Technical Report
Test Procedure Analysis and Recommendations
     for Supersonic Transport Aircraft
           Emission Measurement
             November,1976
              Prepared by:
            Gary F. Austin
  Standards Development and Support Branch
    Emission Control Technology Division
Office of Mobile Source Air Pollution Control
     Office of Air and Waste Management
    U.S. Environmental Protection Agency

-------
           Technical Report
Test Procedure Analysis and Recommendations
     for Supersonic Transport Aircraft
           Emission Measurement
             November,1976
              Prepared by:
            Gary F. Austin
  Standards Development and Support Branch
    Emission Control Technology Division
Office of Mobile Source Air Pollution Control
     Office of Air and Waste Management
    U.S. Environmental Protection Agency.

-------
                                                      AC 76-04
                           Technical Report
              Test Procedure Analysis and Recommendations
                   for Supersonic Transport Aircraft
                         Emission Measurement
                          November, 1976
                           Prepared by:
                         Gary F. Austin
                                Notice

     Technical support reports for regulatory action do not necessarily
represent the final EPA decision on regulatory issues.   They are in-
tended to present a technical analysis of an issue and recommendations
resulting from the assumptions and constraints of that analysis.  Agency
policy considerations or data received subsequent to the date of release
of this report may alter the recommendations reached.  Readers are
cautioned to seek the latest analysis from EPA before using the in-
formation contained herein.
               Standards Development and Support Branch
                 Emission Control Technology Division
             Office of Mobile Source Air Pollution Control
                  Office of Air and Waste Management
                 U.S. Environmental Protection Agency

-------
Abstract

     Determination of the emission levels from gas turbine engines
operating in the afterburning mode has proven to be an extremely dif-
ficult task.  The promulgated procedure for emission measurement from
non-afterburning engines is not applicable because the exhaust gas
reactions at the sample extraction location (the exhaust nozzle exit)
are incomplete.  ".       r

     The Air Force Aero Propulsion Laboratory at Wright Patterson Air
Force Base, Ohio recently provided a recommended procedure    for
emission measurement from an afterburning engine through a contracted
program with The General Electric Co.  This recommended procedure
provided two methods of constitutent determination, the Far and Near
Plume Methods.  The Far Plume Method determines the emission levels down
stream of the\ exhaust where the exhaust gas reactions are complete.  The
Near Plume Method determines the emissions level at the exhaust nozzle
and through the use of a plume model the actual emission levels are
determined.

     This report provides analysis of two techniques for the measurement
of exhaust emissions from afterburning engines and presents the Air
Force Far Plume Method as the preferred method but in limited area test
environments the Near Plume Method is an acceptable alternative of de-
termining the emission levels from afterburning gas turbine engines.
PrepafeH by
                                                 Approved
                                        Project Manager - Aircraft
               /',
_
    Approved       .
Branch Chief, SDSB

Distribution

D. Alexander
J. DeKany
C. Gray
W. Houtman
G. Austin
E. Stork
M. Williams
G. Kittredge
                                                 Approved
                                           ision Director, ECTD

-------
                      Table of Contents

                                                            Page

1.0  Introduction                                             1

2.0  Summary and Recommendations                              1

3.0  Background                                               2

4.0  Authority and Requirements                               3

5.0  The Air Force Recommended Procedure                      4

     5.1   The Far Plume Method                               4
     5.2   The Near Plume Method                              5

6.0  System Analysis                                          6'<

7.0  The Near Plume Method Analytical Model                   7

8.0  Procedure Recommendation                                 8

References                                                    9

Appendix

     1.  The Air Force Recommended Procedure

          A.  The Far Plume Method
          B.  The Near Plume Method

-------
1.0  Introduction

     Development of emission measurement techniques to date have been
concerned with non-afterburning aircraft engines.  On August 16, 1976
the EPA promulgated standards for the "Control of Air Pollution from
Aircraft and Aircraft Engines, Supersonic Aircraft", 40 CFR Part 87.
This document addressed the emission standards for supersonic after-
burning aircraft engines but did not specify a measurement procedure.

     The technique necessary for the measurement of emissions from
afterburning engines differs from the non-afterburning engines only in
the sample probe design, the sample extraction location and the required
number of samples.  For non-afterburning turbine engines, temperatures
at the exhaust nozzle are typically less than 1200°F,  At such tempera-
tures, chemical reactions no longer proceed at an appreciable rate and
the measured emission levels correspond to those actually ejected into
the atmosphere.  In the case of afterburning gas turbine engines, exhaust
temperatures can reach 3500°F and chemical reactions can occur for a
considerable distance downstream of the exhaust plane.  To obtain the
true emissions levels for afterburning turbine engines, the plume should
be sampled at either a location far enough removed from the engine that
the gases have been cooled to a temperature where reactions have ceased
or at a location where the actual emission levels can be accurately
predicted through the use of modeling techniques.  The far removed
procedure, however, requires a considerable clear area aft of the engine
and thus would generally necessitate an outdoor test facility.

     Since existing emission measurement procedures are not applicable
for engines operating in the afterburning mode the Air Force undertook a
contracted program to provide the definition, development, and demon-
stration of a emission measurement technique for afterburning engines.
The result of this program is the presentation of a recommended procedure.
This procedure is similar to the Society of Automotive Engines, Aerospace
Recommended Practice 1256, for the measurement of emissions from non-
afterburning engines.

2.0  Summary and Recommendations

A.   In afterburning engines the exhaust plume reaction can continue to
     occur at a considerable distance downstream from the nozzle exit.
     Exhaust gas measurements should be taken at a point far enough down
     stream that the reactions are complete or at a location where the
     actual engine emission levels can be predicted with the use of a
     model.

B.   Pollutant constituents of concern are unburned hydrocarbons, carbon
     monoxide, and the oxides of nitrogen.

-------
                                   -2-
C.   The emission measurement sampling system proposed in the Air Force
     procedure is basically the same as that required by the EPA in 40
     CFR Part 87.

D.   The calibration and instrument checks required by the Air Force
     Procedure are basically the same as for the EPA procedure.

E.   The Far Plume Method is the most accurate of the two methods
     considered although it does require a large open area down stream
     of the engine.

F.   The emission calculation procedure for the Far Plume Method consists
     of data verification, a test for sample representativeness, and
     calculation of the emission indices or the emission flow rates.

G.   The Near Plume Method is an acceptable method of determining the
     emission levels.  This method requires the use of a model to
     predict the actual constituent levels since the exhaust gas reactions
     are incomplete at the sampling point.

H.   The Near Plume Method uses a computer model, "PLUNOD", to predict
     the emission levels.  This model uses known chemical kinetic
     reactions and turbulent gas mixing data in its calculations.

I.   Accuracy of the Near Plume Method is best at the minimum afterburn-
     ing power setting and is acceptable at the maximum afterburning
     power setting.

J.   Both methods indicated that NOx was not generated or consumed at
     any power setting in the afterburning exhaust plume.

     The emission measurement procedure presented by the Air Force in
its study of the "Development of.Emissions Measurement Techniques for
Afterburning Turbine Engines,"    should be adopted in its basic form
for the measurement of exhaust emissions for commercial afterburning
engines.  This procedure will provide a basis emission measurement
technique which with only slight modification will fulfill the require-
ments established by the Environmental Protection Agency for the mea-
surement of emissions from afterburning aircraft engines.

3.0  Background

     The pollutant constitutents of concern emitted from an afterburning
gas turbine engine are the same as for a non-afterbruning engine.  These
are unburned hydrocarbons (HC), carbon monoxide (CO), and the oxides of
nitrogen (NOx) and smoke.  Smoke,  a carbonatious particulate, is considered
an aesthetic pollutant due to its light abscuration effect on the
exhaust plume.

-------
                                -3-


     CO and HC emissions generally are products of inefficient combus-
tion.  At high afterburning power level, where the fuel to air mixture
may be greater than stoichiometric, very high levels of CO may exist.
In these regions, temperatures are too high for HC to persist.  The
general tendency in afterburning engines, thus, is that the CO concen-
trations are highest in regions of highest temperatures, while both CO
and HC may exist in low temperature regions.

     The chemical kinetic reactions for the formation of NO are reason-
ably well understood and the dominant parameter in NO formation is the
maximum flame temperature.  Flame temperatures are much higher in the
main engine burner and more NO would therefore be expected to be formed
in the main burner than the afterburner.  The kinetics of the NO formation
process are such that there is no significant decrease in the total NOx
in the afterburner.  The basic reason for this is that at the lower
temperatures where thermodynamic equilbrium favors a decrease in overall
NO level, the kinetics are too slow to premit any appreciable change.

     Smoke is predominately generated in the primary zone of the main
engine combustor under very rich and high pressure conditions.  Since
the pressure in an afterburner is comparatively low the smoke generated
is of minor significance.  In fact smoke participates can be partially
consumed in the afterburner combustion process.

     The exhaust plume of an afterburning engine is commonly divided
into three distinct regions as shown in Figure 1.  Region one consists
of a turbulent mixing area where the supersonic nozzle core flow pene-
trates into the parallel subsonic flow originating at the exhaust nozzle.
Region two is the transition region where the supersonic core flow ends
and the flow field undergoes transition until at some distance downstream
the fully developed plume is turbulent free (Region 3).

     The rea'ctive nature of the exhaust plume renders it impractical to
make direct measurements of the exhaust gas constiutents as in the non-
afterburning engine (at the exhaust nozzle).  This necessitates measure-
ment of the exhaust gas constitutents by either determination of the
level at the exhaust nozzle and through the use of a plume model pre-
dicting the actual pollutant contribution to the environment or measuring
the constitutent levels downstream of the nozzle where the reactions are
complete.  A disadvantage of the latter method is the substantial
dilution of the constituents which compromise the accuracy of the
instruments due to instrument sensitivity.

4.0  EPA Authority and Requirements

     Section 231 of the Clean Air Act directs the Administrator to
"establish emission standards applicable to emissions of any air pollutant

-------
                                -4-

 from any class or classes of aircraft or aircraft engines which in his
 judgment cause or contribute to or are likely to cause or contribute to
 air pollution which endangers the public health or welfare."

     EPA regulations for the "Control of Air Pollution from Aircraft and
 Aircraft Engines, Supersonic Aircraft" 40 CFR Part 87, were issued on
 August 16, 1976.  This document established exhaust emission regulatory
 levels for supersonic aircraft engines.  To compliment these regulatory
 standards an emission measurement technique must be established.

 5.0  Air Force Recommended Procedure

     The, Air Force Aero Propulsion Laboratory, Wright Patterson Air
 Force Base Ohio, has provided a recommended procedure (Appendix) for the
 continuous sampling and analysis of gaseous emissions from afterburning
 aircraft gas turbine engines.  The intent of this document is to stan-
 dardize the emission test procedures and equipment necessary for measuring
 carbon monoxide, carbon dioxide, nitric oxide, total oxides of nitrogen,
 and total hydrocarbons from afterburning engines.  Due to the reactive
 nature of the exhaust plume from afterburning engines, special procedures
 are necessary to assure that the measured emissions levels correspond to
 those actually emitted into the surrounding atmosphere.  The procedure
 presented is comprised of two distinct parts.  The Far Plume Method,
 Section 5.1, describes a procedure for use when exhaust gas samples are
 taken at axial stations far removed from the engine exhaust plane.  In
 the case where the required test facilities are not available for use of
 the Far Plume procedure, an alternative procedure the Near Plume Method,
 Section 5.2, is given which involves sampling at the nozzle exit plane.
 The use of the Near Plume Method requires calculation of the actual
 emissions levels utilizing a computer program derived from a reactive
 plume analytical model.

 5.1  The Far Plume Method

     Analysis Equipment - NDIR instruments are specified for CO and CO
measurements, a flame ionization detector for HC, and a chemiluminescence
 analyzer with converter for NO and NOx.  This follows the EPA equipment
 specifications (Federal Register, July 17, 1973).  For the NDIR in-
 struments, zero drift and span drift are specified at 0.5 percent for 1
hour, rather than 1 percent for 2 hours required by the EPA regulations.
Repeatability and noise are specified at 0.5 percent.  The sensitivity
 specified is 0.3 ppm and 0.005 percent for CO and CO   respectively.
These latter requirements although more restrictive than the current EPA
regulations are consistent with the lower concentration levels encountered
with the Far Plume procedure.   For the HC and NO analyzers,  the noise,
 zero drift, and span drift are as specified in the EPA procedure since
 the more sensitive ranges available on these instruments result in total
 instrument sensitivity adequate for the Far Plume procedure.

