EPA-650/2-74-013


   PERFORMANCE SPECIFICATIONS
                  FOR
        STATIONARY-SOURCE
        MONITORING SYSTEMS
FOR GASES  AND VISIBLE  EMISSIONS
                    by

            John S, Nader, Frednc Jaye,
              and hilliap Conner
          Chemistry and Physics Laboratory
            Program Element Mo. 1\A01Q
          U.S. ENVIRONMENTAL PROTECTION AGENCY
         National Environmental Research Center
          Research Triangle Park, N  C. 27711

               January 1974.

-------
                              PREFACE
     This report provides the background and experimental data as a
technical base for the development and formulation of guidelines for
monitoring pollutant emissions from stationary sources.

     References to commercial products in this report directly or by
inference are not to be considered in any sense as an endorsement of
the product by the Government.  Nor does utilization of  any commercial
product to generate the data reported here signify that  this product
necessarily meets either the performance specifications  exemplified
in this report or any guidelines that nay be proposed on the basis of
this report.

     This report has been reviewed by the Environmental  Protection
Agency and approved for publication.  Approval doea not  signify that
the contents are the official guidelines in response to  the needs
stipulated in the standards of Performance of New Stationary Sources.
This report does provide the technical base from which guidelines to
monitors of specific pollutant emissions from selected source industries
will be formulated and officially proposed in the Federal Register.
                                  anley M.  Greenfield,  Ph.D.
                                  Assistant Administrator
                                for Research and Development
                                  111

-------
                           CONTENTS
LIST OF FIGURES	,'	,	   v
LI ST OF TABLES 	, ..  vl
ACKNOWLEDGMENTS 	  vi
INTRODUCTION.	.'....	.	   1
DEVELOPMENT BACKGROUND . „	   5
PERFORMANCE SPECIFICATIONS	,  	   9
  Parameters	   9
  Specifications	  .  14
  Applicability	  15
SUMMARY	  . ....  	  .  19
REFERENCES	  .  	  21
APPENDIX A   S0_ MONITORING CAPABILITY APPLIED TO  A  POWER
  PLANT	  ....  23
APPENDIX B.  NO  MONITORING CAPABILITY APPLIED TO  A
               X
  POWER PLANT	,	  .  . .  27
APPENDIX C   VISIBLE EMISSIONS MONITORING  CAPABILITY
  APPLIED TO A POWER PLANT.	  29
APPENDIX D   DUKE POWER PLANT  STUDY........	  43
APPENDIX E.  EXAMPLE:  PERFORMANCE SPECIFICATIONS  AND
  SPECIFICATION TEST PROCEDURES FOR MONITORS  OF
  POLLUTANT GAS EMISSIONS  FROM STATIONARY  SOURCES....  ..  49
APPENDIX F,  EXAMPLE:  PERFORMANCE SPECIFICATIONS  AND
  SPECIFICATION TEST PROCEDURES FOR TRANSMISSOMETER
  SYSTEMS FOR CONTINUOUS MEASUREMENT OF THE OPACITY  OF
  STACK EFFLUENTS		61
                                 IV

-------
                        LIST  OF  FIGURES
Figure                                                    Page


 A-l)  Flue Gas Diversion Systeip, ...... „ ...............   25

 A- 2  Gas Diversion and Instrument Installation Design,,,   26

 C-l  Particle Extinction Coefficients for Various
        Aerosols,  Calculated from Mi e Theory ............   30

 C-2  Out-of- Stack Transnitta*>ce of Three Colors of
        Light as a Function of In -Stack  Tran snuttar.ee
        of White Light for an Experimental White (Oil)
        Plume ..... »,....., ......... ...... ........ . ......   32

 C-3  Qut-of-Stack Transmit tance of Three Colors of
        Light as a Function o£ tn-Stack  Transmit tance
        of White Light for an Experimental Black (Carbon)
        Plume .......... ,,,, ......... , ____ . ____ ..........   33

 C-4  Calculated Error in True Light  Transmit tance at
        2 Degrees  Detector Angle of View .... ....... ,  ....   34

 C-5  Calculated Error in True Light  Transmittanee at
        20 Degrees Detector Angle of  View,...  ........  „   j^
 C-6  In-Stack Transiittance of a Coal -Fired  Power
        Plant Emission for Various Transiissometer
        Light Projection and Detection  Angles.....  .....   35

 C-7  In-Stack Transnittance of a Coal-Fired  Power Plant
        Emission at Various Light Projection  Angles and
        Wavelengths ..... .... .'.„,„.„.„.„<,,,,  . ..... . ....   37
 C-8  Out-of-Stack Transmittance of a Coal -Fired  Power
        Plant Emssion as a Function of In-Stack
        Transmittance. ..........................  ...  ..
 C~9  Transmissoneter  with  Collimating

 C-10  Effluent Transroittance at Stack  Exit  as  a
        Function of In-Stack  Transmittance and
        Ratio of Stack Exit Diameter to
        Transmissometer Path length,. ..........

-------
                        LIST OF TABLES
Table
    "Performance  Specifications for Pollutant  Gas
       Emissions  from Stationary Sources
     Performance  Specifications for Transmissometer
       Systems  for Measurement of Visible Emissions
       from Stationary Sources ...... .....
     Design Specifications fear Transmissometer Systems for
       Measurement  of Visible Emissions from Stationary
       Sources ........... ............. ........... . ......   15
A-l  SO  Monitoring System Test Results
B-l  NO  Monitoring System Test Results
                                                         28

D-l  S02  and NOX Monitoring System Test Results..	   44
                   ACKNOWLEDGMENTS

       He express  our  appreciation to James Hcrnolva  and Mike
  Barnes of the Stationary Source Emissions Measurement Staff
  in acknowledgment of their providing the measurement data on
  the Duke Power Plant Study.

-------
            PERFORMANCE  SPECIFICATIONS
                               FOR
   STATIONARY-SOURCE  MONITORING  SYSTEMS
        FOR  GASES  AND  VISIBLE  EMISSIONS

                          INTRODUCTION
     The phrase continuous monitoring instrumentation conveys dif-
ferent meanings to various users of air pollution control equipment, de-
pending upon the nature of the application, or the design of the system,
or ooth.  The basic concept of this type of instrumentation is,  however,
that the functional operation of the instrument  system is automatic
and more or less continuous in tine.   Differences in design, or
application or both iray be such that either the  system or a component
operates in a cycle having a tine period ranging from fractions  of a
second (relatively instantaneous) to tens of minutes or pore.
     This report is intended to provide technical information and
discussion that will assist in the formulation of federal regulations
and state and local laws that require monitoring systems.  The document
also furnishes the instrument manufacturers and  users with technical
guidelines on the desired performance requirements of these systems.
     In the discussion of monitoring systens for stationary-source em-
issions, the measurement of pollutant gases, visible effluents,  and other
such substances involves the whole instrument system functioning auto-
matically and pore or less continuously in tine.  The response time of
the system that defines the specific  functioning of t:ie system with

-------
respect to time is prescribed by the performance specifications *




discussed later,  The monitoring systems considered include the major




component operations of sampling, analysis, and data presentation.




The major subsystems performing these operations are defined function-




ally as follows:




     1   Sampl in_g_Interface.   This subsystem performs one or more of




         the following operations:  delineation, acquisition,  trans-




         portation or conditioning of a sairple of the source effluent




         or protection of the analyzer from the hostile effects of




         the sample or the source environment.




     2"  Analyser.  This subsystem (often treated as a whole and in-




         dependent measurement system for analytical purposes)  senses




         the pollutant substance or parameter and generates a signal




         output that is a function of the pollutant concentrations,




     3,  D; ata Presentation.  This subsystem provides a display of the




         output signal in concentration units or other specified units





     Monitoring systems are used in engineering and research studies




to evaluate performance of control equipment, in surveys of enissions




for input to dispersion rrodels, in testing for compliance with certain




emission standards, and in enforcement programs.  New Source Perform-




ance Standards (NSPS) proposed and promulgated by the Environmental




Protection Agency (EPA) require the installation and operation of




inonitoring systems for specific pollutants (sulfur dioxide, nitrogen




oxides, visible emissions) emitted from specified affected facilities in




Category I and II sources     It is anticipated that similar require-




ments for monitoring additional source categories (III, IV, and others)

-------
will be proposed and promulgated in the near future   State Imple-




mentation Plans may also require monitoring of existing sources,




The New Source Performance Standards propose guidelines on the




selection and operation of the specified rnonitonng systems, and




the State Implementation Plans will likely follow a similar pro-




cedure.




     The performance specifications presented in this document provide




technical guidelines for the application of iromtoring systems to




specific pollutants from selected source industries   No constraints




have oeen placed on the sampling approach, the analytical scheme,




the system design, or the development of hardware.   Test procedures




are presented to determine whether or not any given monitoring system




can meet the prescribed performance specifications

-------
Page Intentionally Blank

-------
                    DEVELOPMENT  BACKGROUND



     Of the many possible performance  parameters  for continuous -mon-



itors (relating, for example,  to the numerous  applications and the



various purposes for which data are  generated), those were delineated



that are essential to reliable performance  and to the generation of



valid data with a minimum of downtime  and within  a reasonable level



of accuracy, compatible with the current availability of commercial



systens and with the intended  utilization of the data.  The-determining



guidelines in the choice of the parameters  and the development of these



specifications were the need to provide data to ("flag") the operator



or control officer on (1) malfunctions of control equipment,  (2) ver-



ification of acceptable emission control operation, and  (3) relative



levels of pollutants.  Since these parameters  have a degree of flexibility,



the values specified and test  procedures prescribed can be adjusted



to meet the specific needs of  a given  source-pollutant combination.



Furthermore, other parameters  may be added  or  some of the present



ones modified or deleted to meet special needs.



     Sorne of the parameters selected have undergone changes in definition



and specification values over  the period of tinie that the specifications



have been under development and evaluation.  The parameters are generally



accuracy,  calibration error, zero drift, calibration drift, repeatability,



response time, and operational period.  These  are discussed more fully



in the context of their definitions and the specifications given later.



     In 1971, EPA supported several Contract programs to evaluate


                                                                 (2)
commercially available monitoring systeirs for  sulfur dioxide (SO,)     and

-------
nitrogen oxides (NO V   in power-generating plants fired by fossil fuels.



Previously," over a nuipber of years, EPA and the former Edison Electric



Institute* had conducted cooperative studies of visible emissions from


                                                     '    (4)" '  ''    '  ' -
power plants and the rnstrunental measurement of opacity.     These



studies provided a substantial technical data base for the drafting



of a set of performance specifications and test procedures to verify



the specifications.(Summaries of three of these studies are given in



Appendices A, B, and C, illustrating the data relating to the perfor-

       j  '               ~ '     >,~~''      •  f      E<      ,  -
mance specifications; complete'details are provided in the reports



referenced.



     A draft of the performance specifications and test procedures for


monitors of SO^, NO , and visible emissions was prepared in early



1972 and circulated for review and comment within EPA and externally


to a representative cross section of instrument manufacturers, in-



dustrial users, and local governmental agencies   The comments re-



ceived were incorporated (where applicable) in a revised document,



dated September 30, 1972.



     In early 1973, at a power plant of the Duke Power Company, EPA



conducted research studies to evaluate (airong other objectives) the



specifications and test procedures contained in the draft of September



30, 1972.  (A summary of this study as it relates to the performance



specifications is given in Appendix D.)



     This study showed that all the performance specifications of



the September 30 draft were adequate, wit1! the exception of the



accuracy specification..  The range of values extended beyond the



specified *_ 5 percent maximum for the mean to as much as 8 3 percent.
   Currently a part of the Electric Power Research Institute.



