No. FAA-RD-71-22
             STODY  OF  VISIBLE  EXHAUST  SMOKE
                 FROH9 AIRCRAFT JET  ENGINES
                   John Stockham and Howard  Betz
                        III  Research  Institute
              10  West 35th Street, Chicago, Illinois 60616
                             I ONE 1971
          Availability is unlimited.  Document may be released to the
          National Technical Information Service, Springfield, Virginia
          22151, for sale to the public.
                             Prepared for

                              OF TRANSPORTATION
                  FEDERAL HIATIB3  ADMINISTRATION
                  Systems Research  & Development Service
                                  D. C., 20580

-------
The contents of this report reflect the views of the
contractor, which is responsible for the facts and
the accuracy of the data presented herein, and
do not necessarily reflect the official views or
policy of the FAA or Department of Transportation.
This report does not constitute a standard,
specification, or regulation.

-------
                                                 TECHNICAL REPORT STANDARD TITLE
1. Report No. 2. Government Accession No.
FAA-RD-71-22
4. Till* and Subtitle
Study of Visible Exhaust Smoke From
Aircraft Jet Engines
7. Author'.)
John Stockham and Howard Betz
9. Performing Orgonizotion Home and Address
IIT Research Institute
10 West 35th Street
Chicago, Illinois 60616
12. Sponsoring Agency Name and Address
Federal Aviation Administration
Systems Research & Development Service
Washington, B.C. 20590
& National Air Pollution Control Administratio;
3. Recipient's Catalog No.
5. Report Date
June 1971
6. Performing Organization Code
8. Performing Organization Report No.
FAA-NA-71-2U
10. Work Unit No.
11. Conlroct or Grrmt No.(502- 306- ' 	 . !
DOT-FA69WA-2208
13. Type o( "Report and Period Cohered
Final Report
14. Sponsoring Agency Code
1
15. Supplementary Notes
NONE
16.
       t  xhe objective of this  study was to relate  the visibility of
inflight jet exhaust to the SAE  smoke number.  A method  based on photo-
graphic  photometry was developed for measuring the optical density of
smoke  plumes.   This method was related to visibility and to the smoke
number through transmissometer measurements and visibility theory.  A
portable transmissometer, capable of operating over a wide range of
optical  path lengths and under varying ambient light conditions was
fabricated for use on this study.  The mathematical expression relatir.~
the  transmission measurements  to the smoke number was derived.  Liminal
visibility requirements of smoke trails, developed from  light scat-
tering theory, correlated with actual visual observations  and the trar.s-
raissometer and photometry measurements.  Test results, with the engines
investigated,  indicate that SAE  smoke numbers below 23 were associated
with invisible exhaust plumes.   Samples of the exhaust smoke showed the
particles to be composed of lacy agglomerates.  At the nozzle, the
geometric median particle diameter was 0.052 \j.m.  At a distance of
10 nozzle diameters the geometric median particle diameter was 0.13 utr.
at cruise condition.
 17. Key Words
   Pollution
   Smoke
   Jet Aircraft
                                    18. Distribution Statement
                                    Availability is unlimited.  Document
                                     may be released to the National
                                    Technical Information Service,
                                    Springfield, Virginia  22151, for
                                     sale to the public.
 19. Security Clo»»if. (of this report)

I    Unclassified
                          20. Security Clossif. (of this poge)

                             Unclassified
                                                    21. No. of Poges

                                                      75
22. Price
Form DOT f 1700.7 (8-69)

-------
                           PREFACE

This report was prepared by the IIT Research Institute for the
Federal Aviation Administration and the National Air Pollution
Control Administration.  The work effort was o part of a progrem
of the Aircraft Division, Systems Research and Development
Service, Federal Aviation Administration, and the Division of
Motor Vehicle Research and Development, National Air Pollution
Control Administration.

The work was administered under the direction of Mr. G. R. Slushei
who served as project manager.  Provision of the facilities,
conduct of the tests end collection and reduction of the dc'ta for
the SAE smoke numbers were furnished by the Propulsion Section,
Aircraft Branch, Test and Evaluation Division, National Aviation
Facilities Experimental Center, Atlantic City, New Jersey.
                               111

-------
                      TABLF OF CONTENTS

                                                       Page

INTRODUCTION                                             1

     (1.)  Purpose                                       1
     ( 2.)  Background                                    1

DISCUSSION                                               3

     (3.)  Modeling Studies                              3
           (3.1)  Light-Scattering Theory                3
           (3.2)  Visibility Requirements                3
           (3.3)  Theory Relating the Smoke Number
                  and Optical Transmission of Jet
                  Exhaust Flumes                         9
                  (3.3.1)  Reflectivity  of Soot
                           Deposition Filter Paper      11
                  (3.3.2)  Optical Transmission  of
                           Smoke Plumes                  12
                  (3.3.3)  Relating DR to DT and
                           Operating Parameters          13
                  (3.3.U)  Relating Smoke Number to
                           Filter Density               IS

     (U.)  Static Tests of Engines and Aircraft          16
           (U.I)  J-57-P-37A Engine Mounted in the
                  Test Cell                             16
           (U.2)  Light Transmission of  J-57 Engine
                  Exhaust for Various Path Lengths      17
           (U.3)  JT-12-6 Engine Mounted in Wind Tunnel 20
           (U.U)  Tied-Dovm Aircraft Tests              20
                  (U.U.I)  F-100 Single-Engine
                           Aircraft                     20
                  (U.U.2)  Convair 880 Aircraft          2U
                  (U.U.3)  Multiple Engines of Convair
                           880 Aircraft                  2U
                  (U.U.U)  Lockheed Jetstar Aircraft    30
           (U.5)  Discussion and Summary of Results
                  of Static Engine and Aircraft  Tests   30

     (5.)  In-Flight Observations                       32
           (5.1)  Flights Evaluated at NAFEC            32
                  (5.1.1)  F-100 and F-105 Aircraft
                           In-Flight Tests              32
                  (5.1.2)  Convair 880 Aircraft
                           In-Flight Tests              39
                  (5.1.3)  Lockheed Jetstar Aircraft
                           In-Flight Tests              39

-------
                TABLE OF CONTENTS  (Continued)

                                                        Page

            (5.2)  Observations of Aircraft  In-Flight
                   at Chicago O'Hare Airport             39
            (5.3)  Discussion and  Summary of In-Flight
                   Tests                                 39

CONCLUSIONS                                              US

REFERENCES                                               U6

APPENDIX A  Transmissometer  (4 pages)                  1-1

APPENDIX B  Photographic Instrumentation (7  pages)      2-1

APPENDIX C  Sampling Jet Exhaust for Particular
            Matter  (U pages)                           3-1

APPENDIX D  SAE Standard Method for Smoke
            Measurements  (3 pages).                    U-l
                                VI

-------
                   LIST OF ILLUSTRATIONS

Figure                .                                 Page

   1    Visibility of Line Targets                        8

   2    Electron Photomicrographs  of Jet  Engine
        Exhaust Particles                               10

   3    Smoke Instrumentation                           23

   4    Microdensitometer Trace of  Number 1  Engine
        of Convair 880 Aircraft                         26

   5    Number 1 Engine of the Convair 880 Aircraft      27

   6    Microdensitometer Traces from Tied-Down
        Convair 880 Aircraft                            28

   7    Plot of Smoke Number vs Optical
        Transmission                                    33

   8    NAFEC/Atlantic City Airport                     35

   9    Photographs of F-100 and Convair  880 Aircraft
        In Flight                                       36

  10    Photographs of Jet Exhaust  Trails During
        Takeoff at O'Hare Airport                        U3

 1.1    Transmissometer                                1-2

 1.2    Optical Schematic                              1-3

 2.1    Typical H & D Curve                            2-3

 2.2    Illustration of Use of Photographic
        Photometry                                     2-6

 3.1    Exhaust Particulate Sampling Apparatus          3-2

 3.2    IITRI Smoke Sampler in the  Engine Test Cell     3-3

 U.I    F-100 Aircraft with Sampling Probes             U-2

 U.2    General Electric Smoke Sampling Console         U-3
                             Vll

-------
                       LIST OF TABLES

Table                                                  Page

  1     Relationship Between Ringelmann  Number,
        Light Transmission, and Particle Concentra-
        tion of Turbojet Engine Exhaust                   5

  2     Particle Size Distribution of Exhaust  from
        J-57 Engine                                      7

  3     Smoke Numbers and Transmission Data  of Jet
        Smoke - J-57 Engine                             18

  4     Light Transmission as a Function of  Path Length 19

  5     Smoke Numbers and Transmission Data  of Jet
        Smoke-JT-12-6 Engine Mounted  in  Wind Tunnel      21

  6     Smoke Quality from a Tied-Down F--100
        Aircraft                                        22

  7     Smoke Quality from the Number 1  Engine of a
        Tied-Down CV-880 Aircraft                       25

  8     CV-880 Aircraft Tied-Down Multiple Engine
        Observation                                     29

  9     Smoke Numbers from the Number 4  Engine of a
        Tied-Down Jetstar Aircraft                      31

 10     Summary of Smoke Number and Transmission
        for Static Tests                                34

 11     Observations on the F-100 Aircraft In-Flight
        at NAFEC                                        37

. 12     Observations on the F-105 Aircraft In-Flight
,        at NAFEC                                        38

 13     Observations on the Convair 880  Aircraft
        In-Flight at NAFEC                              40

 14     Photographic Measurement of Jet  Smoke  Trails
        from Aircraft In-Flight at O'Hare Airport and
        Visual Estimates of Transmission                41

3.1     Size Distribution and Concentration  of Exhaust
        Particles - Royco Particle Counter-J-57-P-37A   3-5

3.2     Size Distribution and Concentration  of Exhaust
        Particles - Royco Particle Counter JT12-6       3-6

                             viii

-------
                         INTRODUCTION

         (1.)  Purpose

         The purpose of this research program was to develop
techniques for the objective determination of visibility of
jet exhaust smoke and no relate these techniques 1:0 the smoke
number obtained for engines mounted in test cells.  l^se of the
established relationships should enable the prediction of in-
flight visibility of smoke trails from test cell data.  A
theoretical model of smoke trail visibility using the physical
and optical properties of the smoke particles was included in
the research effort.

         (2.)  Background

         Smoke trails from aircraft turbine engines are a
symbol of air pollution.  Elimination of these smoke trails
would erase a source of pollution ccmp.iairr.s and improve visi-
bility in and around jetports.  Engine smoke is currently
determined by a filter stain technique described by The
Society of Automotive Engineers (Reference 1). The results are
expressed as a numerical value known as the SAE smoke number.
This technique is applied to engine test cell operations.  A
need exists for an objective measure of the visibility of smoke
trails of aircraft in-flight, especially at take-off and
approach power conditions, and to relate this measurement to
the SAE smoke number.  The only currently accepted method for
specifying the visual quality of smoke, trails from aircraft
in-flight is the Ringelmann method (Reference 2).  This is a
subjective measurement.  Results are highly dependent: upon the
conditions under which the measurements are performed ^Refer-
ence 3).