-------
                                   -5-


     Sampling Equipment - The sample extraction probe is designed with
the probe temperatures specified such as to permit either hot water or
steam cooling.  In order to specify sampling locations, both axial and
radial, the sampling location is scaled according to the engine nozzle
diameter.  A minimum of 11 sampling points are required, approximately
equally spaced across a diameter.  This number of samples permits mean-
ingful statistical treatment of the data.  If fewer samples were extracted
the results would have correspondingly reduced measurement accuracy.

     Since very long sample lines would probably be involved with this
procedure, no maximum sample line length is specified.  The requirement
for a maximum gas transport time of 10 seconds effectively limits the
sample line length to a realistic value.

     Equipment Layout - The equipment layout is similar to that specified
in EPA regulations.

     Instrument Routines - Instrument routines are similar to those
specified in EPA Regulations.  Checks in addition to 'those required by
the EPA regulations are specified for zero and span drift, repeatability
and noise level.  Thermal converter efficiency check is as specified in
the EPA procedure, but is required only monthly.

     Test Procedures - Testing using this procedure is limited to wind
velocities of less than 5 mph crosswind and 1 mph tailwind.  The test
procedures specified are otherwise similar to the EPA regulations.

     Calculation of Results - CO and CO- are first corrected to true
sample moisture level, in case a dryer was used.  A linear fit of each
pollutant versus C0_ is then made by the method of least squares.  If
the slope of this linear fit line is greater than 10, then the cor-
responding correlation coefficient is required to be greater than 0.95.
A value for the correlation coefficient of a line with a slope of less
than 10 is not given.  Emission: indices are then calculated from the
slope of the linear data fits.   The emission flow rate (Ib/hr) is then
calculated from the overall engine fuel flow and the emission index.
Emission standards can therefore be specified either in terms of emission
indices or the total flow rate of emissions.

5.2  The Near Plume Method

     Analysis Equipment - The specified equipment is basically the same
as required by Far Plume Method.

     Sampling Equipment - The sampling probe is required to be of the
quenching type, and the pressure ratio across the orifice is required to
be at least 5.  Only general guidelines are given in regard to cooling

-------
                                   -6-


of the probe structure, the responsibility for the probe integrity
therefore rests with the probe designer and operator.  Total pressure
measurements are required and permitted to be separated by no more than
0.2 inch from the sampling point.  This permits use of a combination tip
for simultaneous sampling in which the total pressure and gas sample are
obtained from separate orifices.

     The axial sample plane is required to lie within eight inches of
the plane where the exhaust is completely expanded.  A minimum of 22
sampling points are specified to lie across two diameters.  The points
are specified to be approximately equally spaced, and the outer most
located at the edge of the exhaust stream.  The edge of the exhaust
stream is defined as the point where the total pressure of the outermost
sampling point is between 1.05 and 1.10 times the ambient pressure.

     Calculation of Results - All required calculation of results is
accomplished by the plume model.  Assembly of the plume model input
data, however, requires some effort.  The plume model is set up to
accommodate 11 samples or stream tubes.  A manual prqcedure is specified
which effectively averages the data by plotting the data extracted at
the 22 sample points against radial position and drawing a smooth curve
through the data points.  Eleven values are then selected from the
smooth curve for input to the plume model.  These 11 values are selected
to be in centers of 11 equal areas.

     The plume model computes emission levels at 0, 35, and 50 nozzle
radii downstream of the exhaust nozzle.  From this data it can then be
determined at what axial locations the exhaust gas reactions are complete.
If the plume reactions are not complete within this distance, the computer
program is then rerun for an axial distance of 70 nozzle radii downstream.

     The computer program calculates total gas flow and fuel flow, along
with emission indices and total contaminant flow.  The standard may thus
be specified as either contaminant flow or emission index.

     A check of data consistency or representative sampling is required.
It is specified that the calculated fuel flow shall agree with the
metered fuel flow within + 15 percent.

6.0  System Analysis

     The Far Plume Method is the preferred method of determining the
exhaust gas constituent level.  Even though the Far Plume Method de-
termines the level of a dilute low concentration sample this method
provides greater confidence and accuracy than the near plume method.
The Near Plume Method relies upon a computer program model of the chemi-
cal gas kinetic reaction and the turbulent gas mixing to approximate the
emission level from the partically reacted exhaust stream.  As previously
stated the chemical kinetic reactions of NO formation are reasonably

-------
                                  -7-
well understood with the maximum flame temperature being the dominant
parameter in NO production.  The chemical kinetic reactions for the
formation of HC and CO are not well understood and at this time the
formation trend can be predicted accurately but not the absolute mag-
nitude.

     Both the Far Plume Method and the Near Plume Method use the same
basic emission measurement system which is similar to that promulgated
by the EPA for non-afterburning engines.  The only outstanding difference
is the sampling probe.  For the near plume method it must be capable of
withstanding the severe environment in the afterburner flow path where
local total temperatures may reach as high as 3600°F and corresponding
local total pressures in the range of 30 to 40 psia.  This high tempera-
ture at the exhaust plane requires a specially constructed probe capable
of quenching or freezing the chemical reactions at the probe entrance.
This can be done by rapid expansion of the gas to low pressure in the
probe and by cooling of the probe.

7.0  The Near Plume Analytical Model

     The analytical model of the physical and chemical processes occur-
ring in the exhaust plume is formulated and structured from a series of
computer ..programs.  These programs were combined to form a model program
"PLUMOD     of the exhaust plume of an afterburning turbojet or turbofan
engine.  The model represents those features of the gas flow which
influence the consumption of gas constituents in the exhaust such as:

     The time average turbulent mixing of each element of the exhaust
     into the adjacent elements and the mixing of ambient air into the
     hot gas.

     The time varying composition of the gas flow past each point in
     space.

     The generation and decay of gas in the plume.

     The consumption of gaseous contaminants by rate limited chemical
     reactions.

     Input to the model includes data from a probe survey of the engine
exhaust stream, together with properties of the fuel and ambient air and
parameters of the engine cycle.  Based on this input, the model predicts
profiles of velocity, fuel, and contaminant concentrations at various
axial locations in the plume, plus overall residual emissions indices
derived from integration of these profiles.

     This analytical model in its present form does a satisfactory job
of predicting the consumption of emissions in the plume.  The model was
compared with experimental data from only turbojet afterburning, engines,
not mixed-flow turbofan afterburning engines.   When data for turbofan

-------
                                 -8-


engines becomes available, the model should be verified for this type of
augmentor.  In contrast to some analytical modeling efforts which pro-
ceed from relatively simple models to more complex and sophisticated
treatments, this model began from a very sophisticated initial structure
and simplifications were incorporated as work progressed.

     Accuracy of the analytical model predictions is illustrated in
figures 2, 3, and 4 where the model predicts Emission Indices for Min.,
Mid., and Maximum afterburning.  These figures compare the emission
indices calculated from the model to the actual test data.  A comparison
and observations concluded from the two methods are presented as follows:

     Neither method indicated that NOx consumed or generated in the
     plume at any power setting.

     Agreement between the two methods of analysis is excellent at all
     axial stations for the minimum afterburning power conditions.
                                                      '            \
     The predicted overall residual level of HC is only about half of
     the measured level for mid afterburning power setting.

     At the intermediate stations for the mid afterburning power setting
     the model predicts more rapid consumption of both CO and HC than
     was measured.

     At the maximum afterburning power setting the model predicted
     quenching of the CO consumption reaction earlier than the data
     indicated.

     The predicted HC consumption was to very low levels although not to
     the complete elimination as indicated by the data for the maximum
     afterburning power setting.

8.0  Procedure Recommendation
     The,basic procedure described and recommended by the Air Force, in
TR-75-52   , as a method of determining the emissions from aircraft gas
turbine engines is recommended for use in determining emission levels
from commercial afterburning engines.  This procedure as previously
described presents two techniques for emission measurement, the Near
Plume Method and the Far Plume Method.  The Far Plume Method is the
preferred method but in limited area test environments the Near Plume
Method is an acceptable alternative.

These procedures are applicable to afterburning engines operating only in
the afterburning mode.  The regulatory emission measurement procedure
for commerical afterburning aircraft engines would also include a test
procedure for afterburning engines operating in the non-afterburning
mode.

-------
                                   -9-

References

1.  Lyon, T.F., Colloy, W.C., Kenworthly, M.J., and Bahr, D.W.,
    "Development of Emission Measurement Techniques For Afterburning
    Turbine Engines", USAF Aero Propulsion Laboratory, Report No.
    AFAPL-TR-75-52, October 1975 (2 Supplements).

    Supplement 1 - Engine Emission Test Data

    Supplement 2 - Afterburner Plume Computer Program Users Manual

2.  Austin, G.F., "Sources of Variability and Inaccuracies in Aircraft
    Gas Turbine Emission Measurements" EPA, Office of Air and Water
    Programs, ECTD, Report No. AC-76-02, February 1976.

-------
                                                                                    -T
 Parallel
Supersonic

.'.j-^le
 Flow
                                                           Transition Region
Subsonic Mixing Layer
                  Supersonic
                 Mixing Layer
 Fully Developed
Turbulent Free Jet
                                 Figure  1     Supersonic  Turbulent Jet.

-------
  40
  30
  10
 p
pi
|j|

13
              •O
              O
              O,
                                  CO
         x = 0
              7.5
                                   15
                                      30
            60  Ft
  10
a  8
 X
 0)
 •c

 £  4

 c
 o
 •H
                                  HC
         x = 0
              7.5
                          15
30
                                                          60 Ft
  10
                                   NO,,
          x = 0
               7.5
                          15
                                               30
                                                          60  Ft
Figure  2   Plume Model Predictions of Overall Emissions Indices

            Compared with  actual   Test Data - J79-15 Engine, Min.

            A/B.

-------
   70
   60
  50
  40
  30
  20
  10
 
-------
  100
   80
   60
   40
   20
           si
PI
H
M
  rt
01

73

O
                         	1
                                 CO
         x = 0
         7.5
                                    15
                                               30
                                                           60  Ft
D

h


J2
01
•n

C
          x = 0
         7.5
                                HC
                                    15
                                               30
                                                           60  Ft
                               NO,
          x = 0
        7.5
                                   15
                                               30
                                                          60  Ft
Figure  4    Plume Model Predictions of Overall  Emissions Indices

             Compared with  actual  Test Data -  J79-15 Engine,

             Max.  A/B.

-------
APPENDIX

Extracted From:     AFAPL-TR-75-52
                    Development of Emissions Measurement Techniques
                    for Afterburning Turbine Engines
                    Air Force Aero Propulsion Laboratory
                    Wright Patterson Air Force Base, Ohio  45433

SECTIONS

This procedure is divided into the following sections:

PART A - FAR PLUME METHOD

Al.,  Definitions of Terms

A2.  Analysis Equipment

A3.  Sampling Equipment

A4.  Equipment Layout

A5.  Instrument Routines

A6.  Reference Gases

A7.  Test Procedure

A8.  Minimum Information to be Recorded

A9.  Calculation of Results


PART B - NEAR PLUME METHOD

Bl.  Definitions of Terms

B2.  Analysis Equipment

B3.  Sampling Equipment

B4.  Equipment Layout

B5.  Instrument Routines

B6.  Reference Gases

B7.  Test Procedure

B8.  Minimum Information to be Recorded

B9.  Calculation of Results

-------
PART A.    FAR PLUME METHOD (MEASUREMENT PROCEDURE FOR SAMPLING AT AXIAL
          STATIONS FAR REMOVED FROM NOZZLE EXIT PLANE).

M.   DEFINITIONS

     Al.l .     Aircraft Gas Turbine Engine;  A turboprop,  turbofan,  or
               turbo jet aircraft engine.

     A1.2      Engine Exhaust;  Flow of material from an engine as a
               result of the combustion of fuel and air.

