A Study of The Entrainment and Turb Ulence in A Plane Buoyant Jet

A STUDY OF
THE ENTRAINMENT AND TURB ULENCE

IN A PLANE BUOYANT JET
by
Nikolas E. Kotsovinos
W. M. Keck Laboratory of Hydraulics and Water Resources

Division of Engineering and Applied Science
CALIFORNIA INSTITUTE OF TECHNOLOGY
Pasadena, California
Report No. KH-R-32 August 1975

A STUDY OF THE ENTRAINMENT AND TURBULENCE
IN A PLANE BUOYfu~T JET
by
Nikolas E. Kotsovinos
Project Supervisor:
E. John List
Associate Professor of
Environmental Engineering
Supported by
National Science Foundation
Grant No. ENG75-02985
W. M. Keck Laboratory of Hydraulics and Water Resources

Division of Engineering and Applied Science
California Institute of Technology
Pasadena, California
Report No. KH-R-32 August 1975

-i-
ACKNOWLEDGMENTS
I wish to express my deepest appreciation and gratitude to my
advisor, Professor Ericson John List, whose inspired guidance, insight,
perspicacious criticism and enthusiasm were invaluable to this research.
I am greatly indebted to Professors Norman H. Brooks and Donald
Coles, whose advice and interest in this research continuously stimulated
and inspired me. I am grateful to Dr. Robert C. Y. Koh for numerous
discussions, and to Dr. Jorg Imberger for his personal interest, encourage-
ment and friendship during the incipient period of my graduate studies.
I wish also to express my deepest respect and appreciation to Professor
Vito Vanoni for his kind interest and encouragement throughout my
research.
I gratefully acknowledge that the enthusiastic teaching of Professor
ThamrsXanthopoulos provided me with the necessary stimulation to pursue
graduate studies.
I thank the California Institute of Technology for providing me with
financial support through Graduate Teaching and Research Assistantships,
and for its superb research environment facilities. Financial assistance
provided by the National Science Foundation, under Grant ENG75-02985, is
greatly appreciated.
I am indebted to Mr. Elton F. Daly and Mr. Joseph Fontana for their
invaluable assistance in the construction of the laboratory equipment,
and Mr. Bob Shultz for the preparation of the graphs. I thank also Mrs.
Joan Mathews for her patience in typing this manuscript.
Finally, I thank my wife, Kathy, whose love, support and understanding
were essential during the course of this work.

-ii-
This report was submitted by the writer in May 1975 as a thesis
with the same title to the California Institute of Technology in
partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
-iii-
ABSTRACT
The entrainment and mixing processes in a two-dimensional vertical
turbulent buoyant (heated) jet in its transition state from a pure jet
to a pure plume have been studied. The ambient fluid is of uniform
density and non-flowing except for the flow induced by the jet. Density
variations are assumed small.
The equations of motion integrated across the jet have been care-
fully examined and it has been found that the kinematic buoyancy flux of
a heated plume and the kinematic momentum flux of a pure jet are not in
general conserved. It has been proven that the flow in a two-dimensional
pure jet is not self-preserving.
A systematic set of experiments was carried out to examine turbulent
buoyant jet behavior for a wide range of initial Richardson numbers (or
densimetric Froude numbers). Values of the Richardson number, which
describes the relative importance of buoyancy in a jet, extended from
the value appropriate for a pure jet (zero) to that appropriate for a
plume (approximately 0.6). The buoyant jet temperature and velocity
fields were measured using calibrated fast response thermistors and a
laser Doppler velocimeter respectively. The velocity and temperature
data obtained were recorded magnetically in digital form and subsequently
processed to extract both mean and fluctuating values of temperature and
velocity.
The structure of the mean flow (including the spreading rate of the
mean velocity and temperature profiles, velocity and temperature distribu-
tion along jet axis, and the heat flux profile), the turbulence structure
-iv-
(including the profile of turbulence intensity and turbulent heat
transfer, probability density distribution of temperature and velocity,
skewness and flatness factor of temperature fluctuations) and the large
scale motions (intermittency, profile of maximum and minimum temperature,
frequency of crossing of hot/cold, cold/hot interface) of a buoyant jet
were investigated as a function of the jet Richardson number. It was
determined that the turbulent heat transfer and the turbulent intensity
increase with increasing the Richardson number. The spreading rate of
the transverse mean velocity and temperature profiles were found to be
independent of the Richardson number. The turbulent buoyancy flux in a
fully developed buoyant jet has been found to be a significant fraction
(38%) of the axial buoyancy flux.

-v-
TABLE OF CONTENTS
Part Title
NOMENCLATURE 1
1. INTRODUCTION 7
1.1 Introductory note 7

1.2 A review of previous theories 7
Entrainment coefficient approach 12
Dimensional analysis 17
1.3 A review of previous experiments 18
I. Pure jet 18
II. Pure plume 22
III. Buoyant jet 24
1.4 Scope of this work 24
2. THE MECHANICS OF JETS 25

2.0 Basic definitions and assumptions 25
2.1 The mechanics of the two-dimensional pure jet 30
2.2 The mechanics of the two-dimensional pure plume 39
2.3 The mechanics of the two-dimensional buoyant jet 46
2.4 Numerical solution of the buoyant jet problem 52
3. EXPERIMENTAL PROBLEMS AND OBJECTIVES 53

3.0 Introduction 53
3.1 Temperature measurements 53
Thermistors 53
AID sampling rate 58
3.2 Velocity measurements 62
I. Description of the laser Doppler velocimeter 62
II. Derivation of the Doppler shift frequency formula 64
III. Signal processing 68
3.3 Velocity-temperature cross correlations measurements 71
3.4 Experimental set-up 73
3.5 Data collection and processing 77
Program I 81
Program II 86
-vi-
Part Title
4. TEMPERATURE MEASUREMENT RESULTS 88

4.0 Basic experimental parameters and initial data 88
4.1 Mean temperature 90
4.2 Turbulence intensity 106
4.3 Cross sectional profile of maximum and minimum
temperature 120
4.4 Intermittency 125
4.5 Frequency of crossing of the hot/cold interface 130
4.6 The frequency distribution of temperature 135
4.7 Skewness and flatness factor 135
5. VELOCITY MEASUREMENT RESULTS 142

5.0 Basic experimental parameters and initial data 142
5.1 Mean velocity profile 144
5.2 On the conservation of the momentum and on the
decay of the centerline velocity of a pure jet 157
5.3 Centerline velocity of a plume 166
5.4 The frequency distribution of the velocity 167
5.5 Turbulence intensity 167
6. VELOCITY-TEMPERATURE RESULTS 174

6.0 Basic experimental parameters-initial data 174
6.1 Mean velocity and mean temperature profile 175
6.2 Profiles of uT and u'T' 184
7. COMPARISON OF THE THEORY WITH THE EXPERIMENTAL RESULTS 203

7.0 Introduction 203
7.1 Comparison of the theory with the experimental results 203
7.2 Comparison of the theory with the experimental results
of temperature 204
8. ON THE SELF-PRESERVATION OF A PURE JET 210
9. SUMMARY OF MAIN RESULTS AND CONCLUSIONS 228

9.0 Introductory note 228
9.1 Summary of the main results 228
I. The mean flow structure 228
II. The turbulence structure 231
-vii-
Part Title
III. The structure of the large eddies 232

9.2 Conclusions 234
APPENDIX A Pictures of the flow 236
APPENDIX B Mean temperature profiles for all the

experimental runs 244
APPENDIX C Test of the two-dimensionality of the flow 288
REFERENCES 300
-viii-
LIST OF FIGURES
Figure
1. 2.1 Geometry of a buoyant jet. 8
2.0.1 Definition of jet boundary. 29
2.1.1 Geometry of a jet. 30
2.1. 2 Flow towards a line distribution of sinks of

strength given by Eq. 2.1.10. 36
2.1. 3 Schematic of the assumed induced flow and

virtual origin of a real plane jet. 36
2.2.1 Thermal expansion coefficient of water as a

function of temperature. 42
2.2.2 Induced flow towards a line plume on a

horizontal plane. 43
3.1.1 Close photograph of the thermistor probe.

One subdivision is one millimeter. 55
3.1. 2 The bridge circuit used for the measurement

of the temperature. 56
3.1. 3 Typical thermistor calibration curve. 57
3.1.4 Mean temperature as a function of sampling

rate (length of sample 50 sec). 60
3.1.5 Mean temperature as a function of sampling

time (sample rate 10 samples/second). 60
3.2.1.a Optics of laser Doppler velocimeter. 63
3.2.1.b Signal processing of laser Doppler velocimeter. 63
3.2.2 Scattering light from a moving particle. 65
3.2.3 Spectrum analysis of the signal given by Eq. 3.2.4 67
3.2.4 Schematic of a spectrum analyser. 70
3.3.1 Schematic for temperature-velocity measurement. 72
3.3.2 Schematic for the measurement of the correlation uT. 74
3.4.1.b Cross section of the chamber of the emerging jet. 75

-ix-
LIST OF FIGURES (Cont'd)
Figure
3.4.2 Overall experimental set-up. 76
3.4.3 First instrumental set-up. 78
3.4.4.a Front view of the second instrumental set-up. 79
3.4.4.b Plane view of the second instrumental set-up. 80
3.5.1 Typical intermittent signal. 84
4.1.l.a Mean temperature profile for a turbulent

jet at x/D = 10.0. 91
4.1.l.b Mean temperature profile for a turbulent

jet at x/D = 20.0. 91
4.1.1.c Mean temperature profile for a turbulent

jet at x/D = 30.4. 92
4.1.l.d Mean temperature profile for a turbulent

jet at x/D = 51.0. 92
4.1.2 Non-dimensional temperature half-width of a

turbulent jet plotted against non-dimensional
distance along the jet axis. 93
4.1. 3 Non-dimensional mean temperature profile of a

jet plotted against non-dimensional distance
from jet axis. 94
4.1.4.a Mean temperature profile for a turbulent plume

at x/D = 6.0. 95
4.1.4.b Mean temperature profile for a turbulent plume

at x/D = 14.0. 95
4.1.4.c Mean temperature profile for a turbulent plume

at x/D = 30.2. 96
4.1.4.d Mean temperature profile for a turbulent plume

at x/D = 36.0. 96
4.1. 5 Temperature half-width of a plume as a function of

distance from plume orifice. 97
4.1.6 Non-dimensional mean temperature profile for a plume. 100
4.1. 7 Temperature distribution along jet axis. 101

-x-

Figure
4.1. 8 Temperature distribution along plume axis. 102
4.1. 9 Temperature distribution along plume axis. 103
4.2.1 Profile of the intensity of turbulent fluctuations

of plume temperature plotted against non-dimensional
distance from jet axis. 107

of jet temperature against non-dimensional distance
from plume axis. 108
4.2.3 Intensity of turbulent fluctuations of temperature

on jet axis as a function of non-dimensional
distance from jet orifice and initial Richardson
number. 112
4.2.4 Intensity of turbulent fluctuations of temperature

on jet axis normalized by the local mean temperature
as a function of non-dimensional distance from jet
orifice. 113
4.2.5 Sketch of an intermittent temperature signal. 117
4.2.6 Profiles of turbulence intensities T,2 , T'2

hot'
T~;ld for a turbulent plume. 118
4.2.7 Profiles of turbulence intensities T,2, Th~t'

and T~;ld for a turbulent plume. 119
4.3.1 Maximum and mlnlmum temperature for a turbulent jet

as a function of distance from jet axis. 122
4.3.2 Maximum and minimum temperature for a turbulent plume

as a function of distance from plume axis. 123
4.3.3 Distribution of maximum and minimum temperature along

jet axis as a function of initial Richardson number. 124
4.4.1 Temperature intermittency for a turbulent jet plotted

against non-dimensional distance from jet axis. 126
4.4.2 Temperature intermittency for a turbulent plume

plotted against non-dimensional distance from plume
axis. 127
4.4.3 Profile of the velocity intermittency for a turbulent

jet (based on data by Ranum [40]). 128
-xi-

Figure
4.4.4 Temperature intermittency for a turbulent plume

plotted against non-dimensional distance from
plume axis. 129
4.5.1 Profile of frequency of hot/cold, cold/hot interface

crossing for a turbulent plume. 131
4.5.2 Profile of frequency of hot/cold, cold/hot interface

crossing for a turbulent jet. 132
4.5.3 Profile of frequency of hot/cold, cold/hot, interface

crossing for a turbulent jet. 132
4.5.4 The Strouhal number, based on the peak frequency of

hot/cold, cold/hot interface crossing profile against
distance from jet orifice. 134
4.5.5 Average duration of hot bursts in a turbulent plume. 136
4.5.6 Average duration of hot bursts in a turbulent jet. 136
4.6.1.a Probability density of the temperature of a

turbulent plume at y/x' = 0.005. 137
4.6.1.b Probability density of the temperature of a