         Two techniques were developed to accomplish the
objectives of this  program.  The first involved a transmisso-
meter to measure the light a r.r. er.ua tier ci r.he ?.mckef and the
second was photographic photometry.  A special two-path trans-
missometer was developed for measuring the optical attenuation
of smoke from test cell engines and engines of tied-down air-
craft.  The transmissometer is described in Appendix A.
Photographic photometry was developed f~-r eva.";u.-ji:ing the visi-
bility of smoke from aircraft in-flight and f'ro-n tied-down
aircraft.  This procedure is described in Appendix 3.

-------
         The following measurements were made during this
program:

         (1)  Photographic photometry measurements and Ringelmann
              observations of in-flight jet aircraft exhaust
              were madp at: Chicago's CTHare Airport and at the
              National Aviation Facilities Experimental Center
              (NAFEC), Atlantic City. New Jersey.  Aircraft
              studied included a variety of commercial jets and
              the F-100, F-105, CV-880, and the Lockheed Jetstar.

         (2)  Photographic photometry and transmissometer mea-
              surements, Ringelmann observations, and the filter
              stain technique were a.1.1 used to evaluate smoke
              emissions from tied-down aircraft at NAFEC.
              Aircraft used were the F-10CL CV-880,. and the
              Jetstar.  NAFEC personnel obtained the SAE smoke
              numbers.

         (3)  Transmissometer measurements and SAE smoke numbers
              were made on test; engines.  One: a J-57 engine.
              was mounted on a static, sea-level, open-air test
              stand;  the other, a JT-12 engine was mounted as
              a flight installation in a test wind tunnel.
              Samples of exhaust smoke were obtained during
              these tests using a specia?uly constructed sampling
              rig described in Appendix C.

         Quantitative data from the instrumental techniques were
correlated with the subjective Ringelmann observations and visi-
bility theory.  The correlation established the relationship
between the smoke number, the optical attenuation,, photographic
photometry, and the visual appearence of jet smoke.

-------
                          DISCUSSION

         (3.)  Modeling Studies

               (3.1)  Light-Scattering Theory

              The visibility of a smoke plume is a function of
the optical and physical properties of the particles comprising
the plume, the width of the plume, and the contrast between the
plume and its background.

              Where a smoke particle is small in relation to the
wavelength of incident light, the total attenuation of light, i.e.,
extinction, E, is given by tne expression:

             E =	24^ A	
                   999       999
                 (n  + n k) + 4(rr -• n k^ -1)

where,

        2ur
    tt =1T

    X = wavelength of light, urn

    r = particle radius, pm

 n, k = functions of the refractive index, r\, of the particle.

The refractive index, n.. of a particle is a complex number when
the particle both absorbs and scatters the incident light.  In
these instances, the refractive index is equal to n(l -• ik),
where n is the coefficient that defines the scattered light
and nk the coefficient that defines the absorbed light.  The
term i is the operator V-l.  The complex refractive index of
carbon, the major particulate component; of turbojet engine
exhaust (Reference 4), for incident light of wavelength 0.490 um
is 1.59 (1-1.051).  Thus, n = 1.59 and k = 1.05.

              The extinction of light is due to two separate
and distinct phenomena; light scattering, S, and light absorp-
tion, A.  The phenomena are defined as follows:

                    / :  9      2
                 Q  4 .  2.   , .
             s = 8 a   n.._^_A

                   3   r]2 + 2 -                            (2)

             A = E - S

-------
         Substituting the best available information into
Equations (1) and (2), the following values for E, S, and A
result:

         E = 0.9

         S = 0.1
         A = 0.8

         Thus, the extinction of light by turbojet engine smoke
is due to 117o scattering of light and 89% absorption of light.

         The attenuation of light by a smoke cloud, where the
cloud is so diffuse that complications due to particle inter-
actions are not encountered, is given by the expression:
where,
         I  = the intensity of the incident light

         I  = the intensity of the transmitted light

         C  = number concentration of smoke particles

         E  = extinction coefficient

         r  = particle radius

         L  = depth of cloud intervening between I  and I.

The ratio I/I  is the transmission, T, of light through the
smoke .       °

         Thus ,

         2.303 log T = -nr2CLE                            (4)

         Equation (4) illustrates the dependence of light
t.ransmittance through a smoke plume on the length of the viewing
path.  Thus, the position of the observer in relation to the
smoke plume is important to plume visibility.

         If the path length, the particle size, and the extinc-
tion coefficient are known, it is possible to calculate the
number of particles per unit volume of gas that will produce
various levels of light attenuation.  The data in Table 1 is an
example.  The path. length chosen is the diameter of the J-57

-------
                           TABLE 1

           RELATIONSHIP BETWEEN RINGELMANN NUMBER,
               LIGHT TRANSMISSION, AND PARTICLE
           CONCENTRATION OF TURBOJET ENGINE EXHAUST
Percent
Transmission
98
95
90
80
60
40
20
Ringelmann
Number
_ _
1/4
1/2
1
2
3
4
Particle Concentration
at Engine Nozzle
Millions of Particles/cm3 (1>
17
50 • 1
100 i
t


(1)  Particle concentration data is calculated from equation
4 using a path length of 56 cm and a particle size radius of
0.026 urn.  The path length is the nozzle diameter of the J-57
engine and the particle size is the geometric median particle
radius of the exhaust particles at the nozzle under cruise
conditions, Table 2.

-------
turbojet engine exhaust nozzle, 56 cm.  The geometric median
particle diameter of exhaust smoke at the nozzle of the J-57
engine is 0.052 wm, Table 2.  Thegdata in Table 1 show that
particle concentrations of 1 x 10  particles/cm^ corresoond to
transmission values of 90% or an equivalent Ringelmann number
of 1/2.
         (3.2)  Visibility Requirements

         The visibility requirements for jet smoke trails is
similar to those of line targets such as wires, poles, and
antennas.  Two basic factors are involved in line target visi-
bility.  The target must be wide enough to be resolved by the
unaided eye and must have sufficient contrast from the back-
ground to be distinguished.  Tests by Douglas (Reference 5) show
flagpoles and radio towers can be resolved if the angle sub-
tended at the eye of the viewer is greater than 3 or 4 sec . of
arc.  Assuming a nondif fusing smoke track and no attenuation by
the intervening atmosphere, a 2 f t . smoke track could be seen
at a distance of 20 miles.  Obviously, the angle subtended is
not a factor limiting the visibility of jet smoke trails.
The relationship between angle subtended and contrast is shown
in Figure 1.  These, data (Reference 6) were developed in the
IITRI laboratory under ideal conditions; a black line target
(contrast = -1) subtending only 0.6 sec. of arc was visible.

         Contrast, C , between a smoke plume of luminance, B ,
viewed against an extended background of luminance B, , is given
by the expression:
                   n
             r  =  P
                     - B,
         From this equation, a totally light absorbing line
target has a contrast of -1 and a black smoke plume will have
a contrast ranging from 0 to -1 depending on the amount of
background light transmitted through the plurce.  For a plume
that scatters a negligible amount of light such as a carbon
particle smoke (Reference 7)

             Cp = T - 1                                   (6)


where T is the transmission.

-------
                           TABLE 2
            PARTICLE SIZE DISTRIBUTIONS OF EXHAUST
                   FROM THE J-57 ENGINE (1)
Sampler Location,
Engine Diameters
from Engine Nozzle.
0
i

2-1/2

10


. Engine
Thrust
Setting
Approach
75% Norm.
Cruise
Approach
75% Norm.
Cruise
Approach
75% Norm.
Cruise

Particle Size Distribution
Geometric
Median Dia.
um dg
0.053
0.052
0.084
0.084
0.076
0.096
0.13
Geometric
Standard Devia
tiOn °n;
1.63
1.46
1.33
1.40
1.51
1.38
1.40
i
(1)   Data obtained from electron microscope samples collected
by the IITRI sampler described in Appendix C.

-------
2.0 —
0.4


0.2


0.0
                 Visible Region
                               Visible Region
Invisible
Region
      I     I     I
   -101234    56

                   Brightness  Contrast
                                         8
          FIGURE 1.   VISIBILITY OF  LINE  TARGETS.
                             8

-------
         Middleton (Reference 8) discusses the liminal contrast
for visibility of a variety of target shapes, sizes, and
luminance levels.  For targets and luminance levels that approach
conditions under which jet engine exhaust plumes are usually
viewed the liminal contrast ranges from about 1 to 57,.  The
visual range of objects in natural light is usually based on
the assumption that the liminal contrast is 2?0.  On this basis,
the liminal contrast value of 27> was selected for jet exhaust
trails.  Substituting 27, into Equation (6) indicates that jet
smoke with transmission values of 98% or higher will be invisi-
ble.

         In conclusion, the modeling studies show that carbon
smokes attenuate light mainly by absorption.  This factor
permits the use of a simple expression to relate contrast and
transmission measurements to liminal visibility.  This rela-
tionship indicates that jet smoke must transmit more than 987,
of the incident light to be invisible.  Observing the J-57
engine at the nozzle and perpendicular to the exhaust flow,
theory predicts that a smoke with a 987, transmission will have

a particle concentration of 1.7 x 10  particles/cm^.

         The mass concentration of the smoke particles can be
computed from number concentration data .if the particle density
is known.  Because of the lacy agglommerated structure of turbo-
jet exhaust particles, as shown in Figure 2. handbook densities
cannot be used.  If the value of 0.04 g/cm3 reported bv Horveth
and Charlson (Reference 9) for carbon floes is used, the mass
concentration of the smoke producing a transmitted value of
                T
987o is 0.05 mg/m .  Values found in the literature for exhaust
                        o
particulates are 36 mg/m  for a PW-JT8D turbofan engine (Refer-
                              O
ence 10) and 0.5 and 27.0 mg/m  for an unstated turbojet engine
at approach and takeoff power. (Reference 11)

         (3.3)  Theory Relating the Smoke Number and Optical
                Transmission of Jet Exhaust Plumes

         It is the purpose of this theoretical analysis to
relate the SAE smoke number, the.mass flow of an engine, and
the dimensions of the nozzle, to the optical attenuation of the
smoke plume.  The first relationship considered is the effect
of particle concentration in the exhaust gas on the staining
of the filter paper.

-------



                  at engine nozzle

                              .*  *
                      •
   2-1/2 nozzle diameters from the engine nozzle
                                 ^

                                 
-------
              (3.3.1)  Reflectivity of Soot Deposit on
                       Filter Paper

              The reflectivity of soot deposit on filter, paper
may be represented by:
                              i
             lR-V°                                    (7>
where,
         I  = incident light flux
         IR = reflected light flux
         m  = total mass of gas through the filter
         Q  = number of particles per unit mass of gas
         K, = specific attenuation coefficient
         a  = area of filter paper
         b  = a number expressing the exponential
              retention of tne filter.
Equation (7) can be restated as:
                                b
             R = _R = 10  1  a                            (8)
where,
         R = absolute reflectivity of the stained spot.
Taking the logarithm of Equation (8) gives:


             Iog10£ - DR = Kl Ab                      (9)
                               \a /                       \ /

where DR is the optical reflection density of the soot deposit,
The validity of this equation may be tested as follows:
         Taking the logarithm of Equation (9) we have:
             log DD = b(log m + log Q - log a) + log K,   (10)
                  K                                   J-
In a given smoke sampling system, the area, a, of the filter
and K-, are constants.
                                11

-------
         Equation (10) shows that if Q is constant at a given
power setting and different masses of gas are filtered, a
straight line will result when log DR is plotted against log

m.  The slope of this line will be b.  If the power setting is
changed, resulting in a different value of Q, another straight
line with slope b will result, providing Q remains constant as
m is varied.  Data for a variety of power settings, therefore,
yield a family of parallel straight lines.  These lines are
offset because of the different particle concentrations, Q,
produced at the different power settings.  Plots of this type
(Reference 12) give a series of  parallel straight lines.  Data
presented in  Reference 13  were replotted and resulted in
parallel straight lines foj_- approach, cruise and takeoff power.
(The line for idle was poorly defined by the data points and
appeared to have a slightly different slope.  This is not
surprising since the smoke numbers for this low power setting
are less accurate.)  The importance of the relationship is that
the slope of the lines is dependent on the filtering character-
istics of the filter.  Thus, the filter must be specified.
Changes in filter specifications will be difficult to accommo-
date.