     A1.3      Exhaust Emissions:  Substances ejected into the atmosphere
               from the exhaust discharge  nozzle of an aircraft engine.

     A1.4      Particulates;  Solid exhaust emissions.

     A1.5      Smoke;  Matter in exhaust emissions which obscures light
               transmission.

     A1.6      Augmentor;   A.device or method used to obtain thrust  in
               addition to that provided by normal operation of the  main
               engine.

     A1.7      Afterburning Gas Turbine Engine;  A gas turbine engine in
               which thrust augmentation is provided by injection and
               combustion of additional fuel in an afterburner.  The
               afterburner is located between the turbine and the exhaust
               nozzle, the term "afterburner" generally applies to a
               turbojet engine.  If the engine is a turbofan type,
               thrust augmentation may be  obtained by burning in the fan
               stream (fanburner or ductburner) or in the combined core
               stream and fan stream (mixedflow augmentor).

     A1.8      Pollutant;   Objectionable exhaust emission.

     A1.9      P lung;  Region downstream of engine exhaust plane where
               exhaust gases mix with the  ambient air.

     A1.10     Total Hydrocarbons (abbreviated HC):  The total of hydrocarbons
               of all classes and molecular weights in the engine
               exhaust.

     Al.ll     Oxides of Nitrogen (abbreviated NOx):  The total of
               oxides of nitrogen in the engine exhaust.  The total  NOx
               value is calculated as equivalent N00.

-------
     A1.12     Flame lonization Detector;  A hydrogen-air diffusion flame
               detector that produces a signal nominally proportional to
               the mass flow rate of hydrocarbons entering the flame per
               unit of time, generally assumed responsive to the number of
               carbon  atoms entering the flame.

     A1.13     Nondispersive Infrared Analyzer;  An instrument that selec-
               tively measures, specific .components by absorption of infrared
               energy.

     A1.14     Chemiluminescence Analyzer;  An instrument in which the
               intensity of light produced by the Chemiluminescence of the
               reaction of nitric oxide with ozone is proportional to the
               concentration of nitric oxide.  Conversion of N0£ to NO prior
               to entering the analyzer permits the determination of both
               species.

     A1.15     Interference;  Instrument  response due to components other
               than the gas that is to be measured.

     A1.16     Calibrating Gas;  Gas of known concentration used to establish
               instrument response.

     A1.17     Span G.as;  A calibrating gas used routinely to check instru-
               ment response.

     A1.18     Zero Gas;  A calibrating gas used routinely to check instru-
               ment zero.

     A1.19     Concentration;  The volume fraction of the component of
               Interest in the gas mixture, expressed as volume percentage
               or as parts per million.
A2.  ANALYSIS EQUIPMENT

     A2.1  NDIR Instruments;  Nondispersive infrared (NDIR) analyzer shall be
used for the continuous monitoring of carbon monoxide (CO) and carbon dioxide
(C02> in the turbine exhaust.

     The NDIR instruments operate on the principle of differential energy
absorption from parallel beams of infrared energy.  The energy is transmitted
to a differential detector through parallel cells, one containing a reference
gas, and the other, sample gas.  The detector, charged with the component to
be measured, transduces the optical signal to an electric signal.  The elec-
trical signal thus generated is amplified and continuously recorded.

-------
          A2.1.1  Instrument Performance Specifications;

          Response Time (electrical)  - 90% full scale response in 0.5 second
          or less.
          Zero Drift - Less than ± 0.5% of full scale in 1 hour on most sensi-
          tive range.
          Span Drift - Less than ± 0.5% of full scale in 1 hour on most sensi-
          tive range.
          Repeatability - Within ± 0.5% of full scale.
          Noise - Less than ± 1.0% of full scale on most sensitive range.
          Sample Cell Temperature - Minimum 50° C (122° F) maintained within
          ± 2° C (3.6° F).
          A2.1.2  Range And Accuracy;

                         Ranee
          Carbon
          Monoxide
          Carbon
          Dioxide
0 to 100 ppm
0 to 500 ppm
0 to 1,000 ppm

0 to 1%
0 to 2%
0 to 5%
Accuracy Excluding Interferences

± 2% of full scale
± 1% of full scale
± 1% of full scale

± 1% of full scale
± 1% of full scale
± 1% of full scale
          A2.1.3  Sensitivity;

          CO Sensitivity (on most Sensitive range)  - 0.3 ppm
          C02 Sensitivity (on most Sensitive range)  - 0.005%
          A2.1.4  NDIR Cells;  All NDIR instruments shall be equipped with
cells of suitable length to measure concentrations within the above ranges to
the indicated accuracy.  Range changes may be accomplished by use of stacked
sample cells and/or changes in the electronic circuitry.
          A2.1.5  ^Interference;  Interferences from water vapor, carbon
dioxide, and carbon monoxide shall be determined on the most sensitive instru-
ment range.  Response of CO instruments shall be less than 5% of full scale
for 2.5% C02 or 4% water vapor.  Optical filters are the preferred method of
discrimination.  In some cases a cold trap or drying agent may be necessary
to reduce water content below the level at which its interference is accep-
table.

-------
     A2.2  Total Hydrocarbon Analyzer;  The measurement of total hydrocarbon
is made by an analyzer using a flame ionization detector (FID).   With this
type detector an ionization current, proportional to the mass rate of hydro-
carbon  entering a hydrogen flame is established between two electrodes.
This ionization current is measured using an electrometer amplifier and is
continuously recorded.    .
          A2.2.1  General Design Specifications;  The analyzer shall be
fitted with a constant temperature oven housing the detector and sample-
handling components.  It shall maintain temperature within ± 2° C of the set
point, which shall be within the range 155 to 165° C (311-329° F).

     The detector and sample handling components shall be suitable for con-
tinuous operation at temperatures to 200° C (392° F).


          A2.2.2  Instrument Performance Specifications!

          Response Time (electrical) - 90% of full scale in 0.5 second or
          less.
          Noise - Less than ± 1.0% of full scale on most sensitive range.
          Repeatability - Within ± 1.0% of full scale.
          Zero Drift - Less than ± 1% of full scale in 4 hours on all ranges,
          Span Drift - Less than ± 1% of full scale in 2 hours.
          Linearity - Response with propane in air shall be linear within
          ±2% over the range of 0 to 500 ppmC.
          A2.2.3  Range and Accuracy

              Range                     Accuracy

          0 to 10 ppmC        ± 5% of full scale with propane
                              calibration gas.
          0 to 100 ppmC       ± 2% of full scale with propane
                              calibration gas.
          0 to 500 ppmC       ± 1% of full scale with propane
                              calibration gas.


          A2.2.4  Sensitivity

          HC Sensitivity (on most sensitive range) - 0.1 ppm

-------
     A2.3  Chemiluminescence Analyzer

          A2.3.1  General Instrument Description;  A chemiluminescence
analyzer with thermal converter shall be used for measuring nitric oxide (NO)
and total oxides of nitrogen (NOX).   The chemiluminescence method utilizes
the principle that NO reacts with ozone (03) to give nitrogen dioxide (N0£)
and oxygen (02).  Approximately 10 percent of the N0£ is electronically ex-
cited.  The transition of excited N02 to the ground state yields a light
emission (600-2600 nanometer region) at low pressures.  The detectable region
of this emission depends on the PM-tube/optical filter being used in the
detector.  The intensity of this emission is proportional to the mass flow
rate of NO into the reactor.  The light emission can be measured utilizing
a photomultiplier tube and associated electronics.

     The method also utilizes the principle that N02 thermally decomposes
to NO (2N02 .-*• 2NO + 02).  A thermal converter unit designed to provide
essentially complete conversion of N02 to NO is included as a part of the
chemiluminescence analyzer package.   If the sample is passed through the
converter prior to entering the chemiluminescence analyzer, an NOX reading
(NO + N02)  is obtained.  If the converter is bypassed, only the NO portion
is indicated.
          A2.3.2  Instrument Performance Specifications

          Response time (electrical) - 90% of full scale in 0.5 second or
          less.
          Noise - Less than 1% of full scale.
          Repeatability - ± 1% of full scale.
          Zero drift - Less than ± 1% of full scale in 2 hours.
          Span drift - Less than ± 1% of full scale in 2 hours.
          Linearity - Linear to within ± 2% of full scale on all ranges.
          Accuracy - ± 1% of full scale on all ranges.
          A2.3.3  Range and Accuracy

                              Accuracy
          0 to 10 ppm         ± 5% of full scale
          0 to 100 ppm        ± 2% of full scale
          A2.3.4  Sensitivity

          NO sensitivity (on most sensitive range) - 0.1 ppm

-------
A3   SAMPLING EQUIPMENT

     A3.1  Sampling Probe;

          A3.1.1  Des.ign Concept;   The  sampling probe shall be constructed
so that individual samples  may be withdrawn  at various  locations across a
diameter of the plume.   Mixed samples are  not permitted.  Either a single-
element movable probe or a  multielement rake may be used.
          A3.1.2  Probe Material;   The  parts  of  the probe wetted by the
sample gas shall be of stainless steel.   Other materials may be used in con-
tact with the sample gas if it is  demonstrated that the material does not
alter the composition of the sample.
          A3.1.3  Probe Temperatures;   The  sample  line within the probe
shall be maintained at a temperature between  160 and  327° F.
     A3.2  Sampling Locations.   Both radial  and  axial sampling locations
depend upon the size of the engine.   In order  to arrive at a common dimen-
sion, referred to the particular engine to be  tested, there is herein defined
a nozzle exit diameter, and all sampling locations  are referred to this
dimension.  The nozzle exit diameter is for  the  maximum engine power condition
and either may be obtained by actual measurement or may be calculated from
engine operating conditions.   The calculated nozzle exit diameter is obtained
by complete expansion of the total engine flow to ambient pressure at the
maximum power condition.


          A3.2.1  Axial Sampling Station. The axial sampling plane shall be
no less than 20 nor more than 25 nozzle exit diameters from the nozzle exit
plane as shown below.  At this  sampling plane, there shall be an unobstructed
area at least four nozzle exit  diameters in  radial  distance about the
projected engine centerline.   (See sketch below.)
           Engine    ^            Engine Centerline             Probe
  4 Dia.  Min.
      I                h	 20-25 Dia.
              / / / / /A/ / /// / / r / f i / / ///// r/i / / rr / / / v / / /

                   Nozzle                                       A
-------
          A3.2.2  Radial Sampling Locations.  A minimum of 11 sampling
points shall be used.  These sampling points should be approximately equally
spaced across the plume diameter, with one sample located on the plume center-
line.  If the sampling system is such that the sampling points cannot lie
in a straight line (across a diameter), then a minimum of five samples shall
be taken in each of two opposite sampling quadrants, with one sample taken at
the plume centerline.

          The outermost sampling points shall be at least four but no more
than five exit nozzle diameters from the plume centerline.  The sample at the
center of the plume shall be taken at a distance no greater than 0.6 exit
nozzle diameters from the true projected engine centerline.
     A3.3  Sample Transfer.  The sample shall be transferred from the probe
to the analytical instruments through a heated sample line of either stain-
less steel or Teflon of 0.18 to 0.32-inch ID.   The sample line shall be
maintained at a temperature of 300 ± 27° F.

     Sample line length should be as short as possible, consistent with the
test setup.  Suitable noncontaminating sample pumps are required to maintain
the proper sample flow rate and to provide adequate sample pressure at the
instruments.  The total sample flow rate shall be such that the sample gas
is transported from the probe inlet to the analyzer inlet in less than 10
seconds.
A4.  EQUIPMENT LAYOUT

     A schematic diagram of the emissions measuring system is shown in
Figure Bl.  Additional components such as instruments, thermocouples,  valves,
solenoids, pumps, and switches may be used to provide additional information
and coordinate the functions of the component systems.  Parallel installation
of CO and C02 analyzers is an acceptable alternative.  No desiccants,  dryers,
water traps or related equipment may be used to treat the sample flowing to
the NOX analyzer.  The NOX instrument configuration must be such that  water
condensation is avoided throughout the instrument.