4.6.1.c Probability density of the temperature of a

4.6.1.d Probability density of the temperature of a

4.7.1 Skewness factor of temperature fluctuations of a

buoyant jet as a function of distance from the jet
axis. 139
4.7.2 Flatness factor of temperature fluctuations of a

buoyant jet as a function of distance from jet axis. 140
5.1.1.a Mean velocity profile for a turbulent jet

at X/D = 20.8. 145
5.1.1.b Mean velocity profile for a turbulent jet

at X/D = 33.3. 145
5.1.1.c Mean velocity profile for a turbulent jet

at X/D = 37.0. 146
5.1.1.d Mean velocity profile for a turbulent jet

at X/D = 54.2. 146
-xii-

Figure
5.1.1.e Mean velocity profile for a turbulent jet

at X/D = 58.75. 147
5.1.1.£ Mean velocity profile for a turbulent jet

at X/D = 74.50. 147
5.1.2 Non-dimensional velocity half-width for a

turbulent jet as a function of distance from
jet orifice. 148
5.1. 3 ~on-dimensional mean velocity profile for a

turbulent jet. 149
5.1.4.a Mean velocity profile for a turbulent plume

at X/D = 9.0. 150
5.1.4.b Mean velocity profile for a turbulent plume

at X/D = 14.0. 150
5.1.4.c Mean velocity profile for a turbulent plume

at X/D = 19.0. 151
5.1.4.d Mean velocity profile for a turbulent plume

at X/D = 28.8. 151
5.1.4.e Mean velocity profile for a turbulent plume

at X/D = 33.6. 152
5.1.4.f Mean velocity profile for a turbulent plume

at X/D = 39.1. 152
5.1. 5 Non-dimensional velocity half-width of a turbulent

plume plotted against non-dimensional distance
from jet orifice. 153
5.1. 6 Non-dimensional mean veloci_ty profile for a

turbulent plume. 156
5.2.1 Experimental values for the kinematic momentum

flux m(x) of a turbulent jet (R = 0) as a function
of distance from jet orifice. 0 159
5.2.2 Velocity distribution along jet axis (R

o
= 0). 161
5.2.3 Kinematic momentum flux m(x) of a two-dimensional

turbulent jet (R = 0) as a function of distancE
£ /D and jet g~ometry. 164
ill a x
5.2.4 Schematic or the ilv:l'.lced flow by 2 two dimenslonal

pure jet. 165
-xiii-

Figure
5.3.1 Velocity distribution along plume axis. 168
5.4.1 Probability density of the velocity for a turbulent

plume as a function of distance from jet axis. 169
5.4.2 Probability density of the velocity of a turbulent

jet as a function of distance from jet axis. 170

of the velocity in a turbulent jet. 172

of the velocity in a turbulent plume. 173
6.1.1.a Non-dimensional mean velocity profile in a

turbulent jet (R = 0). 176
o
6.1.1.b Non-dimensional mean temperature profile in a
turbulent jet (R = 0). 176
o
6.1.2.a Velocity half-width of a jet (R ~ 0) as a
function of distance along the ~et. 177
6.1.2.b Temperature half-width of a jet (R ~ 0) as a

function of distance along the jet? 177
6.1.3.a Non-dimensional mean velocity profile in a

6.1.3.b . Non-dimensional mean temperature profile in a

6.1.4.a Velocity half-width of a plume as a function of

distance along the plume. 179
6.1.4.b Temperature half-width of a plume as a function

of distance along the plume. 179
6.1.5 Turbulent heat flux in a buoyant jet normalized by

the heat flux at the jet orifice, as a function of
distance along the jet. 185
6.1.6 Turbulent heat flux in a buoyant jet, normalized by

the heat flux at the jet orifice, as a function of
the local Richardson number. 186
6.2.1 Schematic of the fluctuations of the temperature

T'(x,y,t) and vertical velocity u'(x,y,t); (a) plume
(b) jet. 188
-xiv-

Figure
6.2.2.a Profile of the correlation pc R(~x,O) in a

turbulent jet at x/D = 20.8. P 191
6.2.2.b Profile of the correlation pc R(~x,O) in a

turbulent jet at x/D = 37.0. P 19l
6.2.2.c Profile of the correlation pc R(~x,O) in a

6.2.2.d Profile of the correlation pc R(~x,O) in a

6.2.3 Profile of the correlation pc R(~x,O) in a

turbulent plume at x/D = 19.0~ 193
6.2.4 Heat flux half-width for a turbulent jet as a

function of distance along the jet axis. 194
6.2.5 Non-dimensional heat flux profile in a turbulent

jet. 195
6.2.6 Profile of turbulent heat flux in a jet. 198
6.2.7 Profile of turbulent heat flux in a plume. 199
6.2.8 Jet correlation coefficient of velocity and

temperature as a function of distance from the
jet axis. 201
6.2.9 Plume correlation coefficient of velocity and

temperature as a function of distance from the
jet axis. 202
7.1.1 Comparison of the model with the experimental

results for the kinematic mass flux. 205
7.1. 2 Comparison of the model with the experimental

results of the kinematic momentum flux. 206
7.1. 3 Comparison of the model with the experimental

results for the kinematic buoyancy flux. 207
7.2.1 Comparison of the model with the experimental

results for the mean temperature distribution on
the jet axis. 209
8.1 Hypothesis concerning the nonlinear spread of the

mean velocity profile of a turbulent jet. 212
8.2 Correlation of spreading rate K and non-dimensional

lu
virtual origin K ' 213
2u
-xv-

Figure
8.3 Spreading rate Kl as a function of non-dimensional

distance t /D. u 214
max
8.4 Non-dimensional virtual origin K2 as a function of
non-dimensional distance t /D. u 215
max
8.5 Nonlinear spread of half-width b u of mean velocity
profile of a two-dimensional turbulent pure jet. 217
8.6 Correlation of spreading rate KIT and non-dimensional

virtual origin K . 218
2T
8.7 Spreading rate KIT of mean temperature profile as a
function of distance t /D. 219
max
8.8 Non-dimensional virtual origin K2T as a function of
distance t /D. 220
max
8.9 Nonlinear spread of temperature half-width b of a two-
T
dimensional pure plume. 221
8.10 Velocity ratio u 2 /liM(x) as a nonlinear function of non-

dimensional distgnce x/D. 223
8.11 Velocity ratio u2 /u (x) as a nonlinear function of non-

M
dimensional dist~nce x/D. Data from several investiga-
tors. 224
8.12 Correlation of coefficients C and C ' 225

lu 2u
8.13 Coefficient Cl u as a function of non-dimensional
distance t /D. 226
max
8.14 Coefficient C2u as a function of non-dimensional
distance 9., /n. 227
max
A.l Short time (0.01 seconds) exposure photograph of
a turbulent two-dimensional plume (R ~ 0.5). 237
a
A.2 Short time (0.01 seconds) exposure photograph of
a turbulent two-dimensional plume (R ~ 0.5). 238
a
A.3 Medium time (2 seconds) exposure photograph of a
turbulent two-dimensional plume (R ~ 0.5). 239
a
A.4 Short time (0.01 seconds) exposure photograph of a
turbulent two-dimensional jet (R ~ 2500, R = 0). 240
e a
A.5 Medium time (2 seconds) exposure photograph of a
turbulent two-dimensional jet (R ~ 2500, R = 0). 241
e a
-xvi-

Figure
A.6 Hypothetical instantaneous picture of a plume. 242
A.7 Schematic to explain the production of large waves

in a plume. 243
C.l Distribution of mean temperature in a direction

parallel to the slot. 290
C.2 Distribution of mean temperature in a direction



C.5 Distribution of turbulence intensity in a direction

C.6 Distribution of turbulence intensity in a direction

C.7 Distribution of intermittency in a direction parallel

to the slot. 296
C.8 Distribution of maximum and minimum temperature (see

Section 4.3) in a direction parallel to slot. 297
C.9 Distribution of fR (see Section 4.5) in a direction

C.IO Schematic of the induced flow profile between the two

confining walls. 299
-xvii-
LIST OF TABLES
Table
1.3.1 Review of previous experiments for a two-

dimensional pure jet. 20
3.1.1 Thermistor characteristics. 53
4.0.1 Basic experimental parameters and initial data

of the temperature measurements. 89
4.1.1 Linear approximation of the growth of the

temperature half-width and the centerline
dilution. 98
4.1. 2 Experimental values for temperature half-width and

temperature distribution along jet axis for
turbulent jets and plumes. 105
4.2.1 Tabulated experimental values for the intensity

of turbulent fluctuations of temperature along jet
axis for turbulent jets and plumes. 114
5.0.1 Basic experimental parameters and initial data

of the velocity measurements. 143
5.1.1 Linear approximation of the growth of the velocity

half-width. 154
5.1.2 Tabulated experimental values for volume, kinematic

momentum and buoyancy fluxes for turbulent jets
and plumes. 158
5.2.1 The kinematic momentum flux of a plane pure jet. 163
6.0.1 Basic parameters of velocity-temperature measurement. 174
6.1.1 Velocity temperature basic results. 180

-1-
NOMENCLATURE
Arabic Notation
b b (x), half-width of the time averaged mean velocity

u u
profile (see Eq. 1.2.l5.a)
bT(x), half-width of the time averaged mean temperature

or concentration profile (see Eq. 1.2.l5.b)
B(x) boundary of the jet (see Eq. 2.0.3)
C invariant of a buoyant jet (C ~ 0.55, see Eq. 2.0.1

and Section 5.1)
coefficients for the distribution of the mean velocity

along the jet axis (see Eq. 1.3.3)
coefficients for the distribution of the mean temperature

along the jet axis (see Eq. 1.3.4)
c specific heat of water, 1 cal/gr °c

p
D = slot thickness
C/2, experimental constant (see Eq. 2.1.10)
frequency of the laser beam
frequency (in Hz) of the hot/cold, cold/hot interface

crossings (see Eq. 3.5.3)
g gravitational acceleration (981 cm/sec 2 )
G
r
= Grashof number, Sx 3 /v (see Section 2.2)
h heR), buoyancy function (see Eq. 2.3.6)
= coefficients (see Eq. 2.3.10)
H
o
= heat flux at the orifice of the jet, auT D
000
transport of heat by mean flow (see Eq. 6.1.5)
turbulent heat transfer (see Eq. 6.1.6)
= see Eq. 6.2.4
l(x,y,t) characterizes the presence (1 = 1) or absence (1 = 0)

of the temperature field of the jet at a point (x,y)
at time t (see Section 2.0)
-2-
K1T spreading rate of the mean temperature profile (see

Eq. 4.1.2)
K2T non-dimensional virtual origin of the mean temperature

profile (see Eq. 4.1.2)
K spreading rate of the mean velocity profile (see

lu
Eq. 5.1.2)
K non-dimensional virtual origin of the mean velocity

2u
profile (see Eq. 5.1.2)
~ distance of the closest measurement to the jet orifice

min
(see Figure 5.2.4)
~ distance of the most distant (from jet orifice)

max
measurement (see Figure 5.2.4)
m m(x), kinematic momentum flux (see Eq. 2.1.20)
m kinematic momentum flux at the jet orifice, u 2 D

0 o
Max T Max T(x,y,t), maximum temperature excess at a point (x,y)
during the sampling time T (see Section 4.3)
~x TM ~x T(x,O,t)
Min T Min T(x,y,t), mlnlmum temperature excess at a point (x,y)

during the sampling time T (see Section 4.3)
Min TM Min T(x,O,t)
p p(x,y,t), instantaneous pressure deviation above the

hydrostatic ambient pressure
P p(x,y), time averaged mean pressure
q q(R), entrainment function (see Eq. 2.3.7)
qo,ql = coefficients (see Eq. 2.3.10)
R
e
= Reynolds number
R Richardson number at the jet orifice (for a pure jet

0
R ~
o
° and for a pure plume R ~ 0.62, see Section 2.3)
0
R R(x), local Richardson number (see Eq. 2.3.1)
R Richardson number of a pure plume (R ~ 0.62, see

P Section 5.1) p
-3-
R(t,x,O) see Eq. 6.2.3
value of R(t,x,O) on the jet axis
flux Richardson number (see Eq. 2.2.16)
T(x,y,t), instantaneous temperature excess (above the

ambient) at a point (x,y) at time t
T T(x,y), time averaged mean temperature excess (above

the ambient) at a point (x,y)
T ambient temperature
a
T (x) cross sectional mean excess temperature (see Eq. 2.3.3)
c
TM(X), time averaged mean temperature on the jet axis
T' T'(x,y,t), turbulent component of the temperature,
T' = T - T
= T'2(x,y), intensity of turbulent fluctuations of
temperature (see Eq. 4.2.2)
T'2 Th 2 (x,y), turbulence intensity associated with the

hot
sm~£l eddies (see Eq. 4.2.12)
T,2 T~~ld(x,y), turbulence intensity associated with the
cold
large eddies (see Eq. 4.2.11)
T excess (above the ambient) temperature at the jet orifice

o
u u(x,y,t), instantaneous vertical velocity at a point (x,y)
-u u(x,y), time averaged mean vertical velocity at a
point (x,y)
= ~(x), time averaged mean vertical velocity on the

jet axis.
u' = u'(x,y,t), turbulent component of the vertical velocity,

u' = u - U
u'2(x,y), turbulence intensity
u velocity at the orifice of the jet

o
v v(x,y,t), instantaneous horizontal velocity at a point
(x,y)
-4-
v = v(x,y,t), time averaged mean horizontal velocity

VI vl(x,y,t), turbulent component of the horizontal
velocity, VI = V - V
x coordinate axis in vertical direction, same direction

as gravitation (see Figure 2.1.1)
x = distance along jet axis (i.e. X = x)

x virtual origin (see Figure 2.1. 3)
o
virtual origin of the mean temperature profile (see
Section 4.1)
x virtual origin of the mean velocity profile (see

ou
Section 5.1)
Xl distance from the virtual origin along jet axis
Xl same as Xl
y = coordinate axis in horizontal direction (see Figure

2.1.1)
y distance from jet axis (i.e. Y y)
Greek Notation
a.
o
= thermal expansion coefficient at the jet orifice (see
Eq. 2.2.5)
= a.(x) , thermal expansion coefficient (see Section 2.3,

Eq. 2.3.3)
a. universal entrainment coefficient

e
a. entrainment coefficient for a pure jet
j
a. entrainment coefficient for a pure plume
p
So = kinematic buoyancy flux at the jet orifice (see Eq. 2.2.7)
S = Sex), kinematic buoyancy flux (see Eq. 2.3.9)
y = y(x,y), intermittency (or intermittency factor) of the

temperature signal (see Section 2.0.1 for definition)
8(x) Dirac delta function

-5-
ou tn2/Kiu' experimental coefficient
~p(x,y) = p(x,y)-p , density deficiency

a
~p
o
n = non-dimensional distance from the jet axis, n y/x'
llO volume flux at the jet orifice, u D

o
ll(x), volume flux (see Eq. 2.1.9)
\l kinematic viscosity at the jet orifice

o
p p(x,y), time averaged mean density
density of the ambient fluid
= density at the orifice of the jet
PM(x) = p(x,O), time averaged mean density on the jet

axis
(J experimental constant (see Eq. 2.2.13)

u
experimental constant (see Eq. 2.2.14)
normalized turbulence intensity (see Eq. 4.2.7)
= sampling time (see section 3.1)
angle between the y-axis and the induced flow

streamlines (see Figure 2.1.2)
~(x,y), stream function of the induced flow (see

Figure 2.1.2)
Subscripts
a = quantities of the ambient fluid
m,M quantities on the jet axis
0 = quantities at the jet orifice