              (3.3.2)  Optical Transmission of Smoke Plume

              The transmission of light through a smoke cloud
may be represented by the expression:

             IT = IQ 10"C L K2                            (11)


where,
         I  = incident flux
         IT = transmitted flux

         Ko = specific absorption coefficient
         L  = optical path length

         C  = concentration of particles/unit volume of gas
              in the plume.

Equation (11) may be rewritten as:


             T = -I. = 10"C L K2                           (12)
                 I
                              12

-------
vhere,

         T = transmission

         or:

         Iog10 1 = DT = C L K2                            (13)


where D™ is the optical transmission density of the plume.

Equations (12) and (13) are universally accepted expressions
and need no test for validity.

             (3.3.3) Relating DR to D™ and Operating Parameters

             Neither Equation (9) nor Equation (13) are parti-
cularly useful for the study of engine exhaust because they
contain the quantities Q and C.  Neither of these expressions
are measured directly by the filter stain technique or by the
transmissometer.  However, the reflection density of the filter
stain can be related to the transmission density of the plume
at the engine nozzle through the use of engine parameters and
measured values.  Thus, theory may be tested by experimentation

             If each unit mass of gas and unit volume of gas
issuing from the nozzle of an engine at fixed operating condi-
tion contains the same number of particles (this is not always
strictly true, for at times the smoke issues in a series of
puffs) Equation (14) can be written:

             N = MQ                                       (14)

where,

         M = mass of gas per unit time passing through the
             engine

         N = number of particles per unit time in the exhaust.

             The particle concentration/unit volume, C, is:


                                                          (15)

where,

         V = gas velocity (or particle velocity)
         A = area of the nozzle.
                               13

-------
Substituting this value for C in Equation  (13) gives
DT =
 1
                    L K
                 AV
(16)
Rearranging the above equation to solve for Q gives:
            Q - DT (-43L)
                 1  MLK2
                                             (17)
Substituting this value for Q into Equation  (9) gives:

                                b
         n
         °
                           AV
            n  = K
            DR   Kl   T a MLK2
(18)
Equation (18) relates the filter stain reflection density to
the transmission density of the plume through the mass flow
M, the area A of the nozzle, and the velocity V of the gas
flow at the nozzle.  This equation may be simplified if the
relationship between velocity and mass flow is substituted.
reasonable assumption for velocity in subsonic flow is:
V = K,
                   M
                                                          (19)
where K-, is the reciprocal of the gas density,
for V in Equation  (18) gives:
            DR = K
         D
          T a L K
                                    Substituting
(20)
Equation (20) establishes the relationship between the filter
stain density and the transmission density of the plume.  The
optical path length L is the diameter of the nozzle.  The
optical density is also measured at the nozzle.  The mass flow
M of the engine has been eliminated by the use of Equation  (19)
                              14

-------
        Taking the log of Equation (20) gives:


        log DR = b (log DT - log L + log m - log a + log K-j


                 - log K2) + log Kx                      (21)


A simple means for testing the validity of Equation (21) would
be to plot the reflection density of the filter stain DR against

the transmission density DT of the plume on log-log paper for

a series of values of Q resulting from running the engine at
different power settings.  The quantities m and a are held
constant in the smoke sampling system as the power setting is
varied.  All other parameters in the equation are constant
including b which is related to the retention of the filter
material.  If the equation is valid a straight line with slope
b  should result.  By substituting into Equations (20) and (21)
known values of all parameters and measured values of plume
transmission and corresponding values of filter stain reflection
density it should be possible to evaluate the constants and
predict the optical transmission from the filter stain for any
engine in terms of DD and the nozzle diameter L.
                    K

             (3.3.4)  Relating Smoke Number to Filter Density

             The smoke number is defined as:

            	        R
            SN = (1 - -5^)100                             (22)
where,

        R  = diffuse reflectance of the smoke spot
        R  = diffuse reflectance of clean filter paper.

Restating Equation (21), we have:

                   100                                   (23)
            Rs   100-SN

        or,

            log1Q ^ = DR = 2 - Iog10 (100-SN)           (24)
                               15

-------
where,

        DR = reflective density of the smoke stains.

       ^Equation  (24) gives the relationship between smoke
number "SN and reflection density, DR>  Hence by using this
value and Equations (20) or (21) it should be possible to pre-
dict the transmission of the smoke plume as a function of SN
and nozzle diameter L.

        (4.)  Static Tests of Engines and Aircraft

        Static engine and aircraft tests were performed at
NAFEC by both IITRI and FAA personnel.  A number of different
engines and aircraft were studied.  These were: 1) a J-57-P-
37A Pratt and Whitney Engine mounted on a Static, Sea Level,
Open Air Test Stand; 2) a JT-12-6 engine mounted as a flight
installation in a test wind tunnel; and 3) several aircraft
tied-down on the National Guard run-up area at the NAFEC/Atlantic
City Airport.  These aircraft were an F-100, a CV-880,and a
Lockheed Jetstar.

        Various types of instrumentation were applied on each
of the tests so that correlations between the results were
possible.

             (4.1)  J-57-P-37A Engine Mounted in the Test Cell

             The J-57 engine instrumentation consisted of an
IITRI transmissometer to obtain smoke transmission data, the
G.E. smoke sampler  (Appendix D) to obtain smoke numbers, and
the IITRI smoke sampler to obtain particle size data and smoke
numbers.

             Table  3 summarizes the data obtained.  The trans-
mission value reported is the one-way transmission as deter-
mined by the transmissometer and was obtained by taking the
square root of the  two-way measured-transmission value.  The
SAE smoke number is calculated as described in Appendix D.
The other smoke numbers reported are variants of the SAE smoke
numbers.  They were obtained by substituting a Millipore filter
for the standard Whatman filter or using a white backing rather
than a black backing when measuring the reflectance of the
filter deposits.  For all power settings, no visual detection
of smoke could be made by looking across the exhaust at 90° to
                             16

-------
the jet nozzle.  However, the irregular dark background of the
test cell was not  conducive to visible observations.   Smoke
could be seen only by looking along the plume path.   It appeared
as puffs with low obscuration values,  even at the higher power
settings.

         Data in Table 3 show  that smoke numbers collected
through the IITRI probe are nearly identical to those  collected
through the NAFEC probe when both are  located at the nozzle.
Since the NAFEC probe was designed for isokinetic sampling, it
is concluded that isokinetic sampling  is not required  for smoke
number determination.  Since both probes gave similar  results
at the nozzle, the differences between the smoke numbers obtained
as the IITRI probe was moved downstream from the nozzle reflect
the effect of plume dilution and expansion.

              (4.2)  Light Transmission of J-57 Engine Exhaust
                     for Various Path  Lengths

              Visual observations of smoke plumes with both
tied-down aircraft and aircraft in-flight indicated  that the
visibility of the smoke was dependent  upon the viewing geometry.
It was observed that a plume may be invisible when viewed at
right angles to the plume axis, but visible as the plume was
observed at other angles.  The phenomena is due to the increase
in the attenuation caused by the longer optical path through
the plume.  To illustrate this phenomena under controlled con-
ditions, experiments with the J-57 test cell engine  were con-
ducted.  The transmissometer was used  to measure the light
attenuated by the plume at four angles, 90°, 45°, 30°, and 19.5°,to
the plume axis.  These correspond—to 1, 1.4, 2 and 3 --times the
normal (90°) path length through the plume.  The engine was
operated at a series of selected power settings for  each angle.
The results of these measurements are  summarized in  Table 4.

         A distinct decrease in transmission as the  path length
through the smoke is increased is noted.  If the visibility
threshold is taken to be 98% transmission (see page  9  ) then
a smoke which is invisible when viewed at 90° may become clearly
visible when viewed at other angles.  The results are not pre-
cise, because at the smaller angles (30° and 19.5°) it was neces-
sary to place the corner reflector approximately 30  and 50 ft.
downstream from the nozzle of the engine to protect  it from the
blast.  The plume had become much wider at this point.  Never-
theless, the principle of increased absorption with  path length
is clearly demonstrated.
                               17

-------
                                                                   TABLE  3
Test Number
IITRI NAFEC
~
  I
  8
  9
  2
  3
  4
  5
 10
 11
 12
 13
 J4

(1)

(2)
(3)
(4)
(5)

(6)
       31
       33
       35
       37
       23
       25
       27
       29
       39
       41
       43
       45
       47
Distance in
Nozzle^Dia.
From Engine
To IITRI Probe
nnd Transmisso.
     0
     0
     0
     0
     2-1/2
     2-1/2
     2-1/2
     2-1/2
    10
    1C
    10
    10
    10


SMOKE
NUMBERS AND TRANSMISSION DATA OF JET
SMOKE




J-57 ENGINE MOUNTED IN TEST CELL
Smoke Numbers (l>
G. E. Smoke Sampler
Whatman Filters^2)
Engine
Gas Flow
Lbs/Sec
53
' 94
114
145
53
94
120
148
53
94
114
145
168
Engine
Thrust V
Setting E
Idle
Approach
757. Norm,
Cruise
Idle
Approach
757. Norm,
Cruise
Idle
Approach
757, Norm
Cruise
Takeoff
NAFEC
Ihite
tackinj
21
36
. 49
55
19
35
. 50
58
21
35
. 48
55
58
Probe
Black (t
> Backing^
12
27
. 40
47
11
26
42
51
14
27
40
48
51
IITRI
, , White
4;Backin
16
34
48
54
19
33
42
--
10
22
34
39
--
Probe
Black
£ Backing
25
41
47
11
24
34
--
6
15
. 25
31
--
Millipore Filters^
NAFEC
White
Backing
10
26
38
49
10
24
43
54
10
24
43
48
-.-
Probe
Black
j Backi ng
10
25
37
48
9
24
42
53
10
23
42
46
--
IITRI
White
Backing
6
21
41
47
8
20
31
--
4
13
21
28
--
Probe
Black
j Backing
6
18
40
46
8
18
30
--
4
12
20
27
--
IITRI Sampler

Millipore Fllcer8(3) IITRI
IITRI Probe
White ,,,.
Backing13-'
9
68
94
>100
6
30
59
58
0
9
26
40
40
Transmissi
Percent
Transmiss
99.3
98.4
96.3
95.6
99.0
97.3
95.8
96.0
99.5
98.1
97.0
95.0
89.4(6)
     The NAFEC probe was always located at the engine nozzle.  The IITRI probe was located at the distances from the engine nozzle given
     in column 3.  All G. E. smoke sampler data were collected and analyzed by NAFEC personnel.
     Whatman filter data were obtained by interpolating reflectance readings to a filtration density of 0.3 ft /in .
     Millipore fiiLei: date were obtained by interpolating reflectance readings to & filtration density of 0.0565 ft /in  .
     The SAE smoke number.
                                                                                                                                      3    2
     Because the IITRI sampler filtered diluted exhaust it was necessary to extrapolate the data to a filtration density of 0.0565 ft /in
     at thrust settings other than idle.
     This value appears to be erroneous.  The value of 94.27, reported in Table 4 appears more reasonable.

