. A/D TIME OUT
NO



YES



IF YES ANSWERS ARE INDICATED FOR ANY OF THE ABOVE QUESTIONS, A JOB
REQUISITION SHOULD BE INITIATED
VI.  COMMENTS:  J.R. No.
                               was initiated
ROUTE:  Original to Operating Engr.

-------
                               EXAMPLE C

                                     STATION
                  OPACITY MONITOR QUALITY ASSURANCE PROCEDURES

                                           COMPANY,       STATION

                       LSI RM41 OPACITY MONITORING SYSTEM



                           OVERVIEW OF QA PROCEDURES

     A proposed quality assurance program has been developed for the
                   - Co.              Station opacity monitoring system.  The
specific QA procedures have been developed to be compatible with the
Station (1) opacity monitoring instrumentation and monitoring system
configuration, (2) data recording device, (3) effluent handling system, and (4)
management and organizational structure.  The proposed QA program will be
field-tested during implementation, reviewed and evaluated periodically, and
revised as necessary over a one-year period.  Through this process, it is
expected that QA procedures will be developed and demonstrated which are both
adequate for maintaining high levels of data quality and cost effective in
term's of necessary time and material resource expenditures.

     The following elements are included in the proposed       Station opacity
monitor quality assurance program.

     (1) Daily Log. Daily Check Instructions - The Daily Log  is to be completed
         by Operations personnel in the boiler control room.  Step-by-step
         Daily Check Instructions are provided for completing the Daily Log.
         The Daily Check  and Daily Log do not require extensive time to
         complete, nor do they require that the person performing the procedure
         be intimately familiar with the opacity monitoring instrumentation.
         The Daily Check  and Daily Log provide for identification of monitoring
         problems and  initiation of corrective action.

     (2) Corrective Action Log and Instructions -  The Corrective Action
         Instructions  and Log  are used when  adjustment,  repairs, and/or other
         non-routine  corrective  action is necessary  as indicated by  the Daily
         Checks.  The Corrective Action  procedures are to  be  utilized  by
         technical  personnel  experienced  in  resolving problems with  the
         monitoring  systems.   The  specific  corrective action  procedures are
         relatively  extensive, but will  provide  adequate documentation for
         future refinement  of QA  procedures  and  demonstration of their
         effectiveness.   The  corrective  action procedures  are utilized only on
         an "as necessary"  basis.

      (3) Periodic QA Check and Instructions  - The Periodic QA check  is intended
         to be performed in conjunction with the opacity monitor routine
         preventive maintenance program performed three  times per  year at the
                Station.   The Periodic QA Check  procedures provide  for checks of
         monitoring system components and operating status which are
         unfeasible,  impractical,  and unnecessary on a daily basis.

-------
(4) The documentation of the QA program will be reviewed periodically to
    determine if modification to the proposed procedures are
    appropriate.  Such modifications may be made as additional experience
    and data are obtained.

(5) An attempt will be made to develop a practical method for conducting
    an annual clear-path check of the       Station opacity monitoring
    system.  If successful, this method will be utilized in conjunction
    with a performance audit of each monitor to be conducted by station
    personnel during the project and a performance audit to be conducted
    at the end of the one-year study by the Pilot Project Staff.

-------
                                         .        COMPANY
                                      STATION

                          DAILY  CHECK INSTRUCTIONS

                          OPACITY MONITORING SYSTEM
                          (Lear Siegler, Inc. RM41)

     The following are step-by-step instructions for conducting the Daily
Check of the       Station opacity monitoring systems.  Criteria are provided
for determining when corrective action should be initiated.  The person
conducting the Daily Check of each monitoring system should complete all
blanks on the Daily Log as indicated below.

I. GENERAL INFORMATION

   o Enter name of person performing check, the date (month, day, year), and
     the time of day (24-hour clock) that the check is begun.

   o Hours Boiler Down - Enter the number of hours the boiler was not in
     operation (i.e., not combusting fuel)  during the preceding 24-hour
     period.

   o Hours Monitor Down - Enter the number of hours the monitor did not
     provide a record of effluent opacity during the preceding 24-hour
     period.  Describe monitor downtime in "Part IV Comments" (i.e. , cause of
     outage, time began, time ended, and any corrective action taken to
     return monitor to service.)

II. FAULT LAMPS

   o Examine the fault lamps on the Control Unit.  Check the appropriate box
     on the Daily Log to indicate the status of each fault lamp (check "YES"
     if lamp is illuminated).

III. PANEL METER DATA

   o Rotate the Measurement switch to the "REF" position and record the
     reference current value (ma) indicated on the 0 to 30 scale.

   o Rotate the Measurement switch to the "COMP" position and record the zero
     compensation level. (The panel meter readout is in units of optical
     density, OD.)  Return the Measurement switch to the "100% OPACITY"
     postion.

   o Check the appropriate boxes on the Daily Log to indicate whether the
     reference and zero compensation values are outside of acceptable limits.

IV.  STRIP CHART DATA

   o Examine the strip chart to determine the zero and span responses of the
     most recent monitor calibration.  Record the zero and span calibration
     values on the Daily Log in units of % opacity.

-------
   o Check the appropriate boxes on the Daily Log to indicate whether the
     zero and span values are outside acceptable limits.

     Note: IF YES ANSWERS ARE INDICATED FOR ANY OF THE QUESTIONS ON THE DAILY
           LOG, CORRECTIVE ACTION SHOULD BE INITIATED AS SOON AS POSSIBLE.