-6-
Superscripts
time averaged mean value
fluctuating component
Mathematical Symbols
A
exp A e , exponential function
£n2 = Naperian logarithm of 2 ~ 0.693 .... )
IT 3.141. .....
is proportional to
E{X} expectation of variable X

-7-
INTRODUCTION
1.1 INTRODUCTORY NOTE
The interest in the mechanics of buoyant jets has been accelerated
by the increased demand for shore-sited power plants which use cold
ocean water for cooling. Such plants discharge warm water back to the
ocean and it has been suggested that a fast dilution of the warm effluent
will minimize the disruption of the ecological balance. Rapid dilution
can be accomplished with a multiport diffuser structure in which jets
from numerous ports spaced along the ocean outfall merge and form a two-
dimensional buoyant jet. Hence the understanding of the entrainment
and mixing process in a two-dimensional buoyant jet is necessary for the
optimal solution of the warm water disposal problem and for an effective
design of the diffuser.
Apart from this particular problem, the mechanics of jets is
important to a wide range of problems, from jet propulsion of airplanes
and submarines, to jet pumps. From a more fundamental point of view,
the investigation of two-dimensional buoyant jets will contribute to the
understanding of free turbulence phenomena.
1.2 A REVIEW OF PREVIOUS THEORIES
Previous theories of two-dimensional pure jets, pure plumes, and
buoyant jets will be described briefly with a view to examining the
principal assumptions and results of each theory.
The flow configuration and geometry are shown diagrammatically in
Figure (1.2.1). The ambient fluid will be assumed to be of uniform

-8-
density p and motionless, except for flows induced by the jets. The
a
difference between the time averaged local fluid density at any point,
p(x,y), and the ambient density, p , will be small, so that

a
-(p(x,y) - p )/p «I,

a a
in the buoyant jet. Thus, there will be little error in using the
ambient density p to replace the local density p(x,y) in the

a
description of the inertia forces. However, the difference
p(x,y) - p will be important in the description of the gravitational

a
body forces (Boussinesq approximation).
It is further assumed that the flow is fully turbulent and that
the viscous and molecular transport of momentum and heat respectively
can be neglected relative to the turbulent transports.
Mean Velocity
Oi stri bution
UM /2
~--~----+---~----~~
Uo, po
/
(y, V)
Figure 1. 2 .1. Geometry of a buoyant jet

-9-
Under these conditions, the time-averaged Boussinesq equations for the
two-dimensional incompressible motion of a turbulent fluid with density
variations are:
(i) conservation of volume flux
~+~ o, (1.2.1)
ax ay
(ii) conservation of momentum
- au - au (1.2.2)
u-+ v-
ax ay
- av - av
u-+ v-
_ l:..-.lE. _ a~ aV"2'
-- -" (1.2.3)
ax ay p ay ax ay
a
where u', and v' are the deviations in velocity from the time-averaged mean
values of u and v respectively, and p the deviation from hydrostatic pressure.
(iii) conservation of tracer
-3c
u-+v-=-
-3c aU'e' aV'e' (1.2.4)
ax ay ax ay
where c is the time-averaged tracer concentration and c' the deviation
from this mean value. An equation must also be provided for the
relation between the density and the tracer concentration. The linear
form
ex.
o
c(x,y) (1.2.5)
is assumed, where ex. is taken to be constant.

o
The above system of five equations (1.2.1) to (1.2.5) has ten
-10-
unknowns u, v, u'2, c, c'2, v'2, p, u'c', v'c', p, and therefore cannot
be solved even if the appropriate boundary conditions were provided and
this is the fundamental problem of turbulent fluid mechanics.
An extremely large number of methods of dealing with this general
problem has been proposed but the existing work on two-dimensional free
shear flows (jets, plumes, wakes, etc.) can be classified into one of
two schools of thought. The first school tries to derive a constitutive
equation, i.e., a relation (not necessarily linear) between the Reynolds
stress tensor and the deformation tensor of the fluid. The early
workers in this school assumed, in addition, self-preservation for the
mean and fluctuating quantities (see, for example, Tollmien [45], G. I.
Taylor [46], or the textbook by Schlichting [47j, pp. 590-613). More
recent workers use more complex constitutive equations and solve
numerically the appropriate system of equations (see, for example,
Saffman [48]). The validity of this approach has been criticized tY a

growing number of investigators, including Laufer [49] and Coles [50].
Batchelor [51] also pointed out the defects of the early models of this
school.
A second group of investigators integrates the equations of motion
across the jet and derives conservation integrals for the mass, momentum
and buoyancy fluxes (see Corrsin and Uberoi [52], Morton et al [20],
Brooks and Koh [53], List and Imberger [23]). For a two-dimensional
buoyant jet in uniform and still environment this approach leads to
equations of the form:
(i) conservation of volume flux

-11-
~x JOO -;-(x,y)dy -2 lim v(x,y) (1.2.6)

-00 y-++oo
(ii) conservation of momentum
d Joo (-;-2 + U'2 + .E.)dy =_Joo p(x,y)-P a gdy-2lim [-;- v + U'V ' ]
dx -00 Pa -00 Pa y-++oo
(1.2.7)
(iii) conservation of tracer
OO
J [-;- c + [c v +
d
U I C '] dy = -2 lim C I V '] (1.2.8)
dx -00 y-++oo
The following assumptions are then generally made:
(i) The integration is assumed to be extended to infinity, hence it is
argued that
lim [-;- v+ U'V ' ] 0

y-++oo
lim [c v + C I V I ] 0
y-++oo
(ii) It is assumed that the turbulent tracer transfer u'c' is negligible
in comparison with u c.
(iii) It is assumed that ~ + .E. «-;-2. This has been justified by the
P
experimental work of Miller and Comings [12].
These assumptions then lead to the following set of equations:
(i) conservation of mass
~x Joo -;-(x,y)dy -2 lim v (x,y) , (1. 2.9)

-00 y-++oo
-12-
(ii) conservation of kinematic momentum flux
p(x,y)-P
a
~x J u2 (x,y)dy =-J
00 00 - - - - - gdy , (1.2.10)
-00 _00 Pa
(iii) conservation of kinematic buoyancy flux (from Eq. (1.2.8) and
Eq. (1. 2 . 5) )
d Joo p(x,y)-P a
dx gudy o . (1.2.11)
-00 Pa
A more accurate description of the basic jet mechanics than repre-
sented by the above equations and assumptions will be presented in
Chapter 2, where the deficiencies of Eq. (1.2.10) and Eq. (1.2.11) will
be discussed in more detail.
The Eqs. (1.2.9), (1.2.10) and (1.2.11) have provided the basic
framework for almost all the previous investigations of jets, plumes and
buoyant jets. From this point, two further approaches have been used to
obtain the basic jet behavior.
I. Entrainment Coefficient Approach
The entrainment coefficient approach (Brooks and Koh [53], Morton
et al [20], Morton [24], Fan [58], Fan and Brooks [22]) is based on the
hypothesis of G. I. Taylor [20] that the entrainment should be propor-
tional to some characteristic velocity of the jet, i.e.
-lim 2V(x,y) (1.2.12)

y-++oo
where the entrainment coefficient a was assumed by Morton [24] to be a

e
universal constant with the same value for jets, plumes, and
-13-
buoyant jets. Similarity of the velocity and temperature profiles has
also been assumed and, on the basis of the experimental evidence, they
have been given by the Gaussian curves
~(x,y) (1.2.13)
= (p (x)-p ) exp[-~n2 (y/b )2] , (1.2.14)

M a T
where bu(x) and bT(x) are the characteristic lengths of the velocity
and temperature profiles respectively, and are defined by the relations
~(x,±b (x» = ~(x,0)/2 , (1.2.l5.a)

u
p(x,0)/2 (1.2.l5.b)
and where ~(x) = ~(x,O) and PM(x) = p(x,O). One further assumption has
been that
bT(x)
b (x)
(1.2.16)
u
where A is supposed a universal constant.
Under these assumptions the system of equations (1.2.9), (1.2.10)
and (1.2.11) becomes
- 1/2
20'.e ~(~n2/TT) (1.2.17)
(1.2.18)
-14-
(1.2.19)
where ~PM(x) = PM(x) - Pa ' The system of the ordinary differential
Equations (1.2.17), (1.2.18), and (1.2.19) with appropriate initial
conditions can be solved numerically (see Fan and Brooks [22]) to find
the evolution of ~(x), bu(x) , and ~PM(x).
It seems worthy to examine closely the two particular cases of a
pure jet and of a pure plume, where analytical solutions can be obtained
for the equations (1.2.17), (1.2.18) and (1.2.19). For a pure jet
p(x,y) = Pa , hence the kinematic momentum flux is conserved, i.e.
(1.2.20)
where u the exit velocity and D is the width of the jet at the orifice.
o
Then the system of Equations (1.2.17) and (1.2.20) can be solved
explicitly to give
(16 £n2/1T)1/2 Ct. (1.2.21)

J
where K a constant and Ct is the entrainment coefficient for a pure

1u j
jet.
The equation (1.2.21) can be written
b
u
-=
D
The decay of the centerline velocity is then given by Eq. (1.2.20) as

-15-
U)2 - ~
( TIM
o
~n2
_ rr
(D
x + (1.2.22)
and the centerline dilution of a passive tracer is given by
- )2 = (lz
To
-- --- 1 x
Klu(-D + K2u ) , (1.2.23)
1+(:~)2
(
'EM tn2
where To and TM(x) are the excess tracer concentrations above ambient
at the jet exit, and on the jet axis, respectively.
For a plume it is easy to verify that the system of Equations
(1.2.17), (1.2.18) and (1.2.19) admits the following solution (see Lee
and Emmons [27J)
~(x) A S 1/3 "" constant ,

o 0 (1.2.24)
LlPM(X) -1
Alx (1.2.25)
(po -p)
and
db
--- =
u ( 4 £ n2 /rr ) 1/ 2 a (1.2.26)
dx p
where AO,A , are constants which can be easily determined. a is the

l p
entrainment coefficient for a plume and
-16-
the initial kinematic buoyancy flux if po is the density of the jet at
the orifice.
Since the basic assumption of this approach is that the entrain-
ment coefficient is constant, i.e.
a., a. a.
J p e
it is concluded from (1.2.21) and (1.2.26) that the jet half-width b

u
should be twice the plume half-width. This is indicated also in the
numerical solution of Fan and Brooks ([22], Figure l7(e». Brooks [67]
adopted the value a.

e
= 0.14 since he believed that the flows of interest
to him were more plume-like than jet-like, hence
db
u
dx = Klu 0.26 for a jet,
and
db
u
0.13 for a plume.
dx
The experimental results of this investigation for pure jets and
plumes have shown that the coefficient K is approximately the same

lu
for both jets and plumes (see Section 5.1 and Chapters 8 and 9). The
assumption of a universal entrainment coefficient is therefore incorrect
and in reality a.
a., J
J 2
-17-
II. Dimensional Analysis
Dimensional analysis can be used to find the basic characteristics
of the jets and plumes. For a pure jet in uniform environment, Eq.
(1.2.10) implies that the kinematic momentum flux is conserved, i.e.
constant.
Dimensional analysis then gives for a pure jet (Landau and Lifshitz [63])
b ,... x
u
-1
x
For a plume,dimensional analysis gives (see Batchelor [54])
~,... S~/3 ,... constant ,
b (x) ,...x
u
-1
,...x
List and Imberger [23] used dimensional analysis to write the integrated
equations of motion in the form
d].l m
q(R,C) (1.2.27)
dx ].l
dm
-=
dx
l:!f heR C) (1.2.28)
m '
dS
-= 0 (1.2.29)
dx
-18-
where II
C
(mx) 1/2
and ]J JOO u(x,y) dy, m

_00 -00
00 P-P
S = -J --
P
a
g udy
_00
a
Their approach was to find the leading terms of the functions q(R,C)
and h(R,C). Their basic idea has been extended in this investigation
(see Section 2.3).
1.3 REVIEW OF PREVIOUS EXPERIMENTS
1. Pure jet
Usually an experimenter measures the mean velocity and concentra-
tion profile at different distances from the exit, he plots the growth
of the half-widths of mean velocity and temperature profile, the
centerline decay of mean velocity and the centerline dilution and
finds that
b
u
D
(1.3.1)
(1.3.2)
(1.3.3)
(1.3.4)
-19-
The coefficients Klu ' K2u ' KIT' K2T , Clu ' C2u ' CIT' C2T are then
usually empirically determined. Alternatively some investigators report
the fitting of their mean velocity results in the form
u(x,y)
= e (1.3.5)
~ (x,y)
m
without determining the location of any virtual origin. In this case
the coefficient K can be calculated by

lu
K =~ Q,n2
lu cS
u
Table (1.3.1) summarizes the principal parameters and results for the
mean velocity and mean temperature found by other investigators. The
coefficients K ' K ' C ' C for Forthmann's [3] experiments were

lu 2u lu 2u
calculated from his graphs by the present investigator. The experimental
investigation of Heskestad [5] was good but the mean velocity at the jet
exit, u , was not reported. However, it can be estimated from the

o
reported Reynolds number, so the coefficients C and C were calculated
lu 2u
by the present investigator using an inferred value for u , (they were
o
also calculated by Flora et al [4]). Mih's [11] experiments were carried
out by discharging the jet into a diffusion chamber which was too small
to ensure a "free" jet. To overcome this problem, he pumped water out of
the diffusion chamber which produced a current in the chamber, thereby vio-
lating the assumption that the ambient flow be a jet-induced flow only.
The centerline velocity decay (Figure 7, [llJ) is apparently incorrect
because all of his experimental points indicate that he measured more

Experimenter X R
Aspect e K K C C KIT Kn CIT C Reference
from Ii to Ratio x 10 3 lu Zu lu Zu n
Miller 0 40 40 20 0.096 1.15 0.181 -0.62