-------
               Measure Gas Temperature at  these Outlet  Points
               Minimum of 122°  F Maintained within ± 4° F,  Except
               when Optional Water Trap is Used.
                                            Span
                                       Zero   ,'(J^s
                         5 Liters/min
 Sample
 Probe
Heated FID
320 ± 9° F
     -Probe Coolant
     160-327° F
                         Zero
i	
                                        Unrestricted Outlet
                                        Flow Rate 2 Liters/min
                                        Nominal
                                                                      Flowmeter
                                                                      Flow Rate 2 Liters/min Nominal
                                                                         Unrestricted Outlet
                                                       Optional
                                                    Moisture Removal
                                                 Filters      Span
                                          Pump
                                     Recorder
  Span
Note: The Sample Line Between the Converter
      and the NOX Analyzer and the Converter
      Bypass Sample Line shall be Heated,  if
      Necessary, to Prevent Water Condensation

      rz.~ Z—~ Sampling Line
               Heated to 300 ± 27° F to the FID.
               Heated to 130 ± 9° F Downstream of
               the FID.
                                           •=•=
                                                           Pump
                                             Recorder

                                   Dump 20 Liters/min to be
                                   Consistent with the Line
                                   Residence Time Specified
                               Figure Bl.   Sampling System Schematic.

-------
A5.  INSTRUMENT ROUTINES

     A5.1  NDIR Instruments;   Following the instrument manufacturer's instruc-
tions for startup of instruments, the following minimum requirements shall be
adhered to:
          A5.1.1  Monthly Routine;

          (1)  Check detector tuning, following manufacturer's prescribed
               routine.

          (2)  Set instrument zero using dry nitrogen.

          (3)  Using previous gain setting check calibration curves using
               calibration gas with nominal concentrations of 30, 60, and
               90% of each range used.  Use the same gas flow rate through
               instruments during calibration as when sampling exhaust.  Any
               response value differing from the previous value by more than
               ± 3% of the previous value at the same gain setting may reflect
               some problem in the instrument system, and a thorough instru-
               ment check should be made.  Confirm or reestablish calibration
               curves for each range.  Log gain reading.

          (4)  Check response of interference gases as  called out in A2.1.4.
               If unacceptable, determine cause and correct — detector
               replacement may be indicated.

          (5)  Prior to each testing period, a check of the instrument zero
               and span drift, repeatability and noise level shall be made
               on the most sensitive instrument range to insure that it
               conforms with the instrument performance specifications.
          A5.1.2  Daily Routine;

          (1)  If analyzer power is not left on continuously, allow 2 hours
               for warmup.  (If daily use is anticipated, it is recommended
               that analyzer be left on continuously.)

          (2)  Replace or clean filters.

          (3)  Check system for leaks.

          (4)  Check detector tuning and record reading.  If the reading
               changes by more than ± 3% from the previous value, instrument
               readjustment is indicated.  For the following tests the temper-
               ature of zero and span gas in the instrument cells shall be
               within ± 2° C (± 3.6° F) of typical sample gas temperature

-------
          measured at the outlet of the sample cell, and gas flow rate
          through the instruments shall be the same for zero and span
          gas as for sample gas.

     (5)  Zero the Instrument on dry nitrogen.  If there Is a significant
          change In setting of Zero control, determine the cause and
          correct.

     (6)  Using span gas to give 75 to 90% full-scale deflection, check
          the response of the instrument on each range using the gain
          setting from the previous use.  If the reading differs from
          the previous value by more than 3%, an instrument problem may
          be indicated.  Check and correct as necessary.  If instrument
          reading is within ± 3% of previous value, adjust gain control
          to produce proper instrument output.  Log gain setting at
          final adjustment.

     (7)  Check zero with dry nitrogen and repeat step 6 if necessary.

     (8)  Zero and span shall be checked before and after each test, and
          at approximately one-hour intervals during the test.
A5.2  Total Hydrocarbon Analyzer

     A5.2.1  Initial Alignment;

          A5.2.1..1  Optimization of Detector Response:

          (1)  Follow manufacturer's instructions for instrument startup
               and basic operating adjustment.  Fuel shall be 60% helium,
               40% hydrogen containing less than 0.1 ppmC hydrocarbon.
               Air shall be "hydrocarbon-free" grade containing less than
               0.1 ppmC.

          (2)  Set oven temperature at 160° C ± 5° C (320° F ± 9° F) and
               allow at least one-half hour after oven reaches tempera-
               ture for the system to equilibrate.  The temperature is
               to be maintained at set point ± 2° C (± 3.6° F).

          (3)  Introduce a mixture of propane in air at a propane concen-
               tration of about 500 ppmC.  Vary the fuel flow to burner
               and determine the peak response.  A change in zero may
               result from a change in fuel flow; therefore, the instru-
               ment zero should be checked at each fuel flow rate.
               Select an operating flow rate that will give near maxi-
               mum response and the least variation in response with
               minor fuel flow variations.

-------
                                               Optimum
                                       Fuel Flow
               (4)  To determine the optimum airflow,  use fuel flow setting
                    determined above and vary airflow.  A typical curve for
                    response versus airflow is shown below:
                                              Optimum
                                         Air Flow
               After the optimum flow settings have been determined, these
               flows are to be measured and recorded for future reference.
               A5.2.1.2  Oxygen Effect;  Check the response of the detector
with varied concentrations of oxygen in the sample following steps outlined
below; this test shall be made with oven temperature at the set point and
with gas flow to the detector at optimum conditions, as determined in
A5.2.1.1.

               (1)  Introduce nitrogen (N2) zero gas and zero analyzer;
                    check zero using hydrocarbon-free air; the zero should
                    be the same.

               (2)  The following blends of propane shall be used to deter-
                    mine the effect of oxygen (02) in the sample:

                         Propane in N2
                         Propane in 90% N2 + 10% 02

                         Propane in air

                    The volume concentration of propane in the mixture reach-
                    ing the detector should be about 500 ppmC, and the con-
                    centration of both the 02 and hydrocarbon should be known
                    within ± 1% of the absolute value.  The zero should be
                    checked after each mixture is measured.  If the zero has
                    changed, then the test shall be repeated.

-------
                    The response to propane in air shall not differ by more
                    than 3% from the response to propane in the 10%-02/90%-N2
                    mixture, nor differ by more than 5% from the response to
                    propane in nitrogen.

                    If these specifications cannot be met by changing the
                    sample flow rate or burner parameters, such as airflow
                    and/or fuel flow rate, it is recommended that the detec-
                    tor be replaced.
               A5.2.1.3  Linearity and Relative Response;

               (1)  With analyzer optimized per AS.2.1.1, the instrument
                    linearity shall be checked for the range 0 to 100 and
                    0 to 500 ppmC in air at nominal concentrations of 50 and
                    95% full scale of each range.  The deviation of a best
                    fit curve from a least-squares best-fit straight line
                    should not exceed 2% of the value at any point.  If this
                    specification is met, concentration values may be calcu-
                    lated by use of a single calibration factor.  If the
                    deviation exceeds 2% at any point, concentration values
                    shall be read from a calibration curve prepared during
                    this alignment procedure.

               (2)  A comparison of response to the different classes of
                    compounds shall be made using (individually) propylene,
                    toluene, and n-hexane, each at 20 to 50 ppmC concentra-
                    tion in nitrogen.  If the response to any one differs by
                    more than 5% from the average of the three, check instru-
                    ment operating parameters.  Reducing sample flow rate
                    improves uniformity of response.
          A5.2.2  Routine At Three-Month Intervals;  These checks are to be
made at three-month intervals or more frequently should there by any question
regarding the accuracy of the hydrocarbon measurements:

          (1)  Check for and correct any leaks in system.

          (2)  Check and optimize burner flows (air, fuel, and sample) as
               required by criteria of AS.2.1.1.

          (3)  Check 02 effect as outlined in A5.2.1.2.

          (4)  Check response of propylene, toluene, and n-hexane as outlined
               in A5.2.1.3.

          (5)  Check linearity as outlined in AS.2.1.3.

-------
          (6)   Prior to each testing period,  a check of the instrument zero
               and span drift,  repeatability, and noise level shall be made
               on the most sensitive instrument range to insure that it con-
               forms with the instrument performance specifications.  Zero
               instability may  be caused by HC condensation in the Zero gas
               cylinder.  A molecular sieve trap has been found effective  in
               removing HC from the zero gas.
          A5.2.3  Daily Routine

          (1)   Clean or replace filters.

          (2)   Check instrument for leaks.

          (3)   Check instrument temperatures.

          (4)   Ascertain that all flows to detector are correct.

          (5)   Check zero with zero gas.

          (6)   The response using blends  of propane in air shall  be checked
               on each range:

                    For range                          Use	

                    0 to 10 ppmC        7 to 10 ppmC propane in air
                    0 to 100 ppmC       70 to 100 ppmC propane in air
                    0 to 500 ppmC       350 to 500 ppmC propane in air

          If the response differs from the last previous check value by more
          than 3% of the value logged during the last prior day's use, an
          instrument problem may be indicated.

          A zero and span gas check shall be made before and after each test
          and at approximately one-hour intervals during the test.  If the
          cumulative changes exceed 3% during the day, an instrument problem
          may be indicated.
     A5.3  Chemiluminescence Analyzer;   Follow the instrument manufacturer's
instructions for startup of instrument.

-------
          A5.3.1  Thermal Converter Efficiency  Check;  Check the NOX to NO
converter efficiency by the following procedure.  Use the apparatus described
and illustrated below:

                                                 C"
                       NO  Converter Efficiency  Detector
                                    Ozonator
  >2 or Alr          cb
    s»ppiy      T    v/.
      a—cxj-y
    NO/N2
    Supply
                                                                 Converter
                                                                   Inlet
                                                                 Connector
          (a)   Attach  the NO/N2 supply (75-100 ppm)  at C2,  the 02 supply at
               Cj_,  and the analyzer inlet connection to the efficiency
               detector at Cj.  If lower concentrations of  NO are used, air
               may  be  used in place of 02 to facilitate better control of
               the  N02 generated during step (d).

          (b)   With the efficiency detector variac off, place the NOX con-
               verter  in bypass mode and close valve V3. Open valve MV2
               until sufficient flow and stable readings are obtained at the
               analyzer.  Zero and span the analyzer output to indicate the
               value of the NO concentration being used. Record this concen-
               tration.

          (c)   Open valve V3 (on/off flow control solenoid  valve for ©2) and
               adjust  valve MV1 (02 supply metering valve)  to blend enough
               G£ to lower the NO concentration (b)  about 10%.  Record this
               concentration.

-------
(d)  Turn on the ozonator and increase its supply voltage until
     the NO concentration of (c)  is reduced to about 20% of (b) .
     N02 is now being formed from the N0+03 reaction.  There must
     always be at least 10% unreacted NO at this point.   Record
     this concentration.

(e)  When a stable reading has been obtained from (d) ,  place the
     NOX converter in the convert mode.  The analyzer will now
     indicate the total NOX concentration.  Record this concentra-
     tion.
                                               \
(f)  Turn off the ozonator and allow the analyzer reading to
     stabilize.  The mixture NO+02 is still passing through the
     converter.  This reading is  the total NOX concentration of
     the dilute NO span gas used  at step (c) .   Record this concen-
     tration.

(g)  Close valve V3.  The NO concentration should be equal to or
     greater than the reading of  (d) indicating whether the NO
     contains any N02.

     Calculate the efficiency of  the NOx converter by substituting
     the concentrations obtained  during the test into the following
     equation.
               % Eff .  =     ~     X 100%
                        (f)  - (d;

     The efficiency of the converter should be greater than 90
     percent.  Adjusting the converter temperature may be needed to
     maximize the efficiency.

(h)  If the converter efficiency is not greater than 90 percent, the
     cause of the inefficiency shall be determined and corrected
     before the instrument is used.

(i)  The converter efficiency shall be checked at least monthly.
A5.3.2  Monthly Routine

(1)  Adjust analyzer to optimize performance.

(2)  Set instrument zero using zero grade nitrogen.

(3)  Calibrate the NOX analyzer with nitric oxide (nitrogen diluent)
     gases having nominal concentrations of 50 and 95% of full
     scale on each range used.  Use the same gas flow rate through
     the instrument during calibration as when sampling exhaust.
     Log zero and gain settings.