                                   6

-------
The conclusion was to revise the specification for the accuracy to



be the sum of the mean and the 95 percent confidence interval.  This



provided greater flexibility in the specification by prescribing the



combined rather than the individual values,  In addition, the September



30 draft was revised to incorporate similar changes in the specifica-



tions of the other parameters and in some of the definitions deemed more



appropriate and consistent with the test procedures.



     The decision to revise the specifications was based on a



real and significant difference between the contract studies (TRW and



Monsanto) and the in-house study (Duke plant).  The former were to



provide background information for the development of the specifications



and test procedures.   In the latter, the proposed specifications and test



procedures were assessed, and their applicability to a selected



source-pollutant combination was validated.

-------
Page Intentionally Blank

-------
                   PERFORMANCE SPECIFICATIONS

     Tile technical background discussed earlier  and  summarized in the
Appendices irakes it possible to identify parameters  that adequately
delineate the desired performance of a  measurement system as it applies
to monitoring emissions fron stationary sources.  This background also
provides a technical basis for a set of specifications that are realistic
and meaningfully ronsistant with the current  availability of commercial
stack-monitoring systens for specific pollutants  fror selected industries
(pollutant-source combination).

PARAMETERS
     The parameters that adequately delineate the desired performance  of
a measurement system as it applies to munitonrg  missions from stationary
sources are accuracy, calibration error, zero  and caliaratien drift,
repeatability, response tme and operational  period.  Accuracy (relative)
is defined as the degree of coirectiess with  which the reasurement
system yields the value of gas concentration  of  a sample relative to
the value given by a defined reference  method.  This accuracy is expressed
in terms of error that is the difference between  the oaired concentration
measurements.  The error is expressed as a  percentage of the itean value
determined by the defined reference method.   The  absolute error is the
combination of the error in the iponitoring  measurement and whatever error
exists in the reference measurement.  It is possible that the absolute
accuracy of the monitor may prove to be better than  that of the reference
method.  Theoretically, however, the assumption  is made  that the
reference method is one that provides the best available accuracy.
This is not always the case in practice.
                                  9

-------
     Furthermore, within a given measurenent system, the error con-




tributed by the sampling operation tends to he greater than the




error from the analytical operation and becomes trie limiting factor




in the absolute accuracy of the measurement systen as a uhole.  The



contained effect of sampling in both the monitoring and the reference




measurements can result in a significant impact on the relative




accuracy,



     For x'lsible-ernissions monitoring systems, relative accuracy is




not specified.  The reference method could be an instrument system




based on first principles in its design or a set of design specifications




that would implicitly involve a specified error that is predictable




from theoretical considerations   To designate an instrument system




as a reference method for monitoring systems that would require,  in




effect, the sane design criteria by definition ivas deemed impractical




and. a circuitous approach.  The decision rfas to use the design criteria




for reference purposes fron the"beginning.  The design specifications




used in  this docunent introdjce a riajon-rm er^or nf S nercent in opacity




/or narticle sizes that are 10  micrors i"n Hip-nster,  The error decreases



with smaller sizes, approaching zero error at 0.3-micron sized particles.




     Calibration error is defined as the difference between the pol-




lutant concentration indicated by the measurement svstem and the




known concentration of the test gas mixture,:   This type of error is




measured by the use of a known concentration of the gas pollutant in




some clean, dry gas stream, usually air or nitrogen.   Frequently,




calibration is executed by introducing the calibration test gas into




the analyzer and by-passing the sampling interface.  This performance




specification requires that the entire measurement systen be included



                                  10

-------
in the test procedure.  This means introducing the calibration test


gas at the sampling interface upstream of the analyzer.  Thus, the


sampling interface is provided the same opportunity to have an impact


on calibration as on the source measurement.  In the case of optical


in-situ -(nonextractive) systems,,  calibration test gas could be caused


to flow through a sample cell located in the optical path.  An alternative


would be a cell with the calibration test gas sealed in the cell;


however, the long-term stability of the calibration test gas in the


sealed cell would need to be verified, possibly presenting practical pro-


blems.


   - Zero drift is defined as the change in measurement system output


over a stated period of tine of- normal continuous operation, when the


pollutant concentration at the time of the measurements is zero.  Cal-
               rf                                            •*

ibration drift is defined as the change in measurement system output


over a stated period of tine of normal continuous operation when the


pollutant concentration at the time of the measurements is the same


known upscale value.  Zero and calibration drift are critical parameters


that have a direct effect on calibration error and ultimately on the


accuracy of the data output.  Both short and long term drifts are


important considerations.  Short-term instability can lead to Mis-


interpretation of variations in emission concentrations; long-term


instability presents a problen in interpreting long-term trends in


degradation of control equipment performance.  The 24-hour cycle for


zero and calibration check and adjustment is stipulated to provide   i=,


consistency and optimum utilization of data for evaluating day-to-


day and month-to-month functioning of the source emission control


facilities, or pollutant emission levels or both.


                                  11

-------
      Repeatability is  defined  as'a  measure  of  the measurement  system's

 ability to give the same output reading(s)  upon  repeated  measurements

 of the same pollutant  concentration(s).   \  separate  specification  is not

 given because it is Implicitly covered  by the  confidence  interval  included

 in the calibration error.

      Response time is  defined  as the tine interval from a  step  change

 in"pollutant  concentration at  the input to  the measurement system  to
                                                      \
 the time  at which  95 percent of the corresponding final value_ is

 reached as displayed on  the measurement system data  presentation device,

 The response  time  of a measurement  system is strongly influenced by

 the sampling  approach  used.  Its impact on  the data  output is  also

.largely related to the combination  of process  operation and control

 equipment  ceing monitored.  Generally,,  in a process  involving  relatively

 long term changes,  such  as a power  plant  that  -nay require  a half hour

 to go from half to full  load,  a response  time  of 10  minutes may be

 acceptable for monitoring purposes.  On the other hand, in batch-type

 processes such as  nonfcrrous smelters,  dramatic  changes in emissions

 occur within ninutes,  and a short response  time  is more appropriate.

 Extractive sampling techniques that involve several  hundred feet of

 sampling lines obviously introduce  a more significant delay in

 measurement response than an in-situ method IP which no sample  ex-

 traction is done and the response can be  instantaneous.

      The nature of the data output  requirement and the rate of  data

 output also play a significant role in  specifying a  response  time.

 In visible-emissions  (opacity) regulations  on  power  plants, for ex-

 ample, an opacity of 40  percent  is  allowed  for 2 minutes  per  hour,  In

 this case, a response  time  less  than  1  minute  would  be required.   Slow


                                  12

-------
rates of data output such as one data 'point per 5 to  15 minute in-




tervals for steady-typo processes will'allow some ireasure'iert systems




to nonitor several emission points on  a'"time-snared" basis   An




essential point  in the overall consideration of response time ib that




some trade-off may be required between the roture and/or rate o£ data




output and cost.  The cost impact will depend largely upon the neasnre-




mcat technique, the process, and the equipment utilized.




     The operational period is a minimum period of time over which a




measurement systen is expected to operate within certain performance




specifications without unscheduled maintenance, repair, or adjustment,




     The location of the monitoring system on a stationary source depends




laigely on the puipose for whi'ch the data is being collected   For mon-




itoring pollutant concentration so as to indicate malfunction or im-




proper maintenance of control equipment or excess i\e emissions,  a




measurement system can be located at an> point where reasonabl)  re-




presentative sampling of the emission  concentration cap be rnade.  Gen-




eral guicelines include fl) consideration of convenience and acce<=s-




abilitv for maintenance of the monitoring system, and {2j location



downstream of any polljtion control equipnent and upstream of j




dilution or other process affecting the emission eonccntmtion.   It




'nay be necessary to conduct preliminary nieasuremcrits traversing the



stack diameter to determine the nature of the concentration profile and




to establish a representative sampling location for the find installation




of the ironitoring systen.




     Stack monitoring systems that utilize different sampling approaches




are comirercially available; tfiey include extractive, in-s-it'j, and re-




mote techniques.  '  The performance specifications provide no constraints




                                 13

-------
on the sampling interface or any other major subsystem (see Introduction)

within the monitoring system.  Many source monitors that require ex-

tractive sampling are available as analyzers exclusive of the ''interface"

subsystem.  Much infornation is available on the performance of the

analyzer in the laboratory under controlled conditions and on field

applications.  The sampling interface has a significant impact on per-

formance reliability and is specifically related to each analyzer-source

combination.  Data on the various combinations,  including the inteiface_

as a total system, are very limited or nonexistent

SPECIFICATIONS

     The performance parameters for gas-pollutants and visiole emissions

that have been defined are assigned values as shown in Tables 1 and

2, respectively.  Table 3 shows design specifications that are tech-

nically discussed in Appendix C,

          Table 1.  PERFORMANCE SPECIFICATIONS FOR POLLU1ANT
                GAS EMISSIONS FROM STATIONARY SOURCES
          Parameter          .               Specification

  Accuracy (relative)           520% of mean reference valuea

  Calibration error            < 51 of each test  gas  value3
  Zero drift (2 hr)            < 2% of emission  standard3

- Zero drift (24 hr)            1 41 of emission  standard3

  Calibration drift (2 hr)      1 2% of emission  standdrd3

  Calibration drift (24 hr)     < Si of emission  itandaida

  Response time                                 _  10 mm

  Operational period                             ]68 hr

   Absolute mean value •*• 95  percent confidence interval,

                                   14

-------
           Table  2.   PERFORMANCE  SPECIFICATIONS  FOR TRANSMISSOMETER
                 SYSTEMS  FOR MEASUREMENT  OF  VISIBLE EMISSIONS
                           FROM STATIONARY SOURCES
                   Parameter                   Specification
          Calibration  error  "      '   ±10%  of  test  filter  value3
           Zero drift  (24 nr)
           Calibration drift  (24  lir)
           Response time
           Operational period
£.10%  of  emssion  standard3
5.10%  of  emission  standarda
           < 10 sec
           168 hr
           Absolute n>ean value  +  95 percent  confidence  interval.
              Table 3   DESIGN SPECIFICATIONS FOR TRANSMISSOMETER
     SYSTEMS FOR MEASUREMENT OF VISIBLE EMISSIONS FROM STATIONARY SOURCES
Parameter
        Specification
Spectral response               .         Peak and mean response within 500 to
                                         600 nrr; less than 10 % of peak response
                                         outside of 400 to 700 nn.
tagle of view                   '        " 5 degrees maximum (total angle)
Angle of projection             '         5 degrees maximum f total angle)
                                I
aThe relative response of a transmissoraeter to radiation of different wavelengths.
     maximum (total) angle of radiation that is seen by the photo-detector
 assembly of an optical transmissometer .'
cThe maximum (total) angle of radiation that is projected by the lamp assembly
 of an optical transmissorr.eter.

                                                                      3
 APPLICABILITY
      The performance  parameters  and  their  specified vilues  in  Tables
 1  and 2  mast be  considered  in  the  context  of their  application before
                                   IS

-------
they can be utilized with any real weaning.  Furthermore, test pro-




cedures* must be established to verify that a monitoring system does




indeed meet the specifications   The test procedures must be con-




sidered not only Li the context of the measurement system's application




but also, just as importantly, in relation to how the data generated by




the system is to be used.  In fact, the anticipated utilization of




the data is implicit in the development of the specification values




cited in Tables 1, 2, and 3,  As stated earlier, the guideline used




in the development of tnese performance specifications has been largely




a middle-of-the-road approach that would pemit optimum utilization of




commercially available monitoring systems; that is, systems that will




have the potential to meet many needs adequately.