V. COMMENTS

   o Describe any problems observed during the performance of the Daily Check
     and/or any other apparent problems which may affect monitor performance.

   O ENTER THE TIME OF DAY THAT THE DAILY CHECK IS COMPLETED. (Part I)

-------
                          Opacity Monitoring System

                                  DAILY LOG
  I.   GENERAL INFORMATION

      Name:
Date:
                                                   STATION
Time Start:
Time Complete:
      Hours Boiler Down:

 II.   FAULT LAMPS
             Hours Monitor Down:
FAULT LAMPS ON?
Filter
Shutter
Ref
Windows
Over Range
NO





YES





III.  PANEL METER DATA
      "REF" value (ma):

Does
Does
"REF"
"ZERO
value
COMP"
exceed
exceed
acceptable
acceptable
range
range
(17
(+
.9-22
.018
.2
OD)
ma)? .
7 ;
NO


YES


IV.  STRIP CHART DATA

     Zero Calibration (% opacity):
     Span Calibration (% opacity):

Does zero value exceed acceptable limits of j
Does span value exceed acceptable limits (.32
h 2.0% opacity?
.5 +_ 2.0% opacity?.
NO


YES


     IF YES ANSWERS ARE INDICATED FOR ANY OF THE ABOVE QUESTIONS, CORRECTIVE
     ACTION SHOULD BE INITIATED AS SOON AS POSSIBLE.
  V.  COMMENTS:

-------
                                         .  COMPANY,        STATION


                 PERIODIC  QUALITY  ASSURANCE  CHECK INSTRUCTIONS
                         LSI  OPACITY MONITORING  SYSTEM


     The folLowing  are  step-by-step instructions for  performing  the Periodic
QA Checks of the      .  opacity monitoring  system.   Initially, the Periodic QA
Checks are to be performed in conjunction  with the  routine  opacity monitoring
system maintenance  program performed at  least once  every  four months.  The
person performing the QA Checks  should complete  all blanks  on the Periodic QA
Check Log.

  I. GENERAL INFORMATION

      o  Enter the.  name of the person performing the  check, the  date (month,
         day, year) , and the  time  of day (24-hour clock)  that the check is
         begun.                                                  -W,

 II. CONTROL UNIT CALIBRATION CHECK DATA

     (1) Rotate the Measurement  switch to  the "REF" position and record the
         reference  current value (ma) on the Log.   Return the Measurement
         switch to  the  "100% Opacity" position.

     (2) Initiate a manual calibration by  depressing  the  "OPERATE/GAL"
         switch.  Record the zero  calibration responses (% opacity) of the
         panel meter and strip chart recorder.

     (3) Rotate the Measurement  switch to  the "COMP"  position and record the
         zero compensation value indicated by the panel meter (OD).  Return
         the Measurement switch  to the "100% Opacity" position.

     (4) Initiate a span calibration check by depressing  the "ZERO/SPAN"
         switch, and record the panel meter and strip chart span check
         responses  (% opacity) on  the Log.

     (5) Record the minimum 6-minute  average opacity indicated by the chart
         recorder for the hour period preceding the QA check.

III. TRANSMISSOMETER CHECK/SERVICE

     (1) At the monitoring location, check the AGC LED (Automatic Gain Control
         light emitting diode) on  the right hand side of  the transceiver,  and
         record whether it is "ON" or "OFF."  If the AGC  is not illuminated,
         repairs to the monitoring system must be completed before continuing.
         (A Corrective Action Log Sheet  should be filled  out.)

     (2) Alignment Check - Remove the plastic cover from the mode switch on
         the transceiver and turn  the switch to the "ALIGN" position.
         Determine  the optical alignment status of the transceiver and
         retroreflector by looking through the "bull's eye" and  observing

-------
    whether the image is within the circular target  (acceptable),  or
    outside the circular target (unacceptable).   Indicate the position of
    the light beam on the diagram,  and check the appropriate box for
    alignment status on the QA Log.

    Open the retoreflector, and determine if the light beam appears to be
    centered within the port.  Indicate the apparent position of the
    light beam on the diagram and check the appropriate box for
    retroreflector alignment status on the QA Log.

    If the optical alignment is unacceptable, realign the opacity
    monitoring system in accordance with the manufacturer's
    instructions.  If a shift in the baseline opacity occurs after
    realignment, note the magnitude of the change which was observed in
    "Part V COMMENTS."

(3) Inspect and service, as necessary, the purge-air blowers, air
    filters, and shutter mechanism as per the manufacturer's  ,
    instructions.  Note whether the blower system status is acceptable,
    and describe any corrective action taken on the QA Log.

(4) Retroreflector Cleaning/Check - Record the average current value (ma)
    corresponding to the double-pass transmittance indicated by the J-Box
    meter.  Also, record the exact time to allow subsequent determination
    of the "before cleaning" effluent opacity from the chart record (or
    have an assistant note the panel meter opacity value if two-way
    communications are available).  Release the retroreflector latches,
    swing the retroreflector open, and remove all accumulated particulate
    within the optical path of the monitor.  Clean the retroreflector
    surfaces according to the manufacturer's instructions.  Close and
    secure the retroreflector.

    Record the average transmittance indicated by the J-Box meter and the
    "post  cleaning" exact time (or effluent opacity level).  Wait at
    least  two full integration periods (12 minutes) before performing
    step (5) unless an assistant records real time data.