-------
          (4)  Prior to each testing period, a check of the instrument zero
               and span drift, repeatability, and noise level shall be made
               on the most sensitive instrument range to insure that it
               conforms with the instrument performance specifications.
          A5.3.3  Daily Routine

          (1)  If analyzer power is not left on continuously, allow two
               hours for warmup.

          (2)  Clean or replace filters.

          (3)  Check system for leaks.

          (4)  Ascertain that flow to detector is correct.

          (5)  Check zero with zero-grade nitrogen.

          (6)  Zero and span shall be checked before and after each test
               and also at approximately one-hour intervals during the test
A6.  REFERENCE GASES

     A6.1  Mixture Composition;  Reference gases for carbon monoxide and
carbon dioxide shall be prepared using nitrogen as the diluent.  They may
be blended singly or as dual component mixtures.  Nitric oxide reference gas
shall be blended in nitrogen.  Hydrocarbon reference gas shall be propane
in air.  Zero gas shall be nitrogen, or optionally high purity air as
specified in A6.4.
     A6.2  Calibration Gases;  Calibration gases shall be certified by the
vendor as accurate within ± 1%.
     A6.3  Span Gases;  Span gases shall be supplied by the vendor to a
stated accuracy within ± 2%.
     A6.4  Zero Gas;  Nitrogen zero gas shall be minimum 99.998% N2 with less
than 1 ppm CO.  This gas shall be used to zero the CO, C0£ and NO analyzer.
     Zero-grade air shall not exceed 0.1 ppm hydrocarbon.  This gas shall
be used to zero the HC analyzer.  Zero-grade air includes artificial air
consisting of a blend of N2 and Q£ with Q£ concentration between 18 and 21
mole percent.

-------
A7.  TEST PROCEDURF

     A7.1  Test Layout;  Set up engine, sampling equipment, and analysis
equipment as specified in Sections A3 and AA.
     A7.2  Fuel;  The fuel used shall be as specified by the engine manu-
facturer.  The carbon-to-hydrogen ratio shall be determined; this parameter
is required in the calculation of results (Section A9).   The emissions levels
determined by this procedure may be a function of the type of fuel used, and,
therefore, the type of fuel shall be included as an integral part of the test
data, as specified in Section A8.
     A7.3  Ambient Conditions

          A?.3.1  Ambient Temperature, Pressure, and Humidity.   Changes in
ambient temperature, pressure, and humidity can cause changes in emissions
levels both through direct changes in comb ustor conditions and through
changes in engine operating parameters.  Since generally accepted methods are
not currently available for correcting test data to standard conditions,
extremes of ambient conditions should be avoided.  Ambient temperature, pres-
sure, and humidity shall be measured for the test record (Section A8), but
these data are not required for calculation purposes.


          A7.3.2  Wind Velocity and Direction.  The wind velocity and direc-
tion shall be recorded, and the crosswind and tailwind components shall be
calculated.  The crosswind velocity component shall not exceed 5 mph during
the test.  The tailwind component shall not exceed 1 mph.
          A7.3.3  Ambient Air Composition.  Unusually high concentrations
of CO, HC, and CC>2 in the ambient air should be avoided since high values
can adversely effect data accuracy.  For comparison purposes, standard air
contains 300 ppm C02, and the EPA ambient air quality standards are 9 ppm,
0 24 ppm, and 0.05 ppm for CO, HC, and N02 respectively.  Unusually high
concentrations may indicate abnormal conditions such as exhaust gas reinges-
tion, fuel spillage, or additional sources of these emissions in the test
area.  It is suggested that an ambient air sample be obtained with the engine
running before obtaining emissions data at each power setting.


     A7 .it  Instrument Calibration.  Calibrate exhaust analysis instruments
before and after each test period using daily procedures given in Section
A5.

-------
     A7.5  Test  Sequence

          (a)   Start engine and adjust to desired power setting, allow
               adequate time for stabilization.

          (b)   Measure concentrations of CO, C02, HC, NO, and NOX at 11
               radial sampling locations as specified in Section A3.2.2.

          (c)   The engine may then be stabilized at another power setting
               and measurements made as in (b)  above.  Repeat until test
               series is complete.
A8   MINIMUM INFOKMATION TO BE RECORDED.

     The following information, as applicable, shall form a part of the
permanent record for each test.
     A8.1  General;

          (a)  Facility performing test and location.

          (b)  Individual responsible for conduct of test.

          (c)  Test number, reading number, etc.

          (d)  Date.

          (e)  Time.

          (f)  Fuel type, fuel specification, additives, H/C ratio and
               method of determination.

          (g)  Ambient Conditions:  temperature, pressure, humidity,
               wind velocity, and direction.

          (h)  Engine mounting position and height.

          (i)  Test procedure designation.

          (j)  Exceptions, if any, to this procedure.


     A8.2  Engine Description

          (a)  Manufacturer

          (b)  Model number, serial number

-------
     (c)  Time since overhaul and other pertinent maintenance
          information.

     (d)  Nozzle exit diameter (per Section A3.2) and method of
          determination.
A8.3  Engine Operating Data;

     (a)  Nominal power setting, throttle angle.

     (b)  Rotational speed:  N^, N2-

     (c)  Fuel flow (main engine and afterburner).

     (d)  Airflow and method of determination.

     (e)  Compressor discharge temperature and method of determination,

     (f)  Compressor discharge pressure or EPR.

     (g)  Exhaust nozzle position.

     (h)  Thrust.


A8.4  Exhaust Sampling Data;

     (a)  Axial sampling location.

     (b)  Radial sampling location (distance from projected engine
          centerline).

     (c)  Concentrations of CO, C02, HC, NO, and NOX at each sampling
          location.

     (d)  Sample line temperature.

     (e)  Probe coolant temperature.

-------
A9   CALCULATION OF RESULTS

     A9.1  General Calculations Procedure.  Calculation of results Involves
the following steps :

          (a)  Correction of measured concentrations to actual or wet
               concentrations of CO, C02 , HC, NO, and NOX.

          (b)  Calculation of the slope of a linear fit, by the method of
               least squares, of the concentration of each pollutant plotted
               against the concentration of C02 .

          (c)  Calculation of emission indices (Ib per 1000 Ib of fuel)
               for CO, HC, NO, and NOX from the slopes found in (b) above.

          (d)  Calculation of emission flow rate (Ib per hr) from emission
               indices found in (c) above and the total engine fuel flow
               rate.


     A9 . 2  Symbols

          (CO)    = Actual (wet) concentration of CO in exhaust, ppm.

          (C02)   = Actual (wet) concentration of C0£ in exhaust , % .
          (HC)    = Actual (wet) concentration of hydrocarbon in exhaust,
                    expressed as ppm equivalent methane.

          (NO)    = Actual (wet) concentration of NO in exhaust, ppm.

          (NOX)   = Actual (wet) concentration of NOX in exhaust, ppm.

          (CO)^, (C02) etc. = Dry concentration of CO, C02 , etc.

          (c°)sd» (C02)sd» etc' = Semi-dry (0.602% moisture) concentrations
                    of CO, C02, etc.

          a, b    ° Constants in linear curve fit relationship.

          y, x    = Variables in linear curve fit relationship;
                    y represents pollutant concentrations (wet) in ppm;
                    x represents C02 concentration (wet) in %.

          Y-L, X£  = Actual (wet) concentration of pollutant (Y-L) and C02 (X
                    at each sampling point.

          m       = Total number of sampling points (i = 1 to m) .

          r       = Correlation coefficient.

-------
          WFE = Total engine (mainburner plus afterburner) fuel flow rate,
                Ib per hr.

          EIZ = Emission index of pollutant Z, Ib per 1000 Ib fuel

          W2  = Emission flow rate of pollutant Z, Ib per hour

          n   = hydrogen to carbon atomic ratio of fuel

          K
-------
          „  . v    200 - h (1 + n/2)	
          "w "  d   200 - h + K,hn  /  _ 1.006 (CO).^                 (B2)
                                         10
where (C0)8(j and (C02)S(j are the measured semi-dry concentrations.  Then,
          Concentration (wet) = Ky x 1.006 x Concentration (semi-dry).
     A9.A  Calculation of the Slope of Emissions vs. C02 by the Method
           of Least Squares

          At a given engine power setting, the linearity of CO, HC, NO, and
NOX with respect to C02 shall be determined, by analysis of the measurements
collected at the various probe sampling locations, in the following manner.

          (a)  Using the method of least squares, a polynominal of the form:

                         y - a + b x                                  (B3)

               shall be fitted to each pollutant (CO, HC, NO, NOjj) , in turn,
               where y represents the wet concentration (ppm) of the pollu-
               tant being analyzed, and x is the wet concentration (%) of
               C02.  Hie constants a and b are determined by the well-known
               least square relationships (Reference 45) :

                        EX EX Y -EY EX 2
                          111
                    a - - x
                         (EX±)  -

                        EX.EY  - mEX.Y
                    b -   1  * - ^                               (B5)
                         (EX±) - mEX^


          Xi and Y£ represent the concentrations of X and Y at the ith samp-
ling location of the particular test point, and the summations are over the
total number (m) of sampling locations at which gas samples were extracted.

          The slope, b, of the linear fit is proportional to the emission
index.  The intercepts are related to the ambient concentrations.  Note that
the Y intercept, a, should be no greater than the ambient pollutant level,
and the x intercept, - a/b, should be no greater than the ambient CC>2 level.
Ambient level is taken to be concentrations in the local air with which the
engine exhaust mixes.
                                                                      sn » \
                                                                      (B4)

-------
          (b) The appropriateness of the linear fit of each constituent
versus C02 shall be determined by calculation of the correlation coefficient
(r) for each constituent, defined by the equation (Reference 45):
                                                                      (B6)
                                         2 ]
          Values of r near 1.00 indicate good data consistency over the plume
diameter, that plume reactions have ceased, and that plume mixing is complete.
If the slope, b, is greater than about 10, then r should be greater than
0.99 if measurements are carefully made.  For smaller values of b, r is
Influenced by instrument sensitivity, and low values may result.

          Low values of r may be an indication that plume reactions are not
complete.  If b is greater than 10 and r is less than 0.95, the measurements
shall be repeated at a sampling station farther downstream from the engine.
The new sampling station shall be at a distance from the engine exhaust plane
1.3 to 1.4 times the distance of the previous measurements.

          (c) The following standard deviations shall be computed and
reported for each pollutant as a statistical measure of the degree of error
associated with each term in the curve fit (Reference 46):
                Z(Y  - a-bX.)2
                   1     -i—                                        (B7)
          °a - »y /   --.--  ?   -—7."                              (B8)
                                                                      (B9)
     A9.5  Calculation of the Emission Indices of CO. HC, NO,  NOX

          The emissions indices (lb/1000 Ib fuel) of CO, HC, NO and NOX at
a given test condition shall be determined from the values of b (Section
A9.4a, Equation B3) obtained for each pollutant using the following
equations :
          El   =       2-801 (                                        ,B10v
            CO   (Me + nMH)  /  + b(X) + b^                           

                            \       10*

-------
          «BC
                  1 +
                         104
                          4.601 (bNO)
                                     10

          -.    .          4.601 (X)
            N0x
     A9.6  Calculation of Emissions Flow Rates for CO.  HC. NO.  NOy

          The emissions flow rates (Ib/hr)  of CO,  HC, NO, NOX shall be
determined from the equation:

          Wz = .001 (EIZ) (WPE)                                       (B14)

where z represents CO, HC, NO, and NOX.

-------
     PART B.   NEAR PLUME METHOD (MEASUREMENT PROCEDURE FOR SAMPLING AT
              NOZZLE EXIT PLANE)
Bl.   DEFINITIONS


     Bl.l      Aircraft Gas Turbine Engine;  A turboprop, turbofan, or
               turbojet aircraft engine.

     B1.2      Engine Exhaust;   Flow of material from an engine as a result
               of the combustion of fuel and oxidizer.

     Bl.3      Exhaust Emissions;   Substances ejected into the atmosphere
               from the exhaust discharge nozzle of an aircraft engine.

     Bl 4      Particulates;  Solid exhaust emissions.

     B1.5      Smoke;  Matter in exhaust emissions which obscures light
               transmission.