     Appendices E and F are examples of how tne specifications in




Tables 1, 2, and 3 na> be applied in conjunction with test procedures




as guidelines in the assessment or selection of monitoring systems




for pollutant gas emissions and visible emissions, respectively.  It




is rietessarv to use the teguiatory language in the examples given to




naintain the integrity of tie values and definitions of the parameters




 and the  test procedures  for the sake of  internal consistency because




of tne  interaction  among  these  factors




     The guidelines in appendices E and  F  are  potentially  applicable




to the  following needs




      1.  Criteiia tlipt provide  consistent  and  valid data relevant




         ro measurement needs and ensure the generation of data of




         knoun sualit\,  toiv-jtible uitn  specified use
  specified in -\ppendices E and F




                               16

-------
     2.  Criteria that provide clearly defined objectives for in-




         strument development and manufacture.




     3.  Performance parameters that permit maximum flexibility n




         design and fabrication of the system hardware, including




         electronic equipment,  for mnimum cost and with irininum




         complexity,




     4.  Criteria for the user to assess available systems on a common




        1 basis and to make appropriate selection for various applications




         that may include monitoring process operation, control operation,




         anc emission levels.





     It is possible to move toward tighter or looser specifications




than those given in the examples  shown.   Whether or rot it is necessary




or desirable to do so depends upon the specific pollutant-ndustry com-




bination, the intended utilization of the data, and other factors, such



as economic considerations and  experience.  The constraints on moving




toward tighter specifications are not the lack of present technological




capability but rather the cost impact, current availability of commercial




systems, and practical experience witn a given pollutant-industry com-




bination.  Technical data on the applicability of specifications to




various pollutant-industry combinations are gradually being developed




     The guidelines m the appendices are supported technically by




actual application to the pollutant-industry combinations for SO-, NO ,



and visible emissions in coal-fired power plants based on the develop-




ment background.  This experience enables an understanding of the appli-




cability of the guidelines.  It provides the proper basis for considerirg




any changes in the specification values and/or test procedures as may be






                                 17

-------
dictated by special needs in the utilization of the data for these same




pollutant-industry combinations.  The parameters delineated should apply



in general.  The specification values and test procedures exemplified



possibly can be made generally applicable to other pollutant-industry



combinations.  Any projected application to other conbinations is more




or less valid, depending on the intended or end usage'of the data generated



and on whether or not the applicability of the guidelines to the new



combinations has been demonstrated,



                                   4
                                  18

-------
                                SUMMARY




     'Parameters for evaluation  of  the performance of continuous monitors



for th'e measurement of emissions" from stationary sources have been de-



lineated and defined.   Field  testing programs have been conducted to




provide technical background  on the availability and applicability of



commercially available source monitoring  instrumentation.  The technical



data generated by these field studies are shown as the basis for the de-



velopment of performance specifications  and of  test procedures by



which these specifications can  be verified.



     Examples are given of how  performance  specifications may be applied



in conjunction with test procedures as  guidelines in the assessment or



selection of monitoring systems for pollutant gas emissions and visible



emissions from stationary sources   Applicability of these guidelines



was discussed within the constraints of  specified pollutant-industry



combinations and utilizations of the data generated.
                                  19

-------
Page Intentionally Blank

-------
                             REFERENCES
1,  Standards of Performance for New Stationary  Sources,  Federal
    Register, Vol.  36,  No.  247,  Deceniber  23,  1971;  Vol.  38, No.
    Ill, June 11, 1973; Vol. 38, No. 84,  May  2,  1973.

2.  Jaye, F, C.  Monitoring  Instrumentation  for the  Measurement of
    Sulfur Dioxide  in Stationary Source Emissions    TRW  Contract
    No. EHSD 71-73.   NTIS PB 220-202.   Office of Research and Monitoring,
    U. S. Environmental Protection Agency,  February 1973.   124 p.

3.  Snyder, A. D. et al. Instrumentation  for  the Determination
    of Nitrogen Oxides  Content of Stationary  Source Emissions.  Vol.  I  and
    Vol. II.  Monsanto Contract  No. EHSD  71-30.   NTIS  PB 204-877 and  NTIS
    PB 209-190,   EPA.  January 1972.

4.  Conner, W, 0. and J. R. Hodkinson.  Optical  Properties  and
    Visual Effects  of Smoke-Stack Plumes,   EPA,  Research Triangle
    Park, N. C.   Publication Number AP-30.  May  1972.  89 p.

5,  Nader, J. S, Developments in Sampling and Analysis Instrumentation
    for Stationary Sources.  J.  Air Pollut. Contr   Assoc.  23:
    589-591, July 1973.

6.  Hodkinson, J, R.  The Optical Measurement of Aerosols.  In.
    Aerosol Science, Davies, C.  M  (ed.). New York,  Academic
    Press, 1966.  p. 289.

7.  Ensor, D, S, and M. J,  Pilat.  The  Effect of Particle Size
    Distribution on Light Transmittance Measurement.  Amer. Ind.
    Hyg. Assoc.  J.   32: 287-292, May 1971.

8.  Peterson, C. M  and M.  Tornaides, In-Stack Transmittance
    Techniques for Measuring Opacities  of Particulate Emissions
    from Stationary Sources.  Environmental Research Corporation.
    Contract No. 68-02-0309.  NTIS PS  212-741.   EPA.  Research
    Triangle Park,  N  C. April  1972   87 p.

9.  Hoirolya, J.  B.  A Review of Available  Techniques  for  Coupling
    Continuous Gaseous  Pollution Monitors to  Source Emissions.
    EPA, (Presented at  165th National American Chemical  Society
    Meeting   Dallas, Texas.  April 8-13, 1973   12  p.)
                                  21

-------
Page Intentionally Blank

-------
         APPENDIX  A.   $02  MONITORING  CAPABILITY



                  APPLIED  TO A POWER  PLANT^2)




     In September 1970,  the Environmental Protection Agency contracted



with TRW Systems Group,  Redor.do  Beach, California, to perform the



required field tests to  evaluate commercially available instrument-



ation for monitoring SO- emissions  from a coal-burning power plant.



This program included  an identification of commercially available



instrumentation, selection of instrumentation for the field test,



design of the field  test, and operation of the field test  site.



     TRW surveyed the  instrument marketplace to identify manufacturers,



vendors, and developers  of SO. measurement instrumentation.   Eighteen



manufacturers and/or vendors responded, and 17 companies indicated an



active program in the  development of SO, measurement units.   The survey



revealed that,  as of January 1,  1971, 12 SO- source monitors were com- -



mercially available.   The detection and analysis scheme consisted of



one of the five following principles   spectroscopic, electrochemical,



conductometric, coulometric, and mass spectrometric.   The  spectroscopic



technique included UV/visible absorption, nondispersive and  dispersive



IR/visible absorption.   After this  information was received,  additional



companies announced  the  sale of  new instruments, which operated on one



of the five principles mentioned above, except that one used flame stim-



ulated emission spectroscopy (flame photometric).



     The contractor  initiated the field test study with six  SO, mon-



itors that were available for delivery in accordance with  the  required



schedule.  They utilized the first four of the five analytical principles



given above.




                                  23

-------
     The field tests were conducted at a coal-fired steam generating
plant.  The unit employed in the SO  monitor test burned coal with
                                   2
approximately 0.5 percent sulfur that produced an effluent SO   level
of approximately 250-300 ppm.  The results are summarized in Table
A-l for the test set up shown in Figures A-l and A-2.
             Table A-l.   S02 MONITORING SYSTEM TEST RESULTS
 Test parameter
          Value range
 Accuracy
 Calibration error
 Zero drift (2 hr)
 Zero, drift (24 hr)
 Calibration drift (24 hr)
 Repeatability
-4.0 % +_ 14% to 0.5% +_ 6.3%
 0.34 +_ 0.05% to 2.9% +_ 31
 0.1% to 3%
-2.2% +_ 0.1% to 3.2 +_ 3%
-0.4% +_ O.S to 1.1% +_ 2.1%
-6.1% to 0.6%       .  :
 a All  values are reported as mean values -^ 95 percent confidence intervals as
   percentage of 1000 ppai span.

   Accuracy expressed as mean deviation +_ 95 percent confidence interval of
   instrument readings versus barium-thorin titration Method 6, Federal
   Register, December 23, 1971.

 c Error expressed as mean difference +_ 95 percent confidence interval between
   least squares calibration curves (5 points) and single point (1080 ppra) span
   calibration when measured with three other calibration gases at 190, 660,
   and 880 ppip.
                                    24

-------
     SAMPLE MANIFOLD
,W!TH PROBE INSERTION PORTS
SAMPLE LINE (7.62-CENTIMETER
HEAT-TRACED ALUMINUM PIPE)
            HIGH-VOLUiE    OPTIONAL ACCESS
              BLOWER,.       FOR FILTER,   / SAMPLE PORT-^     DUCT
                         7 82 CENTIMETER /
                           RETURN LINE

                                                      RETURN PORT
                         .3
                     Figure A-l   Flue gas diversion system
                                    25

-------
 LEGEND
 0 PROBE
   DUCT
   STACK   '
   HEAT-TRACED SAMPLE LINE
   STACK GAS RETURN LINE
   BLOWER
   SAMPLE MANIFOLD (SEE(T
   VALVE TIMERS
 8 BLOWBACK COMPRESSED AIR SYSTEM
 9 STANDBY BLOWBACK N2
10,11 ZERO AND SPAN GASES
12 CALIBRATION GASES
13,16 VACUUM PUMPS
14 WET CHEMISTRY
15 TYPICAL INSTRUMENT
17 DIGITAL PRINTER
                                   Figure A-2  Gas diversion and instrument installation design
    'SINGLE OR MANIFOLD
    OPTIONAL OUTLETS
 ©OPTIONAL IN-LINE
    FILTER SYSTEM
 A   '    VALVE ~
-*"*- ELECTRIC ACTUATOR
 rt MANUAL VALVE
 9 PRESSURE GAUGE
 f PRESSURE REGULATOR

-------
         APPENDIX B.   NOX MONITORING CAPABILITY



                  APPLIED TO A  POWER  PLANT




     In-December 1970,  the Environmental Protection Agency'con-
       l-U


tracted-with Monsanto Research  Corporation, Dayton,  Ohio, to per-

        t

form the required field  tests to evaluate commercially available



instrumentation for  nonitcring  NO  emissions from a coal-burning



power plant.  This program included  (1)  identification of the man-



ufacturers and users of N0x-monitoring equipment to determine the



state-of-the-art, (2) perfonr-ance of laboratory tests of suitable



monitoring equipment, and (3) performance of field testing at a



coal-fired power plant.   This approach was taken (as opposed to



the approach for SO.  monitoring) because, at the time of program



inception,  it was determined that NO  monitoring was ill-defined



and in a less advanced state than SO, monitoring.



     Monsanto obtained information from 85 organizations regarding



the manufacture or use of  NO  monitoring instrumentation.  As  of



January 15,  1971, nine NO   source monitors were commercially available.



The detection and analysis  scheme consisted of one of the following



principles:   spectroscopic, electrochemical,  and mass spectrometnc.



As in the  case of SO-, the  spectroscopic technique included  UR/visible



absorption,  nondispersive  and dispersive UV/IR visible absorption.