-------
                    TABLE 4
LIGHT TRANSMISSION AS A FUNCTION OF PATH LENGTH
          J-57 Engine in Test  Cell
: Engine
Thrust
Setting
Idle
Approach
75% Norm.
Cruise
. Takeoff
j
Percent Transmission
Relative
1
98.2
97.0
95.3
94.3
94.2
Optical
1.4
99.0
97.0
94.5
92.4
91.6
Path Through
2
98.9
96.4
92.0
89.0
88.0
Plume ^
3
97.8
93.0
85.8
82.2
79.6
'(1)  Relative paths correspond to angles of 90
     and 19.5°  to the plume axis.
                                           45°  30
                        19

-------
             (4.3)  JT-12-6 Engine Mounted in Wind Tunnel

             The JT-]2-6 engine instrumentation consisted of
the IITRI transmissometer,  GE smoke sampler, and IITRI smoke
sampler.  No visual or photographic observations of smoke could
be made.  The transmissometer was located at a distance of
7-1/2 nozzle diameters downstream from the engine nozzle and
viewed the smoke plume through windows located on opposite sides
of the tunnel.  The IITRI smoke sampling probe was also placed
7-1/2 nozzle diameters from the engine.  It was connected to
the dilution apparatus by a heated sampling line.  This was
necessary because personnel safety considerations prevented
the placement of the dilution apparatus just outside the wind
tunnel at the point where the probe was placed.

             Table 5 summarizes the data obtained on the JT-12-6
engine.  The JT-12-6 engine gave transmission values exceeding
98% at all power settings.   Based on the modeling study, this
engine should not produce a visible smoke.  The SAE smoke
numbers were less than 25 at all power settings.

             (4.4)  Tied-Down Aircraft Tests

                    (4.4.1)  F-100 Single-Engine Aircraft

             An F-100 aircraft, operated by the New Jersey
National Guard, was obtained for study.  This aircraft has a
J-57-P21A engine, similar to the engine tested in the test cell.
The plane was tied-down at  the National Guard run-up area at
the NAFEC/Atlantic City Airport.  The exhaust was analyzed with
the IITRI transmissometer,  the GE smoke sampler, by photographic
photometry,  and by visual observations.  A photograph of the
instrumentation is shown in Figure 3.  The test results are
summarized in Table 6.

             The smoke numbers of the F-100 engine are similar
to those obtained with the  test cell engine.  The smoke number
at cruise is higher than at takeoff.  This inconsistency is
due to the movement of the  aircraft at takeoff power.  The
nose wheel depresses and the tail of the aircraft rises at
takeoff power; thus, the location of the sampling probe in
relation to the engine nozzle is disturbed.  The data show a
a distinct correlation between transmission and photographic
                             20

-------
                                                                  TABLE 5
                                               SMOKE NUMBERS AND TRANSMISSION DATA OF JET SMOKE
                                                   JT-12-6 ENGINE MOUNTED  IN WIND TUNNEL
                                                                            Smoke Numb
                                                                                      ers'  '
Test Numt.
IITRI
WT- 1 , 2
WT-4,5,7
WT-6,8,9
,er
TIA"FEC
77,78
80,81
82,83
Engine
Cas Flow
Lbs/Scc
31
46
56
Engine
Thrust
Setting
Approach
Cruise
Takeoff
NAFEC 1
White
Backing
6
29
'IJ

C.
Whatman Filters*2'
Probe
Black „,
Backing^ '
21
25
IITRI
1 Backing
5
18
E. Smoke
Probe
Black
Backing
t,
10
12
Sampler
Ml
lllipore Filters*3'
NAFECTrobe
White
Backing
2
21
30
Black
Backing
3
?0
. 29
1 1 TB I 1
White
Backing
1
R
11
probe
Black
Backing
2
7
10
IITRI Sampler
Milllpore Filtei
IITRI Probe
White ,,.
Backing13''
J
n
17
r* •' IITRI (6)
Transmissometer v"'
Percent
Transmission
99.1
98.3 i
98.1 '•
(1)   The NAFEC probe w«»  always  located  at  the  12  o'clock  position  at  the  engine nozzle.  The  IITRI probe was always  In  the  center
     of the wind tunnel  7-1/2  nozzle  diameters  from the  engine.  All C.  E.  smoke sampler data  were collected and analyzed  by NAFEC  personnel.
(2)   Sec note 2, Table  3.
(3)   See note 3, Table  3.
(ft)   The SAE smoke number.
(5)   See note 5. Table  3.
(6)   Transmlssomett-r located 7-1/2  nozzle  diameters from engine  nozzle.  Path  length was the diameter  of the wind  tunnel at  this  location.

-------
                                            TABLE 6
                          SMOKE QUALITY FROM A TIED-DOWN F-100 AIRCRAFT
                                         J-57-P21A ENGINE
Engine
Thrust
Setting
Idle
Approach
Cruise
Takeoff
Engine
Gas
Flow
Lbs/Sec
53 .
94
137
159
IITRI
Transmissometer.. ^
70 Transmission^ '
Photometry ,,v
70 Transmission^ '
98 100
97
95
94
100
<100
91-94
Smoke Number
Whatman Filter
Black ^W01116
Backing^ 'Backing
9 15
23 31
54 62
48<5> 58<5>
Visual Transmission
Across Plume. % (3)
At Well Behind
Nozzle Aircraft
Clear Clear
Clear Clear
Clear 90-95^
Clear 90-95^
CD
     The  transmissometer  viewed  the  exhaust at the nozzle and at right angles to  the  direction
     of plume  travel.  The  photographs were taken similarly by lying on the ground and  photo-
     graphing  the  exhaust against a  sky background.
     SAE  smoke number.
     Viewed against  the normal background of the sky horizon.
     Smoke  is  not  continuous  but billowy.
(5)   Low  smoke numbers are  attributed to the movement of the aircraft at takeoff  power.
(2)
(3)

-------
                 -
                                                 "
•

                                                        .«—
                                                   fete   :,,;,
                                                 :w&*
                                                                   •

                                •
        ' S'«**fe                      ,5    jf^
                   •i^t.«
           FIGURE 3.  SMOKE  INSTRUMENTATION.

-------
 photometry.   Also,  smoke with  transmission values higher  than
 9770  and  smoke number  values  less  than 23 are  invisible.   Smokes
 with transmissions  less than 97%  and smoke numbers above  23
 are  visible.

               (4.4.2)  Convair  880 Aircraft

               The left outboard engine of a CV-880 aircraft,
 operated by  the  FAA,  was studied  with the same methods used for
 the  F-100 aircraft.   The engine was a CJ805-3B.  Results
 are  summarized in Table 7.  The data show excellent agreement
 between  the  transmission values obtained by the transmissometer
 and  by photographic photometry.   Plumes with  transmission values
 of 98% and SAE smoke  number of  23 or less were not visible.

               Conditions were  ideal for photographic photometry.
 The  plume was well  above the horizon, and the  sky density  below
 the  plume was essentially  the  same as above,  For these reasons,
 a takeoff microdensitometer  trace is shown in Figure 4.   The
 values presented in Table  7, as wel-l as the other tables  re-
 porting  photometry  data, are for  the maximum  decrease in  optical
 density.   For the illustrated  example, this value is in the
 center of the plume.  A positive  print of the negative from
 which the microdensitometer trace was made is shown in Figure 5.

               (4.4.3)  Multiple Engines of Convair 880 Aircraft

               Multiple engines  of the Convair 880 aircraft were
 tested using  the IITRI transmissometer, photographic photometry
 and  visual estimates  of obscuration.  These results are summar-
 ized in  Table 8.  In  most  cases,  the transmission measured
 photographically is less than  that measured with the transmisso-
 meter.   This  is  apparently due  to ground interference with the
 plume.   The  height  of the  transmissometer was approximately that
.of the engines.  The  photographs, however, show the plume down
 to the horizon line which  is well below the line of sight of
 the  transmissomrter.  The  microdensitometer traces, Figure 6,
 show a continuous increase in  the density down to the horizon
 line.  Since  the obscuration values given in  the tables are
 the  maximum  obscuration values  (at the ground line) they  will
 be higher than the  transmissometer values because the trans-
 missometer was fixed  at the approximate average exhaust height
 with the engines.  At the  time  the test was made the influence
 of engine tilt and ground  dirt  was not fully  recognized.
                              24

-------
                                                TABLE  7

                  SMOKE QUALITY FROM THE NUMBER 1  ENGINE  OF  A  TIED-DOWN CV-880 AIRCRAFT

                                             CJ805-3B  ENGINE
Engine
Thrust
Setting
Idle
Approach
Cruise
Takeoff

Engine
Gas
Flow
Lbs/Sec
42
124
147
163

IITRI
Transmissometer
Reading m
% Transmission^ '
98
89
87
87

Photographic
Photometry ,~^
7o Transmission^- '
96
87
85
86

Smoke Number
Whatman Filter
Black m White
Backing^ ' Backing
23 32
60 67
' 69 73
65<6) 70(6)

Visual Transmission
Across Plume. %
At Well Behind
Nozzle Aircraft
100 100
90-95 80-90^3^
90-95 75-85(4)
90-95 75-85(5)

to
   (1)  See note 1,  Table 6.

   (2)  SAE smoke number.

   (3)  Smoke is not continuous  but  billowy.   The  smoke puffs  have  a  transmission of 80%.
        Integrating  the clear sky  and  the  puffs  yield  an estimated  visual  transmission  of
        80-90%.   Viewing the  smoke from in front of  the engine,  the transmission  varies
        from 40-60%.

   (4)  Integrated puffs and  background sky give transmission  values  of  75-85%.   When
        viewed from  in  front  of  the  engine the transmission  is 40%.

   (5)  Smoke is more uniform in composition,  not  so puffy.

   (6)  Low smoke numbers are attributed to movement of the  aircraft  at  takeoff power.

-------
        Decreasing Density
                _ . • r-r	j- -
                    -  Ground Surface
FIGURE 4 . MICRODENSITOMETER TRACE OF
         NUMBER 1 ENGINE OF CONVAIR 880
         AIRCRAFT - TAKEOFF POWER.
                   26

-------
N
-J



                             -S ,


                                                    akaflJJL... f'-- .'•.. _ Jifete.,.
                                                                                   ,, ,,.„.>,,,,«,,.?--.•.• l^;*>l?'r*wS*T*?>
                                                                                                       _;
                     FIGURE 5.   NUMBER 1  ENGINE OP CONVAIR  880 AIRCRAFT.   TAKEOFF POWER.