     B1.6      Augmentor;  A device or method used to obtain thrust in
               addition to that provided by normal operation of the main
               engine.

     Bl.7      Afterburning Gas Turbine Engine;  A gas turbine engine in
               which thrust augmentation is provided by injection and
               combustion of additional fuel in an afterburner.  The after-
               burner is located between the turbine and the exhaust nozzle.
               The term "afterburner" generally applies to a turbojet engine.
               If the engine is a turbofan type, thrust augmentation may be
               obtained by burning in the fan stream (fanburner or ductburner)
               or in the combined core stream and fan stream (mixed-flow
               a.igmentor).

     B1.8      Pollutant;  Objectionable exhaust emission.

     Bi.9      Plume;  Total external engine exhaust including ambient
               air with which the exhaust mixes.

     Bl iO     Total Hydrocarbons (abbreviated HC):  The total of hydrocarbons
               of all classes and molecular weights in the engine exhaust.
                                                          i
     Bl.ll     Oxides of Nitrogen (abbreviated NOX):  The total of oxides of
               nitrogen in the engine exhaust.  The total NOX value is calcu-
               lated as equivalent NC^.

     Bl 1.2     Flame lonization Detector;  A hydrogen-air diffusion flame
               detector that produces a signal nominally proportional to the
               mass flow rate of hydrocarbons entering the flame per unit of
               time, generally assumed responsive to the number of carbon atoms
               entering the flame.

-------
     B1.13     Nondispersive Infrared Analyzer;  An instrument that selectively
               measures specific components by absorption of infrared energy.

     Bl. 14     Chemiluminescence Analyzer;  An instrument in which the in-
               tensity of light produced by the chemilumlnescence of the
               reaction of nitric oxide  with ozone is proportional to the
               concentration of nitric oxide.  Conversion of N(>2 to NO prior
               to entering the analyzer permits the determination of both
               species.

     Bl.15     Interference;  Instrument response due to components other
               than the gas that is to be measured.

     Bl.16     Calibrating Gas;  Gas of known concentration used to establish
               instrument response.

     Bl.17     Span Gas;  A calibrating gas used routinely to check instru-
               ment response.

     B1.18     Zero Gas;  A calibrating gas used routinely to check instru-
               ment zero.

     Bl.19     Concentration;  The volume fraction of the component of
               interest in the gas mixture, expressed as volume percentage
               or as parts per million.
B2.  ANALYSIS EQUIPMENT

     B2.1      NDIR Instruments;  Nondispersive infrared  (NDIR) analyzers
               shall be used for the continuous monitoring of carbon monoxide
               (CO) and carbon dioxide (C02) in the turbine exhaust.

     The NDIR instruments operate on the principle of differential energy
absorption from parallel beams of infrared energy.  The energy is transmitted
to a differential detector through parallel cells, one containing a reference
gas, and the other, sample gas.  The detector, charged with the component to
be measured, transduces the optical signal to an electric signal.  The elec-
trical signal thus generated is amplified and continuously recorded.


          B2.1.1  Instrument Performance Specifications;

          Response Time (electrical) - 90% full scale response in 0.5 second
          or less.
          Zero Drift - Less than ± 1% of full scale in 2 hours on most sensi-
          tive range.
          Span Drift - Less than ± 1% of full scale in 2 hours on most sensi-
          tive range.
          Repeatability - Within ± 1.0% of full scale.

-------
          Noise - Less than ± 1.0% of full scale on most sensitive range.
          Sample Cell Temperature - Minimum 50° C (122° F) maintained within
          ± 2° C (3.6° F).

          B2.1.2  Range and Accuracy

                         Range          Accuracy Excluding Interferences

          Carbon    0 to 100 ppm               ± 2% of full scale
          Monoxide  0 to 500 ppm               ± 1% of full scale
                    0 to 2,500 ppm             ± 1% of full scale
                    0 to 20,000 ppm            ± 1% of full scale

          Carbon    0 to 2%                    ± 1% of full scale
          Dioxide   0 to 5%                    ± 1% of full scale
                    0 to 15%                   ± 1% of full scale


          B2.1.3  NDIR Cells;  All NDIR instruments shall be equipped with
cells of suitable length to measure concentrations within the above ranges to
the indicated accuracy.  Range changes may be accomplished by use of stacked
sample cells and/or changes in the electronic circuitry.


          B2.1.4  Interferences;  Interferences from water vapor, carbon
dioxide, and carbon monoxide shall be determined on the most sensitive instru-
ment range.  Response of CO instruments shall be less than 5% of full scale
for 2.5% C02, or 4% water vapor.  Optical filters are the preferred method
of discrimination.  In some cases a cold trap or drying agent may be necessary
to reduce water content below the level at which its interference is accept-
able.


     B2.2  Total Hydrocarbon Analyzer;  The measurement of total hydrocarbon
is made by an analyzer using a flame ionization detector (FID).  With this
type detector an ionization current, proportional to the mass rate of hydro-
carbon entering a hydrogen flame is established between two electrodes.  This
ionization current is measured using an electrometer amplifier and is con-
tinuously recorded.


          B2.2.1  General Design Specifications;  The analyzer shall be fitted
with a constant temperature oven housing the detector and sample-handling com-
ponents.  It shall maintain temperature within ± 2° C of the set point, which
shall be within the range 155 to 165° C (311-329° F).

     The detector and sample handling components shall be suitable for con-
tinuous operation at temperatures to 200° C (392° F).

-------
          B2.2.2  Instrument Performance Specifications;

          Response Time (electrical) - 90% of full scale in 0.5 second or less.
          Noise - Less than ± 1.0% of full scale on most sensitive range.
          Repeatability - Within ± 1.0% of full scale.
          Zero Drift - Less than ± 1% of full scale in 4 hours on all ranges.
          Span Drift - Less than ± 1% of full scale in 2 hours.
          Linearity - Response with propane in air shall be linear within
          ± 2% over the range of 0 to 2,000 ppmC.


          B2.2.3  Range and Accuracy

               Range                       Accuracy

          0 to 10 ppmC:            ± 5% of full scale with propane
                                   calibration gas.
          0 to 100 ppmC:           ± 27. of full scale with propane
                                   calibration gas.
          0 to 1,000 ppmC:         ± 1% of full scale with propane
                                   calibration gas.
          0 to 10,000 ppmC:        ± 1% of full scale with propane
                                   calibration gas.


     B2.3  Chemiluminescence Analyzer

          B2.3.1  General Instrument Description;  A chendluminescence
analyzer with thermal converter shall be used for measuring nitric oxide (HO)
and total oxides of nitrogen (NOX).  The Chemiluminescence method utilizes
the principle that NO reacts with ozone (03) to give nitrogen dioxide (N02)
and oxygen (02).  Approximately 10 percent of the N0£ is electronically
excited.  The transition of excited N02 to the ground state yields a light
emission (600-2600 nanometer region) at low pressures.  The detectable region
of this emission depends on the PM-tube/optical filter being used in the
detector.  The intensity of this emission is proportional to the mass flow
rate of NO into the reactor.  The light emission can be measured utilizing a
photomultiplier tube and associated electronics.

     The method also utilizes the principle that N02 thermally decomposes to
NO (2N02 •> 2NO + 02).  A thermal converter unit designed to provide essentially
complete conversion of N02 to NO is included as a part of the Chemiluminescence
analyzer package.  If the sample is passed through the converter prior to
entering the Chemiluminescence analyzer, an NOg reading (NO + N02) is obtained.
If the converter is bypassed, only the NO portion is indicated.

-------
          B2.3.2  Instrument Performance Specifications

          Response time (electrical) - 90% of full scale in 0.5 second or less.
          Noise - Less than 1% of full scale.
          Repeatability - ± 1% of full scale.
          Zero drift - Less than ± 1% of full scale in 2 hours.
          Span drift - Less than ± 1% of full scale in 2 hours.
          Linearity - Linear to within ± 2% of full scale on all ranges.
          Accuracy - ± 1% of full scale on all ranges.
B3.  SAMPLING EQUIPMENT

     B3.1  Sampling Probe;

          B3.1.1  Design Concept;  Local exhaust gas temperature at the nozzle
exit plane of afterburning engines may be as high as 3500° F with correspond-
ing total pressures 30 to 40 psia.  Extremely careful design of the probe
coolant passages is required to remove the heat resulting from impingement of
the hot gases on the probe surface.  The coolant must have good heat transfer
properties (water is preferred), and high coolant velocities must be maintained,
especially at the probe leading edge.

     In order to prevent continued chemical reaction within the probe, a
quenching-type probe is required.  Quenching of the reactions is accomplished
by adequate cooling of the tip and by expansion of the gas flow across the
orifice.

     In order to properly mass weight the various samples, impact pressure
must be measured at the sampling point.  Local mass flow is subsequently cal-
culated from the local impact pressure, static (ambient) pressure, and total
temperature (calculated from the gas composition).  The mass weighting implies
that individual samples must be taken.  Due to potential nonhomogeneities in
the exhaust stream, both radially and circumferentially, a relatively large
number of samples are required.  Either a movable probe (single element) or a
fixed rake (multiple element) may be used.  In either case, accurate position-
ing of the probe is required.


          B3.1.2  Probe Material;  The parts of the probe wetted by the sample
gas, except for the probe tip, should be of stainless steel.  Other material
may be used in contact with the sample gas, if it is demonstrated that the
material does not alter the composition of the sample.  To assure adequate
cooling, the probe tip shall be of copper (AMS 4500).


          B3.1.3  Probe Temperature;  The sample line within the probe shall be
maintained at a temperature between 130 and 327* F.

-------
          B3.1.4  Probe Tip Design;  The probe tip orifice shall be sized so
as to give adequate sample flow under critical flow conditions.  A short
expansion section directly following shall enlarge the flow passage to a
minimum diameter of three times the orifice diameter.  A typical probe tip
design is shown below:
                                                  P9G
                                                  Internal Sample Line
                                                  Stainless Steel
                                        Probe Body
                  Probe Tip
                  Copper
          B3.1.5  Total Pressure Measurement;  Total pressure shall be mea-
sured within 0.2 inch of the sampling location.  This permits a combination
probe tip design in which separate orifices are provided for the gas sample
and impact pressure measurements.  Alternatively, both pressure measurement
and gas sample may be obtained from a single orifice, in which case the
measurements cannot be made simultaneously.
     B3.2  Sampling Locations

          B3.2.1  Axial Sampling Plane;  The axial sampling plane shall be
within eight inches of the plane at which the exhaust is completely expanded.
Care should be taken so that adequate clearance exists between the sampling
probe and exhaust nozzle for every position of the exhaust nozzle during the
test.
          B3.2.2  Radial Sampling Locations;  A minimum of 22 sampling points
shall be used for each test condition.  A minimum of five sampling points
shall be located in each of four quadrants, with two sampling points located

-------
near the engine centerline.  Sampling points in adjacent quadrants shall be
separated by at least 60 degrees angular displacement.

     Adjacent sampling points along each diameter should be equally spaced.
In order to accurately locate the edge of the exhaust stream, the impact
pressure at the outermost sampling point shall be between 1.05 P^^ and
1.10 P   , where P  ,  is the ambient pressure.
      amb         amb


     B3.3  Sample Transfer;  The sample shall be transferred from the probe
to the analytical instruments through a heated sample line of either stain-
less steel or teflon of 0.18 to 0.32-inch ID.  The sample lines shall be
maintained at a temperature of 300 ± 27° F from the probe to each analytical
instrument.

     Sample line length should be as short as possible, consistent with the
test setup.  Suitable noncontaminating sample pumps shall be used to maintain
a partial vacuum within the probe so that the pressure ratio across the probe
orifice Is no less than five.  The total sample flow rate shall be such that
the sample gas is transported from the probe inlet to the analyzer inlet in
less than ten seconds.
B4.  EQUIPMENT LAYOUT

     A schematic diagram of the emissions measuring system is shown in Figure
B2.  Additional components such as instruments, thermocouples, valves,
solenoids, pumps, and switches may be used to provide additional information
and coordinate the functions of the component systems.  Parallel installation
of CO and C(>2 analyzers is an acceptable alternative.  No desiccants, dryers,
water traps or related equipment may be used to treat the sample flowing to
the NOx analyzer.  The NO^ instrument configuration must be such that water
condensation is avoided throughout the instrument.
B5.  INSTRUMENT ROUTINES

     B5.1  NDIR Instruments;  Following the instrument manufacturer's instruc-
tions for startup of instruments, the following minimum requirements shall be
adhered to:
          B5.1.1  Monthly Routine;

          (1)  Check detector tuning, following manufacturer's prescribed
               routine.