-- -- -- -- [12]
0.163 1.05 [39]
Hesk~stad 47 155 120 34 0.11 -4.2 0.27 -10.0 -- -- -- -- [ 5]
Zijnen 10 40 20 13 0.096 0.6 0.162 0.6 0.141 0.6 0.25 0.6 [16]
10 35 25 13 0.099 1.7 0.205 1.7 -- -- -- -- [38]
Forthmann 0 25 21 10 0.098 0.33 --
0.18 0.11 -- -- -- [ 3]
Jenkins. 10 60 24 14 0.088 4.5 0.16 4 0.123 4.8 -- -- [8]
10 60 24 14 0.091 3.0 -- -- 0.128 3.6 -- -- [8]
10 60 24 14 0.096 2.5 -- -- 0.137 3.2 -- -- [8]
Knystautas 50 295 50 40 0.106 -1.5 0.18 -1.4 -- -- -- -- [7]
Gartshore 100 100 130 20 0.115 -- -- -- -- -- -- [9]
0.102 I -2~~ 0.164 -2.0 [ 37]
I
N
Flora
o
0 90 50 20 0.109 0.0 -- -- -- -- -- -- [ 4] I
Goldschmidt 15 67 15 16 0.099 0.66 0.207 -5.83 -- -- -- -- [10]
Bradbury 15 70 48 30 0.109 -- -- -- -- -- -- -- [ 2]
Mih 6 300 58 20 0.116 -- 0.175 -- -- -- -- -- [11]
Albertson 1 2000 300 10 0.126 -- 0.192 -- -- -- -- -- [I]
Reichardt 0 100 21 24 0.108 -- -- -- 0.17 -- -- -- [13]
Vu1is 0 15 15 80 0.109 0.64 -- -- -- -- -- I -- [14]
Hetsrolji 10 30 26 8 0.128 -- -- --
-- -- -- -- [6]
Nakaguchi 10 100 133 10 0.106 -0.3 0.189 -0.3 -- -- -- -- [37]
Tail1and 10 100 20 0.10 0.0 -- -- -- -- -- -- [42]
Householder 16 0.091 1.46 0.193 -6.98 -- -- -- -- [4]
[44]
Bicknell 15 125 60 30 0.115 -2.7 0.112 6.7 0.17 -Z.O -- -- [57]
--
Table 1.3.1. Review of previous experiments for a two-dimensional pure jet
-21-
momentum than the jet had originally (of the order 25% at x/D = 15).
The opinion of this investigator is that his coefficients C ' C

lu 2u
should be discounted. He also did not calculate the location of the
virtual origin.
The experiments by Miller et al [12] can be considered of high
quality. They verified experimentally that ~ and ~ are of the same

p
order and of opposite sign. Their experimental results are tabulated
in Miller's Ph.D. thesis ([39], Table 18) from where this writer
calculated C ' C • The writer also calculated the coefficients K '

lu 2u lu
K for the experiments by Vulis et al [14] but these should also
2u
probably be discounted because of the limited accuracy involved in
estimating the half-width from their very small published figures.
The variation of the coefficients K ' KIT' found by Jenkins et al

lu
[8] is, in this investigator's opinion, due to minor instrumentation
problems rather than to the argument presented by them that K ' KIT
lu
increased with increasing the exit temperature. It seems unlikely that
this is the case when the initial Richardson number was of the order
10-
5
(~ pure jet), and the measurements were in the same region (10 to
60 diameters from exit). Jenkins et al [8] claim that the centerline
2
temperature T decays like x- , which is not in agreement with any other
m
theory or experiments. Hestroni et al [6] did not calculate the loca-
tion of the virtual origin and probably this is the reason why their
K is large. Reichardt [13] did not calculate the virtual origin nor
1u
the centerline velocity decay. The writer calculated the coefficients
K ' K2u from Knystautas' [7] experimental results presented in his

lu
Figure 6 and the coefficients C ' C from his Figure 8. The
lu 2u
-22-
coefficients C ' C from Flora et al [4] should probably be discounted

lu 2u
because they calculated the exit velocity from the measured momentum of
the jet at a downstream position and assumed conservation of the kine-
matic momentum flux, an assumption that will be shown to be questionable
(see Chapter 2.1). Bradbury [2] investigated a plane jet in a slow
secondary current hence his coefficients C ' C are not appropriate to

lu 2u
this study. The coefficient K for Tailland et aI's [42] experiments
lu
was calculated by the present investigator from their Figure 3.
The writer calculated the coefficients K ' K ' C ' C from

lu 2u lu 2u
Bicknell's [57] experimental results presented in his Figure 8, and the
coefficients KIT' K2T from his Figures 12, 13 and 14.
Turbulence intensity studies have been reported by Miller [12],
Heskestad [5], Zijnen [17], Bradbury [2], and Vulis [14]. The turbulence
energy balance was investigated by Heskestad [5] and Bradbury [2].
Intermittency measurements were reported by Heskestad [5] and Hannum [40].
II. Pure plume
Rouse et al [21] considered experimentally the pure two-dimensional
plume, simulated in their investigation by a line gas burner. This kind
of experiment is very delicate because small convection currents in the
ambient air usually exist and the assumption of a motionless environment
is most probably violated. Their experimental measurements of velocity
show substantial scatter either because of inadequate instrumentation or
because of existing ambient air currents. They found
u(x,y) u T(x,y) T
m m
-23-
which give K
lu
= 0.147 and KIT = 0.130. They did not know the input
heat flux ~ priori, but calculated it from the measured mean velocity
and temperature profile assuming implicitly that u'T' « u T. It will
be shown later that this is a questionable assumption (see Chapter 6
on temperature-velocity results). Kotsovinos and List [25] measured
directly the entrainment in a pure plume, simulated by a heater in a
water tank. The input heat flux was known and, although the velocity
and temperature profiles were not measured, the results can be combined
with the results of the present investigation to give K

lu
= 0.11.
Ellison et al [26] measured an entrainment coefficient for the half-
plume equal to 0.112, which may be used with (1.2.18), to give K

lu
= 0.105.
Lee's Ph.D. thesis [55J is the basis of the Lee and Emmons [27] publica-
tion. They measured temperature profiles above a line gas burner, from
which the coefficients KIT and K2T can be calculated as 0.130 and 0.9
respectively (see [55], Figure 5-2). Velocities were not measured, but
they tried to estimate the entrainment coefficient a and the growth of

p
the velocity half-width K from temperature measurement alone which, in
lu
principle, is not possible without any implicit or explicit assumptions.
In their case, they assumed implicitly, that the turbulent heat transfer
J"" u'T' dy was negligible with respect to the mean heat transfer J"" u T dy,
-00 -00
which will be shown to be incorrect (see Chapter 6). The conclusion of
their calculations (not measurements) that a

p
= 0.16 and KlT/Klu = 0.88
should therefore be discounted.
The experimental measurements of concentration by Anwar [62] show
substantial scatter. He did not determine the location of the virtual
origin. He suggested KIT = 0.13.

-24-
III. Buoyant jet
To the best of the knowledge of this writer~ no experiments have
been reported concerning a two-dimensional buoyant jet in transition
from a jet to a plume.
1.4 SCOPE OF THIS WORK
The scope of this work is the systematic study of the entrainment
and mixing processes in a two-dimensional buoyant jet in its transition
from a pure jet to a pure plume. For this purpose an organized set of
experiments over a wide range of initial Richardson number has been
carried out to improve the understanding of the buoyant jet mechanics
and to make possible the development of a theoretical model which is
based on correct physical assumptions. The practical outcome of this
investigation is to provide a definite input for the design of two-
dimensional buoyant jet type diffusers. From a broader point of view~
the attitude is to study the jet problem in a rigorous way with the hope
that the understanding of the fundamental laws of free turbulence flows
will be improved.
-25-
2. THE MECHANICS OF JETS
2.0 BASIC DEFINITIONS AND ASSUMPTIONS
The two-dimensional turbulent pure jet, pure plume and buoyant jet
problem will be studied. The fluid (in this investigation, water) is
assumed to be incompressible. The ambient fluid is assumed to be motion-
less, except for the movements induced by the jet, and of uniform density
p. The pure jet, plume, or buoyant jet is assumed to be turbulent and

a
therefore viscosity and molecular diffusivity will be neglected. Density
variations will be assumed small, so that the density difference is
important only in the gravitational body force term in the equations of
motion.
A pure jet is defined as a source of kinematic momentum flux, m

o
and/or kinematic mass flux ~ . a pure plume is a source of kinematic
0'
buoyancy flux 8 only. A buoyant jet is a source of kinematic mass flux

o
~ , kinematic momentum flux m , and kinematic buoyancy flux
0 0 0
8. Using
dimensional analysis. it can shown that each buoyant jet has effectively
two characteristic length scales. One, ~2/m , defines the importance of

o 0
2 3
the initial volume flux (or kinematic mass flux), and the other m /8 /
o 0
defines the importance of the initial momentum flux relative to the
buoyancy. The ratio of these two scales is the non-dimensional number
which expresses the ratio of the distance at which kinematic momentum
flux is important relative to kinematic buoyancy flux, to that at which

-26-
kinematic momentum flux is important relative to kinematic mass flux.
The non-dimensional number, R, is called the Richardson number of the
jet. The Richardson number, calculated at the jet origin, assuming
uniform velocity and temperature profiles, can be written
~ g D
R Po
o
i.e., the Richardson number is the square of the inverse of the densi-
metric Froude number. For a pure plume, the local Richardson number
~ 3 (x) 6
R
P m3 (x)
will be shown (see Chapters 2.2 and 5.1) to be a constant (NO.6), while
for a pure jet it is equal to zero. A buoyant jet with an exit Richardson
number R « R , can be considered to be a pure jet for a distance

p o
2 3
x < m /6 / (where x is the distance from exit). However, since the
o 0
local Richardson number continuously increases, eventually, for

2 3
x » m /6 / , lt "h
" Wl"11 approxlmate t e constant va 1 ue 0 f t h e p 1 ume
o 0
Richardson number R. The conclusion from this discussion is that a
p
buoyant jet is always in transition towards a pure plume.
Another important invariant of a buoyant jet is the non-dimensional
ratio C defined by (see Section 5.1)
~(x)
C (2.0.1)
1m (x) (x-x o )
where x is the location of a virtual origin. The invariant C, which is

o
-27-
a measure of the angle expansion of the buoyant jet, has been found to be
almost constant at the same value (0.55) for jets, plumes, and buoyant
jets. The virtual origin is then defined by applying Eq. 2.0.1 at the
orifice of the jet, i.e., at x = 0 so that
D
x - ~ ~ - 3.3 D . (2.0.2)
o C
This definition of the virtual origin is quite different from the virtual
origins x ,x T of the mean velocity and temperature profiles respec-

ou 0
tively defined from the relations
x
ou
= -K2u D
The virtual origin defined by Eq. 2.0.2 is always upstream of the jet
orifice; the experimentally determined virtual origins x ' xoT were

ou
found (see Table 1.3.1) to be upstream or downstream of the jet orifice.
This present study will explain that these variations are due to the non-
self-preserving character of the jet.
In its initial stage of development, (5-6 characteristic lengths
for a pure jet), a plane jet consists of two plane mixing layers. After
the two mixing layers have merged for some distance, the jet becomes
fully developed. Viewed instantaneously an irregular and sharp interface
separates the jet fluid which is turbulent from the ambient fluid which
is irrotational. If the jet or ambient fluid has a tracer, this sharp
interface also divides fluid of different concentration (see Coles [43]).

-28-
The difference that can be seen in a short exposure photograph and time
exposure photograph of a jet is dramatic (see Werle [36] plates A, B).
In a time exposure of a jet, a fictitious interface (a plane) appears
which separates fluid that is sometimes turbulent from fluid that is
never turbulent. This plane will be called the boundary of the jet.
It seems necessary to clarify the idea of the boundary of a jet.
For this purpose, first the intermittency YA(x,y) of a quantity A
of jet will be defined as
YA(x,y)
where
if quantity A is present at
X,y at time t
IA(X,y,t)
otherwise
The vorticity of the fluid or the presence of tracer is usually
used to characterize the presence of the jet at any point (x,y) at time
t. The boundaries of the jet can then be defined by the two lines
y =± B(x) , (2.0.3)
where 2B(x) is the width of the intermittency profile y(x,y) when Y ~ O.
The experimental determination of Y for values close to zero is difficult
and depends on the chosen threshold level for the quantity A. Although
an estimation of B(x) for practical purposes can be obtained (see Figure
4.4.1 and 4.4.2), this estimate must necessarily be a function of the
threshold level. The profile of maximum temperature can also be used to
estimate B(x) (see Section 4.3).