-------
Idle Power
Approach Power
Cruise Power
Take-Off Power
       FIGURE 6. HICRODENSITOMETER TRACES FROM TIED-DOWN
                 CONVAIR 880 AIRCRAFT - ENGINES 1,  2,  and 3
                 OPERATING.
                                28

-------
                                              TABLE JB
                       CV-880 AIRCRAFT TIED-DOVfN MUI.TIPLE KNGINE OBSERVATION

Engine :
Number in>.-.v
Operation^ '
1
1
1
1
1-2
1-2
: 1-2
;
1-2
; 1-2-3
1-2-3
1-2-3
1-2-3
1-2-3-4
1-2-3-4
1-2-3-4
1-2-3-4

f Engine
Thrust
Setting
Idle
Approach
Cruise
Takeoff
Idle
Approach
Cruise
Takeoff
Idle
Approach
Cruise
Takeoff
Idle
Approach
Cruise
Takeoff

IITRI Transmissometer |
% Transmission (2) I
99
91
89
89
98
81
74
71
95
65
52
43
94
61 1
i
38 !
	 30 	 __j

Photographic i
Photometry
7<> Transmission
96
87-94
85
86
91
63
49
49
95
49
38
37
94
44
28
38 0)

Visual Transmissions
7o Transmission
100
80-95
75-85
75-85
100
	
	
i
100
85
80-85
75-80
100
70-75
70-75
70-75
(1)   The number 1 engine is the left outboard engine.
(2)   Transmissometer located past the tail of the aircraft at the nozzle as in previous test
(3)   It is believed that the power was being reduced when the photograph was taken.

-------
All values obtained with either technique show a similar varia-
tion with power  setting and number of engines.

              The  results of the multiengine study were largely
nullified because  of ground dirt becoming airborne during the
tests.  The aircraft was positioned at the run-up area with
engines 3 and 4  exhausting over the ground rather than concrete.
The blown ground dirt and the downward tilt of the plume is the
reason why the microdensitometer traces continue to increase
in density down  to the horizon.  However, the effect of com-
bining smoke plumes should be predictable from consideration of
increased path length.

               (4.4.A)  Lockheed Jetstar Aircraft

              A  Jetstar, owned by the FAA, was observed in the
same manner as the Convair 880 aircraft.  This aircraft's engine
is similar to the  JT-12-6 engine studied in the wind tunnel.
No smoke was visible at any angle of viewing and no smoke
could be detected  photographically.

              No transmission data were obtained with the 1ITR1
transmissometer  because of an amplifier malfunction.  Repairs
could not be completed in time to make the tests.  Smoke numbers
for the Number 4 engine were measured with the GE smoke sampler
and the results  are shown in Table 9.

              Results show that probe location must be studied
if representative  smoke numbers are to be obtained.  Also, a
single probe location may not be a representative location for
all power settings.  For example, the 3 o'clock probe position
gave smoke at takeoff averaging about 3070 lower than the 6 o'clock
position.  At approach, the 3 o clock position gave smoke numbers
25-30% higher than the 6 o'clock position.  The SAF. smoke
numbers were below 23 at the 3 o'clock position for all thrust
settings.  The SAE smoke number at the 6 o'clock position
reached 35 at takeoff thrust.

               (4.5)  Discussion and Summary of Results of
                     Static Engine and Aircraft Tests

             .This discussion will show the empirical rela-
tionship between the visibility, luminance, and smoke number.
Table 10 is a collection of data taken from Tables 33 5, 6, 7
and 9, arranged  to show more clearly the comparison of smoke
                               30

-------
                           TABLE 9



            SMOKE NUMBERS FROM THE  NUMBER 4  ENGINE
OF A TIED-DOWN JETSTAR AIRCRAFT

I Engine
' Thrust
Setting
Approach
Cruise
Takeoff
i




Smoke Number 1
Probe at
6 O'clock
Backing
Black(l) White
8.6
30.0
i 35.0
1
12.0
37.0
42.5
Probe at 3 0
'Clock
Backing
Black(l) White
10.7 15
23.0 30
23.0 31
.5
.9
i
.7
(1)   SAE Smoke Number
                              31

-------
number, transmission and visibility.  Figure 7 is a plot of the
measured transmission against the smoke number.  These values
of transmission were measured at right angles at the nozzle on
a single engine.  As discussed on page 17 ,  we assume that smoke
with transmission values greater than 98% will be invisible.
In Figure 7 a line at 98% transmission is shown to divide the
plot into two parts.  Above this line, the exhaust is invisible;
and below the line the smoke is visible.   It will be seen that
SAE smoke numbers less than 23 are invisible.

         A comparison between transmission measurements made
with the transmissometer and with photographic photometry,
Table 10, indicates that both techniques  lead to similar values
within experimental error.  Hence, photographic photometry is
a valid measure of smoke plume transmission.  Photographic
photometry should therefore result in true transmission values
when applied to aircraft in flight,

         A comparison of Columns 4 and 5  with Column 6 of Table
10   show  a   reasonable agreement between the transmission
measurements and visual estimates of transmission.

         (5.)  In Flight Observations

              (5.1)  Flights Evaluated at NAFEC

              A number of different aircraft were studied
in-flight at NAFEC/Atlantic City Airport, Figure 8.  Photographs
and visual estimates of the obscuring power of the smoke were
made.  The aircraft were observed during  practice landing
operations on runway 13-31.  The observers were stationed at
the junction of taxiways G and F.  Aircraft maneuvers consisted
of touch and go landings and takeoff or low altitude fly-bys
at about 500 feet.

              (5.1.1)  F-100 and F-105 Aircraft In-Flight Tests

              An F-100 aircraft made a number of low altitude
fly-bys at several power settings with and without afterburners.
A typical photograph is shown in Figure 9.  Results of the.
visual observations and the photographic  photometry measurements
are given in Table 11.  Two passes by an  F-105 aircraft are
reported in Table 12.
                              32

-------
c
o
•H
w
CO
s
u
H
g
o
u
0>
PL,
     100
      oo
96
94
      92
90
      88
      86
              tfl
                   J.U. i
                   IE!
                  im
                             4-i—
                             tfri-
                             rtrr
                          33
-B
                                               ritf
       0    10   20   30   40    50   60   70   80


                 SAE Smoke Number



               OJ57 in test cell

               XJT-12 in wind tunnel

               QF-100 aircraft  tied-down

               VCV-880 aircraft #1  engine tied-down





               (All Data Taken  from Table 10)
       FIGURE  7.
             PLOT  OF SMOKE NUMBER vs
             OPTICAL TRANSMISSION.
                          33

-------
                        SUMMARY OF SMOKE NUMBER
AND TRANSMISSION FOR STATIC TESTS


I
Engine > Engine
Thrust
Setting
SAE m
Smoke *> '
Number

Transmission,
IITRI
Transmis-
someter
J-57 on Idle 16 i 99.3
Test Stand Approach 27 98-4
75° Norm. 41 96.3
'
Cruise
i Takeoff
JT-12 in ' Approach
' Cruise
Tunnel

F-100
Tied-Down


CV-880
Tied-Down
#1 Engine


Jet-Star
#4 Engine

Takeoff
Idle
Approach
Cruise
Takeoff
Idle
Approach
Cruise
Takeoff
Approach

49 95.6
51 94.2(2)
4 99.1
21 98.3
25 98.1
9 98
23 97
54
48
23
60
69
65
9.8(3)
Cruise 26.5
95
94
98
89
87
87
__
--
Takeoff 29
i • I
(1) Smoke number for black backing and Whatman #4
(2) No data was obtained for takeoff power at the
Photu.
Trans-
mission
..
--
—

--
--
—
—
—
100
100
<100
91-94
96
87
85
86
100
100
100

filters.
time the

% |
Visual i
Estim. i
Trans . '
„
--
-.

—
--
--
--
--
100
100
90-95
90-95
100
80-90
75-85
75-85
100
100 i
100
i

other
(3)
test data was obtained (Table 3) due to the high noise level and
vibration causing a malfunction of the voltmeter.  The value shown
here is from Table 4.

Smoke numbers for Jetstar are averages from black backing values
at the two probe positions.  See Table 9.
                                   34

-------
                                          SCALl
FIG. 8 NAFEC/ATLANTIC CITY AIRPORT, ATLANTIC CITY, NEW JERSEY
                                   35

-------


                                          ••'•
                          »

                                                             .
  F-100


    .
          v-  .^r.ss
          •


           -
                 .
CV-880

                                                     -
  CV-880
CV-880
FIGURE 9.  PHOTOGRAPHS OF F-100 AND CONVAIR 880

               AIRCRAFT  IN FLIGHT.
                          36

-------
         OBSERVATIONS ON THE F-100 AIRCRAFT IN FLIGHT AT NAFEC

(l\
Pass v '
Number
1
2
3
4
5
5
6
7
.8
: 8
9
9
: 10
10
11
11
12
12
: 13
13
14
14

Thrust
Setting
Takeoff(5)
Takeoff
Military
Military
Cruise
Cruise
Military
Military;. •
Military
Military
Military
Military
Military
Military
Military
Military
Cruise
Cruise
Cruise
Cruise
Cruise
Cruise

Viewing*- '
Angle
90°
90°
90°
90°
90°
70° (6)
90°
90°
45° (7)
90°
30° (7)
90°
20- 30° (7)
90°
30° (7)
90°
20° (7)
90°
10- 15° (7)
90°
10° (6)
90°


Percent Light Transmission Through Plume
Photographic (3)
100
100
94
93
96
92
96
95
89
95
86
96
89
94
90
95
87
95
80
95
90
95
Visual (4)
100
100
90
90
90
90
85-90
85-90
70-80
70-80
50
80
50
80
50
80
60
85
60
85
60
85
(1)   No.'s 1-7  on one film and 8-14 on other film taken on different days.
(2)   Viewing angle -  angle made by line of sight and plume axis.
(3)   Transmission measured by photographic photometry (Appendix B).
(4)   Transmission as  visually determined at time of flight.
(5)   With afterburner.   All other data are without afterburner.
(6)   Aircraft flying  away from observer, after  landing exercise.
(7)   Aircraft approaching observer during landing or fly-by exercise.
                                   37

-------
                         TABLE 12
    UbbEKVATIONS ON THE F-105 AIRCRAFT IN FLIGHT AT NAFEC
L
! . . .
Pass Estimated
Number ! Thrust
i Setting
I
1
1 i Military
2 j Military


i • 	 	 -------
Viewing Percent Light Transmission
Angle Through Plume
Photographic(l)
90° 96
90° 95
Visual(2)
85-90
85-90
i
(1)   Transmission measured by photographic photometry (Appendix B)



(2)   Transmission is visually determined at time of flight.
                               33

-------
              (5.1.2)  Convair 880 Aircraft In-Flight Tests

              Similar tests were made with the FAA Convair 880
aircraft.  The data is summarized in Table 13.  The data pre-
sent an outstanding example of how smoke density increases as
the smoke path is increased due to the position of the observer
in relationship to the smoke track.  When viewed at right angles
visual transmission values of about 80-9070 are typical.  When
viewed at 45°, visual transmission is on the order of 50% or
less.  Figure 9 shows some photographs of the CV-880 aircraft
in-flight.  As seen from the photographs the air during this
series of flights seemed excessively turbulent.

              (5.1.3)  Lockheed Jetstar Aircraft In-Flight Tests

              No smoke was visible to the eye and no smoke was
detectable from photographs of the Jetstar.  This aircraft's
plume appeared clear under all conditions of viewing.