          (2)  Set instrument zero using dry nitrogen.

-------
                  Measure Gas Temperature at these Outlet Points
                  Minimum of 122°  F Maintained within ± 4° F,  Except
                  when Optional Water Trap is Used.    L.
                                                                  flrestrjr.tert Outlec.
                                                                FJyw Rat> 2 Liters/min
                                                                Nominal
           Engine
           Exhaust
           Nozzle
                                                                             Unrestricted Outlet
                                                           Optional
                                                        Moisture Removal
                                                                        Liters/mln Nominal
Probe Coolant
130-180° F
      Note: The Sample Line Between the Converter
            and the NOX Analyzer and the Converter
            Bypass Sample Line shall be Heated,  if
            Necessary, to Prevent Water Condensation
                                                  Pump
          Recorder
Dump 20 Liters/min to be
Consistent with the Line
Residence Time Specified
 Figure B2.
ir.'Z.-S." Sampling Line
        Heated to 300  ± ?7° F to the FID.
        Heated to 130  ± 9° F Downstream of
        the FID.
 Sampling  System Schematic. Instrument Layout  is  the  Same  as  in Figure  Bl.  Major
 Difference is the Probe Pump.

-------
(3)  Using previous gain setting check calibration curves using
     calibration gas with nominal concentrations of 30, 60, and
     90% of each range used.  Use the same gas flow rate through
     instruments during calibration as when sampling exhaust.  Any
     response value differing from the previous value by more than
     ± 3% of the previous value at the same gain setting may reflect
     some problem in the instrument system, and a thorough instru-
     ment check should be made.  Confirm or reestablish calibration
     curves for each range.  Log gain reading.

(4)  Check response of interference gases as called out in B2.1.4.
     If unacceptable, determine cause and correct — detector
     replacement may be indicated.
B5.1.2  Daily Routine;

(1)  If analyzer power is not left on continuously, allow 2 hours
     for warmup.  (If daily use is anticipated, it is recommended
     that analyzer be left on continuously.)

(2)  Replace or clean filters.

(3)  Check system for leaks.

(4)  Check detector tuning and record reading.  If the reading
     changes by more than ± 3% from the previous value, instrument
     readjustment is indicated.   For the following tests the temper-
     ature of zero and span gas in the instrument cells shall be
     within ± 2° C (± 3.6° F) of typical sample gas temperature
     measured at the outlet of the sample cell, and gas flow rate
     through the instruments shall be the same for zero and span
     gas as for sample gas.

(5)  Zero the instrument on dry nitrogen.  If there is a significant
     change in setting of Zero control, determine the cause and
     correct.

(6)  Using span gas to give 75 to 90% full-scale deflection, check
     the response of the instrument on each range using the gain
     setting from the previous use.  If the reading differs from
     the previous value by more than 3%, an instrument problem may
     be indicated.  Check and correct as necessary.  If instrument
     reading is within ± 3% of previous value, adjust gain control
     to produce proper instrument output.  Log gain setting at
     final adjustment.

(7)  Check zero with dry nitrogen and repeat step 6 if necessary.

(8)  Zero and span shall be checked before and after each test, and
     at approximately one-hour intervals during the test.

-------
B5.2  Total Hydrocarbon Analyzer
       l
     BS.v 2.1  Initial Alignment;

          B5.2.1.1  Optimization of Detector Response;

          (1)  Follow manufacturer's instructions for instrument startup
               and basic operating adjustment.  Fuel shall be 60% helium,
               40% hydrogen containing less than 0.1 ppmC hydrocarbon.
               Air shall be "hydrocarbon-free" grade containing less
               than 0.1 ppmC.

          (2)  Set oven temperature at 160° C ± 5° C (320° F ± 9° F) and
               allow at least one-half hour after oven reaches tempera-
               ture for the system to equilibrate.  The temperature is
               to be maintained at set point ± 2° C (± 3.6° F).

          (3)  Introduce a mixture of propane in air at a propane con-
               centration of about 500 ppmC.  Vary the fuel flow to
               burner and determine the peak response.  A change in zero
               may result from a change in fuel flow; therefore, the
               instrument zero should be checked at each fuel flow rate.
               Select an operating flow rate that will give near maximum
               response and the least variation in response with minor
               fuel flow variations.
                                           Optimum
                                  Fuel Flow
          (4)   To determine the optimum airflow, use fuel flow setting
               determined above and vary airflow.   A typical curve for
               response versus airflow is shown below:
                                           Optimum
                                    Air Flow
          After the optimum flow settings have been determined, these
          flows are to be measured and recorded for future reference.

-------
               B5.2.1.2  Oxygen Effect;  Check the response of the detector
with varied concentrations of oxygen in the sample following steps outlined
below; this test shall be made with oven temperature at the set point and
with gas flow to the detector at optimum conditions, as determined in
B5.2.1.1.

               (1)  Introduce nitrogen (N2) zero gas and zero analyzer; check
                    zero using hydrocarbon-free air; the zero should be the
                    same.

               (2)  The following blends of propane shall be used to determine
                    the effect of oxygen (02) in the sample:


                              Propane in N2

                              Propane in 90% N2 + 10% 02

                              Propane in air


                    The volume concentration of propane in the mixture reach-
                    ing the detector should be about 500 ppmC, and the con-
                    centration of both the 02 and the hydrocarbon should be
                    known within ± 1% of the absolute value.  The zero should
                    be checked after each mixture is measured.  If the zero
                    has changed, then the test shall be repeated.

                    The response to propane in air shall not differ by more
                    than 3% from the response to propane in the 10%-02/90%-N2
                    mixture, nor differ by more than 5% from the response to
                    propane in nitrogen.

                    If these specifications cannot be met by changing the
                    sample flow rate or burner parameters, such as airflow
                    and/or fuel flow rate, it is recommended that the detector
                    be replaced.


               B5.2.1.3  Linearity and Relative Response;

               (1)  With analyzer optimized per B5.2.1.1, the instrument
                    linearity shall be checked for the range 0 to 1,000 and
                    0 to 10,000 ppmC in air at nominal concentrations of 50
                    and 95% full scale of each range.  The deviation of a
                    best fit curve from a least-squares best-fit straight
                    line should not exceed 2% of the value at any point.  If
                    this specification is met, concentration values may be
                    calculated by use of a single calibration factor.  If the
                    deviation exceeds 2% at any point, concentration values
                    shall be read from a calibration curve prepared during
                    this alignment procedure.

-------
               (2)  A comparison of response  to  the different  classes  of
                    compounds shall be made using  (individually) propylene,
                    toluene, and n-hexane, each  at 20  to  50 ppmC concentra-
                    tion in nitrogen.  If the response to any  one  differs  by
                    more than 5% from the average of the  three, check  instru-
                    ment operating parameters.   Reducing  sample flow rate
                    Improves uniformity of response.
          B5.2.2  Routine At Three-Month Intervals;  These  checks  are  to be
made at three-month intervals or more frequently should  there be any question
regarding the accuracy of the hydrocarbon measurements:

          (1)  Check for and correct any leaks in system.

          (2)  Check and optimize burner flows (air, fuel,  and sample)  as
               required by criteria of B5.2.1.1.

          (3)  Check 02 effect as outlined in B5.2.1.2.
                       r
          (4)  Check response of propylene,  toluene, and n-hexane  as outlined
               in B5.2.1.3.

          (5)  Check linearity as outlined in B5.2.1.3.
          B5.2.3  Daily Routine

          (1)  Clean or replace filters.

          (2)  Check instrument for  leaks.

          (3)  Check instrument temperatures.

          (4)  Ascertain  that  all  flows  to  detector  are  correct.

          (5)  Check zero with zero  gas.

          (6)  The response  using  blends  of propane  in air  shall  be checked
               on each range:

                    For range           	Use	
                    0  to  10  ppmC         7  to  10 ppmC  propane  in air
                    0  to  100 ppmC        70 to 100  ppmC  propane  in air
                    0  to  1,000 ppmC      700 to 1,000  ppmC propane in air

           If  the  response differs  from  the last previous  check  value by more
           than  3% of the  value logged during  the last prior day's use,  an
           instrument problem may be  indicated.

-------
          A zero and span gas check shall be made before and after each test
          and at approximately one-hour intervals during the test.  If the
          cumulative changes exceed 3% during the day, an instrument problem
          may be indicated.
     B5.3  Chemi luminescence Analyzer;
instructions for startup of instrument.
                                        Follow the instrument manufacturer's
          B5.3.1  Thermal Converter Efficiency Check;   Check the NOx to NO
converter efficiency by the following procedure.   Use  the apparatus described
and illustrated below:
                     NOV Converter Efficiency Detector
                                    Ozonator
  03 or Air
    Supply
                             .AAAAAAJUUU
                                                                  Converter
                                                                    Inlet
                                                                  Connector
          (a)   Attach  the  NO/N2 supply  (150-250 ppm) at C2,  the 02 supply
               at  C]_,  and  the analyzer  inlet connection to the efficiency
               detector  at C3.  If  lower  concentrations of NO are used, air
               may be  used in place of  02  to facilitate better control of
               the N02 generated during step (d).

          (b)   With  the  efficiency  detector variac off, place the NOx con~
               verter  in bypass mode and  close valve V3.  Open valve MV2
               until sufficient flow and  stable readings are obtained at the
               analyzer.   Zero and  span the analyzer output  to indicate the
               value of  the NO concentration being used.  Record this con-
               centration.

-------
(c)  Open valve V3 (on/off flow control solenoid valve for 02) and
     adjust valve MV1 (02 supply metering valve) to blend enough
     02 to lower the NO concentration (b) about 10%.  Record this
     concentration.

(d)  Turn on the ozonator and increase its supply voltage until the
     NO concentration of (c) is reduced to about 20% of (b).
     is now being formed from the NOK>3 reaction.  There must
     always be at least 10% unreacted NO at this point.  Record
     this concentration.

(e)  When a stable reading has been obtained from (d), place the
     NOx converter in the convert mode.   The analyzer will now
     indicate the total NOx concentration.  Record this concen-
     tration.

(f)  Turn off the ozonator and allow the analyzer reading to sta-
     bilize.  The mixture NOH)£ is still passing through the con-
     verter.  This reading is the total NOx concentration of the
     dilute NO span gas used at step (c).  Record this concentra-
     tion.

(g)  Close valve V3.   The NO concentration should be equal to or
     greater than the reading of (d) indicating whether the NO
     contains any N02>

     Calculate the efficiency of the NOx converter by substituting
     the concentrations obtained during the test into the following
     equation.

          % Eff. =  ft. ~ ffi  X 100%
                    (f) - (d)

     The efficiency of the converter should be greater than 90 per-
     cent.  Adjusting the converter temperature may be needed to
     maximize the efficiency.

(h)  If the converter efficiency is not greater than 90 percent,
     the cause of the inefficiency shall be determined and corrected
     before the instrument is used.

(i)  The converter efficiency shall be checked at least monthly.
B5.3.2  Monthly Routine

(1)  Adjust analyzer to optimize performance.

(2)  Set instrument zero using zero grade nitrogen.

-------
          (3)  Calibrate the NOx analyzer with nitric oxide (nitrogen diluent)
               gases having nominal concentrations of 50 and 95% of full scale
               on each range used.  Use the sane gas flow rate through the
               instrument during calibration as when sampling exhaust.  Log
               zero and gain settings.
          B5.3.3  Daily Routine

          (1)  If analyzer power is not left on continuously, allow two
               hours for warmup.

          (2)  Clean or replace filters.

          (3)  Check system for leaks.

          (4)  Ascertain that flow to detector is correct.

          (5)  Check zero with zero grade nitrogen.