-29-
Therefore, given any threshold level by which to determine the inter-
mittency, the boundaries of the jet y = ± B(x) describe an inherent

property of the jet, which is quite different from the width y = ± b(x),
which is defined by
T(x,o)
T(x, ± b(x)) (2.0.4)
N
N can be any real number larger than one but it usually has the value
two, in which case b(x) is called the half-width.
t=-B(x) y(X,y»o
y (X, y) = 0 y(X,y)=o
Figure 2.0.1. Definition of jet boundary

-30-
2.1 THE MECHANICS OF THE TWO-DIMENSIONAL PURE JET
A pure jet is defined here as a steady source which delivers kine-
matic mass flux ~ , and kinematic momentum flux TIl , through a slot of
o 0
thickness D to an ambient fluid. The geometry and the axes of coordinates
of Figure 2.1.1, indicating a jet out of a wall, will be considered.
(X, U)
Y = bu ( X )
y=-B(x) tJ
Mean Velocity Profile
( y, V )
Figure 2.1.1. Geometry of a jet
The jet will be assumed turbulent, which according to Sato et al [66]
requires a Reynolds number at the exit greater than 500. The equations
are the
-31-
continuity equation
(2.1.1)
and momentum equations,
a [u2 + u' 2 + .E.] + ~ [~v + u' v' ] a , (2.1.2)

ax p ay
a . (2.1.3)
The turbulent fluctuations are zero at the boundaries of the jet, so that
u'v' (x, ± B(x» = ~(x, ± B(x» = -;t2(x, ± B(x» a (2.1.4)
It is not possible to solve the Reynolds equations with the present
state of the art of free turbulence; the Equations (2.1.1), (2.1.2),
(2.1.3) are therefore integrated across the jet using the boundary
conditions (2.1.4) to give
d IB(x) u(x,y) dy + 2V"(x, B(x» = 2 dB (x) u(x,B(x» (2.1.5)

dx dx
-B(x)
~x IB(x) [u2 + U
l2
+ £j dy + [u v]B(x)= 2 dB(x) [u2 (x,B(x» +~] (2.1.6)
-B(x) P -B(x) dx p
~x IB(x) (U V" + u'v') dy - 2 dB(x) u(x,B(x» V"(x,B(x» =a

-B(x) dx
-32-
Now consider each of the terms in equation 2.1.6 in turn. The term
d J B(x) (li2 + u'2 +.E.) dy

dx -B(x) P
is what Benjamin [68] has termed the flow force of the jet and represents
the rate of change of the total flux of momentum and the pressure force
along the jet. Miller and Comings [12] (Figures 7 and 8), and Bradbury
[2] (Figures 8, 9, 10, and 11) have investigated the contributions from
turbulent momentum flux and the pressure, and have shown that these
contributions are approximately equal in magnitude and opposite in sign.
Their experimental results indicate a weak positive pressure gradient on
the jet axis and this will tend to reduce the momentum flux of the jet.
We will therefore ignore these contributions to the integral. The second
term
[u V]B(X)
-B(x)
represents the flux of axial momentum into the jet from the outer flow
entrained into the jet.
Morton et al [19] and Corrsin and Uberoi [52] ignored this term by
assuming that axial component of velocity u(x) vanished outside the jet.
The term on the right-hand side
2 dB(x) [u2 (x,B(X) +.E. (x,B(X»]

dx P
represents the flux of vertical momentum and the component of axial

-33-
pressure force that act on the sloping side of the jet as defined by
y = ± B(x). The pressure component is small and, as Miller and Comings
[12] show, represents a decelerating component of force and it will be
ignored in the subsequent analysis because it is difficult to characterize.
However, recognizing that it is a decelerating component, we can recognize
its influence. The momentum flux term is also small but can easily be
incorporated into the analysis later as will be shown.
We believe that an improvement can be achieved by examining the
error involved in neglecting the above terms associated with the induced
flow field, and which have been ignored by previous investigators. The
procedure will be to assume first that the terms usually neglected are
negligible. From this we will obtain an approximate solution, and then
we will determine the correction to this approximation. The equations
(2.1.5), (2.1.6) are now written as
d).l
-2v(x, B (x» ,
dx
dm
dx
m 6(x)
0
, (2.1.8)
where
).I (x) = JB(x) u(x,y) dy ,

-B(x)
m(x) = J B(x) u2 dy (2.1.9)

-B(x)
and mo is the kinematic momentum flux at the exit of the jet. From
-34-
(2.1.8) m(x) m so that using dimensional analysis

o
d~ = f(m,x) f(m ,x) x

-1/2
(2.1.10)
dx 0,
where El is a constant to be determined experimentally, (E

l
= 0.28, as
will be shown in Section 5.1). Hence the kinematic mass flux increases
+1/2
as x and a flow in the region where the intermittency is zero must be
induced towards the jet. The ambient fluid therefore sees the jet as a
. 1/2 -1/2
distribution of planar slnks of strength Elmo x . The induced flow
is irrotational (since at y = ±oo the vorticity is zero and a slip condi-

tion is assumed). Lippisch [29] has an illustrative picture of such a
flow induced by a jet out of a wall.
Now, for the ideal case of a series of sinks in a straight line from
x =0 to infinity (see Figure 2.1.2) of strength given by Eq. 2.1.10,
the streamlines ~(x,y) of the induced flow, which satisfy the Laplace
equation V2~(x,y) =0 and the boundary conditions
~(O,y) 0
~x (~(x,O) x
-1/2
(2.1.11)
were calculated by G. I. Taylor [28] to be
~(x,y) (2.1.12)
Clearly these streamlines represent parabolas with slope ¢

-35-
y = 0, (see Figure 2.1.2). For the case of a real jet out of a wall, the
flow field depicted schematically in Figure 2.1.3 will be assumed.
Specifically it will be assumed that the jet originates as a source of
kinematic momentum flux (only) m at a distance x = ~ /C 2m ~ 3.3 D

0 0 0 0
upstream from the jet orifice (see Eq. 2.0.2 and Figure 2.1.3), and that
sinks of intensity given by Eq. 2.1.10 are distributed on the xl-axis
from the virtual origin 0' to infinity. The angle ¢ (see Figure 2.1.3)
will be assumed constant at a value rr/4. The existence of the potential
core is neglected in this approximation.
The procedure now will be to consider the momentum Eq. 2.1.6 and to
discuss the terms which were previously neglected. From Figure 2.1.3
u(x, B(x)) tan ¢ vex, B(x))
Hence [uv]B(x) 2 tan ¢ v 2 (x, B(x)) , (2.1.14)

-B(x)
and combining (2.1.14) and (2.1.5), it is seen that
[-u -v]B(x) = -1 tan ¢ (d~)2/(

-- 1 - dB(x) tan ¢) 2 (2.1.15)
-B(x) 2 dx dx
Physically, Eq. 2.1.15 represents the x-component of the momentum of the
induced flow. The momentum equation (2.1.6) now becomes
or
-36-
..p(X,y)
Figure 2.1.2. Flow towards a line distribution of sinks

of strength given by Eq. 2.1.10.
Xl
y=-8(x) y=B(X)
tIll 1111
-u
Virtual_---.lt----~~ y
Ori!;lin
Figure 2.1.3. Schematic of the assumed induced flow and

virtual origin of a real plane jet.
-37-
dm 1 d 2
-dx = m0 o(x) 2 tan ¢ (d~) /(1 - ~ tan </» (2.1.16)
where
dB(x)
~ 0.25 (see Section 4.3).
dx
We observe that the momentum of the induced flow (second term of the
R.H.S. of the Eq. 2.1.16) tends to reduce the kinematic momentum flux
defined by Eq. 2.1.9.
A solution which would predict precisely the kinematic momentum
flux m(x) appears to be very difficult. However, a rough estimation of
the reduction of the kinematic momentum flux due to the opposing momentum
of the induced flow can be obtained using the simplified flow shown
schematically in Figure 2.1.3.
Again using dimensional analysis, we write
djl = E m1/2 x -1/2 (2.1.17)

dx 1
and substitution of (2.1.17) in (2.1.16) gives
dm
m o(x) - l2 E2 tan </> x- 1m/(1 - K_ tan ¢) (2.1.18)
dx o 1 -13
Considering that induced flow exists only for Xl ~ x ' Eq. 2.1.18
o
implies, with x replaced by Xl, that
-38-
E2
m(x)
m
= (~)
x+x
2~1-KB)
o 0
= (x 3 . 3 )0.0522
(2.1.19)
1)+3.3
C
where E1 2~ 0.28 and x is the distance from the jet orifice.
For example, at x/D = lOa, we obtain m(x)/m = 0.83, i.e., the

o
opposing momentum of the induced flow reduces the initial kinematic
momentum flux by 17%.
It will be shown in Section 5.2, where the experimental results of
previous investigators are examined, that this estimation of the reduc-
tion of the initial kinematic momentum flux is conservative and that
most investigators of a pure plane jet out of a wall experienced a much
greater reduction of the kinematic momentum flux. This can be explained
if we assume that the decelerating pressure gradient existing along the
jet axis has a larger negative contribution to the flow force than the
increase resulting from the turbulent momentum flux. However, this
aspect of the problem, although investigated to some extent by Miller
and Comings [12] and Bradbury [2], needs further study.

-39-
2.2 THE MECHANICS OF THE TURBULENT PURE PLUME
The convection flow generated by a steady line source of heat is
called a two-dimensional plume. In the analysis that follows, it is
assumed that the ambient fluid has only the motion induced by the plume
and is of uniform density p , and that density variations from pare

a a
small. According to Forstrom et al [31] a fully turbulent plume is
characterized by a modified Grashof number G greater than 5xl0 9 where

r
G
r
where H is the input heat flux per unit length,

0
a is the thermal expansion coefficient,
x is the distance from the heat source,
g is the gravitational acceleration (981 cm 2 /sec),
Pa is the density of the ambient fluid,

c is the specific heat,
p
v is the kinematic viscosity.
The Reynolds equations of motion integrated across the turbulent plume
are
(i) volume conservation
~x IB(x) ~(x,y) dy = -2V"(x, B(x» + 2d~~x) ~(x, B(x» (2.2.1)

-B(x)
-40-
This equation was derived from integrating the incompressibility equation
and assuming that small volume changes arising from the transfer of mass
by diffusion are neglected (see List [32]).
(ii) kinematic momentum flux
p-p
~x JB(x) (u2+u'2+~) dy=_JB(x)(_ _
a)gdy+ [uv]B(x) (2.2.2)
-B(x) P -B(x) Pa -B(x)
(iii) conservation of heat flux
J B (x) (u T+ u' T') dy (2.2.3)

-B(x)
where T(x,y) is the excess (above the ambient) temperature and p is the
pressure deviation from the hydrostatic ambient pressure.
It will be shown that the experimental results of this investigation
indicate that u'T' cannot be neglected (see Chapter 6). Density varia-
tions are related with the temperature variations by the equation
1 dp(8)
p(8) d8
-a(8) (2.2.4)
where p(8) is the density at temperature 8, (OC),
a(8) is the thermal expansion coefficient (l/oC)
at temperature 8.
For pure water and for SoC <8< SO°C the following algebraic
expression was obtained using values of a given by Batchelor [33]

-41-
a(e) (-0.773 + 0.19 e -0.0027 e 2
+ 0.000021 e 3 ) x 10- 4 (2.2.5)
(see also Figure 2.2.1). The kinematic buoyancy flux S is defined as
S = _JB(x) L (p-p )u dy ,
-B(x) Pa a
or for small temperature increments,
B(x)
S= J-B(x) ag Tu dy (2.2.6)
where T is the excess temperature increment above ambient. It is obvious
from (2.2.6) that the kinematic buoyancy flux cannot be conserved since
it depends, implicitly, through the thermal expansion coefficient, on the
temperature field of the plume. However, in order to derive the basic
features of the two-dimensional plume, the thermal expansion coefficient
will be assumed constant, which leads to
agH
B(x) o
S = ag J-B(x) uT dy = - - (2.2.7)
i.e., the kinematic buoyancy flux is conserved if a is constant.
The basic character of the line pure plume can be found
using dimensional analysis. In particular, it can be shown that
(2.2.8)
o
THERMAL EXPANSleN CeEFFICIENT (WATER)
'.'~
~ "~I--~~--~--~----~--~--~~--~--~----~--~
~o
W
0 0
u~
•
ZO
D
t-t C'l
I
eno
2('1")
o
r-I
0:" H
o
(L0 .IJ
tJ
X co
W rx.
o <1.1 I
.l::-
C"\I r-I
....J. co N
I
0 tJ
(J)
0:
:E
a:
WO
I~
l-o
o
o I ~~~ ______ ~ ______ ~ ______ ~ ______ ~ ________ ~ ______ ~ ______ ~ ______ ~ ______ ~
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0
TEMPERRTURE IN O(
Figure 2.2.1. Thermal expansion coefficient of water as a function of
temperature.
-43-
dm
-= (2.2.9)
dx
where E , E2 are experimental constants, and

1
~(x)
IB(x) u(x,y) dy ,
-B(x)
m(x) IB(x) u 2 dy
-B(x)
The conclusion from (2.2.8) is that the ambient fluid sees the two-
1/3
dimensional plume as a distribution of sinks of constant strength E 8 .
1
Hence for a plume on a half horizontal plane the induced flow will have
streamlines orthogonal to the axis of the plume as in Figure (2.2.2) and
an entrainment velocity equal to E, 81 / 3 , i.e., independent of x.

2
(X,u)
• •
• .... dt --
dx
E,
I
f33
• 2
•
(y, V)
Figure 2.2.2. Induced flow towards a line plume on a horizontal plane

-44-
This observation was used by Kotsovinos and List [25] to measure
the entrainment velocity using a flow visualization technique. The
second term of the right-hand side of Equation (2.2.2) is then zero. It
can then be shown from Equations (2.2.8) and (2.2.9) that
d~ -1/2 1/2 -1/2

dx = (€1€2 ) m x (2.2.10)
which is quite similar to the relation (2.1.17) for a pure jet, except
that here the kinematic momentum flux, m, increases as x. Two
important non-dimensional ratios can be obtained from (2.2.8) and (2.2.9):
]..l(x) €l
cp = - - = constant , (2.2.11)
Im(x)(x-x )
0
rr;
~38 €l 3
R
P
--3 = (-)
m
constant . (2.2.12)
€2
The non-dimensional ratio C is a measure of the plume angle and R is

p p
the Richardson number of the plume. Again using dimensional analysis,
two useful relations for the centerline velocity and temperature can be
obtained:
cr 81 / 3 , (2.2.13)
~ u
82/ 3
TM = ag(x-xo)cr (2.2.14)
T
-45-
where au' aT are experimental constants, and Xo locates the virtual
origin. It will be shown that the experimental results of this investiga-
tion give
a ~ 1.66 and
u
(see Section 5.3 and 4.1 respectively).
The turbulence-energy equation is in this case
(I) (II) (III)

--- --
= -u'v' 3u + --- -:"'T'"
ag u'T' _(U'L_V'L) --
~ 3u
3y 3x
(IV)
3 p' q' 2
- -- v'(-- + ---) - Dissipation (2.2.15)
3y p 2
where
Each term of (2.2.15) represents a rate of change of the turbulent energy
per unit mass, specifically:
The total time derivative of the turbulence kinetic energy (left-
hand side) is equal to production of turbulent energy by shearing stress
(I), plus production by buoyancy forces (II), plus production by normal
stress (III), plus diffusion by turbulence (IV), plus the dissipation
per unit of mass by the turbulent motion (see [34], [35]). The ratio of
buoyant production of turbulent energy to shearing stress production of
turbulent kinetic energy is called the flux Richardson number R (see [69])
f
-46-
ag~
(2.2.16)
- , - , elu
u v -
ely
From Equation (2.2.13) Sl/3 and T ~ S2/3, hence u' and T' are very
well correlated, i.e., u'T' > O.