(5) Transceiver Cleaning/Check - Record the average effluent
    transmittance indicated by the J-Box meter and the exact time (or
    average effluent opacity).  Release the transceiver latches, swing
    the transceiver open, and remove all accumulated particulate matter
    from the optical path of the monitor.  Clean both the transceiver
    window and the zero reflector according to the manufacturer's
    instructions.  Close and secure the transceiver.

    Record the average effluent transmittance indicated by the J-Box
    meter  and the exact time (or the "post-cleaning" effluent opacity).

(6) Return to the control  unit/data recorder station and obtain the
    average  opacity values  indicated by the 6-minute chart recorder
    corresponding to  the times recorded on the data sheet  (i.e. before
    and  after  alignment  adjustments, before  and  after cleaning  of the
    retroreflector, and before and after cleaning of the transceiver).

-------
        Record all data on the QA Log.  (This step is not necessary if an
        assistant records real-time opacity values during the QA Check
        activities.)

    (7) Note periods of cleaning on the strip chart record to ensure that
        these periods will not be later mistaken for excess emissions.  Reset
        alarms if activated during QA Checks.  Initiate zero calibration to
        reset zero compensation and record the post-QA Check zero
        compensation level (OD) on the QA Log.

    (8) Note that the periodic QA Check was performed in the "Part V
        COMMENTS" section of the Daily Log.

IV. FINAL MEASUREMENTS

    (1) Upon completion of all the above steps, record the minimum 6-minute
        average effluent opacity value displayed on the strip chart for the
        hour period following completion of all adjustments, repairs, and
        service for the monitor.

    (2) Initiate a manual calibration by depressing the "OPERATE/CAL" switch.
        Record the final zero compensation value using the procedure detailed
        in II (3) above.

 V. COMMENTS

    All observations regarding monitor performance should be explained.

    ENTER TIME OF DAY PERIODIC QA CHECK IS COMPLETED (PART I).

-------
  I. Name:
Opacity Monitoring System

    PERIODIC QA CHECK

          COMPANY

         Date:
                                                                 STATIONS
Time Start:	
Time Complete:
 II. CALIBRATION CHECK DATA

Reference Current (ma)
Zero value (% opacity)
Zero compensation (OD)
Span value (% opacity)
Panel Meter




Chart Recorder
///////

/ / / / / / /

     Minimum 6-min. opacity value (hour preceding QA Check):_

III. TRANSMISSOMETER CHECK/SERVICE

i AGC ON?
YES

NO

                                                        See instructions
                                                        if AGC is not on.
     2. Alignment Status (Note position of Light Beam on Diagram)

          Transceiver  -v--4--?—                     Retroref lector
Alignment Acceptable?
Transceiver
Retroref lector
YES


NO


                                                        See instructions if
                                                        alignment is not acceptable.
     3.
Blower Status OK?
Transceiver
Retroref lector
YES


NO


4.
5.

Before retroref lector cleaning
After retroref lector cleaning
Before transceiver cleaning
After transceiver cleaning
TIME




STRIP CHART
(% opacity)




J-BOX
(% transmittance)




 IV. FINAL MEASUREMENTS

     1.  Minimum 6-min. opacity value (hour following QA Check)
     2.  Final Zero Compensation value (panel meter, OD):	
  V. COMMENTS:

-------
                                            CO.
    STATION
                        CORRECTIVE ACTION INSTRUCTIONS
                         LSI OPACITY MONITORING SYSTEM

     The following are step-by-step instructions for completing the Corrective
Action Log when adjustments and/or repairs of the opacity monitoring system
are necessary.  The person performing the corrective action should complete
all blanks on the Corrective Action Log as indicated below.
I.  GENERAL INFORMATION

   o Enter the name of person performing repairs or adjustments, the date
     (month, day, year), and the time of day (24-hour clock) that the
     corrective action is initiated.

II. SYSTEM/MONITOR FAULTS

   o Enter "ON" or "OFF" for each fault lamp on the Log.

   o Monitor calibration and completion of Part III of the Log is not required
     if "FILTER" or "SHUTTER" problems initiate the corrective action.
     However, if the "REF," "WINDOW," or "OVER RANGE" fault lamps are
     illuminated, the Control Unit data listed in Part III must be obtained
     prior to adjustment of the monitor.
   o Record the time when the fault  is  corrected.
     action taken in the "Part IV COMMENTS".
Describe all corrective
III. CONTROL UNIT

    (1)  Rotate the Measurement  switch  to  "REF"  position  and record the panel
         meter ma  reading  (0 - 30  scale).

         If the "REF"  is not within  acceptable range  (green band of panel meter:
         17.9 - 22.2 ma),  make necessary adjustments,  and enter post-adjustment
         "REF" ma  reading  on the Log.   (If no adjustment  is required, enter
         "NA.")

    (2)  Rotate the Measurement  switch  to  the "100% OPACITY" position.  Depress
         the  "OPERATE/CAL" switch  and record the monitoring system zero check
         responses indicated by  both the panel meter  and  the data recorder
         (% opacity).

    (3)  Rotate the Measurement  switch  to  the "COMP"  position, and record the
         zero compensation level displayed by the  panel meter  in units of optical
         density (OD).

-------
    (4)  Cleaning of the transceiver  and  retroreflector  optics  and/or  other
        adjustments are necessary when:

        (a) zero check response  exceeds +_ 2% opacity, or
        (b) zero compensation exceeds +_  .018 optical density.

        If no adjustment is necessary, enter "NA" on "Adjusted Zero Value" and
        "Adjusted Zero Comp" lines of the Log.