In the study,  Monsanto used seven NO  monitors, *hich utilized  the



first two  of the three analytical principles  mentioned above.         *•*•



     The NO   monitor  study  included both laboratory  and  field test   -'•;



programs at  a coal-fired steam generating plant.  The tests  at  this



source showed S00 levels of 1800-2400 npm SO,  and NO  levels of
                2                    -r    2        x



                                  27

-------
250-350 ppn, of which about 10 ppm were NO .   Table B-l summarizes



the performance of several monitors in the laboratory and field



tests.
               Table 3-1,  NO  MONITORING SYSTEM TEST RESULTS2
                             x
  Test parameters
           Value range
  Accuracy                           '         -2 .3% ^ 4.5% to -7.4% +_ 4.3%



  Calibration error0                   '       -10% +_ 0.4% to 9 4% +5%



  Zero drift (2 hr)                  !          0.2% + 1,1% to -1.5% -4.4%
  Zero drift (24 hr)



  Calibration drift- (24 hr)
 0.4% *_ 0.5% to 0.61 +_ 7 .n



-6.6% + 4.3% to 3.7% +8.2%
  Repeatability                      ,  "       <2%
  Response time
                                               10 mm
  A
    All values reported as nean values as percentage of 500 ppm span.




    Accuracy expressed as mean deviation +_ 95% confidence interval of



    instrument readings versus phenol disulfonic acid analysis Method ">,



    Federal Register, August 17, 1971. (Field test,)




  c Error expressed as mean deviation +_ 95 percent confidence interval of



    instrument readings versus standard calibration gas.
                                   28

-------
APPENDIX C.   VISIBLE  EMISSIONS  MONITORING CAPABILITY



                  APPLIED TO  A  POWER  PLANT





      Visible  emissions are measured In  terms of the amount  of light  atten-



 uated or transmitted by the emissions in  the visible portion of the



 spectrum relative to the incident light.  The percentage of light



 attenuated is defined as the opacity of the emissions.  The light  trans-



 mitted is  the ttransnittance of the emissions.  The sun of the trans-



 mitted light  and the attenuated light constitutes the total incident



 light.  Therefore, a plume that does not  attenuate any incident light



 is invisible  and will have a transmittance of 100 percent and an



 opacity of 0,  A plume that attenuates  all the incident light is



 said to be 100 percent opaque;  it will  have an opacity of 100 per-



 cent and a transmittance of 0.   An instrument that measures and mon-



 itors opacity or transmittance  is referred to as a transmissometer.



 Instruments that monitor transmittance  of an effluent within the



 stack are  available commercially as in-stack transmssoneters.



 As many as 23 manufacturers have been identified as producers of



 in-stack monitors, and more are appearing on the scene.



      Two inportant optical characteristics of transmissometers  must



 3e specified  to obtain similar  performance from instruments:   the



 operating  wavelength and light  collimation of the instruments.   The



 operating  wavelength is important because fine particalates attenuate



 shorter wavelength radiation more than  longer wavelengths.   Proper
                                   29

-------
 selection  of the operating wavelength of  the transmissometer will

 also minimize interference due  to absorption and radiation from

 gases and  water vapor  in the emissions.   The light  collimation  is

 important  because measurement of the true transmittance and opacity

 of aerosols  requires the exclusion of light scattered by the aerosols

 from the measurement.   Collimation requires an optical design that

 limits the light viewing and projection angles of the transmissometer.

 No restriction on viewing and projection  angles results in instru-

 ments with poor sensitivity and performance,

      The effect of wavelength and particle size on  the transmittance

 of smoke can be illustrated by  examining  the theoretical particle

 extinction efficiency  curves reported by  Hodkinson    (Figure C-l).
                          PARTICLE-SIZE PARAMETER: a = « d/ X
                         5                  IB
15
i    2
                                   CURVE a - Tss>ispssEf«r WMODISPEPSE SPHEDK i»«133
                                   CURVE B-TMNSPfPEITMOIWDIbPERSE SPHERES i- -I i
                                   CJPVEC-AeSORBINt M€TOOI$?£P5E SPHERES •", SJ -Bk&i|T!TAL EXT"!CT!t)ii)
                                   CJ»VEO = AeSORBlN6«'aWt)iS'EPSE S'HERtS ft 53 -OSoi'SCSTTEffiNo tor, a01EliT
                                   CJPV£E'S6?CRilNC»OtOC!S»W5t SPHERES r-^ H -U66i'SBSOI>!'TION COfPONES1
      0,05          1          15          2          25          3
               AREA-MEAN PARTICLE DIAMETER FOR 0 52 micrometer WAVELENGTH, imEr
-------
The particle extinction efficiency factor Q depends on the particle




refractive index relative to the surrounding nedium, its shape, ard




its size relative to the wavelength, usually expressed as a = nd/>,




where d is the particle diameter ,tacacitv  of an aeiosol with particle  size parameter




greater than  6  (nean dianeter  of  about 1 inn.ion in  0 5  nicron  light)




will not geneially be a function  of transpissometer wavelength, and,




if t^e aerosol  15 h.ghl)  absorbing, the opacity can be  independent

-------
of wavelength when the particle  size  parameter exceeds  3  (mean di-



ameter of  about 0.5 niicron in 0.5  micron light).  At  smaller par-



ticle sizes,  the opacity is wavelength dependent since  the  volume



extinction  coefficient is proportional to A - n, 0
-------
                 100
                  80
              o
              z
                  60
              S  40
                 20
                  "0       20      40      80      80     100
                         IN-STACK TRANSMIT!ANCE, peicent

                Figure C-3  Out-of-stack transmittance of
                three colors of light as a function of in-stack
                transnittance of white light for an experimental
                black (carbon) plume
     The effect  of the detector angle of view  can be illustrated by .

use of the  theoretical results reported by  Ensor and Pilat.  (?)  These"

results were  used  to calculate the error in  the true light transpittance

as a function of the true lignt transmittance  at various particle size

distributions  for  detector angles of 2  and  20   (Figures C-4 and C-5).

The erroi E   in  the true light transnittance is defined as:
                        ( measured -   true)      „,.
                    t  -           T
                                  true


Figure C-6  shows  that for a 2-degree detector angle of view the

error in the  true transmittance is in general less than 5 percent

in the 40  to  100  percent transmittance  range for an aerosol with

particles  of  size geometric standard deviation 3 and particle mass

mean diameter less than 10 microns,  Figure  C-5 shows that for a

                                  33

-------
   20
   15
   10
LU
3
IE
O
IE
CC
LU
  IPARTICLE REFRACTIVE INDEX-i 5
         LIGHT WAVELENGTH =1
GEOMETRIC STANDARD DEVIATION
  OF PARTICLE SIZE DISTRIBUTIONS
  PARTICLE MASS MEAN DIAMETER=d ~
      LIGHT SOURCE EMITTANCE
                   ANGLE=0 degrees
            20       40      60
            TRUE TRANSWTTANCE, percent

    Figure C-4   Calculated error in true light
    transmittance at 2 degrees detector angle
    of view
                 PARTICLE REFRACTIVE INDEX =1 5
                  LIGHT WAVELENGTH =0 55*m
                  PARTICLE SIZE STANDARD
                      DEVIATION -3         —
                    LIGHT SOURCE E1MITTANCE
                     ANGLE «0 degrees
            20      40      60      80
              TRUE TRANSMITTANCE, percent
 Figure C-5  Calculated error in true  'ight
 vansnvttance a' 20 degrees detector  angle
 of view
                     34

-------

                     10    20 .    30    40     50    60
                        OETECTION.ANGLE, degrees

            Figure C-6  In-'stack transnittartce of a coal-
            fired power p'ant emssfon for various trans-
            missometer light projection and detection angles
20-degree  detector angle of view, the error in true  transmittance

can range up to  30 percent  in the 40 to 100 percent  transmittanee

range for the  sane aerosol.  ,In general, the error associated with

a given detector viewing angle increases with increased particle

pean diameter, and,,at  a given'particle mean diameter, decreases

with increasing particle size  geometric standard .deviation (increasing

polydispersity of  the particle size).

     An experimental study  of the effect of the collimating angles

of the detector and light source of a transmissometer on the measure-

ment of the opacity of  emissions from a coal-fired' steam generator

                                        (8) *         "
was conducted  by Peterson and Tomaides,  ; They also examined the
               j
opacity of this emission at  several wavelengths within the visible

light region and considered  the correlation between  the in-stack

opacity measurement and the  opacity of the plume.

     To conduct the study,  the investigators developed an experimental

transmissometer with variable detector and lamp collimating angles


                                  35

-------
and with provision for inserting interference filters.  The trans-




missometer was installed alongside a reference transmissometer in




the stack of a coal-fired electric power generating plant.  The




reference transmssoiieter was well collimated and had a photopic




spectral response.




     The stack was of cylindrical steel construction, approximately




3 meters in dianeter.  The top of the stack was 89 meters above ground




level and 55,5 meters above the boiler room roof, through which




it protruded.  The stack carried the effluent from a 120- megawatt




boiler, equipped with electrostatic precipitators rated at 13 per-




cent efficient,   The experimental and reference transmissoirieters




were located in the stack 15 2 and 18.9 meters above the roof,




respectively.  The opacity of the emission was controlled during




the study by varying the precipitator voltage,




     The collamating angle tests (Figure C-6) show that the size



of the collimating angles of the detector and light source produce



similar errors and both must be constructed as small as practical




to minimize the error in the measured transauttanee and opacity.




The trar.smittance measurements at four different wavelengths within




the visible spectrum showed that the transmittance of the emission




increased with each decrease in wavelength (Figure C-7).  This




result indicates that the particle size was large enough to place




the particle extinction coefficient beyond the first maxima of




the extinction efficiency curve.  The ir.-stack and plume opacity




measurements showed good correlation (pigure C-8).  The in-stack



measurements were with the reference transmissometer, which had




a 3-degree angle of vietv and near 0-degree light projection





                                36

-------
  20
            FILTER^
            o  fl 436/iia
            o  0 436^15 ,
            A  0
S.
      _ RECEIVER ANGLE -5 degrees
            PROJECTOR ANGLE, degrees

 Figure C-7   In-stack transmittance of coa>-
 fired power plant emission at various light
 projection angles and wavelengths
  20       40       SO       80

IN-STACK TRANSMITTANCE, perceni
                                           loo
Figure C-B   Out-o*-stack transmittance of
coal-fired power plant emssion as a function
of in-stack transmittance
                    37

-------
angle.  The out-of-stack plune transrnittance measurements were



made with a narrow 1/2 degree angle telephotometer by the method



                                                         f41
of contrasting targets described by Conner and Hodkinson.  '




     Clearly, the wavelength response of the transmissometer used



for compliance to opacity regulations must measure the opacity of




the emissions to visible radiation since the regulation is based




on the visibility of the emissions.  This is best accomplished by




restricting the spectral response of the transmissometer to the

                                                 )


visible.  Interferences caused by the absorption and emission of




radiation by hot_ gases are also at a minimum in  this region of the




spectrum.  This specification, can be raet by proper selection of




detector and lamp or detector-lamp-filter combination.  Most




commercial instruments can easily meet this specification.



      In  theory, the collimation angles for the  light source  (pro-




jection  angle) and for the detector  (detection  angle) should be



0 to  avoid the detection of scattered light.  This is an ideal




that  cannot be met in practice because the execution of any  in-



strument design will require  a finite collimation angle.  Associated




with  the finite collimation angle  is,a systematic error in  the



transmission that is a function of the particle size and' particle




size  distribution in1 the particulate emission.   There are   ' '



limiting" factors on how small a collmation angle, one can




can specify.  As one approaches smaller  collimation-angles,-  in-




strument cost increases and stability and optical1alignment  problems




are generated.  In practice,  a compromise is necessary in which  a


                              /                      '

moderate restriction on the collimating  angles  for the detector




and light  source is specified.  Accordingly, an angle of 5  degrees




                                38

-------
(total angle) or  less  is  recommended.   Empirical data on the emission;,

of a coal-fired steam  generatorW  indicate that a transnnssoineter with

5-degree colligating angles  measure the opacity about 5 percent low

relative to  the true value  (Figure  C-6)   This error would approach

0 for the same transraissometer applied to j source with a particle

size distribution  shifted to smaller mass mean diameters (towards

submicron sizes,  Figures  C-4 and C-5).