              (5.2)  Observations of Aircraft In-Flight at
                     Chicago's O'Hare Airport

              Table 14 summarizes the results of photographs
of commercial aircraft in-flight taken at O'Hare Airport on
September 25, October 23, and November 26, 1969.  Most of
the photographs were taken of aircraft during takeoff but s.
few, as noted, were taken of landing aircraft.  Typical photo-
graphs are shown in Figure 10.  The location of the observer
relative to the plume path was far from ideal.  In most instances
the observer and camera were located at an undesirably long
distance from the flight path.

              (5.3)  Discussion and Summary of In-Flight Tests

              The flights observed at NAFEC show that the visual
estimates of transmission are slightly lower (5-1070 lower) than
the photographic measurement.  This discrepancy did not appear
in the tied-down tests previously described and is probably due
to the subjecture effect of a small plume viewed against a large
expanse of light sky.  Photographic transmission measurements
made at 90° to the plume axis are always higher than those made
at lesser angles.  A similar trend may be seen in the visual
estimates.  Cere was taken to evaluate visually the change in
transmission with the viewing angle at NAFEC.  The importance
of this factor was not fully recognized in the earlier tests
at O'Hare Airport.
                              39

-------
                          TABLE 13
OBSERVATIONS ON THE
CONVAIR 880

AIRCRAIT IN FLIGHT AT NAFEC

Pass
Number

Thrust
Setting
1 Takeoff
i
1 Takeoff
1 ' Takeoff
2 ; Approach
2
2
2
3
3
3
3
3
4
Approach
Approach
Approach
Approach
Approach
Takeoff
Takeoff
Takeoff
Approach
(1) Transmission

Viewing
Angle
90°
45° (5)
0°(5)
45° (4)
90°
5- 10° (5)
0° (5)
5-10° (4)
10- 15° (4)
90°
45° (5)
20° (5)
0-5° (4)


Percent Light Transmission
Through Plume
Photographic (1)
70 & 82(3)
45
52 & 70(3)
62 & 83(3)
81
44
-70
65 & o 7
70
70
44
52
44
Visual (2) |
50-70
50-70
I 50-70
50
85-90
i
t
	 I
	
80
60-80
80
20
30
50
measured by Photographic Photometry (Appendix B)
(2) Visual estimate of transmission made at timr
(3) Two distinct
of flight.
trails appear on photograph-transmission given
     for each trail.

     Aircraft approaching observer during landing or fly-by
     exercise.

'(5)  Aircraft flying away from observer after landing exercise.
                              40

-------
PHOTOGRAPHIC MEASUREMENT OF JET SMOKE

TRAILS
FROM AIRCRAFT IN FLIGHT AT

O'HARE AIRPORT AND VISUAL ESTIMATES OF TRANSMISSION

Type and
Operational
Mode of
Aircraft(l)
727


727


DC9


727

727
727
: 727*
; 727
707
737
! 727
: 727
i
i
; 707

1 - -.
I 2 engine
Distance
Behind
Nozzle(2)
at nozzle
25 ft.
100 ft.
at nozzle
75-100 ft.
100-200 ft.
at nozzle
50 ft.
150 ft.
at nozzle
100 ft.
at nozzle
at nozzle
at nozzle
at nozzle
at nozzle
at nozzle
at nozzle
at nozzle
100 ft.

at nozzle
100 ft.
200 ft.
at nozzle
I Number
! of
Plumes (3)
Percent Light Transmission
Through Plume
Photographic (4) Visual(5)
1 67
1 ; 69
1 i 83
3 83,80,84
25



2 76,82
1
1
. 1
1
1
' 1
; 1
1
1
1
1
1
2
1
1

1
1
1
1
76
58
65
73
50
73
68
70
76
74
86
100
74,83
68
73

86

50


40

35
10
30
35
70
45
35
30


55
86
89
77
35
                  41

-------
Type and
Operational
Mode of
Aircraft(l)
2 engine
2 engine
2 engine*
DC9
DC9
727*
i
707*
727
i
i
i
' Distance
Behind
Nozzle(2)
at nozzle
at nozzle
at nozzle
at nozzle
at nozzle
at nozzle
j at nozzle
[ at nozzle
!
Number
of
Plume (3)
1
1
1
1
1
1
1
1
Percent Light Transmission '
Through Plume
Photographic (4) Visual (5)
93
76
73
76
75
63
93
73
60
20
25
20
20
10
85
15-25
(1)   Asterisk identifies those aircraft landing  at  O'Hare.   All  other
     aircraft were observed during takeoff.   All aircraft are  jet
     aircraft.

(2)   Estimated distance behind engine  nozzle that the  photographic
     transmission measurements were performed.

(3)   Number of smoke   plumes at point  where  photographic transmission
     measurements were performed.

(4)   Transmission measured by photographic photometry  (Appendix  B).
     The values are for a 90° viewing  angle.

(5)   Transmission as  visually determined at  time of light.   The
     effect of optical path length was not fully appreciated at  the
     time these measurements were  performed.  While the visual measure-
     ment was made "on mark" from  the  camera operator  in effect  the
     visual observer  was mentally  measuring  the  smoke  density  up until
     'the time the photographs was  taken. The visual transmission values
     therefore are not instantaneous but, intregated over a  short
     period of time.
                                   42

-------





                              ..
                                                              •V
                                    ,

                                                        ,-


727 Jet - Ringelmenn Number 4-1/4     707 Jet - Ringelmarm Number 3/U
          September 25, 1969.                   September 25,  1969.
              FIGURE 10.  PHOTOGRAPHS OF JET EXHAUST

                          TRAILS DURING TAKOFF AT O'HARE

                          AIRPORT.
                               43

-------
        At O'Hare, a much greater difference exists between the
photographic and visual estimates.  The Ringelmann reader
always gave a reading equivalent to 0.75 to 1 Ringelmann
numbers higher than the photographic method.  These measure-
ments were made with no previous experience on aircraft smokes
and it appears that the observer tends to exaggerate the black-
ness of the trail against the expanse of sky.  Also, the visual
measurements reported were made on command from the camera
operator at the time the photographs were taken.  Because the
smoke observer was mentally measuring the smoke density up
until the time the photographs were taken the value reported
is an  integrated rather than an instantaneous reading.  The
photographic results essentially represent a plume viewing
angle of 90°.  However, the visual observations show the
influence of viewing the smoke through greater optical path
lengths prior to taking the photograph.  For later measurements,
the observer became more critical and was aware of great dif-
ferences for various viewing angles on the same smoke trail.
There appears to be no reason for demoting the photographic
values in Table 14 because they agree with transmissometer data
obtained under controlled conditions.  Hence, the apparent
difference in transmission as reported by the visual observer
at NAFEC and O'Hare is due to the increased experience of the
observer.  It should be noted that the photographic results
were not obtained until several weeks after the observations
were made, so they could not influence the observer.
                             44

-------
                         CONCLUSIONS

          It is concluded that:

          1.  Consideretions of turbojet exhaust visibility
indicate that siuoke plumes with transmission values greater
than 9S7o will be invisible.

          2.  Correlations between the SAE smoke TV.;trb«*r and
transmissoinettn- datp indicate thpt for the engine* investigated,
smoke numbers less than 23 ere associated with invisible plumes.

          3.  Photographic photometry provides a means for
evaluating the optical density of smoke trails from in-flight
turbojet aircraft.

          ^.  A transmissometer, of the type designed for use on
this program, provides a means for evaluating the optical density
of turbojet engines mounted in test cells and on tied-down
aircraft.  Transmission data is less sensitive than the SAE
method of smoke measurement.

          5.  A Ringelmann smoke observer tends to overestimate
the obscuration properties of jet smoke.  This is believed due
to several factors.  Among these factors are the relative
crudeness of the Ringelmann scale, the wide field of vision,
the accentuated visibility of line targets, end the increase
in the optical path length through smoke, when the plune is
viewed et angles other than 90°.

          6.  Attempts to evaluate the effect of multiple plumes
were nullified by ground dirt stirred up by the engine blest.
However, the effect should be predictable based on theoretical
considerations, as indicated by angular measurements performed
on a single engine exhaust plume.

          7.  Jet engine smoke is composed of lacy agglomerates.
The geometric median agglomerated particle size of the exhaust
from the J-57 engine operated et cruise power is 0.052 urn.  The
geometric standard deviation is 1.^6.  At 10 nozzle diameters
from the nozzle the geometric median size of the agglomerates is
0.13 urn.

-------
                           REFERENCES

     Aerospace Recommended Practice,  ARP 1179,  Smoke Measurement
     Techniques, Society of Automotive  Engineers  Inc.,
     2 Pennsylvania Plaza, New York,  New York,  10001, Issued May U,

     Bureau of Mines Information Circular 7718.   The Ringelmann
     Smoke Chart, U.S. Department of  Interior,  August
 3.  Marks, L., Mech. Engr . ,  J59,  681,  1937.

 4.  Sawyer, R. , Astronautics and Aeronautics,  8,  62,  1970.

 5.  Douglas, D., National Bureau of Standards  Memo, May  20,  1954.

 6.  Langer, G., "The Development of Chemical Smoke Tracking
     Aids,"  IITRI Final Report C086,  1957.

 7.  Conner, W., Hodkinson,  J., "Optical  Properties and Visual
      Effect of Smoke Stack Plume,'1' PHS Publ. 999-AP-30,  1967.

 8.  Middleton, W., "Vision Through the Atmosphere," University
     of Toronto Press, Toronto, Canada, 1952.

 9.  Horvath, H., Charlson,  R., Amer .  Ind. Hyg .  Assoc. J.,  30,
     500,  1969.                                           ~~

10.  Sallee, P. G., "A Status Report on Jet  Exhaust Emissions,"
     Air Transport Assoc.  of America,  1000 Connecticut Ave . ,
     Washington, D.C. 20036,  Dec. 11,  1967.

11.  Dix.  D., Bastress, E.,  "Nature and Control of Aircraft
     Engine Exhaust Emissions," Report #1134-1,  Northern
     Research and Engineering Corp., 219  Vassar St., Cambridge,
   .  Mass. 02139, Nov. 1968.   Report prepared for  National Air
     Pollution Control Administration, Contract Number PH22-68-27.

12.  Shaf fernocker , W. and Stanforth,  G.  M.,  Smoke Measurement
     Techniques ,  SAE Air Transportation  Meeting,  Paper number
     680346, April 29, 1968,  New  York.

13.  Slusher, G., "Review of  Investigations  on  Aircraft Pollution,"
     FAA Program Review, NAFEC May 1970.

14.  Mack, J. E. and Martin,  M. J., "Tine  Photographic Process"
     McGraw-Hill, pp. 202-204.

15.  Hardy, Arthur, G., andPerrin,  F.  H.,   "The Principles of
     Optics,"  McGraw-Hill Inc.,  1932, pp. 212-213.
                               46

-------
                 APPENDIX A,  TRANSMISSOMETER

         A transmissometer,  Figure 1.1,  was designed and con-
structed to measure the optical transmission of smoke plumes
from jet engines under varying ambient light conditions.  This
transmissometer is readily portable and  can operate over optical
path lengths ranging from a  few feet to  several hundred feet
without impairing its effectiveness.