        Describe all adjustments and/or  corrective action in "Part IV
        COMMENTS."  After all adjustments are completed, repeat  steps  (2) and
        (3) above and record the post-adjustment zero responses  and zero
        compensation values on the Log.

   (5)  With Measurement switch on "100%  OPACITY" and the monitor in zero check
        mode, depress "ZERO/SPAN" switch.  Record the monitoring system span
        check response displayed on  both  the panel meter and the chart recorder
        (% opacity).

   (6)  Adjustment of the monitor is necessary if either of the span responses
        of the monitor exceeds -h 2%  opacity from the correct value.

        If no adjustment is necessary, enter "NA" on the "Adjusted Span Value"
        line of the Log.  Describe all adjustments and/or corrective action in
        "Part IV COMMENTS."  After all adjustments are  completed, repeat step
        (5) above and record the post-adjustment span response on the Log.

IV. COMMENTS

    The reason for initiating corrective  action should  be stated, and all
    repairs and/or adjustments performed  as a result of the above procedures or
    as a result of other monitor malfunctions should be described.   Sufficient
    explanation should be provided to determine what was done and what effect it
    had on monitor performance.

    ENTER TIME OF DAY ALL CORRECTIVE ACTION IS COMPLETED.  (Part I)

-------
                             Opacity Monitoring System

                               CORRECTIVE ACTION LOG

                                       COMPANY
 I. GENERAL INFORMATION
    Name:
Date:
                          STATION:
Time Start:	
Time Complete:
II. SYSTEM/MONITOR FAULTS
FAULT LAMPS ON?
FILTER
SHUTTER
REF
WINDOWS
OVER RANGE
NO





YES





CORR1
Date





5CTED
Time





Note:  Monitor calibration and completion of Part III below is not required if
       "FILTER" or "SHUTTER" problems initiate Correction Action.

"REF" value (ma)
Adjusted "REF" value (ma)
ZERO value (% opacity)
ZERO comp (OD)
Adjusted ZERO value (% opacity)
Adjusted ZERO comp (OD)
SPAN value (% opacity)
Adjusted SPAN value (% opacity)
PANEL METER








CHART RECORD
/ / / / / s
/ / / / /

/ / / / /

/////,


Note:  (ma) and  (OD) values  from  panel meter; all other  data from strip chart
       recorder.
IV. COMMENTS:  (Describe adjustments  to monitor  and/or  all  other  corrective action)

-------
                                                - COMPANY
                                         STATION

                           DAILY  CHECK INSTRUCTIONS

                           OPACITY MONITORING SYSTEM
                          (Contraves Goerz Model 400)

     The  following  are  step-by-step instructions  for  conducting the Daily
Check  of  the            Station  opacity monitoring systems.   Criteria are
provided  for  determining  when corrective action should be initiated.   The
person conducting the Daily  Check of each monitoring  system should  complete
all  blanks  on the Daily Log  as  indicated below.

I. GENERAL  INFORMATION

   o Enter  name of  person  performing check,  the date  (month,  day, year),  and
     the  time of day (24-hour clock)  that the  check is begun.

   0 Hours  Boiler Down  - Enter  the  number of hours  the boiler was not  in
     operation (i.e., not  combusting fuel) during the  preceding 24-hour
     period.

   0 Hours  Monitor Down - Enter the  number of  hours the monitor did not
     provide  a record of effluent opacity during  the preceding  24-hour
     period.   Describe  monitor downtime  in "Part  IV COMMENTS" (i.e. , cause  of
     outage,  time began, time ended,  and  any corrective action  taken to
     return monitor to  service.)

II. FAULT LAMPS

   o Examine  the fault  lamps on the  Control Unit.   Check  the  appropriate  box
     on the Daily Log to indicate the status of each fault  lamp  (check "YES"
     if lamp  is illuminated).

III.  STRIP  CHART DATA

   o Examine  the strip  chart to determine the zero  and  span responses  of  the
     most recent automatic monitor calibration.  Record the zero and span
     calibration values on the Daily Log  in units of %  opacity.

   o  Check  the appropriate boxes on the Daily Log to indicate whether  the
     zero and span values are outside acceptable limits.

     Note;   IF YES ANSWERS ARE  INDICATED FOR ANY OF THE QUESTIONS ON  THE
             DAILY LOG,  CORRECTIVE ACTION SHOULD BE INITIATED AS SOON AS
             POSSIBLE.

 IV.  COMMENTS

   o  Describe any problems observed during the performance of the Daily Check
     and/or any other apparent  problems which may affect monitor performance.

   o  ENTER THE TIME OF  DAY THAT  THE DAILY CHECK IS COMPLETED. (Part I)

-------
                                EXAMPLE D

                                       STATION
                  OPACITY MONITOR QUALITY ASSURANCE PROCEDURES

                                        COMPANY,           STATION

              CONTRAVES GOERZ MODEL 400 OPACITY MONITORING SYSTEM



                           OVERVIEW OF OA PROCEDURES

     A proposed quality assurance program has been developed for the
                Co., Lake Road Station opacity monitoring system.  The specific
OA procedures have been developed to be  compatible with the           Station
(1) opacity monitoring instrumentation and monitoring system configuration, (2)
data recording device, (3) effluent handling system, and (4) management and
organizational structure.  The proposed  QA program will be field-tested during
implementation, reviewed  and  evaluated periodically, and revised as necessary
over a one-year period.   Through this process,  it  is expected that QA
procedures will be  developed  and demonstrated which are both adequate for
maintaining high  levels  of data  quality  and cost effective in terms of
necessary time and  material  resource expenditures.