          (6)  Zero and span shall be checked before and after each test and
               also at approximately one-hour intervals during the test.
B6.  REFERENCE GASES

     B6.1  Mixture Composition;  Reference gases for carbon monoxide and carbon
dioxide shall be prepared using nitrogen as the diluent.  They may be blended
singly or as dual component mixtures.  Nitric oxide reference gas shall be
blended in nitrogen.  Hydrocarbon reference gas shall be propane in air.
Zero gas  shall be nitrogen, or optionally high purity air as specified in
B6.4.
     B6.2  Calibration Gases:  Calibration gases shall be certified by the
vendor as accurate within ± 1%.


     B6.3  Span Gases;  Span gases shall be supplied by the vendor to a stated
accuracy within ±2%.


     B6.4  Zero Gas;   Nitrogen zero gas shall be minimum 99.998% N£ with less
than 1 ppm CO.   This  gas shall be used to zero the CO, C02 and NO analyzer.

     Zero-grade air shall not exceed 0.1 ppmC hydrocarbon.  This gas shall be
used to zero the HC analyzer.  Zero-grade air includes artificial air consist-
ing ut a blend of N£  and 02 with 02 concentration between 18 and 21 mole
percent.

-------
B7.  TEST PROCEDURE

     B7.1  Test Layout;  Set up engine, sampling equipment, and analysis
equipment as specified in Sections B3 and B4.
     B7.2  Fuel;  The fuel used shall be as specified by the engine manufac-
turer.  The carbon-to-hydrogen ratio shall be determined and this parameter
is required in the calculation of results (Section B9).  The emissions
determined by this procedure may be a function of the type of fuel used and
so the type of fuel shall be included as an integral part of the test data,
as specified in Section B8.
     B7. 1  Ambient Conditions;  Changes in ambient temperature, pressure,
and humidity can cause changes in emissions levels both through direct
changes in combustor conditions and through changes in engine operating
parameters.  Since generally accepted methods are not currently available
for correcting test data to standard conditions, extremes of ambient condi-
tions should be avoided.  Ambient temperature, pressure, and humidity shall
be measured for the test records (Section B8).
     B7.4  Instrument Calibration;  Calibrate exhaust analysis instruments
before and after each test period using daily procedures given in Section B5.
     B7.5  Test Sequence;

          (a)  Start engine and adjust to desired power setting.  Allow
               adequate time for stabilization.

          (b)  Measure concentrations of CO, C02, HC, NO, NOX, and impact
               pressure on a minimum of 22 radial sampling locations as
               specified in Section B3.2.2.

          (c)  The engine may then be stabilized at another power setting
               and measurements made as in (b) above.  Repeat until test
               series is complete.
B8.  MINIMUM INFORMATION TO BE RECORDED

     The following information, as applicable, shall form a part of the
permanent record for each test.

     B8.1  General

          (a)  Facility performing test and location.

          (b)  Individual responsible for conduct of test.

-------
     (c)  Test number, reading number, etc.

     (d)  Date

     (e)  Time

     (£)  Fuel type, fuel specification, additives, HC ratio, and
          method of determination.

     (g)  Ambient Conditions:  Temperature, pressure, humidity.

     (h)  Test procedure designation.

     (i)  Exceptions, if any, to this procedure.


B8.2  Engine Description

     (a)  Manufacturer

     (b)  Model number, serial number.

     (c)  Time since overhaul and other pertinent maintenance infor-
          mation.


B8.3  Engine Operating Data

     (a)  Nominal power setting, throttle angle.

     (b)  Rotaticaal speed; NI, N£.

     (c)  Fuel flow (main engine and afterburner).

     (d)  Airflow and method of determination.

     (e)  Compressor discharge temperature and method of determination.

     (f)  Compressor discharge pressure or EPR.

     (g)  Exhaust nozzle position.

     (h)  Thrust.

     (1)  Fuel temperature.

     (j)  Engine bypass ratio.

     (k)  Engine inlet (ram) total temperature.

-------
     B8.4  Exhaust Sampling Data

          (a)  Axial sampling location.

          (b)  Radial sampling location (distance from projected engine
               centerline).

          (c)  Concentrations of CO, CC<2» HC, NO and NOx at each sampling
               location.

          (d)  Probe impact pressure at each sampling location.

          (e)  Sample line temperature.

          (f)  Sample line pressure within probe.

          (g)  Probe coolant temperatures.


B9.  CALCULATION OF RESULTS

     For afterburning engines, chemical reactions can continue in the exhaust
plumes downstream of the nozzle exit plane.  The higher the exhaust tempera-
ture at the exit plane, the greater is the extent of subsequent reactions.
The composition of the exhaust at the nozzle exit plane is thus not represen-
tative of the actual levels of pollutants ejected into the surrounding
atmosphere, the actual levels being generally less than the levels at the
nozzle exit plane.  The measured emissions levels must thus be corrected for
plume reactions through the use of a computer program derived from a reactive
plume analytical model.


     B9.1  Plume Model Input Data

     The plume model computer program computes mixing and reactions along a
maximum of 11 stream tubes in the plume, but the measurements are made at 22
radial probe positions.  The 22 separate measurements must be reduced to 11
values before Insertion into the plume model.  To accomplish this, the impact
pressure measurements are first plotted against radial position, as shown in
Figure B3, and a smooth, averaged curve is drawn through the data.  The outer
radius (RQ) of the exhaust jet is taken as that radial location at which the
impact pressure equals the ambient pressure.

     Similarly, the CO and C02 concentration data are plotted against radial
position (Figure B4) and smooth, average curves shall be drawn through each
set of data.  The plume model computer program requires that input CO and C(>2
composition data be on a semi-dry basis; that is, with 0.602% moisture in the
sample.  In case the samples were not partially dried (with an ice trap) to
this level, then appropriate correction factors shall be applied.  Suitable
correction factors are given in Section A9.3 of this procedure.

-------
 (0
 a
 3
 in
 H)
 ^
 a,
 a
 £
                                                         Ambient (Static)

                                                         Pressure"" ~"~
     15
                                                                       8     r
                                    Radius, Inches
Figure B3.   Impact Pressure Vs.  Radial Position at Max.  A/B Power Level.  R  is

             Outer Radius of Exhaust.                                         °

-------
   16,000
   14,000 -
        0     1     2     3      4     5     6     7     89
                         3456

                            Radius, Inches
Figure B4.   C02 and CO Concentrations  Vs.  Radial Position
             at  Max.  A/B Power Level. R  is Outer Radius
             of  Exhaust.

-------
     Finally, the HC and NO^ concentration data shall be plotted against
radial position, as shown in Figure B5.  If large variations in HC concen-
trations are noted across the stream, then the HC data shall be plotted on
semilogarithmic coordinate paper, as in Figure B5.

     The exhaust area, as determined by Ro, shall be divided into 11 equal
areas by defining
                                2
                         g - TT                                      (B15)
The 11 radial locations, %, are then selected to be in the center of each of
11 equal areas.  Thus,
                                        -  .                           (B16)
where i = 1, 2, --- , 11.

This equation can be simplified to yield
                         Ri = 0.2132  R0 / (21-1)                      (B17)

where 1 = 1, 2, —, 11.

     The complete list of input data required for the plume model computer
program Is given in Table Bl along with a brief description of each vari-
able.  Note that Item 5, RADJ, is equal to R0 expressed in feet.  Similarly,
Item 13, RADII, is equal to RJL expressed In feet.

     The local gas composition at the 11 selected radial locations (Item 12,
CAROL) must be expressed either in mole fraction, parts per million, or some
unit proportional to mole fractions.  Note that mixed units (e.g., ppm and %)
cannot be entered.  Item 19, SF, is the scale factor appropriate to the units
used.

     Tht emissions indices for NOx  (EIN62C) and for CO (EIC6C) at the  after-
burner inlet are also required for inputs to the plume model.  These values
are normally obtained from previous measurements on the engine at military
power.

     For mixed flow augmentors, in which afterburning fuel is injected into
the mixed stream consisting of core engine exhaust and fan air, the overall
engine bypass ratio (BETA) may be obtained from engine cycle data.  The local
bypass ratio (BL9C) at each probe location is generally obtained from  emis-
sions measurements at military power condition.

-------
    10,000
a
     1000 —rv7
      100
                                                               12
                              Radius,  Inches
      120
                            3456
                                Radius, Inches
8    f 9
    R
Figure B5.  HC and NOX Concentrations Vs. Radial Position at
             Max. A/B Power Level.  Ro is Outer Radius  of  Exhaust.

-------
                      Table Bl.   Input  Data  to  Plume Model Computer Program.
Item
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Variable
Name
PO
TO
HUM
VO
RADJ
T2
FAR5
EIN92C
BETA
T25
FUEL
CAROL
RADII
FT
PS
BL6C
ELI6C
TITLE
SF
Dimension
1
1
1
1
1
1
1
1
1
1
3
4, 11
11
11
11
11
11
20
1
Initial Default
Value Value
14.696 - i
518.69 - i
.00531 - S
0.0 - 1
1.0 - (
518.69 - 1
0.02 - :
20.0 - 1
0.0 - 1
518.69 - 1
2.0 - 1
537.0 - ]
BITS Est. ]
BITS - ]
BITS - I
BITS - 1
BITS PO S
BITS BETA 1
BITS 0.0 1
j
Blank - (
1.0 - £
                                                                                         Description .r
20
             PRINT
                             30
                                            BITS
0.1*RADJ
10*RADJ
BITS
Ambient air static pressure,  psia	
Ambient air static temperature,  *  R.
Specific humidity of ambient  air,
Ib H20/lb dry air.
Flight speed, ft/sec.
Outer radius of exhaust jet,  ft.
Engine inlet (rain) total temperature,  ° R.
Turbine exit fuel-air ratio,
Ib fuel/lb dry air.
Main combustor NOg emission index,
Ib N02/1000 Ib fuel.
Engine bypass ratio, Ib fan air/lb  core air.
Bypass air (fan discharge)  temperature, ° R.
Hydrogen - carbon atom ratio  of  fuel.
Fuel temperature, ° R.
Lover heating value of  fuel,  Btu/lb.
Local gas composition CO, CC<2, HC,  NOX.
Radial locations of probe measurements, ft.
Probe impact pressure,  psia.
Static pressure at probe locations,  psia.
Local bypass ratio at each probe location.
Main combustor CO emission index,
Ib CO/1000 Ib fuel.
Output page heading information
Scale factor for CAROL  data (leave  1.0  if
composition in mole fraction  - set  1E-6 if
in ppm, etc.)
Axial stations (feet) at which output is
to be printed.

-------
     Data item No. 20 (PRINT) contains the axial stations at which the output
is to be printed.  Data shall be calculated and reported for at least three
axial stations as follows:  (1) nozzle exit plane (zero feet aft), (2) 35
times RADJ,  and (3) 50 times RADJ.  Data may be calculated for additional
axial stations in order to follow the course of reactions within the plume.
Reactions are substantially completed at 35 times RADJ.

     The Computer Program User's Manual should be consulted for additional
information and for the complete description of the plume model computer pro-
gram.  The User's Manual has been published as Supplement 2 to AFAPL-TR-75-52.
     B9.2  Plume Model Output

     At each axial station, as designated in the input data, the plume model
computer program calculates various quantities related to mixing and reactions
in the plume along the 11 stream tubes initially selected.  In addition, at
each station, the following overall or integrated values are calculated, and
shall be reported:

               Total Flow, pps:                     Gas Mixture, Fuel
               Emission Indices, lb/1000 Ib fuel:   CO, HC, NOX

               Contaminant Flow, pps:               CO, HC, NC>x


     The overall values shall be examined for internal consistency.  The
following criteria shall apply:

     (a)  Calculated fuel flow shall agree within ± 15% with the metered
          total engine fuel flow.  If the agreement is not within 15%, then
          the plume model input data shall be reassessed.  In particular,
          the curves of concentration versus radial location shall be
          examined to determine that the composition of the samples is
          representative of the actual local composition.

     (b)  Emission indices and contaminant flows at axial station 50 times
          RADJ shall be within 5% of the value calculated for axial stations
          35 times RADJ.  If such is not the case, the computer program shall
          L>e rerun for an axial station of 70 times RADJ.

-------