         A schematic optical diagram of  the instrument is shown
in Figure 1.2.  Light from a ribbon filament lamp is focused by
the condensing lens on the aperture plate and collimated by the
achromatic lens.  A rotating chopper, located between the con-
densing lens and the aperture plate, modulates the collimated
beam.  The beam is directed across the exhaust plume of the
engine under test.  On the opposite side of the plume, the beam
is intercepted by a cube-corner  retroreflector which directs
the light back along the same path to the collimating lens.
Half of the returned beam.is directed to the SGD-100 silicon
detector by the half-transmitting mirror.  Baffles minimize the
amount of light reaching the detector from scattering within
the instrument.  An infrared absorbing and visible transmitting
filter is used to approximate visual response.

         Light can reach the detector in four ways:

         (1)  Light from the main beam is reflected by the
         corner reflector.  This tranverses the smoke plume
         twice and is attenuated by absorption and scattering
         in the plume.  The  signal generated in this manner is
         the signal of interest.

         (2)  Light may reach the detector from ambient light
         either by scattering or by directly entering the col-
         lection optics.

         (3)  Light scattered in the direction of the collecting
         optics by particulate matter in the plume may reach
         the detector.

         (4)  Light scattered within the instrument by lens
         elements, the walls of the tubes, and dust on the opti-
         cal surfaces may reach the detectors.

         A synchronous lock-in voltmeter operating at the chop-
ping frequency is used for signal detection.  Since the ambient
light is not modulated by the chopper, the ambient light gives
no signal.
                               1-1

-------

          -.'1               -• -          ••.•fM»^KMgr  .'jjif^asi'SB^JSWBBfiJF
                                          : T  -'



           1
1
      trti nrrtiiti .. i     ^^N&>^


 ;
     '




      •I  "• -   •----..
     • -J
                ••*-.-.,

:



                                                    ,
        RETROREFLECTOR
TRANSMITTER-RECEIVER
                   FIGURE 1.1  TRAHSMISSOMETER,

-------
                               Light Trap
                                                 If-
            Half
            Transmitting
Aperture   /Mirror
10-200 ft.-

       Jet Plume
Condensing
System

  ibbon Filament
                                              Collimator
                                 Detector
                                 SGD 100A
                                                 Cube
                                                 Corner
                                                 Reflector
                                  Alternate
                                  Chopper
                                  Position
               Chopper
               Position
              FIGURE 1.2 OPTICAL SCHEMATIC (NOT TO SCALE).

-------
         Chopped light from the beam, which is scattered back
into the optics by the aerosol, as in Item 3 above, will appear
as a real signal.  To minimize this signal, the angular field of
view is only 36 minutes of arc.  By placing the chopper at the
corner reflector, as shown in Figure 1,2 as the alternate chopper
position, the signal contribution due to backscatter can be
eliminated.  By alternating the chopper position, the effect of
scattering can be evaluated.  In actual practice, no change in
signal due to backscatter was measured, and for convenience, the
chopper was used at the source position for all measurements.

         Light scattered from within the instrument will give
rise to an unwanted signal.  However, by careful baffling and
inserting a light trap above the half-transmitting mirror, error
from this source was very small.  At distances up to 50 feet,
this signal was less than 0,17, of the signal obtained from a
clear path.  For distances of 150 feet or more, where the un-
wanted signal approached 170, the offset-voltage-adjust on the
voltmeter was used to zero out this unwanted signal.  Thus,
the effect of internally scattered light can be completely
removed from the true signal.

         Errors due to refraction by the plume are also eliminated
by the instrument.  The size of the projected uniform beam is
larger than the retroreflector; hence, the illumination of the
retroreflector remains unchanged by refraction in the plume.
The return beam will retrace the same path as the projected
beam so that its intensity is not affected by refractive index
gradients.

         The system is used to measure the transmission of smoke
plumes in the following manner: the transmitter-receiver and the
cube-corner retroreflector are set-up on opposite sides of
the smoke plume.  The light and chopper are turned on the
beam centered on the retroreflector.  The intensity is adjusted
by a variac to give a reasonable signal.  This signal is the
signal corresponding to 100% transmission.  The system is
allowed to stabilize, and the detected voltage is recorded on a
strip chart recorder.  The engine under test is then started,
and the signals resulting from various power settings are re-
corded.  The two path transmission is calculated by taking the
ratio of the signal to the signal obtained with 1007=. transmission.
The square root of this ratio is the single path transmission
through the plume.
                              1-4

-------
         APPENDIX  B,  PHOTOGRAPHIC  INSTRUMENTATION

         Photographic photometry was chosen as an objective
technique ror quantitatively measuring, the apparent transmission
of smoke plumes from aircraft in flight.  If the luminance of
the smoke trail and the luminance of the background upon which
the smoke trail appears can be measured, then the relative
luminance will give a measure of the apparent transmission of
the plume.   If the plume is composed of light absorbing particles
that scatter light poorly then the value obtained is a true
measure of the transmittance of the smoke.  In actual practice.
it is necessary to use the luminance of the area adjacent to
the smoke plume rather than the area behind.  Unless the sky
is composed of small broken clouds, use of the adjacent area
presents no difficulty.

         While a direct reading photometer (such as small spot
brightness meter or photographic exposure meter) could be used,
no permanent record is available.  Also, since the phenomena is
transient,  it is difficult to obtain reliable readings.  To
overcome this difficulty, it was decided to photograph the
smoke trails and use photographic photometry to recover the
luminance values from the photograph.

         A 35 mm single lens reflex camera with 50 mm. 125 mm,
and 250 mm lenses, was used to photograph the smoke trails.
The longer focal-length lenses gave sufficiently large images
to permit easy measurement of the density of the smoke plumes.
The 125 mm proved most useful for aircraft in-flight.  For
tied-down aircraft, the 50 mm proved best.  For measuring the
density., a David W. Mann #1140 microdensitometer was used.

         The technique of photographic photometry is as follows .
In addition to photographing the subject  whose luminance is to
be measured, a transparent gray-step wedge is also photographed.
When a 35 mm camera is used, the gray-step wedge is simply
photographed on a separate frame from the subject.  The film is
then developed.  Reasonable care is taken to ensure uniform
development.  (If a camera other than a 35 mm is used and the
gray-step wedge is photographed on a separate film, care must
be taken so that is receives exactly the same development time,
temperature, and agitation.)  In order that the film will exhibit
identical response to the gray-step wedge and the subject, the
gray transparent step wedge is photographed x^ith the sky as
illumination.

-------
         A convenient step wedge for calibration of the film
is the Eastman Kodak Step Tablet #3, having 21 density steps
ranging from 0-05 density units to 3.5 density units.  The
range of the step wedge is such that when the proper exposure
is given, some steps are underexposed and others are over-
exposed.

         The density of each step of the original step tablet
is measured on the microdensitometer,  as are the corresponding
steps on the negative.

         Density is defined as:


             D - log10 !> , log10 I                       (2.1.


where, I  = the light incident on the negative,

       T  = the fraction of the light transmitted by the negative,

       I  = the light transmitted.


         If the density of the negative step wedge is plotted
against the logarithm of the exposure the characteristic H & D
curve of the photographic emulsion results (Reference 14) .
(The exposure is the product of time and illuminance of the
film and is commonly given in meter-candle-seconds.)  It is
not necessary to know the absolute value of exposure, but only
the exposure ratio between the various steps of gray.  From
Equation (2.1), it is readily seen that the logarithm of the
transmission ratio, that is, the density difference between
steps of the original gray scale, is proportional to the nega-
tive logarithm of the exposure ratio for the corresponding steps
on the negative.  Consequently, it is  convenient to plot the
data for the H & D curve as shown in Figure 2.1, where the
density steps of the negative are plotted against the corres-
ponding density steps of the original gray scale.-'
*  The density values are measured by a densitometer depend
upon the geometry of the collection optics and the light scat-
tering of the sample as well as upon the absorption of the
sample.  That is, the density measured for a scattering sample
by an instrument accepting a large solid angle will differ from
that measured by an instrument accepting a small solid angle.
The first type of instrument measures "diffuse density" and the
second "specular density." A sample which does not scatter, but
                             2-2

-------
               Exposure Ratio
               0.01
                      0.1
1.0
00
0)
c

o
•I-t
x:
a
CO
S-i
60
o
4J
O
JT
CL


-------
attenuates only by absorption would give the same measurement
regardless of the instrument,  A photographic step wedge does
scatter some light and the density will vary with the instrument
used.  Therefore the density as measured by the densitometer may
be different than the effective density as observed by the camera
and an additional step is required before the measured density
differences may be expressed as log exposure ratios.  The measured
density may be converted to effective density by experiment as
follows.  A series of carefully times camera exposures are made.-
of the step wedge.  These various timed exposures provide a
series of known exposure ratios with which the measured densities
Df the original step wedge may be compared.  In general it is
found that the density values obtained by different instrument.'-
nay be expressed by D  = g D , where D  represents the density
                     o      Q.        5
obtained with a specular instrument (limited acceptance angle)
and D, represents the density as measured by a more diffuse

instrument (large acceptance angle) and g is a constant called
Calliers coefficient (Reference 15,)  Calliers coefficient will
oe a constant depending upon the characteristics of the two
particular instruments.  Such an experiment was performed on the
present program and it was found that density values as measured
by the Mann densitometer corresponded exactly to that of the
camera optics.  Therefore, Calliers constant was unity and no
correction was required.
If the corresponding densities of smoke plume and background
on a negative are measured by a microdensitometer and compared
to the H & D curve for that negative, the exposure ratio and
hence, the transmission of the plume can be determined.

         It is characteristic of photographic film that a
portion of the H & D curve is a straight-line, thus the density
difference divided by the slope of the straight-line gives the
logarithm of the ratio directly.  (A properly exposed negative
takes advantage of this straight-line portion to correctly
reproduce the relative luminances of the scene.)  The slope of
the straight-line portion of the H & D curve (commonly called
"gamma," y) is controlled by the development and is constant
over the entire length of the film.  Hence, a calibration step
wedge need be photographed only once on the film..  In practice,
it is usually photographed at the beginning and end of the film
to check on uniformity of development.

-------
         Figure 2.2 illustrates the steps in determining the
luminance ratio of a smoke cloud to its background.  Figure
2.2a represents a negative obtained by photographing the smoke
cloud;  Figure 2.2b represents the negative obtained by photo-
graphing the gray wedge; and Figure 2.2c represents an H & D
plot of Figure 2.2b.  The density of the steps in the gray-step
negative are plotted against the density of the original step
wedge.  Suppose one wishes to measure the smoke luminance ratio
for 2 points,  a  and  b , in  Figure 2.2a.The density of each
of the points is measured and the corresponding density values
plotted on the H & D curve as shown in Figure 2.2a.  The corres-
ponding original gray step wedge density difference is read from
the abscissa of the curve.  The density difference is equal to
the negative log-,Q of the luminance ratio.