     The following  elements  are  included in the proposed           Station
opacity monitor quality  assurance program.

     (1) Daily Log,  Daily Check  Instructions  -  The Daily  Log  is  to be  completed
         by Operations personnel in the  boiler  control  room.   Step-by-step
         Daily Check Instructions are  provided  for completing the Daily Log.
         The  Daily Check and Daily Log do not require  extensive time to
         complete,  nor do they require that  the person performing the procedure
         be  intimately familiar with the opacity monitoring instrumentation.
         The Daily Check and Daily Log provide for identification of monitoring
         problems and initiation of corrective action.

      (2)  Corrective Action Log and Instructions - The Corrective Action
          Instructions and Log are used when adjustment, repairs, and/or other
          non-routine corrective action  is necessary as indicated by the Daily
          Checks.  The Corrective Action procedures are to be utilized by
          technical personnel experienced in resolving problems with the
          monitoring systems.  The specific corrective action procedures are
          relatively extensive,  but will provide adequate documentation for
          future  refinement of QA procedures and demonstration of their
          effectiveness.  The corrective action procedures are utilized only on
          an  "as  necessary" basis.

      (3) periodic QA  Check  and  Instructions - The Periodic QA check is  intended
          to  be performed in conjunction with the  opacity monitor routine
          preventive maintenance program and  should be  performed  once per  month
          at  the           Station.  The Periodic  QA Check procedures provide
          for checks of  monitoring  system  components and  operating status  which
          are unfeasible, impractical, and unnecessary  on a daily basis.

-------
(4) The documentation of the QA program will be reviewed periodically  to
    determine if modification to the proposed procedures are
    appropriate.  Such modifications may be made as additional experience
    and data are obtained.

(5) An attempt will be made to develop a practical method for conducting
    a clear-path check of the           Station opacity monitoring
    system.  If successful, this method will be utilized in conjunction
    with a performance audit of each monitor to be conducted by station
    personnel during the project and a performance audit to be conducted
    at the end of the one-year study by the Pilot Project Staff.

-------
                          Opacity Monitoring System

                                  DAILY LOG

                                   CO.
  I.  GENERAL INFORMATION

      Name:
      Hours Boiler Down:

 II.  FAULT LAMPS
Date:
III. STRIP CHART DATA

     Zero Calibration (% opacity):
     Span Calibration (% opacity):
                       , STATION, UNIT
Time Start:
Time Complete:
             Hours Monitor Down:
FAULT LAMPS ON?
Stack Power
Dirty Window
NO


YES



Does
Does
zero
span
value
value
exceed
exceed
acceptable
acceptable
limits
limits
of
of
+ 2
+ 2
.0%
.M
opacity?
opacity?
NO


YES


     IF YES ANSWERS ARE INDICATED FOR ANY OF THE ABOVE QUESTIONS, CORRECTIVE
     ACTION SHOULD BE INITIATED AS SOON AS POSSIBLE.
 IV.  COMMENTS:

-------
                                         COMPANY,            STATION
                  PERIODIC QUALITY ASSURANCE CHECK  INSTRUCTIONS
               CONTRAVES GOERZ MODEL 400 OPACITY MONITORING  SYSTEM

 QA Chicks " of 15£ng ^ Ste7rbnSrP lnSJrUCtl°nS f°r Perf°™ing the Periodic
 QA Checks of the         ... Unit #  opacity monitoring system.  Initiallv  th*>
 Periodic QA Checks are to be performed in conjunction with the routine opacity
 monitoring system maintenance program performed at least once a month.  ?he

 ?heck\og!      g     QA ChGCkS Sh°Uld C°mplete aU blanks °n the Periodic QA
   I.  GENERAL INFORMATION

       o   Enter the name of the person performing the check, the date (month,
          day,  year),  and the time of day (24-hour clock) that the check is
          begun.

  II.  CONTROL UNIT CALIBRATION CHECK DATA

      (1)  Rotate  the "MODE"  switch on the control panel to the "ZERO" position

                                    responses  of the panel meter and
     (2) Rotate  "MODE" switch on  the  control  panel  to  the "SPAN"  position,  and
         record  the span check responses  of the, panel  meter and chart recorder
         {.* opacity).

     (3) Record  the minimum 6-minute  average opacity  indicated by  the chart
         recorder for the hour period preceding  the QA check.

III. TRANSMISSOMETER CHECK/SERVICE

     (1) Alignment Check - Determine the optical alignment  status of  the
         transceiver and reflector by looking through the alignment sight and
         observing whether the image is within the circular target
         (acceptable), or outside the circular target (unacceptable).
         *™-Cat? ^ P08*'1™ of the «*l* be*"> °« the diagram, and  check the
         appropriate box for alignment status on the QA Log.

         Open the reflector, and  determine if the light beam appears  to be
         centered within the port.  Indicate the apparent position of the
         light beam on the  diagram and check the appropriate box for reflector
         alignment status on the  QA Log.

         If the optical  alignment is unacceptable,  realign the opacity
         monitoring system  in accordance  with  the manufacturer's
         instructions.   If  a shift in  the baseline  opacity occurs  after
         realignment,  note^the magnitude  of the  change  which was observed in
         Part V  COMMENTS . "

-------
(2) Inspect and service, as necessary,  the  purge-air  blowers,  air
    filters, and shutter mechanism as per the manufacturer's
    instructions.  Note whether  the  blower  system  status  is acceptable,
    and describe any corrective  action  taken on  the QA Log.