     The 5-degree  (naximum)  specification reportedly can be met with

little or no impact on cost.   It is a  good compromise between large errors

introduced at larger angles  and alignment and instability problems in-

troduced at smaller angles.

     Most commercial transmisiOTieters  do not meet this collimation

specification, however, they can ireet  the specification by simply

adding light baffling or  by  adding  tollimating lenses and apertures

to their detector  and  light  souice   A coliimated transipissometer

is shown in Figure C-9.
                   ANGLE OF VIE*
   DETECTOR  (•APERTURE
i  PROJECTION ANGLE
 COMPACT
-FILAMENT
 LAMP '
         COLLIMATWGyBil
         LENS        T
                    LENS
                    CLEANING
                    AIR
                 Figire C-9  Transnissometer with colliTating optics
                                   39

-------
     The design specifications discussed above  are  important  for



performance stability and accuracy, and they 'represent minimal re-



quirements to obtain equivalent performance of  different transmisso-


                                '  •                  '  ,'J  --
meters for measurement of the opacity of stationary, soii-rce emissions,
                                            '         >* £


Most commercial transmissometers need improvement and "testing to



demonstrate acceptability.  The once-a-day minimum  calibration re-



quirement is particularly difficult for most instruments since they



do not generally have in-stack calibration capability./



     Since opacity is dependent upon the depth  of effluent fpathlength)



through which the measurement is made, the in-stack pathlength for



the measurement and the pathlength for the applicable opacity



regulation and instrument readout must be specified  It is recommended



that the pathlength for the measurement include the entire width



of the stack or duct.  An installation that uses a "pathlength less



than the effluent depth would negate tie averaging  advantages of



across the stack iti-situ measurements.  Since the opacity legulation



is based on the opacity of the plume enitted Dy the source, an in-



stack measurement of opacity at a pjthlengtb cifferentltfroui the di--

                          "                      '       lOS'c        '  "
ameter of the plume at the stack exit nust be corrected to the opacity



of the effluent for a pathlength equal to the stack exit diameter to



determine compliance with the regulation.  The  relation for' the cor-



rection is                          A                  -,_   ••          *


                                            log(l-02)
where 1, is the stack exit diametax and 0  is the opacity of the  >



effluent at the stack exit diameter, I  is the  inr,stack pathlength



of the traismissometer and 0, is the nieasuied opacity of the effluent
                            i


within the transmissonieter path.  This relation  is  =hown, Hi figure C-1Q



                                  40                   '  •''

-------
   UJ


   fe
   UJ
        10
2D
       20        40     M


IN STACK TRANSMITTANCE, percent
60   70   10 90  100
Figure C-10. 'Effluent transmtttance at stack exit as a function of in-stack transmittance
and ratio of stack exit diameter to transmissometer pathlength
           O Ji


in terms  of transmittance (transmittance equals 1 minus opacity).


     The  transmissometer output  should permit expanded display of


the in-stack  opacity on a standard  0  to 100 percent readout  scale.


A graph or custom-made scale should be provided with  each  installa-


tion to show the relation between the standard 0' to 100 percent


readout scale and the opacity  of the  effluent for a pathlength


equal to  the  stack exit diameter.   The full-scale measurement should


represent  an  opacity approximately  1.5 times the maximum allowable


opacity of 40 percent for steam  generators.


                                   41

-------
     Caution must be used in the location of the transmissotpeter.


The instrument should be located so that a representative participate


concentration passes through the viewing volume.  For devices using


slotted alignment tubes, care should be taken to ensure that the


design does not interfere with free flow of the effluent through


the entire optical volume.  If the transtnissometer rrust be located


near a bend or obstruction 'where a representative participate con-


centration xs in douDt, the location should be detenuned experi-


mentally.


     Good operational stability requires maintenance of optical  align-


ment and clean optical  surfaces.  Installation and design requirements


are directed toward minirpizing the effects of temperature, vibration,


and other operational conditions that can cause optical misalignment


and. lead to excessive drift.  Installation requirements will


depend largely on the characteristics of the plant operation and


this factor can vary significantly among various industries.  Accord-


ingly, installation costs can vary depending upon the severity of
                                                      O'. '

installation conditions.  The adequacy of the system's lens cleaning
                                                        . u

irechanisp must also be  judged in the context of the specific installation.


It is advisable to observe closely the performance of a system for a


period of time following installation and to take steps, if necessary,


to correct any deficiencies that may develop.


     Automatic ler.s cleaning and calibration and installation will


have an impact on cost.  An alternative to automatic cleaning and


calibration is manual maintenance.  Aside from relative cost con-


sidcratjons, it ib necessary to consider t>ie reliability of manual


naintenance versus automatic operation.  Expenorce to date is in-


adequate t.o asbCbs the  trade-off possibilities and the imuact on cost.


                                  42

-------
        APPENDIX  D.   DUKE  POWER  PLANT STUDY



     EPA conducted a source measurements  study over  a 3-month period


(January thru March of 1973)  at  a pulverized-coal fired power plant of


the Duke PoueV Company in  Charlotte, North Carol ina. <"9J The study had


as one of its prime objectives  the comparison of various sampling


approaches for the measurement  of S0« by  source emission iponitors and


the stability of operation of an opacity  monitoring  system.  The


three different approaches to gas measurement included: across-the-


stack and within-the-stack (in-situ) measurement with UV absorption;


extractive sampling with UV absorption analysis; and remote sensing


with UV correlation spectroscopy.   All these measurements were com-


pared to tne compliance test  method (Method No. 6 of the NSPS).   An


additional objective was to evaluate and  validate the applicability


of the proposed specifications  and test procedures for monitoring


SO,, and visible emissions  on  a  coal-burning power plant.


     Corraiercially available instrument systems that represented


the three measurement approaches  to SO- ipcnitoring were installed

             <.- '  ''.
on the stack of a 150-megawatt  wall-fired boiler.  The stack was


also instrumented with ar  acrosi-the-stack transrcissometcr (opacity


monitor)  for neasuieroent of visible emissions and with instrumental ion


to monitor continuously stack gas  velocity.  The control equipment


consisted of both hot and  cold  electrostatic precipitators containing


a total of 14 stages.   This instrumentation permitted control of


the participate loading in the  stack emissions over a range in finite


increments determined by cutting  off a specific number of precipitator


stages,  A small building  erected at the  base of the stack housed the


instrument control units and  the  digital  data acquisition equipment


                                 43

-------
     The instrument systems were operated continuously after installation

for a 90-day period.  Daring this period, the proposed (September

30, 1972, draft) performance specifications test procedures were ex-

ecuted as prescribed and were evaluated on their applicability.  The

instrumentation was also left unattended for about 60 days for a more

extensive reliability test,  Within tne total period, 65  measurements

by l^S'S Method No. 6 vere made as reference measurements  for correlation

with the corresponding (in tine) instrumental measurements   Measure-i

pent data were also obtained at three levels of particulate loading

ranging from 2 percent to 47 percent opacity, effected by appropriate

cutback on various stages of precipitator control.  Table D-l summarizes

some of the results on the performance tests on the SO  monitoring

systems.



         Table  D-l.   S02 AND NOX MONITORING  SYSTEM TEST RESULTS

                                              Results, %

            Test  parameter               SG2             NOX
       Accuracy3                    5.8  to 32   '   15J ' to 35

       Zero drift  (2 hr)b           0.52 to  i 8     0*'08-to  1.5

       Zero drift  (24 hr)b          i.o  to  i S     0 2flto  3.3

       Calibration drift  (2 hr)b    0,82             1 3  to  6,6

       Calibration drift  f24  hr)    3.6  to  4.8   ,  1,7  to  5.0
        aResult  is  the  sum of  the  absolute mean erior  and  the  95%
         confidence interval as a  percentage of tne mean measurement
         by  Method  6 for  S02 and by Method 7 for KOX.  Method  7  for
         NOX has  run in sets of two.   Sets of  three should  nprove
         the confidence interval and  the  accuracy

         Result  is  the  sum of  the  absolute mean value  and  the  951
         confidence anterval as a  percentage of the emission
         standard (620  ppn for S02 and 525 ppm for NOX)
                                  44

-------
     The study indicated that the specifications and test procedures
                                     j\
were applicable and valid, except that the accuracy specifications


may need to be modified if maxiirum possible utilization is to be

made of available monitoring systens.  The rationale for «uch a con-


sideration is that this test s*as specifically for a selected pollutant-


source combination involving a representative cross section of measure-

ment approaches.  This study was the first real assessment of the


applicability of the proposed specifications and test procedures.


Nonethelessj  sach a consideration would be predicated on the purpose


for monitoring, the required accuracy, and the intended utilization

of the data.

     The evaluation of the specifications and test procedures for


opacity indicated that they were applicable and valid.  Most of the

effort on the opacity evaluation was directed at the zero and calibra-


tion drift parameters as a measure of the system's stability.  As

long as the system was in proper functioning condition, it maintained

zero and calibration staDility that was significantly ipore than

adequate   Zero and calibration drift were less than 0.5 percent and

0.3 percent opacity per week, respectively.  The instrument was un-


attended and checked out on a weekly basis during the 90-day period

At each check the following actions were taken;  the alignment of


the system was verified; all optical surfaces exposed to the emissions


were cleane_d; and the filters of the air purge system (to keep the

optics clean) ^ere checked and cleaned as required.

     The zero drift and calibration drift were not accumulative.  No

optical alignment drift was noted over the 3-month period.  Cleaning


                                   45

-------
of the optical surfaces showed no measurable effect on the zero and




calibration check or on the"opacity measurement.




     During the test, two malfunctiors occurred, which resulted in




excessive instrument drift.  The first malfunction took place when




a plant electrostatic particulate collector developed a leak and




enveloped the transnissoireter in a cloud of participates clogging




the filters of the air purge system used for keeping the optical sur-




faces clean.  The clogged filters reduced the flow of purge air and




permitted the effluent to contaminate the transmissomcter optical




surfaces.  This resulted in an increased drift  of 5 and 3 percent



opacity per week in  the 0 and span, respectively. After the air




filters were  cleaned,  the transnissometer was observed to continue




to drift because of  soiling of optical  surfaces for the next 2 weeks.




The 2 weeks of excessive drift following the air filter clogging




were  apparently due  to contamination  of the air tube  following the




filters.  Normal drift performance returned after the tubes'were




cleaned.           -                              ;'"  '*"'**




     The second malfunction'of the instrument occurrecl!/'during the




last  3 weeks  of operation and appeared  as an erratic  instrument 0




and span check which resulted in an apparent zero and span drift




of 2  5 and  1.0 percent opacit> drift  per week,  respectively.  Ex-




amination of  the instrument at the time of removal from the' stack




showed the  erratic behavior of  the 0  and span check to  be due  to  a




loose  calibration reflector.  It ^ould  be desirable to  treat the




reflector retaining  screws  and any critical hardware  subject to




displacement  by vibration by  some manner to keep them locked in




place.