- - elu
Also u'v' -- < 0, hence the flux
ely
Richardson number R is negative and indicates that the turbulent kinetic
f
energy is increased. The centerline turbulence intensity of a plume is
expected to be larger than the intensity of a pure jet, because for a
pure jet the main production term, the shearing stress, is zero at the
center, while for the plume the production by buoyancy should dominate.
It will be shown that this was verified by the experimental results of
this investigation (see Chapter 4).
2.3 THE MECHANICS OF THE TWO-DIMENSIONAL BUOYANT JET
A buoyant jet is defined as a source of kinematic mass flux ~ ,

o
kinematic momentum flux m and kinematic buoyancy flux S. Using
o 0
dimensional analysis, it can be found that each buoyant jet has effec-
tively two characteristic length scales. One, ~2/m , defines the

o 0
importance of the initial volume flux (or kinematic mass flux), and the
other m IS2/3 defines the importance of the initial momentum flux. As

o 0
noted previously the ratio of these two scales yields the Richardson
number
R (2.3.1)
which relates the distances at which kinematic momentum flux is important
relative to kinematic buoyancy flux, to the distance at which kinematic
momentum flux is important relative to kinematic mass flux.

-47-
For a pure plume, the local Richardson number
R(x) (2.3.1)
was shown (see Chapters 2.2 and 5.1) to be a constant (equal ~ 0.6),
while for a pure jet it is equal to zero. A buoyant jet with an exit
Richardson number R «R can be considered a pure jet for a distance

o p
x < m /82/ 3 (where x is the distance from jet orifice). But given any
o 0
significant distance to develop, any slightly heated jet will eventually
become a plume, because the initial momentum will be a small fraction of
the momentum which will be gained by the continuous action of the buoyancy
forces. The local Richardson number continuously increases, so that for

2/3
x » m /8 it will approximate the constant value of the plume
o 0
Richardson number R. On the basis of this, transition to a plume is

p
assured at any point, x, for which
> 10 •
m
o
The conclusion from this discussion is that a buoyant jet is always in a
transition towards a pure plume.
Further reiterating, it should be noted that another important
invariant of a buoyant jet is the non-dimensional ratio C defined by
~(x)
C (2.3.2)
v'm(x) (x-x)
o
-48-
where x the location of the virtual origin; this invariant C, which is

o
a measure of the angle expansion of the buoyant jet has been found to
have almost the same value (0.55) for jets, plumes and buoyant jets.
The thermal expansion coefficient a which appears in the buoyancy
flux term (Eq. 2.2.6) is, implicitly, a function of the coordinates x,y,
but for simplicity it will be assumed that a is a function only of x,
and that its value at a given cross section is given by
a(x) aCT a + Tc (x»
where T is the ambient temperature, and

a
T is the cross sectional mean excess
c
temperature defined by
H
o
T (x) (2.3.3)
c p c ~(x)
a p
The time-averaged integral equations of motion are Equations (2.2.1),
(2.2.2), and (2.2.3) with initial conditions
~(O) ~o
m(O) m
0
a gH
o 0
B(O) = Bo = ----
pc
o p
-49-
From (2.1.14)
(2.3.4)
where ~, now a function of local Richardson number, ~(R), is the angle
of the induced flow streamlines with the jet boundaries (or approximately
with the jet axis). Using dimensional analysis we can write according
to List and Imberger [23]
-2v(x, B(x))= mq(R)/(l-~ tan~ht , (2.3.5)
and
IB(x) ~ gdy = ~ heR) (2.3.6)

-B(x) p m
The governing equations then become
dJ1 m
dx = ~ q(R) , (2.3.7)
dm =~ heR) tan~ m2 q2(R) (2.3.8)

dx m - -2- J12 (1-~ tan~) ,
a[T
a
+ Tc (x)]gH 0
sex) pc (2.3.9)
p
The objective is to then find the linear part of the functions q(R) and
heR). An expansion of q(R) heR), ~(R) gives

-50-
heR) = h +h R + 0(R2 ) (2.3.10)

o 1
The coefficients to' ¢1 can be calculated immediately considering that
¢(O) = ¢o = rr/4 and ¢(Rp ) = rr/4 + ¢1 Rp = 0 ,
hence
¢(R) = rr /4 (1 - ~ )
p
Using Equations (2.3.7), (2.3.8) and (2.3.9), it can be shown that
dR 3R [ Rh + .f. (2.3.11)
dx = xC 2 q - 2
_dC - _1 [ q _C_ Rh +
2 __ q 2. tancp ]
(2.3.12)
dx - Cx 2 2 4(1-~ tan¢)
where Rand C are given by Eq. 2.3.1 and Eq. 2.3.2 respectively.
For the case of a pure jet R = 0, ¢ = rr/4, and (2.3.12) gives
or
(2.3.13)
-51-
For a pure plume R = Rp , ¢ = 0 and (2.3.11) and (2.3.12) give
o , (2.3.14)
o , (2.3.15)
or h
c2
=-
o R (2.3.16)
P
(2.3.17)
Hence the governing equations become
(2.3.18)
dB
-=
dx
with initial conditions
m(O) = mo
B(O)
-52-
On the basis of the experimental results of this investigation (see
Sections 4.3 and 5.1)
0.146, 0.252 ,
h
o
0.48 , ~ = 0.25
2.4 NUMERICAL SOLUTION OF THE BUOYANT JET PROBLEM
The theoretical analysis developed in the previous Section 2.3
results in a system of ordinary differential equations.
The numerical solution and the stability of such systems is well
established. The solution was obtained by direct step by step integra-
tion using a subroutine "MODDEQ/Differentia1 equation solver" available
at Caltech's computer center. The subroutine was based on the Runge-
Kutta-Gill method with automatic control of truncation error. The
solution of this model which gives the growth of the kinematic mass,
momentum and buoyancy fluxes will be presented in Chapter 7 with the
experimental results from this investigation.

-53-
3. EXPERIMENTAL PROBLEMS AND OBJECTIVES
3.0 INTRODUCTION
The objective of this investigation was to study the mechanics of
a two-dimensional turbulent buoyant jet for a wide range of initial
Richardson numbers. Values of the Richardson number, which describes
the relative importance of buoyancy in a jet, extended from the value
appropriate for a pure jet (zero) to that appropriate for a pure plume
(approximately 0.6). Jet buoyancy was produced by increasing the
temperature of the jet discharge over the temperature of the ambient
fluid and the buoyant jet temperature and velocity fields were measured
using calibrated fast-response thermistors and a laser Doppler veloci-
meter respectively. The velocity and temperature data obtained were
recorded magnetically in digital form and subsequently processed to
extract both mean and fluctuating values of temperature and velocity.
A variety of other data such as intermittencies, frequencies, minima and
maxima, and so forth, were also obtained. Conditional sampling was a
significant feature of the data processing.
3.1 TEMPERATURE MEASUREMENTS
Thermistors
Six small bead thermistors were used for temperature measurements
and their characteristics are given in Table 3.1.1.
Dissipation
Type Bead Constant-Still Time Constant Resistance
Diameter Water at 25°C in Moving Water at 25°C
Small bead
Veco thermistors 0.35 nun 0.75 mw/oC 27 milli- "",200 HI
type 52A26 seconds
Table 3.1.1. Thermistor characteristics.

-54-
Each bead thermistor was insulated and mounted on a stainless steel tube
thereby forming a temperature probe as shown in Figure 3.1.1. The
thermistor temperature response was measured with a bridge circuit as
shown in Figure 3.1.2. The resistors used in the bridge were precision
wire wound resistors (1% tolerance) and an integrated circuit operational
amplifier (741) was used to reduce the impedance of the output signal to
a few ohms to match the input characteristics of the analog-to-digital
converter. The electrical current passing through the thermistor was
approximately 5xlO -6 Amperes, the ohmic dissipation of the thermistor
was approximately 6xlO

-6 Watts, hence the self-heating of the thermistor
due to the ohmic dissipation can be calculated using the dissipation
constant given in Table 3.1.1 'to be less than 0.008 °e. The drift of
the thermistor circuit for an extended period of time (3 hours) was of
the order one millivolt resulting in an abeolute accuracy of the tempera-
ture measurement of the order 0.01 °e.

Each thermistor was individually calibrated and a third order poly-
nomial fitted to the set of calibration points in the least s0uares
sense
T A + A1V + A V2 + A V3
023
where T is the temperature in °e, V the analog OUCpUC from the bridge
in volts (see Fig. 3.1.3), and Ao' AI' A , A3 regression coefficients.

2
Each analog output V was fed into one channel of an eight channel
analog-to-digital data acquisition system (manufactured by Digital Data
Systems) which digitized the information and packed it on magnetic tape

-55-
3.1.1.
--'"'------
Close of the thermistor One
subdivision is one millimeter.
er
R ::: 2 :::
The for the measurement

C)
THERMIST~R CRLIBRRTI~N CURVE
C)
"~i----~--~----~--~'~i----~----~--~--~
LJ')
If)
2. 3
T=17.09+2ij.S7V-13.71V +18.66V
T=TEMPERATURE IN °c
o
o V=VOLT
ou~ ~
~
z
~
12)
a o I
we:> I1"l
V1
.......
a:" I
::>~
t-
a:
a.::
?
w
0-0
Q
Q
~~
wln D
J-C'\I
~
C)
a
If)
......
0.0 0.2 O.Ll 0.6 0.8 1.0 1.2 1.Ll 1.6
VljLT
Figure 3.1.3. Typical thermistor calibration curve.
-58-
compatible with the IBM 360175 high speed digital computer in Booth
Computing Center. The digitizing resolution was one millivolt, and the
sampling rate could be varied from 0.01 to 1600 samples per second. The
thermistor time constant was determined by moving a thermistor quickly
from air into a hot water tank (simulation of a step function) and re-
cording the response of the thermistor. The time required to obtain 63%
of the final reading was found to be 27 msec for one thermistor with
special insulation and 45 msec for the five others.
AID Sampling Rate
An investigation was carried out to determine the lowest sampling
rate and sampling time for the AID converter that would give a time-
averaged temperature T with a standard deviation of ± 0.02 °c from the
expected exact one T . Ideally, if T(x,y,t) is the temperature at any

E
point (x,y) at time t, then
lim 1
T
JT
0
T(x,y,t) dt ,
T-+oo
but what is measured is
1 N
T = -N i=l
~ T(
x,y, ti) '
where
T = sampling time (sec),
N total number of samples during T seconds,

N
= sampling rate (samples/sec),
T
ti T.
=N 1,
•
1 = 1, 2 ,N.
-59-
The question is to determine the minimum sampling time T and the minimum
sampling rate, ~, such that

T
The sampling rate was determined by first forming a temperature file data
at a sampling rate much faster (400 samples/sec) than necessary to deter-
mine a good spectrum, then calculating the mean using every sample,
every second sample and so on, i.e.,
1 M
TM(x,y) = -M L T(x,y,jt.)
i=l 1
where j 1, 2, ......... 100
N = 400 T, the total number of the recorded
samples during the sampling time T.
M the integer part of the quotient N/j, 1~0 <M<N.

i 1,2, •.•.....• N, the first, second, etc., sample
t.= time in seconds of the ith recorded sample

1
t =0
1
T = sampling time for the file in seconds (fixed)
TM(x,y) as a function of the sampling rate M/T is shown in Figure 3.1.4.
From Figure 3.1.4 it can be seen that a reasonable minimum sampling
rate is of the order of 10 samples/sec, which indicates that the cut-off
frequency is probably about 20 Hz. In order to determine the minimum
sampling time (length of record) a 10 minute file was taken at a fixed
sampling rate 10 samples/sec, i.e., an array was generated with elements

-60-
u
o
::~:t
23.03
23.02
23.01
23.00 L--_ _-=--_ _-----IL-_ _ _ _ _---.lL-_ _ _ _ _- - ' - _
10 100 1000
SAMPLING RATE (samples/ sec)
Figure 3.1.4 • Mean temperature as a function of sampling

.. ate (length of sample 50 sec).
TL , o-(L)
26.08
l)
0
w 26.04
0::
::::> 26.00
r
<t
a::
w 25.96
Cl..
~ 25.92
\J.J
I-
25.88
50 100 150 200 250 300

SAMPLING TIME (sec)
Figure 3.1.5. Mean temperature as a function of sampling
time (sample rate 10 samples/second).
-61-
T(x,y,t.) where
1
T(x,y,t.) is the temperature at a point (x,y)

1
in a plume at time t.
1
i 1,2, ........ 6000, the first, second, etc., sample
o
time in seconds of the ith recorded sample,
i. e. , t.
1
= il10
A matrix S was then calculated with elements found by the running mean
SjL as follows
L
SjL (IlL) £~1 T(x,y,t j + t£)
where
L = 50,100,150,200, ..•..... ,6000. (number of samples),
£ = 1,2,3, .......• L,
j 0,10,20,30, ........ ,J,
J+L < 6000,
t1 0,
t. j/10 seconds (determines the beginning of the file),
J
tL = L/10, sampling time in seconds, t.
Then the mean TL of all the means SjL
M
11M 2: S
j=l jL
where M J/10 is the number of the means with sampling time L/10.
-62-
The variance
and the mean TL have been plotted in Figure 3.1.5 versus the sampling
time t . On the basis of Figure 3.1.5 the sampling time was chosen to
L
be 150 to 200 seconds.
3.2 VELOCITY MEASUREMENTS
I. Description of the laser Doppler velocimeter
A laser Doppler velocimeter, with direction sensitivity and capable
of measuring the velocity of a free turbulent flow of high turbulence
intensity was developed and used in this experiment (Figure 3.2.l.a and
b). A Spectra-Physics Model 120, 5 mw Helium-Neon laser was the source
of the laser beam which had wavelength 0.6328 micron (in air). The
intensity of the light of the laser beam has a Gaussian distribution and
the diameter, where the intensity is 1/e 2 (e 2.718 ... ) of the center-
line intensity of 0.65 mm.
The laser beam was first passed through a rotating radial diffraction
grating disk (manufactured by the Dynamics Research Corporation). This
grating disk, which had 2048 lines was mounted directly on the shaft of
a synchronous motor (Hurst Corporation, Model Minibee) rotating at 1800
revolutions per minute (r.p.m.). The zeroth order diffracted laser light
has the frequency of the laser beam f and the kth order diffracted
o
order beam has a frequency
-63-
Stop
Lens
Loser - fo
Spectra
Physics 120
Synchronous
Motor Pholodelector
1800 Rpm
Beomspliller
Difhoction
GratinQ
Disk
Figure 3.2.l.a. Optics of laser Doppler velocirneter.
Pholomulflplier
(RCA 6645)
Preamplifier
Spectrum
AID Recorder *------iO Analyser
Figure 3. 2 .1. b . Signal processing of laser Doppler velocirneter.