(3) Reflector Cleaning/Check - Record the average  effluent opacity value
    indicated by the transceiver meter.  Also, record the exact  time  to
    allow subsequent determination of the "before  cleaning" effluent
    opacity from the chart record (or have  an assistant note the panel
    meter opacity value if two-way communications  are available).
    Release the reflector latches, swing the reflector open, and remove
    all accumulated particulate  matter within the  optical path of the
    monitor.  Clean the reflector surfaces  according to the
    manufacturer's instructions.  Close and secure the reflector.

    Record the average effluent  opacity value indicated by the
    transceiver meter and the "post  cleaning" exact time  (or effluent
    opacity level indicated at the control  unit).  Wait at least two  full
    integration periods (12 minutes) before performing step (A)  unless an
    assistant records real time  data.

(4) Transceiver Cleaning/Check - Record the average effluent opacity
    indicated by the transceiver meter and  the exact time (or effluent
    opacity indicated at the control unit).  Release the  transceiver
    latches, swing the transceiver open, and remove all accumulated
    particulate matter from the  optical path of  the monitor.  Clean both
    the transceiver window and the zero/span chopper according to the
    manufacturer's instructions.  Close and secure the transceiver.

    Record the average effluent  opacity indicated  by the  transceiver
    meter and the "post-cleaning" exact time (or the effluent opacity
    indicated at the control unit).

(5) Install the "zero jig" on the transceiver and  determine the  initial
    response of the monitor (i.e., before any adjustments are performed)
    at a minimum of three test points (e.g., 0% opacity, 30%-70% opacity,
    and 100% opacity).   Adjustments  to the monitor should be performed
    and a Corretive Action Log completed if the monitor response exceeds
    i 3% opacity of the correct  test values.

(6) Return to the control unit/data  recorder station and obtain  the
    6-minute average opacity values  indicated by the chart recorder
    corresponding to the times recorded on  the data sheet (i.e. before
    and after alignment adjustments, before and after cleaning of the
    reflector, and before and after  cleaning of the transceiver).  Record
    all data on the QA Log.  (This step is not necessary if an assistant
    records real-time opacity values indicated at  the control unit during
    the QA Check activities.)

-------
IV. FINAL MEASUREMENTS

    (1) Note periods of cleaning on the strip chart record to ensure that
        these periods will not be later mistaken for excess emissions.  Reset
        alarms if activated during QA Checks.

    (2) Note that the periodic QA Check was performed in the "Part V
        COMMENTS" section of the Daily Log.

    (3) Upon completion of all the ahove steps, record the minimum 6-minute
        average effluent opacity value displayed on the strip chart for the
        hour period following completion of all adjustments, repairs, and
        service for the monitor.

 V.  COMMENTS

    All observations regarding monitor performance should be explained.

    ENTER TIME OF DAY PERIODIC QA CHECK IS COMPLETED (PART I).

-------
  I. Name:
Opacity Monitoring System

    PERIODIC QA CHECK

   COMPANY

         Date:
'   STATION, UNIT #

Time Start:	
Time Complete:	

ZERO value
Adjusted ZERO value
SPAN value
Adjusted SPAN value
% OPACITY
PANEL METER




CHART RECORD




     Minimum 6-min. opacity value (hour preceding QA Check):	

III. TRANSMISSOMETER CHECK/SERVICE

     1. Alignment Status (Note position of Light Beam on Diagram)

          Transceiver    /*\                      Reflector
Alignment Acceptable?
Transceiver
Reflector
YES


NO


                                                        See instructions if
                                                        alignment is not acceptable.
     2.
Blower Status OK?
Transceiver
Reflector
YES


NO


     3.

     4.

Before reflector cleaning
After reflector cleaning
Before transceiver cleaning
After transceiver cleaning
TIME




STRIP CHART
(% opacity)




TRANSCEIVER METER
(% opacity)




     5.Zero Jig Calibration Data
TEST VALUE
% Opacity
0%
100%

MONITOR RESPONSE
% Opacity



DIFFERENCE
% Opacity



 IV. FINAL MEASUREMENTS
     1.  Minimum 6-min. opacity value (hour following QA Check):
  V. COMMENTS:

-------
                                           CO.,

                         CORRECTIVE ACTION INSTRUCTIONS
                 CONTRAVES GOERZ M400 OPACITY MONITORING SYSTEM

 ArMo^ foliowinf/re step-by-step instructions for completing the Corrective
 Action Log when adjustments and/or repairs of the opacity monitoring system

           ar'Th
 al bl«,,           "erformlng ^e corrective action sould
 all blanks on the Corrective Action Log as indicated below.

 I.   GENERAL INFORMATION
    °  fmonTh^H  name °Vers°n P"forming repairs or adjustments, the date
      (month  day,  year),  and  the time of day (24-hour clock) that the
      corrective action is initiated.

II. SYSTEM/MONITOR FAULTS

    o  Enter  "ON"  or "OFF"  for  each fault  lamp on the Log.

    o  If  the "DIRTY WINDOW"  lamp  is illuminated,  note the  minimum 6-minute
      average effluent  opacity which occurs  during  the one hour  period
      preceding  the initiation of corrective action.   After cleaning  both the
      effJuenV    f re^t0r Wind°WS'  n°te the miniraum  6~min«te average
      effluent opacity  which occurs during the following one hour period!
       M            W*en..the fault is corrected.  Describe all corrective
     action taken in the "Part IV COMMENTS".