                                  46

-------
     A study similar to tne S02 work was condacted on NOX monitoring



systems at the Duke Power Plant in Charlotte.  The same stack source




and experimental facilities were used.  The monitoring systems were




operated for a period of about 3 months, September to Novenber 1973




The monitoring systems included across-the-stack UV absorption, extrac-




tive NDIR and cheipi luminescent techniques with permeation and water




condensation interfaces, and extractive visible absorption




techniques   Reference Method No  7 has xun concurrently and at




various tines during the 3-month period for a total of about 50



measurements.  A range of particulate loading was encountered as in




the case of the ,502 study.




     The objectives of this study were to compare the results of




various analytical and sampling approaches to the measurement of NOX




and to evaluate and validate the applicability of the proposed specifi-




cations and test procedures for monitoring ,NOX emissions from a coal-




burning power plant.




     Table D-l summarizes some of the results on the kQx monitoring




system.  The range of values for the test parameters is of the same




order of iragnitude for the NOX monitors as those for the S02 monitors.




The accuracy values for the NOX monitors are based on a test procedure




using sets of two Method 7 measurements.  The test procedure requiring




sets of three measurements should improve the accuracy values.
                                   47

-------
Page Intentionally Blank

-------
                          APPENDIX  E.
                            EXAMPLE:
    PERFORMANCE  SPECIFICATIONS AND  SPECIFICATION
     TEST PROCEDURES FOR MONITORS  OF  POLLUTANT
       GAS  EMISSIONS  FROM  STATIONARY  SOURCES

     Performance specifications  for continuous measurement  syste-ns
for pollutant gases  are given in terms of critical operating para-
meters.  Test procedures are given to test the capability of the
measurement systems  to conform to the performance specifications,
     1.  Principle and Applicability
        1.1  Principle - Pollutant gases are sampled continuously
             in the stack emissions,and the gas concentration is
             analyzed continuously as a function of time.  Sam-
             pling  may include  either the extractive or non-ex-
             tractive (in-situ) approacb.
        1.2  Applicability -  The performance specifications are
             given  for continuous pollutant gas measurement systems
             applied to specific source-pollutant combinations.  The
             following discussion is addressed to 862 and  NOX emissions
             from coal-burning  power plants   Instrument system should
             be capable of operation within performance specifications
             at particulate loadings and in a temperature  range corre-
             sponding to those  of the environment of the installation.
     2   Apparatus
        2,1  Calibration Gas  Mixture - Mixture of a known  concentra-
             tion of the pollutant gas in oxygen-free nitrogen,
                               49

-------
         Nominal con cent rat ions of 30 percent, 60 percent, and


         90 percent of span are recommended.  It is strongly


         recommended that the gas mixture be analyzed by a


         reference method prior to use,


    2.2  Zero Gas - A gas containing no irare than 1 ppm of


         the pollutant gas


    2.3  Equipnent for measurement of pollutant gas concentration


         using the reference method.


    2,4  Strip Chart Recorder - Analog strip chart recorder, in-


         put voltage range compatible with analyzer system out-


         put, full scale (per travel) in two seconds or less.


    2.5  Continuous measurement systen for pollutant gas.


3,  Definitions


    3=1  Measurement System - The total equipment required for


         the determination of a pollutant gas concentration in


         a given source effluent,,  The system consists of three


         major subsystems:


         1.  Sampling Interface - That portion of the measure-


             ment system that performs one or more of the fol-
                                                   ." * * f_ •«*

             lowing operations:  delineation, acquisition, trans-


             portation, and conditioning of a sample of the source


             effluent ox protection of the analyzer from the


             hostile aspects of the sample or source environment.


         2.  Analyzer - That portion of the system which senses


             the pollutant gas and generates a signal output that


             is a function of the pollutant concentration.
                              SO

-------
     3.  Data Presentation - That portion of the measure-


         ment system that provides a display of the output


         signal in terras of concentration units.


3,2  Span - The value of pollutant concentration at which the


     measurement system is set to produce the maximum data


     display output   For the purposes of this method, the


     span shall be set at a pollutant gas concentration of


     1,5 times the emission standard or the pollutant gas


     concentration of interest.


3,3  Accuracy (Relative) - The degree of correctness with


     which the measurement system yields the value of gas

         *    •                                    •    .  ^

     concentration of a sample relative to the value given


     by a defined reference method.  This accuracy is ex-


     pressed in terms of error which is the difference be-


     tween the paired concentration measurements.  The error


     is expressed as a percentage of the reference mean


     value.

    , 1; •  0 H i,       •        ,          '.
3,4  Calibration Erroi - The difference between the pollutant

   vrf:  l                  '
     concentration indicated by the measurement system and


     the known concentration of the test gas mixture.


3,5  Zero Drift - The change in measurement system output


     over a stated period of time of normal continuous


    " operation-when the pollutant concentration at the time


     of the measurements is zero.


3=6  Calibration Drift - The change in measurement system


    "output over a stated period of time of normal contin-


     uous operation when the pollutant concentration at


                          51

-------
         the time of the measurements is the same known up-




         scale value.




    3,7  Repeatability - A measure of the measurement system's




         ability to give the same output reading(s) upon re-




         peated measurements of the sane pollutant concentra-




         tion^) .




    3.8  Response Tine - The time interval from a step change




         in pollutant concentration at the input to the measure-




         ment system to the time at which 95 percent of the




         corresponding final value is reached as display on




         the measurenent system data presentation device,




    3.9  Operational Period - A mini»»um peridd of time over




         which a measurement system is expected to operate




         within certain performance specifications without




         unscheduled maintenance, repair or adjustment.




4.  Measurement System Performance Specifications




        The following performance specifications shall be met




    in order that a measurement system shall be considered




    acceptable under this method.






Parameter                                 Specification



a.  Accuracy (relative)           <_ 20% of mean reference value




b.  Calibration Error             <_ 5% of each test gas value*




c.  Zero Drift  [2 hour)           <_ 2% of emission standard*




d.  Zero Drift  (24 hour)          < 4% of emission standard*
*Absolute irean value * percent confidence interval.
                                52

-------
 Parameter                                 Specification
  e.   Calibration Drift (2 hour)     f.2% of eiission standard*
  f,   Calibration Drift (24 hour)     f.5% of emission standard"1
  g   Response Time                     10 minutes (maximum)
  a.   Operational Period               168 hours
 *Absolute mean value  *  95%  confidence  interval.'

 5.  Specification Test  Procedures
    5 1  Calibration  Error  and Repeatability Test  - Set up and
         calibrate the  complete Tieasurerrent system according to
         the manufacturer's written instructions.  Record the
         readings of  calibration gas concentrations of
         approximately  30, 60, and 90 percent of span.  Make a
         series of five nonconsecutive readings at each con-
         centration (Example:  30 percent, 90 percent, 60 percent
         90 percent,  30 percent, 90 percent, 60 percent, etc.)
         Convert the measurement system output readings to ppm,
/
    5.2  Field test for accuracy (relative), zero drift, eal-
         ibfation drift, and operation period0
         5,2.1  Set up and operate the measurenent system in
                accordance with the manufacturer's written in-
                structions arid drawings.  Operate the  system
                for an  initial 168 hour conditioning period.
                During this period the system should measure
                the pollutant gas content of tae effluent in a
                normal  operational manner,
         5.2.2  After completion of this conditioning  period, the formal
                168 hour performance ar.d operation test period shall
                begin.  The system shall be monitoring the source efflu-
                ent at  all times v.hen not being zeroed, calibrated, or
                                S3

-------
baekpurged.  During this 168 hour test period, make a




minimum of nine (9) pollutant gas concentration measure-




ments using the reference method at intervals of not less




than 1 hour.  For S02, each of the nine tests shall con-




sist of a Method 6 S02 concentration measurement, plus




the instrumental measurement.  For NOX, each of the nine




tests shall consist of a set of three Method 7 MQX con-




centration measurements, plus the instrumental measure-




ment.  In each set of three Method 7 measurements, three




samples should be taken concurrently or within a 3-minute




interval,  The sampling location for the reference method



shall be as prescribed.  The sampling location for the




monitoring method shall  be as close to the location of




the reference method sampling point as conditions will




pemit to demonstrate the specified accuracy.  The re-




ference method data shall be conpared with simultaneously




collected monitoring data for calculation of ,accur%cy.




Before and after each reference method test, record the




values given by both zero and calibration concentrations.




Record the values given by zero and calibration concentra-




tions at two hour intervals until 15 sets of data are ob-




tained.  This two-hour period need not be consecutive but




may not overlap.  Zero and calibration corrections and




adjustments are allowed only at 24 hour intervals or at




such shorter intervals as the manufacturer's,written




instructions specify.  Automatic corrections made by




the measurement system without operator intervention or





                54

-------
                initiation are allowable at any time.  During the entire


                test period, record the values given by zero and cali-


                bration pollutant gas concentrations before and after


          -      adjustment at 24 hour intervals.  Calibration checks


                are wade with one test gas concentration between 70


                and 90 percent of span.


5.3  Field Test for Reponse Time


     5.3.1  This test shall be accomplished using the entire


            measurement system as installed including sanple


            transport lines if used.  Flow rates, line diameters,


            pumping rates, pressures, etc., shall be at the'nominal


            values for normal operation as specified in the manu-


            facturer's written instructions.  In the case of cyclic


            analyzers, the response time test shall include one


            cycle.  This test shall be repeated for each sampling


            point of multi-sampling point systems.


     5.3,2 "Introduce a zero concentration of pollutant gas into the


          - '''measurement system sampling interface or as close to the


            sampling interface as possible.  When the system output


            reading has stabilized, switch quickly to a known concen-


            tration pollutant gas at 70 to 90 percent of span.   Re-


            cord the tiire from concentration switching to final stable
                i

            response.  After the system response has stabilized at  the


            upper lex'el, switch quickly to a zero concentration of  pol-


            lutant gas.  Record the time from concentration switching to
                               55

-------
                    final stable response; perform  this  test  sequence  three  (3)

                    tines   The strip chart recorder  charts or  copies  of  then

                    from this test should be  included in the  data submission

                    with time scales and up and  down  scale values clearly

                    marked.                               >;
                                                          II
6.  Calculation, Data Analysis and Reporting

    6.1  Procedure for Determination of Mean  Values and  Confidence  Intervals

         6,1.1  The TiC'ii value of a data set  is  calculated according to

                equation  E-l.
                           n
                    x = ~ f1  x
                        n f->   1                                   Equation E-l


                where  x. =  i-ndividual "alues               I =  ^um of  tbe  indi-
                      __                                           vidud.1  values
                      x  = mean x'aluc

                      n  = number of data points

         6.1.2  The 95% confidence interval  (two-sided),  i-s 'calr jJated according

                to equdtior  E-2,                       '  .^Jn-'i
                      C.I    =   l-n/2  \ n[Ix  )   -   IJx  )"       Equation  E-2
                                                 n(n-l)

                where Ex. = SUTH of all data points

                    /p~ = square root of the number of data points

                     t,    ,„ = t „_,_ for n samples f^om  a  table  of  percentages
                      l-«/2     975          *                     t-        s,

                                of the t distribution.

                C.I.-r = 95% confidence interval estimate of  the average  mean

                         value

                                       56                   -'"' '

-------
                    Typical Values for  1 - a/2
n
2
3
4
S
6
t.975
12.706
4.303..
3,182
2.776
2.S71
n
7
, 8
9
10
11
t.97S
2.44?
2.365
2.306
2.262
2.228
n
12
13
14
15
16
t.975
2.201
2.179
2,160
2.145
2.131
The values in this table are already corrected for n-1 degrees of freedom

Use n equal to the number of samples as data points,

6,2  Data Analysis and Reporting

     6.2.1  Accuracy (relative) - This calculation uses the reference method

            test data and the measurenent system concentrations recorded

            at the times the reference method tests were run.  Subtract

            the reference method test concentration from the measurement

            system concentration.*  Repeat for all nine test pairs.   Using

            this data,  compute the mean difference and the 95% confidence

            interval using Equations E-l and E-2.  Report the sum of the

            absolute mean value and the 95% confidence interval as a per-

            centage of the mean reference value.