-64-
f ± 61.440 k , (3.2.1)
o
where N is speed of rotation of the radial diffraction
grating (1800 rpm),
M is number of lines on the radial diffraction
grating (2048),
k is the diffraction order.
Thus a rotating diffraction grating provides a convenient way for opti-
cally shifting the frequency of the laser light (Stevenson [64]). It was
observed in this investigation that some laser light of frequency f was

o
also propagated in the direction of the light of frequency fk and this
caused some problems with the frequency demodulation equipment which will
be discussed subsequently. The basic technique for use of the laser
velocimeter is shown diagrammatically in Figures 3.2.l.a and b. The
zeroth fo and first fl order diffracted light was passed through a beam
splitter in order to increase their separation; all other diffracted
light was masked. A lens focused the two beams of frequencies fo and fl
into the flow field and a photodetector (RCA 8645 photomultiplier tube)
was aligned with the beam fl ("reference beam") after it passed through
the flow. The photomultiplier also collected light of frequency f that

o
was scattered by particles (of the order 1 micron) existing in abundance
in the flow field. The heterodyning of the scattered light with the
reference beam formed the basis of the velocity measurement system as
will be described in more detail subsequently.
II. Derivation of the Doppler shift frequency formula

4-
Consider a particle which is moving with velocity U and which is
-65-
-+
irradiated with a laser beam of frequency f in the direction e
o 0
(Figure 3.2.2). Light scattered by the moving particle will have a
frequency f given by
p
-+ -+
e 'U
f f (1 - n _0_) (3.2.2)
p o c
where n is the index of refraction of the medium,
c is velocity of light in a vacuum.
Laser Particle
Light
fo
/ - -
"--es-eo=2sm-
. e-
2 eu
Figure 3.2.2. Scattering light from a moving particle.
-+
Light scattered from the same particle in the direction e will have the
s
frequency (when viewed in the laboratory)
-66-
-+ -+ -+ -+ -+ -+
e 'U e 'U e 'U
f f (1 + n _s_) f (1 - n _0_) (1 + n _s_) =
s p c o c c
-+
nUf 0 - + - + 1
f + - - . (e - e ) + 0(-) ~
o c s 0 c2
~ f + n u sin(e/2) (3.2.3)
o A
-+ -+
where e is the angle between e ' eo in the medium of index
s
of refraction n,
-+
u is component of the vector velocity U in the plane of
-+ -+
e 0' e and perpendicular to their bisectrice,
s
A is the wave length of laser light in a vacuum.
The cathode of the photomultiplier superimposes the following electrical
fields
(i) A (t) sin2ITf t, the scattered light,

s s
(ii) A sin2ITf t, the reference beam,
l l
(iii) A sin2ITf t, unwanted scattered light from the rotating disk,
o 0
where
Assuming that all the above components have the same polarization
direction, the output current, i, from the photomultiplier (which is a
square law detector) is proportional to the total received intensity so
that
+ Ao As (t) cos2IT(f s - f 0 )t + D.C. terms +

-67-
+ terms oscillating at a very high frequency (order
f
o
= 1014 Hz) that exceeds the response of photo-
multiplier. (3.2.4)
A spectrum analysis of the signal given by (3.2.4) will have the
form of Figure 3.2.3.
0-
E
<l:
Frequency
Figure 3.2.3. Spectrum analysis of the signal given by Eq. 3.2.4
For this investigation from Equation 3.2.1 it can be seen that for a
first order scattering beam f1 = fo - 61,440 Hz, and this has been
verified experimentally to be correct within ± 3 Hz. Then from Equation
-68-
-+
3.2.3 the component u of the vector velocity U (Figure 3.2.2) can be
calculated using the linear relation
61440 + n u sin(e/2) (3.2.5)

A
The problem then becomes how to measure fs - fl experimentally. It can
be done in a time-averaged way by using a standard spectrum analyser.
On such an instrument the spectral peaks can easily be determined. How-
ever, instantaneous demodulation of the photomultiplier output signal is
complicated by the presence of the other spectral peaks.
Apart from the demodulation problem which has been introduced by
these additional spectral peaks it can be seen that the presence of the
rotating radial diffraction grating shifts the zero velocity Doppler
frequency away from zero. There are two distinct advantages to this in
that it becomes possible to measure very low velocities and also to gain
directional sensitivity.
III. Signal processing
There are three basic methods for signal processing, (a) spectrum
analysis, (b) frequency tracking, and (c) period timing (counting).
Period timing is proper for flows with a low particle content where
individual signal "bursts" from a single scattering particle are detected
and the time between zero crossings of the signal can be measured. Fre-
quency tracking or phase-lock-loop techniques are convenient for most
applications but are inadequate for flows with wide or rapid excursions
in velocity such as in the intermittent flow that occurs at a jet
boundary. Spectrum analysis, which has been used in this investigation,

-69-
is appropriate for flows with a high turbulence intensity or inter-
mittency.
The function of a spectrum analyser (Figure 3.2.4) will be described
briefly. A local oscillator (V.C.O.) delivers a signal whose frequency
f increases linearly with time between a minimum fL and a maximum f H.

vco
Consider the case wherein the Doppler signal is A cos2rrfDt where fD
is constant (e.g. laminar flow). As the v.c.o. is swept from fL to fH
the output signal Set) from the mixer varies
where (see Magrab et al [56])
A
Set) = z[cos2rr(f vco + f
c
- fD)t + cos2rr(f
vco
+ f c + fD)t]
A
SL(t) = z[cos2rr(f L
+ f
c
- fD)t + cos2rr(f
L
+ f c + fD)t]
A
SH(t) = z[cos2rr(f H
+ f
c
- fD)t + cos2rr(f
H
+ f c + fD)t]
Since the narrow constant bandwidth filter has a center frequency f and
c
a bandwidth ~f there is no output from it except when IfD - f I<~f.
c vco c
Thus the amplitude of the spectrum analyser display would be zero until
the swept frequency equaled that of the input Doppler signal. In the case
of a random input Doppler signal fD(t ) (turbulent flow) there is an

i
output of the constant bandwidth filter only for those input Doppler
frequencies that are centered in the band IfD(t.) - f (t.) I<~f .

1 vco 1 C
-70-
11 Frequency
.. x
Input 1~f e
Doppler BAND PASS
-- -. ---
SignaL S( t )
Acos27Tfot
- MIXER ~ FILTER Y·AXIS
(a mp I itude)
fe + ~ f e
~~
f L < f veo < f H
SWEEP
veo --- tJ::A~OR

Vo(t)
Time
-- ;-
n..... X·AXIS
( fr equency)
Figure 3.2.4. Schematic of a spectrum analyser.

-71-
The available adjustments of a spectrum analyser are:
(i) Bandwidth ~f of Band pass filter (Hz)

c
df
vco
(ii) Sweep rate of V.C.O. frequency (Hz/second),
dt
(iii) Minimum fL and maximum fH of scan (Hz).
Typical settings in the experiments performed were
~f 100 Hz,
c
df
vco
dt = 200 Hz/sec,
55 KHz, f = 75 KHz.
H
A 8553L-RF section, 8552A-IF section Hewlett-Packard spectrum
analyser was used in the investigation.
The Y-axis output (amplitude) and the X-axis output (frequency) of
the spectrum analyser were fed into the A/D data acquisition system (A/D)
described in Section 3.1. Typical sampling time for data acquisition was
10 minutes with a sample rate of 40 samples per second. This resulted in
a reasonably smooth probability density distribution for the velocity.
3.3 VELOCITY-TEMPERATURE CROSS CORRELATIONS MEASUREMENTS
A technique was developed to measure the product
N
uT =! ~ u(x,y,t.)T(x',y,t.)
N i=l ~ ~
where
-72-
u(x' ,y,t.) is the instantaneous velocity at (x' ,y) at time t.

1 1
T(x,y,t.) is the instantaneous temperature at (x,y) at time t ..

1 1
A small diameter (0.35 mm), fast response (time constant 27 msec) ,
bead thermistor placed near the focal volume of the laser Doppler
ve10cimeter (Figure 3.3.1) so that x - x' ~ 2 mm was used to measure
temperature and the spectrum analyser was used to provide an array of
instantaneous but not necessarily continuous, velocity measurements.
Thermi stor
Figure 3.3.1. Schematic for temperature-velocity measurement

-73-
The AID was used to record (i) the instantaneous frequency of the
spectrum analyser scan, f (t.)(X-axis), (ii) the instantaneous ampli-

vco 1
tude A(t.), (Y-axis), of the spectrum analyser, and (iii) the analog
1
signal for the temperature T(t ). Now if fD(t ) is the instantaneous

i i
Doppler frequency at time t., then defining
1
if l'fD - f vco > ~f c

I:
I'
A(t.)
1
if IfD - fvcol < ~fc·
the instantaneous velocity u(t.) is calculated from the instantaneous

1
Doppler frequency fD(t.) only when A(t.)

1 1
= 1. The scheme is depicted in
Figure 3.3.2. It is obvious that this array u(t.) of the instantaneous

1
velocities cannot be used for power spectra of velocity fluctuations,
but it can be used to construct the product uT.
3.4 EXPERIMENTAL SET-UP
The two-dimensional buoyant jet was generated by hot water emerging
from a chamber combining flow straightening elements and then discharging
through a slot 13 cm long and D cm wide (D could vary from 0.2 to 2 cm).
The jet discharged into tank 4.0 by 4.0 meters by one meter deep filled
with tap water. The jet was confined by two plexiglas walls 13 cm apart
in order to maintain the two-dimensionality of the jet (Figure 3.4.l.a
and Figure 3.4.l.b). A constant head tank was used to provide the
necessary pressure to drive the jet flow and a precision bore Flowmeter
(tube No. B6-27-l0/27 manufactured by Fischer Corporation) was used to
regulate the flow with an accuracy of 1%. The overall experimental set-up
is shown schematically in Figure 3.4.2. A first series of experiments

-74-
1.0 •• • ••• • • •
+0-
<t:
a
50 msec
• •
-
...- • • • •
0
\+0- • • •
50 100 msec
if)
w
....J •
<t:
u
if) ..... • •• • •
>-
;:)
• • •
0::
<t:
0::
l- SO 100
(l)
msec
0::
<l
-
.......
I-
50 100 msec
•
• •• •
.......
l- • •
-:3
• •
50 100 msec
Figure 3.3.2. Schematic for the measurement of the

correlation uT.
-75-
Confining Walls
( Plexiglas)
x
100 em
c c
.,L --/
1000II&;,...---- 45 em
Figure 3.4.1.a. Perspective view of the chamber of the

emerging jet.
Flow Straightening
Elements
r--~I--------r--r'" M ova bIe
Thermistor 8em
~---~~
I-- 45 em -l
Figure 3.4.1.b. Cross section of the chamber of the emerging jet.
CONSTANT
HEAD TANK
PUMP
-AIR VALVE
t FLOWMETER
I
-....J
0-
I
RECIRCULATION PUMP
AIR VALVE
HEATER
THERMOSTAT
Figure 3.4.2. Overall experimental set-up.

-77-
were performed in which only temperatures were measured and the instru-
mental set-up for these tests is shown in Figure 3.4.3. Five thermistors
each at a different elevation and located in the middle plane parallel to
the two confining walls were used for temperature measurement. All the
thermistors could be simultaneously traversed across the jet by means of
a single support carriage. The carriage was connected to a potentiometer
and mercury battery so that a voltage was obtained that was proportional
to the position (Y) of the carriage (resolution 0.01 mm). The AID was
used to record the temperature at the exit of the jet, the temperatures
from the five thermistors, and the Y-position. The jet chamber was placed
on the floor of the 4 m x 4 m tank and adjusted to make the plane of the
exit slot horizontal. For the second series of tests in which both
velocities and temperatures were measured, the instrumental set-up used
is shown in Figure 3.4.4.a and b. A laser Doppler velocimeter was used
to measure the velocities and a small thermistor (0.35 mm) about 2 mm
above the focal volume of the laser was used to measure temperature. The
focal volume was located in the jet midway between the two confining walls.
In these experiments the jet chamber and confining walls were mounted on
the carriage and the thermistor and laser held fixed. The laser Doppler
velocimeter was aligned and focused and then the transverse velocity
profile was measured by placing the jet at different Y-positions. The
purpose of the mirror (Figure 3.4.4.a) was to reduce to path of the
laser light into the water tank.
3.5 DATA COLLECTION AND PROCESSING
The analog outputs of the thermistor circuits, spectrum analyser,

x
1.5 Vol t
A
Potential Between A B B Gives
Transverse Position of Carriage
B··r-----------------------------------------~
Thermistor
Circuit
Ly
(5 bridges)
AID I
--.J
00
RECORDER I
~ 100 cm
, /',
I;, " " /
(Temperature at the Origin of the Jet)
Figure 3.4.3. First instrumental set-up.

Thermistor I .. _ _,
Circuit AI D Recorder
Thermistor Carriage
Probe
Spectrum
Ana Iyser
mirror
Preamplifier
I
-....J
\0
I
reference
beam
Digita I
\ I Vol t meter
signal from thermistor
x
Ly
at the exit
Figure 3.4.4.a. Front view of the second instrumental set-up.