III. CALIBRATION DATA/ ADJUSTMENT

    (1)  Rotate the "MODE" switch on the control panel to the "ZERO" position,
         a*™T°rdy    24erVheCk responses of the panel meter and chart
         recorder (% opacity) .

         Cleaning of the transceiver and reflector optics and/or other
         adjustments are necessary when the zero check responses of either the
                     °r./he ^^ reC°rder 6XCeed "± 2% "P^y-  Ascribe 111
                     and/or corrective action in "Part IV COMMENTS."
         rrf            .                  Panel to the <
-------
IV. COMMENTS

    The reason for initiating corrective, action should be stated, and all
    repairs and/or adjustments performed as a result of the above procedures or
    as a result of other monitor malfunctions should be described.  Sufficient
    explanation should be provided to determine what was done and what effect it
    had on monitor performance.

    ENTER TIME OF DAY ALL CORRECTIVE ACTION IS COMPLETED. (Part I)

-------
                              Opacity Monitoring  System

                                CORRECTIVE ACTION LOG

                                 .. COMPANY
  I. GENERAL INFORMATION
     Name:
                                       Date:
 STATION, UNIT #
Time Start:	
Time Complete:
 II. SYSTEM/MONITOR FAULTS
FAULT LAMPS ON?
STACK POWER
DIRTY WINDOW
NO


YES


CORR!
Date


:CTED
Time


III. CALIBRATION DATA/ADJUSTMENT

ZERO value
Adjusted ZERO value
SPAN value
Adjusted SPAN value

"ZERO JIG" DATA:
Initial Values
Adjusted Values
% OPACITY
PANEL METER





0% OPACITY


CHART RECORD





100% OPACITY


 IV.  COMMENTS: (Describe adjustments to monitor and/or all other corrective action)

-------

-------
          APPENDIX B.





QUALITY ASSURANCE DATA SUMMARIES

-------

-------
                                                EXAMPLE A
                                      OPACITY CEMS DAtt.Y CHECK RESULTS
MONTH:
   10
11
12
13
14
15
16
17
18
19
20
21
22
2
2


2
2
2
3
2


2
0
0


0
0
0
0
0


0
  23
                                                                                               MONITOR*
TIME
DAV REQ.
(MIN.)
1
2
3

4
5
6
7
8
9
2
2
2

2
2


2
2
BOILER
DOWN
(HRS.)
0
0
0

0
0


0
0
MONITOR FAULT ZERO
DOWN LAMPS CHART LOGGER
(HRS.)
0
0
0

0
0


0
0

8
9
9

9
9


11


9
9
9

8
8


8
8
SPAN
CHART LOGGER COMMENTS /OPERATOR(1)

61
61
61

61
61


61


61
61
61 Note: Power surge effect-
ing timer and emissions.
62
61


63
62 Note : Chart recorder gear
                                0
                                0
                                0
                                0
                                0
                                0
                                0
9*
9
9*
61*
61
62*
62
broken.
Note: Data logger not print-
ing 0; manual zero/span ok.
Note: Resetcal timer
10
9
9
9*
9
9
8
8
9*
9
61
62
62
61*
62
62
62
62
62*
62
                                                    9
                62
                62
                61
               Note:  Data logger not
               printing.
               Note:  Data logger not
               printing.
24
25
26
27
28
29
30
31

2
2
2


2
—
3

0
0
0


0
0
0

0
0
0


0
0
0

9
9
9


10
10
10

8
8
8


9
9


61
61
61


62
61
61

62
52
62


62
62









Note : Data logger timer, off
Manual zero /span checks.
*  Zero and span responses transposed on data sheet.

-------
                                     EXAMPLE B
 c
01
O-.
o:

1
5S
?i
Xt-
0. *-
si
?s
?s
i-«.
ii
CO-**
 I-l-
p
 !>•
  a
  o
                              1
                                                                              <£.
                                                                              Or
|.t(4 «Mtl/nP30 DAILY LOG OATA i
	 i
EHO flAS'-ZERO CHT_SPAN OAS.S^AN HAiFNCTN FLAGS J«'NO STAHT COHPLETE 'TOT^T'lHE '
Faulfs " , ,
tc ' I
» 1-
"" u
'1
o
u
'•o
u
i
!*•
, e
T
T
f
I <
*
1
I
o a
so
o o-
tntn,
•z -z.
n C O1
<0-»
n
cat
0 0' 0 0. 0 0
1
•» o--c> O' o-o-
& -C -O «O'-O -O
O -3
•C
o o
IT IT
0 0
O O
O 0
0 O
ne-
at)
run
•O •£
in
1
f. a
9 13
_i~
u*

*" °
o o
3~Ur§
u\ir
o o
o o
'
•q
re »
(U
r> r*
n a
^ in
-o .c
iAin
z z
So
OC
x •»

IT If
0 O
O O
n n
a a
* *-
•o -o
in i/i
3 3
u. u.
li
XI

in in
i
i
irt~
0(V|.
in in
>0
00
1
n tn-c n
^ tn 03
« -o -c -c
o —
\nt~
gg
tn
X O"
-c -c
tn tr
u. il
n ji
3 3
r i

=g
*
0 C 0 C-
i° °
ir ir
O 3
0 0
O 0
1 1
a o
in -c
 01
33
uu

o -4-
e o
0 0
1
if
o ci
1
r- r

a a
•t ir

n 01
< 3
uu

\ I
i
o>o i/i -»'in n
c - o o
-> O IM (VI
in in in in
z z z z
o* o- so x
r~ r~ r-. c
-0 -O .0 -C
o o
inm'n.*
o co a
0000
Io o
i l

o a
•C"-

LJ UJ
3 3
UU

G O
n w
o e
O 3IO O
1 1,1 1
f f Cr 0
A (i'o o
eb s±io o
*•?

a a
ceo-
o o
to o
xrr

o c
1 1

o a
e — •
lf\ V
CC CC

i
-J- U>
•» in
c 
•» ir
c <
O O
0.0.