     6.2.2  Calibration Error - Using the data from Section S.I suDtract

            the known value from the value shown by the measurement

            system for each of the 5 readings at each span test concen-

            tration.  Calculate the mean of these differences values

            and the 95% confidence interval according to Equations E-l

            and E-2,  Report the sum of tne absolute mean value and  the 95%

            confidence  interval as a percentage of test gas concentratioa.
*For S02,  subtract the Method 6 values  from the  corresponding  instrumental
 value.   For NOX)  subtract  the mean value  of the set  of  three  Method  7
 values  from the  corresponding instrumental values
                                  57

-------
623  Zero Drift  (2 Hour) - Using the zero concentration values

       measured during the, field test, calculate the mean value and
       the confidence interval using Equations E-l"and E-2,  Report

       the sum of tne absolute mean valje and the confidence inter-

       val as a percentage of the emission standard'M

6.2.4  Zero Drift  - (24  Hour)  - Using the zero concentration values
       treasured every 24 hour's during the field test, calculate the

       differences between the zero point after zero adjustment and the

       zero value 24 hours later just prior to zero adjustment.  Calibrate
       the mean value of these points and the confidence interval

       using Equations E-l and E-2.  Report the sum of the absolute

       mean value and confidence interval as a percentage of the

       emission standard.

6 2.5  Calibration Drift  (2 Hour) - Using the calibration values

       obtained at two-hour intervals daring the field test, cal-

       culate the differences between the readings and the test
                                               _\ ,,   ijr  *
       gas value   These values should be corrected for the cor-

       responding zero drift during that two-nour period.  Calcu-
                                                     * Ft
       late the mean and confidence interval of these corrected

       values using Equations E-l and E-2.  Report' the SUDI of the

       absolute mean value and confidence interval as a percentage

       of the emission standard.

6.2.6  Calibration Drift  (2-1 Hour) - Using the calibration values

       measured every 24 hours during the field test, calculate
       tTe differences between the calibration concentration reading

       after zero and calibration adjustment and the calibration

                                58

-------
       concentration reading 24  hours  later  after  zero  adjustment but
       before  calibration adjustment.   Calculate the  mean  value  of
       these differences and the confidence  interval  using Equations
       E-l and E-2.   Report sum  of the absolute mean  value and confi-
       dence interval as a percentage  of the emission standard.
6.2.7  Response Time - Using the charts fron Section  5.3 calculate
       the time interval from concentration  switching to 95%  of the
       final stable  value for all" up scale and down scale  tests.
       Report the mean of the three up scale test  times and the mean
       of the three  down scale test times.  The two average times
       should not differ by more than  15% of the  slower time.   Report
       the slower time as the system response tire.

6.2.8  Operational Period - During the 168 hour performance and
       operational test period,  the measurement system shall  not
       require any corrective maintenance or repair or replacement
       or adjustment other than  that clearly specified as  required
       in the operation and naintenance manuals as routine and
            -ior  .,
       expected during a one-weel period.  If the  measurement
           *JO '
       system operates within the specified performance parameters
       and does not  require corrective maintenance, repair, replace-
       ment or adjustment other than specified above, during  the 168
       hour test period, the operational period test  will  be  success-
       fully concluded.  Failure of the measurement to meet this
       requirement shall call for repetition of the 168 hour  test
       period.  Portions of the  te«t which were satisfactorily
       completed need not be repeated.  Failure to meet any perfor-
       mance specifications shall call for a repetition of the one

                                59

-------
                week performance test period and that portion of the testing

                which is relat.d to th'e failed STCCif ication.  All maintenance

                and adjustments requires shal- be recorded   Outcut readings

                shall be recorded before and after all adjustments.

7   Surrojeiripntal Refetences
    Experimental Statistics, National Bureau of Standard's, ".Handbook 91,
      1963, P. 3-31, Paragraph 3-3 1.4.
                                     60

-------
                           APPENDIX  F..
                            EXAMPLE:*
      PERFORMANCE  SPECIFICATIONS  AND  SPECIFICATION
     TEST PROCEDURES FOR  TRANSMiSSOMETER SYSTEMS
                FOR CONTINUOUS  MEASUREMENT
            OF ,THE  OPACITY  OF STACK EFFLUENTS
     Specifications for continuous  measurement of visible emissions by
transmissoraetry are given in terrrs  of critical design, perfornanee , ard
installation parameters.   Test procedures are given to test the capability
of the systems  to conform to the performance specifications.
1.   Principle  and Applicability
     1.1  Principle,  Transmissoimetry is a direct measurement of attenuation
         of visible radiation (opacity) by participates in a stack effluent
         Light from a lawp is projected across the stack of a pollution
         source to a light sensor.  The light is attenuated due to absorption
         and scatter by  the particulates in tne effluent.  The percentage
         of light attenuated is defined as the opacity of the emission   <\
         clean stack that does not attenuate light will have a transmittance
         of 100 or an opacit> of 0.   An opaque stack that attenuates  all of
         the light will  have a transmittance of 0 or an opacity of 100 per-
         cent
     1.2  Applicability   The performance specifications are given for trans-
         missometer systems for continuous measurement of the opacity of
         aerosol within  the effluent  (in-situ) of a pollution source.  The
         ipethod is applicable to the  measurement of the opacity of the

*For coal-burning  power plant.
                                 61

-------
          effluent at points in .the plune at the stack exit re-note from



          the measurement point provided the characteristics of the aerosols




          in the effluent are maintained between the measurement point and



          the remote point of interest in the plume.  Watcr^.in condensed




          form is an interference to the measurenent




2.   Apparatus                          ,                 ., i




     2.1  Neutral Density Filters.   Filters with neutral spectral characteris-




          tics  ard known optical densities to visible light.  Filters 5- or




          6-inch square or 6-ineh diameter with nominal optical densities of




          0.1, 0.2,  and 0.3 (20, 37, and 50 percent opacity) are required.




          Although calibrated filters with accuracies reported to within




          3 percent are available,  it is recommended that filter calibrations




          be checked with a well-collimated photopic transiussometer of



          known linearity prior to use.




     2.2  Strip Chart Recorder.  Analog str.ip chart recorder \«ith input



          voltage range and performance characteristics compatible .with




          the measurement system output             ,  r




     2.3  Opacity Measurement System   An in-stack transnussometer (folded or




          single path) with the optical design specifications^designated



          below, associated control units and apparatus to keep optical




          surfaces clean




3.   Definitions




     3.1  Measurement System.  The total equipment required for the continuous




          determination of a pollutant concentration in a source effluent.




          The system consists of three major subsystems
                                     62

-------
 3.1.1   Sanpling Interface - That portion of t.Te measurement system




         that performs ore or more of the.following operations:




         delineation, acquisition, transportation, and conditioning of



         a sample of the source effluent, or protection of the analyzer




         from the effluent




 3,1.2   Analyzer - That portion of the system wnich senses the  pollu-




         tant gas and generates a signal output that is a function of




         the pollutant concentration.




 3.1.3   Signal Processor - That portion of the measurement system that




         processes the analyzer output and provides a display of the




         output' signal in terms of pollutant concentration.




3,2  Span.  The value of pollutant concentration at which the measure-




     ment is set to produce the maximum data display output.   For the




     purpose of this method, the span  shall be set at an opacity of




     50 percent for an effluent depth  equal to the stack exit diameter




     of the source.




3.3  Calibration Error.   The difference between the pollutant concen-




     tration indicated by the measurement systen and the known values




     of a seri'es of test standards.   For this method the test standards




     are a series of calibrated neutral density filters.



3.4  Zero Drift.  The change in measurement system output over a




     stated period of time of normal  continuous operation when the




     pollutant concentration at the  tine c£ the measurements  is  zero,




3.5  Calibration Drift.   The change  in measurement system output over




     a stated period of tine of normal continuous  operation when the




     pollutant concentration at the  time of the -neasurements  is  the




     same known up scale value.




                                  63

-------
 3.6  Repeatability,  A measure of the measurement system's ability

      to give the same output reading(s) upon repeated measurements
                                                                      *
      of the same pollutant concentration(s).

 3.7  Response Tiiie.  The time interval from a step change in pollutant

      concentration in the stack of the source at the input to the

      measurement system to the time -at which 95 percent j>f the cor-

      responding final value is reached as displayed on the measurement

      system data presentation device.

 3.8  Operational Period,   A minimum period of time over which a

      measurement system is expected .to operate within certain per-

      formance specifications without unscheduled maintenance, repair,

      or adjustment.

 3.9  Transmittance.  The fraction of incident light that is trans-

      mitted through an optical medium of interest,

3.10  Opacity.  The fraction of incident light that is attenuated

      by ah optical medium of interest.  Opacity (0) and transniittance

      (T) are related as follows:   .             $-1

                                   0 = 1 - T            •:

3.11  Optical Density.  A logarithmic measure of the amount of light

      that is attenuated by an optical medium of interest.  Optical

      density (D) is related to the transntittance and opacity as

      follows:


                                   D = -log10T

                                   D = -Iog1() (1-0)             '


3.12  Spectral Response.  The relative response of a transmissometer

      to radiation of different wavelengths.

3.13  Angle of View,  The maximum (total) angle of radiation that is


                                 64

-------
          seen by tae photo-detector assemoly of an optical transmissomster.


3 14  Angle of Projection.   The maximum-(total) -angle of radiation that is


      projected by the lamp assembly of an optical  transmissometer


3,15  Pathl'engthV "The depth of effluent in the light beam b'etween the


      receiver ancf'th'e transmitter of the single pass transpissoneter, or the


      depth-of efflnen'C between the transceiver' and reflector of a double pass


      trans'nussometer.'


 4,    Installation Specifications


      4.1  Location.   The transrcissometer must be located across  a section of


          duct or stack that will provide a participate  flow through  the


          optical volume of the transroissameter that is  representative of the


          particulate flow  through tne duct or stack.


          4.1.1   The transmssometer location shall be downstream from all  par-


                  ticulate control equipment,


          4'. 1.2   The transmissometer location shall be located as far from


      •  •          bends and obstructions  as practical.


          4.1.3   The transmissometer that is  located in the duct or stack fol-


                  lowing a bend shall be  installed  in the plane of the bend  where


                 Jnpossible


          4,1.4   The transmissometer should be installed in the most accessible


                  location possible.


          4.1.5   The transmissometer location in  a section of  duct  or stack


                  where a  representative  particulate concentration is in doubt


                  shall  provide an average measure  of the opacity of the effluent


                  in the duct  or stack.   The location shall  be  determined bv ex-
           '             .*•               i

                  amining  the  opacity profile  of the effluent at  a series of


                  positions  across  the duct or stack while  the  plant  is  operating
               'i' -                             ^


                                      65

-------
     at full load.




4,2  Slotted Tube,  Installations that require the use of a slotted




     tube shall use a slotted tube of sufficient size and blackness




     so as not to interfere v»ith the free flov» of effluent through




     the entire optical volume of the transmissometer or reflect




     light into the light detector of the transmissometer.