         Example:  The density of point a is 1.30 (see ordinate),
The density of point b is 0.99.  These two points correspond
to the original wedge density values a', b1 of 0.75 and 1.10
(see dotted lines on Figure 2.2c.)  Therefore:


              , „.   luminance at b   , , n   n -,c   n TC
             -log,n	:	 = 1.10 - 0.75 = 0.35
                 •i-U luminance at a
         or,
                   luminance at b =          Transmission
               °10 luminance at a     to!0
                                    of Smoke = -0.35

Hence, Transmission of Smoke = antilog (-0.35) = 0.45 = 45%

         Because the points a and b both lie on the straight
line portion of the H & D curve, an alternate calculation is
possible.  The density difference between a and b on the
negative is £D = 0.31.  The corresponding density difference on
the original gray scale is:


             AD' = AD x - = 0.31 x —-— =0.35
                        Y          0.885

Hence, — = antilog (-0.35) = 4570 transmission.
       3

-------
           000
     c
     -*
     'x .
DO/0000000000000000
                                54321
     £00000000000000000000 0 0 0



a. Negative of Scene          b. Negative of Gray Stepwedge
      £  2.00

      CO
      &0
      cu
      25

      ««  1,30
      o

      >,  1.00,
      c  0.99
      0)
      Q
            0
                       Gray Step Levels

                        (1-2-3-4-5)
                                 a'0.75
                            .885
                    2.00      1.00       0

                 Density of Original Gray Scale
            c.   H & D Curve of Negative
        FIGURE 2.2  ILLUSTRATION OF USE OF

                   PHOTOGRAPHIC PHOTOMETRY,
                          2-6

-------
         Note that only the luminance ratio between the smoke
and the background, as seen from the observer's viewpoint, is
measured.  If a highly scattering medium exists in the inter-
vening atmosphere between the observer and the plume, the obser
ved luminance ratio will be reduced over the value at the plume
That is, contrast attenuation by the intervening atmosphere
may render the interpretation inaccurate.   This effect operates
regardless of the type of remote observation and a Ringelmann
reader is subject to the same difficulties.  Hence, for relia-
bility, observations must be made at close range and under clear
air conditions.
                              2-7

-------
                   APPENDIX C,  SAMPLING JET
                EXHAUST FOR PARTICULATE MATTER

         A special exhaust particulate sampling apparatus was
used to provide a Dilute sample of the exhaust gas for particle
size determination.  The apparatus is shown in Figure 3,1.
Exhaust is drawn into the sampler by a syphon and diluted almost
immediately with a large volume of cool air.  A tangential
inlet on the diluter facilitates mixing of the exhaust and diluent
air.  The cooled, diluted exhaust flows down the dilution cham-
ber and exits at the bottom.  Near the exit,samples of the
diluted exhaust are collected on membrane filters and electron
microscope grids.  Flows to the apparatus are controlled by
flowmeters.  The syphon was calibrated in the laboratory prior
to testing.  However, the ram air effect of the exhaust in-
validated the calibration and a thermoanemometer was added near
the dilution chamber exit to monitor the total volume of air
flowing in the apparatus.  Knowing this value and the amount
of syphon and dilution air, the volume of jet exhaust drawn
into the apparatus was calculated.  Critical orifices regulated
the sample flow rate through the filters.  Membrane filters
were used.  The instrument is shown in use at the J-57 test
cell in Figure 3.2.  The apparatus, with the exception of the
flowmeters and pump, was placed at various distances behind the
engine test fixture; the sampling probe was always located at
the center of the exhaust plume.  The probe was 3/8 inches in
diameter.

         The important features of the apparatus are enumerated
below:
         1.  Agglomeration problems are circumvented by quickly
         diluting the exhaust.

         2.  Condensation problems are avoided by large volumes
         of cool diluent air.

         3.  Heated sample lines are not necessary.

         4.  Sample line deposition problems are mitigated by
         the use of short lines,  Cleaning is facilitated.

         The apparatus was used to obtain size and structure
data on the exhaust particulates.  Reflectance measurements
were made on the membrane filters and a smoke number based on
a filtration density of 0.0565  ft.3 of exhaust per in.2 of
filter was calculated.  A white background was used during the
reflectance measurement.  The results, therefore, are not the
SAE smoke number.
                              J-l

-------
        Syphon
           X
Exhaust
Sample
fr
                t
      Tangential
      Inlet
                         Dilution
                           Chamber
                     Royco Particle
                     Counter  Sample
                     Inlet
                         Thermo-
                         couple
        Jet Engine  Room
                                 Dilut
                                 Air
                                           To SAE
                                           Smoke Sampler
Syphon
Air
                                      i TJ
                                              Flowmeters
                                     37—^f   Hg
                                     -^	'  Manometer
                           Clean Filtered Air
                                                                    To Royco
                                                                    Counter
                                Cross-Section AA'


                             Thermo Anemometer

                             Restricted Orifices 900 cc/min

                             Solenoid Valves
                                   Control Room
             FIGURE  3.1 EXHAUST PARTICULATE SAMPLING APPARATUS.

                                     3-2

-------
                                       ;
       	       •" "
FIGURE 3.2  IITRI SMOKE SAMPLER IN THE ENGINE
            TEST CELL
                  3-3

-------
         The diluted jet exhaust was also analyzed by a Royco
Model 245 light-scattering particle counter.  A 25 ft, sample
line was used to connect the sampling apparatus to the Royco.
The lengthy sample line was needed because the Royco is sensi-
tive to noise and vibration; therefore, it was located in the
control room.  The Royco provides particle size distributions
for aerosols whose particles range from 0.3|j to about 10(j .   As
shown in Table 2, the vast majority of smoke particles are
below 0.3u.   Also, the concentration of particles, even in  the
diluted exhaust, exceeded the counting capacity of the Royco
at all power settings, except idle.  The Royco data for the
J-57 engine are reported in Table 3.1.

         The IITRI sampler was also used to sample the exhaust
from a JT12-6 engine mounted in the NAFEC wind tunnel.  The
probe was placed in the center of the wind tunnel 7-1/2 nozzle
diameters from the engine.  Due to personnel safety considera-
tions the dilution chamber was located in the control room  and
connected to the probe with a heated sampling line.  Royco
particle size and concentration data for the JT12-6 engine  are
reported in Table 3.2.

         A General Electric smoke console was used by NAFEC
personnel to collect samples from the IITRI sampler probe for
smoke number determination.  A lengthy, heated sample line,
1/4 inches in diameter, was connected to the top of the syphon
for this purpose.  When the smoke number was being determined,
air flows to the syphon and diluter were shut off.  The line
to General Electric smoke console was plugged when the console
was not being used.
                              3-4

-------
                                           TABLE 3.1
SIZE DISTRIBUTION AND CONCENTRATION OF EXHAUST
AS REPORTED BY THE ROYCO PARTICLE COUNTER- J- 57- P-

Test
IITRI
6
7
8
9
2
3
4
5
10
11
12
13
14

Number ;
NAFEC
31
33
35
37
23
25
27
29
39
41
43
45
47

Sampler Location^
Engine Diameters
From Engine Nozzle
0
0
0
0
2-1/2
2-1/2
2-1/2
2-1/2
10
i
Engine
Thrust
Setting
Idle
Approach
75% Norm.
Cruise
Idle
Approach
75% Norm.
Cruise
Idle
10 Approach
10
10
10
75% Norm.
Cruise
Takeoff

0.3-0.6u
TMC(l)
TMC
TMC
TMC
' 465
TMC
TMC
TMC
385
TMC
TMC
TMC
TMC

PARTICLES
37A ENGINE

Thousands of Particles/Ft
of Undiluted
m 0.6-1. 5um
62.2
TMC
TMC
TMC
260
TMC
TMC
TMC
193
TMC
TMC
TMC
TMC

3
*




Engine Exhaust
1.5-3.0um
20.7
TMC
TMC
TMC
103
TMC
TMC
TMC
43.1
201
455
TMC
TMC
3-5um
6.9
61.7
TMC
TMC
2.5
13.5
23.4
TMC
7.2
13.4
36.2
84.4
249
>5um
1.9
6.2
29.8
86.3
1.5
4.5
4.5
38.0
4.8
2.2
2.5
6.5
16.6
Total
_-
--
--
--
832
--
--
--
633
--
--
^ M
(1)   Too many particles  to count,  exceeds  about  500,000 particles/ft.

-------
                                               TABLE  3.2
                        SIZE DISTRIBUTION AND  CONCENTRATION OF EXHAUST PARTICLES
                        AS REPORTED BY THE ROYCO PARTICLE COUNTER-JT12-6 ENGINE
    IITRI  NAFEC
    V?T- 1
    WT- 7
    WT-9
77
Sampler Location,.  j  Engine
Engine Diameters     Thrust
From Engine Nozzle j  Setting
      7-1/2        |  Approach
      7-1/2        j  Cruise
      7-1/2          Takeoff
        Thousands of Particles/Ft.3
         of Undiluted Engine Exhaust
0.3-0.6um  0.6-1.5pm  1.5-3.0|am  3- 5pm  >5u.ic  Tot
   270        111       16.3      1.2    0.7    39
   TMC(l)     TMC       106      35.0    1.4
   TMC        TMC       TMC      90.0    0.3
. x)
    (1)   Too many particles to count,  exceeds  about 500,000 particles/ft.  .

-------
                   APPENDIX D, SAE STANDARD
                 METHOD FOR SMOKE MEASUREMENT

         The Society of Automotive Engineers' Committee E-31
has developed a method for measuring and specifying turbine
engine smoke.  Their efforts are documented in the Aerospace
Recommended Practice 1179 (Reference 1) .   Much of the testing
required for the development of the method was conducted at
NAFEC.  The procedure used at NAFEC (Reference 13) involves
filtering different volumes of exhaust gases and evaluating
the optical density of the resulting stain.  Volumes of gas,

0.1 ft3, 0.2 ft3, 0.4 ft3, and 0.7 ft3, collected in the plane
of the engine nozzle were accurately measured with a wet test
meter.  The graduated series of stains obtained were evaluated
by measuring the optical diffuse reflectance with a reflecto-
meter.  Whatman No, 4 filters are specified, but membrane filters
were also used during this study to obtain comparative data.
The smoke number associated with each flow is calculated from
the following equation:

                            —        R
             Smoke Number = SN = (1 -- -) 100             (4-1)
where, R  = the diffuse reflectance of the smoke deposit on
            the filter,

       R  = the diffuse reflectance of the clean filter.

Since R  and R  are functions of the material used to back the
f ilterssduring"7analysis ,  the backing material must be specified.
ARP No. 1179 specifies a  black backing as the standard but a
white backing was also used during this study.  The reportable
SAE smoke number is the smoke number corresponding to a gas
                3   2
volume of 0.3 ft /in  of  Whatman No. 4 filter paper evaluated
on a black backing.  It is obtained by plotting the smoke
numbers against gas volume per unit area of filter surface
                                                             3   2
(log-log plot) and interpolating the curve obtained to 0.3 ft /in

         The probe for collecting exhaust gas is shown in Figure
4 .1 , installed on an F- 100 aircraft.  The smoke console built
by General Electric and used to filter the smoke sample and
measure the gas volumes,  is shown in Figure 4.2.
                              4-1

-------
.
             -.

                                                                            -- ••*n-9yt

                                                        •• -  .,,


&-_««.
                          \\
                                                                              -


                                                                    \

                                            '      '        • I         I



                                                    .
                                                                         I


                                         •

                                                    .
                                                        a
                                                                                 :^. -:


                                                        .
                 FIGURE
                        F-100 AIRCRAFT WITH SAMPLING PROBES.

-------