0 ^
'?

a o
•C f-
PONtTft 1
COLV I N

P IN I
I
t
in in|
tr a
SA
in mi
ir in
tn tn tn >o>
o in-o o,
n in tn sa
o CT. o-
-T. C - >- V,

c o
1
- 0
c c
in (vi
0 IV
o in
\n o
o -•
•o in
0)
rv,
C7-
•C
VI
>- z
r^ •c 
X 
ej
in
o
o
o
t
O
O
o
o
93
•»
00
1
ED

-------
MONTH:
                                            EXAMPLE C
                                   OPACITY CEMS DAILY CHECK RESULTS

DAY

1
2

3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
*^*!4
23
24
25
26
27
28
29
30
31
TIME FAULT
REQ. LAMPS
(MIN.)

8:00 REF.on
15:30 REF.on
6



7
5
5
5
5


5
5
5
5
5
6

5
5
Dl A fill/

4


?



REFERENCE
CURRENT
(MA)

REF.on
18.3
20



20
20
20
20
20


20.3
20.4

20.3
20.4
20.4


—

20.5


20.5



Zl
METER


-1
-1
0



0
0
0
0
0


1.5
1.5

1.2
1.4
2.2

2
—

2


2



IRQ
CHART


-1
-1
0



0
0
0
0
0


1
1

1
1
2

1.5
—

1


1



ZERO
COMP
(OD)

-0.02

0



0
0
0
0
0.002


-0.005
-0.004

-0.005
0.004
-0.002

-0.002
—

-0.002


-0.002



SF
METER

32
29
32.4
32.4



32
32
32
32
32


33.5
33.1
33
34
33
34

34
—




33



>AN
CHART OPERATOR COMMENTS

32
29 Changed transceiver band
32.4 onRM41.
32.4



32
32
32
32
32


33.5
33
32
33
33
34

34
~




32.4




-------
MONTH:
                                              EXAMPLE  D
                                     OPACITY CEMS DAILY CHECK RESULTS
                                                                                           MONITOR *
DAY
1
2
3
4
5
6
7
8
9
10

It
12
13
14
15
16
17
18
10
20
21
22
23
24
25
26
27
28
29
30
31

TIME
REQ.

-------
MONTH:
UNIT NO.:
DAY
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
26
27
28
29
30
TIME
REQ.
(MIN.)
7
2
2
2
2
2


1
1
1
3
1
3
3
15
—
60
10
3
3
2
3
2
10
10
10
1
3



BOILER
DOWN
(HRS.)
7










10
21
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24




MONITOR FAULT ZERO
DOWN LAMPS METER CHART

-------

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
    68-02-3962
             3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE

RECOMMENDED QUALITY ASSURANCE PROCEDURES FOR OPACITY
CONTINUOUS EMISSION MONITORING SYSTEMS
             5. REPORT DATE
                February 1986
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

  James W.  Peeler
             8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS

   Entropy Environmentalists,  Inc.
   PO Box 12291
   Research Triangle Park, NC   27709
                                                            10. PROGRAM ELEMENT NO.
             11. CONTRACT/GRANT NO.

                   68-02-3962
12. SPONSORING AGENCY NAME AND ADDRESS
  U.  S.  EPA, Stationary  Source Compliance Division
  Waterside Mall
  401 M Street, SW
  Washington, DC  20460	
             13. TYPE OF REPORT AND PERIOD COVERED
             14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
     This  report presents recommended quality assurance (QA) procedures for opacity
 continuous  emission monitoring systems (GEMS's)  installed at electric  utility steam
 generating  stations.  The recommended procedures are intended to provide a simple,
 cost-effective approach to  the development and implementation of opacity GEMS QA
 plans.  The approach described here allows for much flexibility in  the selection of
 monitor-  and source-specific procedures, the establishment of QA control limits, and
 the organization of the quality assurance elements.   This report identifies and
 describes the major elements of a QA plan: (1) daily QA checks, (2) periodic .QAchecks
 and preventive maintenance,(3)  corrective action procedures, and (4) accuracy checks.
 This report also addresses,  the organization and  interaction of the elements of -the QA
 plan, an  approach for optimizing the QA plan, and the assignment of various QA
 responsibilities.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                           c. COSATI Field/Group
          Air Pollution
          Monitoring
Opacity Monitoring Systems;
Continuous EtntLss ion
 Monitoring
Quality Assurance
 Procedures
18. DISTRIBUTION STATEMENT

       Release to Public
19. SECURITY CLASS (ThisReport)
   Unclassified	
21. NO. OF PAGES
      81
                                              20. SECURITY CLASS (Thispage)

                                                Unclassified	
                                                                         22. PRICE
EPA F«m 2220-1 (R«v. 4-77)   PREVIOUS EDITION is OBSOLETE

-------

-------