4.3.  Pathler.gth.  It is recommended that the pathlength or depth




     of effluent for the transiissonieter include the entire width




   •  or diameter of the duct or stack.  Installations using a shorter




     pathlength must -use | extra caution in determining the measure-




     ment location representative of the particulate flow through



     the duct or stack.    ;




4.4  Recorder Output.  The transmissoweter output shall permit ex-




     panded display of the m-stack opacity on a standard 0 to 100




     percent scale.  In addition, a graph shall be provided with




     the installation to show the relation between the standard




     0 to 100 percent readout scale and the opacity of the effluent




     for a pathlength equal to the stack exit diameter^ ,




     The relation fox constructing the graph is:





                 •log(l  - Oi)  =  Ui/lj)  iogC1  -  '°2)  '
     where 1. is the stack exit diameter and 0, is the opacity of




     the effluent at the stack exit diameter, 1  is the in-stack




     pathlength of the transmissometer, and 0_ is the measured




     opacity of the effluent within the transmissometer path.




     The opacity standard is based on the opacity of the effluent




                               66

-------
         at the stack exit diameter,




5.  Optical Design Specifications




    The following optical design specifications shall be met in order that




    measurement "system shall be considered acceptable under this method'








PARAMETER      "-'••<'•.              -             SPECIFICATION




a.  Spectral response                   Peak and mean response within




                                        500 to 600 rnn; less that 10%




           '                             of peak response outside 400 to




                                        700 nm



b.  Angle of vie"w                       5 degrees maximum  (total angle)




c.  Angle of projection                 5 degrees maximum  (total angle)




6.  Design Specification Data and Test Procedures




    6.1  Spectra: Response.  Obtain spectral data for detector, lamp,




         and filter components used in the -transmissometer from




         their respective nanufacturers.




   ' 6.2  "Angle of View,  Set the receiver up as specified by the man-




         ufacturer ';' Draw an arc with radius of 3 meters   Measure the




         receiver response to a small (less than 3 centimeters) non-




         directional light source at 5-centimeter intervals on the




         arc for 26 centimeters on either side of the detector center-




         line.  Repeat the test in the vertical direction.




    6,3  Angle of Projection,.  Set the projector up as specified by




         the ".anuf actarer,  Draw an arc with radius of 3 meters. Using
                                    67

-------
         a small photoelectric light detector (less than 3 centimeters),

         measure the light intensity at 5'-centimeter intervals on the

         arc for 26 centimeters or. either side of the light source

         centerline of projection.  Repeat the test in the vertical

         direction.

7.  Measurement System Performance Specifications

    The following performance specifications shall be met in order

    that a measurement system  shall be  considered acceptable under  this

    method.
PARAMETER


a.  Calibration Error

b.  Zero Drift (24 hr)

c.  Calibration Drift
      (24 hr)

d.  Response Time

e,  Operational Period
      SPECIFICATIONS


^10% of test filter value*

.110% of emission standard*

±10% of emission standard*


10 seconds (maximum)

168 hr
'Absolute mean value + 35 percent confidence interval.

8   Performance Specification Test Procedures

    8.1  Laboratory Test for Calibration Error, Repeatability, and
           Response Tiire-

         8.1.1  Set up and calibrate the ineasurejnent system as specified by the

                manufacturer's written instructions using a 3-meter pathlength.

                Span the instrument for 0 to 50 percent opacity.   Insert a
                                      68

-------
            range of neutral density filter standards in the transTnissometer




            path at the 2,5-meter point.  A minimum of three neutral density




            filters with nominal opacities of 20, 37, and 50 percent cal-



            ibrated within 3 percent must be used   Make a total of five




            nan-consecutive readings for each filter.  Record the




            measurement system output readings in percent opacity.



     8,1.2  Insert the 30 percent filter in the transmissometer path five




            times and record the time required for the system to respond




            to 95 percent of final zero and span values,




8.2  Field Test for Zero Drift, Calibration Drift, and Operational Period



     8.2.1  Set up and operate the measurement system in  accordance with




            the manufacturer's written instructions and drawings.  Operate




            the system for an initial 168 hour conditioning period.  During




            this period the system should measure the opacity of the



            effluent in a normal operational manner.




     8.2,2  After completion of this conditioning period, the formal  168-



            hour operational test period shall begin.   The system shall be



            ir.onitorirg the source effluent at all tiroes when not being




            zeroed or calibrated.   At 24-hour intervals over the specified



            operational period, the zero and span of  the  measurement  system




            shall  be checked according to the manufacturer's written  in-




            structions.  The span value shall not be  less than 1.5 tines




            the emission standard.   After the zero and span check,  clean all



            optical surfaces open to the effluent,  re-align optics  and make




            any necessary adjustments to the calibration  of the  system.  Zero




            and calibration corrections and  adjustments are allowed onl> at




            24-hour intervals  or at such shorter  intervals  as manufacturer's




            written instructions specify.  Automatic  corrections made  bv the




                                     69

-------
               measurement system without operator intervention are

               allowable at any time.  'During the entire test period,  ,

               record the values of zero and span opacities before

               and after cleaning, aligning, and adjusting the system

               at 24-hour intervals.

9,  Calculation, Data Analysis, and Reporting

    9.1  Procedure for Determination of Mean Values and Confidence In-

         teivals,

         9.1.1  The riean value of data set is calculated according

                to equation F-l.


E                                                                   Equation  F-l
                             1
                           i-l
                where x. = individual values      I = sum of the individual
                  •  " i  -•,'.',  ••                         •  values

                x = mean .value
                n = number of data points

         9,1.2  The 95 percent confidence interval (two-sided) is cal-

                culated according to equation F-2.
                              /TT        V           .n fn-1)     Equation F-2

             •  -where  IX  = SUP of all data points

                       /"n = square root of the number of data points

                        tj  ,. = t „„ for n samples from a table of

                        percentages of the t distribution
                                    70

-------
TABLE OF TYPICAL-VALUES FOR t,
                             1  - «/2
n
2
3
4
5
6
t.975
12.700
4.303
3.182
2.776
2.571
n
7
8
9
10
11
'.975
2.447
2.365
2.306
2.262
2.228
n
12
13
14
15
16
t.975
2.201
2.179
2.160
2.145
2.131
        The values in this table are already corrected for n-1  de-


        grees of freedom.   Use nequal tothe number of samples as


        data points.


   9.2  Data Analysis and  Reporting


      "' 9,2.1  Spectral  Response.   Combine the spectral data obtained


               in accordance with  Section 6.1 to determine the  effective


               spectral  response of the transmissometer.   Report effective


               wavelength  respond curve and mean response of the trans-


               missometer,


        9.2.2  Angle  of  View   Using the data obtained in accordance  with


               Section 6.2,  calculate the response of the receiver  as a


               function  of viewing angle in the horizontal and  vertical


               directions  (26 centimeters of arc with a radius  of 3 meters


               equal  5 degrees),  Report relative angle of view curves.


        9,2.3  Angle  of  Projection.   Using the data obtained in accordance


               with Section 6.5, calculate the response of the  photoelectric




                                    71

-------
       detector as a function of projection angle in the horizontal




       and vertical directions.  Report relative angle of projection




       curves.




9.2,4  Calibration Error.  Using the data from Section 8.1 subtract




       the known filter opacity value from the value shown by the -




       neasureraent system for each of the 15 readings.  Calculate the




       mean of the five difference values at each test filter value




       and the 95 percent confidence interval according to Equations




       F-l and F-2.  Report the sum of the absolute mean value and




       the 95 percent confidence interval as a percentage of the




       test filter value,




9.2.5  Zero Drift.  Using the zero concentration values measured




       every 24 hours during the field test (Section 8.2),cal-




       culate the differences between the zero point after




       cleaning, aligning, and adjustment, and the zero value




       24 hours later just prior to cleaning, aligning and




       adjustment.  Calculate the near value of these points and




       the confidence interval using Equations c-l and T-2,   Pe-




       port the sum of the absolute mean value and the 95 percent




       confidence interval as a percentage of the omission standard,




9,2,6  Calibration Drift.  Using the span ^alue measured every




       24. hours during the field test, calculate the differences




       between the span value after cleaning, aligning, and




       adjustment of 0 and spaa, and the span value 24 hours




       later just after cleaning, aligning, and adjustment of 0





                             72

-------
       and before adjustment of span.  Calculate the mean




       value of these points and the confidence interval




       using Equations F-l and F«2«  Report the sum of the




       absolute mean value and the confidence interval as a




       percentage of the emission standard.





9,2.7  Response Time.  Using the data from Section 8.1 calculate




       the time interval from filter insertion to 95 percent of




       the final stable value for all up scale and down scale




       traverses.   Report the mean of the 10 up scale and dovn




       scale test  times.




9.2,8  Operational Period.   During the 168 hour performance




       and operational test period, the measurement system shall




       not require any corrective maintenance or repair or re-




       placement or adjustment  other than that clearly specified




       as required in the operation and maintenance manuals as




       routine and expected during a one-week period   If the




       measurement system operated within the specified per-




       formance parameters  and  did not require corrective




       maintenance, repair, replacenent,  or adjustment other




       than specified above,  during the 168 hour test period,




       the operational period test will be successfully con-




       cluded.   Failure of  the  tneasurenent to meet  this re-




       quiretient shall call for a repetition of the 168 hour




       test period.   Portions of tne test w.iich were  satisfactorily




       completed need not be  repeated.   Failure to  meet any per-




       fomance specifications  shall call for a repetition of





                              73

-------
the one-week performance test period and that portion of



the testing which is related to the failed specification.



All maintenance and adjustments required shall be re-



corded.  Output readings, shall be recorded before and



after all adjustments.
                    74

-------
                                    TECHNICAL REPORT DATA
                            (flense read fas&itcnons on the reverse before cc
 ]  BEPORTNO
   EPA-650/2-74-013
 4 TITLE ANO SUBTITLE
   Performance Specifications for Stationary-Source
  Monitoring Systems  for Gases and Visible Emissions
                                  6 PERFORMING ORGANIZATION CODE
                                  5 REPORT DATI
                                    January 1974
 7 AUTHORiS)
  John S. Nader
  Fredric Jaye
                                                            8- PERFORMING ORGANIZATION REPORT NO
William Conner
 9 FERFQFMINGORG MMIZAT'ON NAME AND ADDRESS
  Chenistry and Physics  Laboratory
  National Environmental Research Center
  Environmental Protection  Agency
  Research Triangle Park, North Carolina 27711
                                  10 PROGRAM ELEMENT NO

                                     1AA010
                                  11  CONTRACT/GRANT NO
 12 SPONSORING AGENCY MAME ANO ADDRESS
                                                            13 TYPE OF REPOHT AND PERIOD COVERED
                                                            14 SPONSORING AGENCY CODE
 IS SUPPLEMENTARY NOTES
 16 ABSTRACT
   The purpose of this report is to provide a technical  basis for the selection of
   stationary-source monitors that are required by Federal,  State, or local regulations
   for emissions.  The document identifies performance parameters, gives specifications
   and details test procedures to verify the specifications.  Examples of the specifica-
   tions and test procedures  are provided for monitoring systems applied to gases  and
   visible emissions.  Technical dataA utilised for the specifications are based on the
   results of laboratory and  field studies.--
                                KEV WORD* AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b IDENTIFIERS/OPEN ENDED TERMS
                                                                         c  COSATI Field/Group
   Emission measurenent
   Stationary source monitoring
   Performance specifications
   Gas  emissions monitoring
   Visible emissions monitoring
 8 OfS~RiBUTJON STATEMENT

   Release unlinited
                     19 SECURITY CLASS (This Report;
                       Unclassified
                                              20 SECURITY CLASS fThis page/
EPA Form 2220 1 (9 73)

-------