L..
~-
4 meters 1 Filter H OSCillos:~~e I
Spectrum
~ IA/ 0 Recorder
Analyser
gloss
4
meters
Preamplifier
mirror
I
not tron spa rent 00
o
I
signal from a fast thermistor
close to focal volume
gloss
Beam Spl itter
Synchronous motor
and grating disk
Loser
Physics Spectra 120
Figure 3.4.4.b. Plan view of the second instrumental set-up.

-81-
and potentiometer were digitized continuously by the A/D data acquisi-
tion system and packed onto a magnetic tape compatible with the IBM
360/75 high speed digital computer in Booth Computing Center. Two main
programs were written to process the recorded information. The first
(Program I) was applicable to the first instrumental set-up where only
temperatures were measured across the jet using the five thermistors at
five different elevations. The second program (Program II) was used for
the second instrumental set-up when the laser Doppler velocimeter was used
to measure velocities and one thermistor was used to measure temperature
close to the focal volume.
Program I
Each of the six thermistors was characterized by its calibration
polynomial
T = A(i) + B(i)V + C(i)V 2 + D(i)V 3 ,
where i 1,2, ...... 6 (six thermistors),
and V is analog signal from bridge circuit,
T is temperature in °c
The coefficients B(i), C(i), D(i), were practically constant for an
extended period of time (3 months) but the coefficient A(i) did change
slightly with the time (order 1 week), possibly due to the ageing of the
thermistors or the temperature variations of the ambient air (affecting
the precision resistors of the bridge). Thus a calibration of the co-
efficient A(i) was necessary prior to each run. The first file of the
-82-
magnetic tape (for every run) represented the analog outputs V. of the
1
six thermistors at ambient water temperature T. Then A(i) was calculated

a
from the relation
A(i) = Ta
- B(i)V. - C(i)V.
l l
- D(i)V. ,
1
i 1,2, ...... 6,
Measurements of the same temperature from the six thermistors had an
accuracy ± 0.01 DC. Each of the next files then contained the condi-
tiona1 sampling of the signal from the six thermistor circuits and the
signal from the potentiometer (transverse position Y). The main steps
of the program were as follows:
1. Read a file from the tape.
2. Calculate the instantaneous temperature
for k 1,2, .•.•.. 5, (five thermistors at five elevations)
and k 6 (thermistor at the exit).
3. Calculate the mean temperature at ~,y,

N
I: Tk(xk,y,t i )
i=l
N
where k 1,2, ...•.. 6 (thermistors at different elevations
and at exit).
-83-
4. Calculate standard deviation of the temperature fluctuations
k = 1,2, ...... 5 .
5. Calculate I(xk,y,t ) where
i
\ 1 i f T(xk,y, t ) > T
i a
+ 0.04 DC ,
-r/ _
.L\Xk,y,L i )
_ ~ \ (3.5.1)
== ~ 0 i f T(xk,y,t ) < T + 0.04 DC ,
i a
where T the temperature of the ambient water.

a
6. Calculate the intermittency Yk(xk,y) where
N
L I(xk,y,t.)
i=l 1
(3.5.2)
N
7. Calculate the intensity of the temperature fluctuations with
respect to the mean T(xk,y) defined in step 3 but subtracting
the contribution to the signal when the temperature is below
the threshold defined in (5) above,
N
1
Th'
2
= -N L [T(xk,y,t.) - -T(xk,y)] 2 I(~ y,t 1·) •
ot i=l 1 k
8. Calculate the intensity of the temperature fluctuations with
respect to the mean T(xk,y) taking into account only the signal
below the threshold,
N
2 1 - 2
T~old = N L [T(xk,y,t.) - T(xk,y)] [1 - I(xk,y,t i )] .
i=l 1
-84-
9. Calculate the frequency of the conditionally sampled inter-
mittent signal (Figure 3.5.1).
u
o
z
w
a::
:::>
l-
e::{ Ambi en t Temperature
a::
w Temperature Level
a..
~
w
I-
TIME
Figure 3.5.1. Typical intermittent signal.
In order to count the threshold crossings, a temperature
level TL was chosen such that TL = Ta + 0.04 °c (where Ta

ambient temperature). Then the product is formed
Z (t.)
1
and defining F(t.) as

1
-85-
i f Z(t.) > 0,
F(t. )
1 =1
° 1
1 i f Z(t.) < 0,
1
the frequency of threshold crossings fR is given as

N
L: F(t.)
1
i=l (3.5.3)
fR =
Another way to find fR is to construct a function
M(xk,y,t ) such that

i
if [I(xk,y,t i ) + I(xk,y,t i +l )] ° or 2 ,
if [I(xk,y,t i ) + I(xk,y,t i +l )] 1 ,
then
10. Determine the maximum temperature Max T,
sampling time .
11. Determine the minimum temperature Min T .
Min T = minimum {T (x ' y, t ),

k k i
° < ti <,} where ,
sampling time.
-86-
12. Determine the skewness factor
where T,3
k
13. Determine the flatness factor
where
(For a Gaussian distribution the flatness factor is 3.)
14. Plot the transverse mean temperature profile.
Program II
This program had all the features of program I with some additions.
Briefly the steps of this program were
1
2
as in program 1 .
14
15. The frequency histogram of temperature was calculated and
plotted.
16. The frequency histogram of the velocity was calculated and
plotted.
17. The mean velocity u(x,y) was calculated from the velocity
histogram.
18. The standard deviation u,2 was calculated from the velocity
-87-
histogram.
19. The product uT was calculated.

-88-
4. TEMPERATURE MEASUREMENT RESULTS
AND DISCUSSION
4.0 BASIC EXPERIMENTAL PARAMETERS AND INITIAL DATA
Complete information regarding the basic parameters and the
initial data of the experimental runs is given in Table 4.0.1. An
explanation of the data columns is as follows:
Column 1 gives the name of the run. The series of runs June or
July, corresponds to the experiments in which only temperature
was measured; while the series PLVT corresponds to the experi-
ments in which both temperatures and velocities were measured.
Column 2 gives the width of the jet (in cm) at the orifice.
Column 3 gives the aspect ratio of the jet at the orifice.
Column 4 gives the exit velocity of the jet.
Column 5 gives the temperature of the ambient water in °C.
Column 6 gives the excess (above ambient) temperature at the jet
orifice in °C.
Column 7 gives the kinematic viscosity (cm 2 /sec multiplied by 10 2 )
at the jet orifice.
Column 8 gives the thermal expansion coefficient (l/oC multiplied
by 1000), calculated at the jet orifice.
Column 9 gives the Reynolds number at the jet orifice
R
e
= u0 D/v 0 .
Column 10 gives the Richardson number calculated at the jet orifice

a gT D
o 0
R =
o li 2
o
-89-
-u -T
Aspect 0
T \) a R R
Run D em 0 a 0 0 e 0
ratio --
sec °c °c x 10 2 x 10 3
em
1 2 3 4 5 6 7 8 9 10
0.90 o.<~ loj3dd u.UUOv,

PlVTZ-" 0_24 54_<J 62.93 3.;U L".bU
5.40 'l.dU 0.ij5 U.,,7 1777. U'/ u.uvuuo
oLVT7-Z 0.24 "",.i.ll 62.93
0.87 v.,o 1702.7" 0. V 1..10 1..1:>
PI vT?-H 0.24 54.2U 04.20 3.d() '.:.2v
,~.25 0.00 IJ.JI1 lo~7.jU J.UUOlL
PLVT2-& 0.24 54. "a 6.:.<;0 6.1l5
0.78 u.31 27-14.2d O.uOO,J
JUH-3 0.50 2b.OO • J.40 II. ;/U 20.0U
20.JU 0.78 u.~1 £7-14.Lo J.uJv9U
JU~ <-1 U.50 26.UU 43.40 11.20
0.78 O.H ,714.2.6 \J. ()UU'~O
J"~ F- 3 0.50 26.0U 4j.40 11.20 20.00
2U.UO 0.7d .1.31 27'14.20 v.uu09J
43.40 11.20
JU~<-3
JU'lf-l
0.50
0.50 I 26.00
26.00 ;0.50 o.oU
6.6U
19.0'
1 '1.85
O.E6
0.86
I U.i.7
u.n
1707.07
1707. b 7
O.vuv94
U.00090
JlJNF-"' 0.50 2b.00 ;0.5d
19.0'> 0.86 u.21 1707.07 u.uuu94
JIIM-l 0.50 10.00 JO.5d 6.60
0.86 u.I.7 1701.0' O.OI.HJl1lt
JHF\c-7 0.50 26.00 Ju.'>8 b.bl) 19.05
O.ijb u.lT !lb7.o7 V.OLlV"1't
JIlf\F-" u.50 2b.UU 30.58 b.00 19.d'>
9.,0 U.5v 0.77 0.31 '>93.61 U.UVlb'l
PI VT1-A 0.24 50.20 I ~.OO
I. J. 2U 0.74 O.J' 61d.17 0.OU22
PtV T 3-& 0-'4 :>4.2:0 19.00 10 ... U
JUl V-1 1.00 13.00 1'>.29 5.10 ,1.40 0.06 J.n 1709."~ U.vu::.8
0.86 O.i.7 d69.95 0.UV,8
JUt V- 3 1.00 u.uo 15.2" 5.10 t:..i.4U
21.4U 0.86 U.27 170"." , O.0u:>ts
J1JLV-1 1.00 13.00 15.29 5.10
,>.Iv ,1.4U 0.80 0.1.7 1109. 'I'> O.U058
Jill V- I 1.00 U.OO 15.29
LI.4U .1.86 v.L7 170 I. 'I'> u.UU?O
Jill Y-.' 1.00 U.ou 15.29 5.10
4.,0 iI.20 0.S8 u.20 1130.93 U.OU'>U
J1JLY-S 1.00 B.OO 1,.29
LI.20 0.88 J. it> 17,b.';J v.UO,u
JlllY-- 1.00 D.OO 15.29 4.511
0.86 volt> 17J6.9, o.uu~v
JIll Y- 5 1.00 lJ.OO 15.29 4.5U 1.1.211
4.,0 ll.I.U 0.B8 u.I.o 1736. 'IJ O.uU5J
JIILY-S 1.00 13.00 15.29
4.'>0 <1.20 0.88 0."::6 1736.93 0.0050
JUl Y-5 1.00 13.uO 15.2"
0.63 0.38 <'407.91 O.U2~7
JlJlV-~5 1.00 lJ.OO 15.2" 18.30 n.00
Id.,v 23.0v 0.63 U. j8 .24v7.91 u.v2.n
Jill V-75 1.00 13.{)0 15.29
IIl.Ju LJ.<lv 0.63 Udd l .. O7.91 0.0297
J1JlV-2S 1.00 13.uo 15.2.9
iJ.OO 0.63 U.3ti 2407.'1 i u.Ol91
JIILY-25 1.00 13.00 1,.29 18.30
0.63 Od8 2~01.91 0.vi.97
JlIlV-2S
J'JNF-~
1.00
0.50
13.00
26.00
15 • .2 9
!l.ll
18. jU
2 J. '>0
.:J.UU
19.10 0.61
0.61
I 0.3~
oJ.J9
-iY,:>.b4
'1'1,>.b4
O.O,Ud
O.UJv8
.JII'<F-6 0.50 21>.00 12.23 23.5Ll 19.10
23.50 1'1.70 0.61 Ll.J'l lj'l~.b4 O.U,O"
JIJJ.>c_", 0.50 2b.OU 1£.2" o.OJOS
26.00 12.2.3 n.,u I ~. 70 0.t.1 U.39 9'1~.o4
JII"< - ~ 0.50 v.0,J8
I ~. 7u U.bl v. J' '-J'1'J.64
J1.I'I<-b 0.50 2".00 !l.U 13.'>O
15.1 0 a.lu 0.b9 O.J'> IJ;8.7, 0.002
JlIlY-IO 1.00 13. UO I. 10
15.10 22.10 Ll.b9 0.J5 IJJd.7J u.Obl
JUlV-ln 1.00 13.0U '1.18
IJ.OO 9.18 15.ILl <2.lv O.b9 0.j,> IJJ8.TJ 0.U0,
JULY-IO 1.00
15.IU 21..10 0.69 U. J'> lJJb.7J O.ub2
JIILV-IO 1.00 IJ.UJ 9.18
15.1U 2'-.lu .1.69 0.;'> 13J8.7J U.Ubi
Jill V-I '1 1.00 U.OO 9.18
4. '>5 II.,J ,1.5U u.75 u.JL 009.0, 0.117
PLVTI-r 1.00 13.00
11., , 21.60 0.74 0.31. OJ. 'J.f.J., u.17d
PLvrl-r. 1.00 lJ.UO 4.~5
II.UO <I..ou 0.75 0.31. bv'1.U' 0.16j
PLVTI-Z 1.00 IJ.00 4.55
4.~5 10.90 i.2.0J 0.75 0.31. ov7.76 0.167
Pl VTl-C 1.00 13.uO
... 55 11.1'> 23.0U 0.73 0.3J 01.3.9, u.170
PlVTI-K 1.00 13.00
II.UU il..OO u.75 0 • ..)£ 0,,9.05 u.lb9
Pl VTl-R 1.00 B.OO 4.55
Pl VTi-H 1.00 13.00 4.55 11.40 2~.OV 0.73 u. J3 027 .17 U.lol
b.11 20.90 <I.JO 0.62 u.39 917.70 Ll.L1o
JUN<-ZR 1.00 lJ.OO
LIdO 0.62 u.3'1 977.70 O.~lb
JUNc-Z8 1.00 13.00 6.11 2 O. ~O
6.11 20.90 1.1.30 0.62 ().~~ 977.70 O.21a
J""F-ZR 1.00 13.00
6.11 20.~0 ,J..JO U.62 U.J'i 977.7u U • .t:!O
Ju~<-~e 1.00 13.UO
20.9U ,I. ~U 0.62 u.3~ 977.70 0.216
JUN<-2~ 1.00 13.00 b.11
21.d j .11.65 0.61 U.4u 99'1.~4 0.2 Jl
JUl~-l 1.00 lJ.OO 6.11
b.ll 21.05 21.b5 0.61 u.4L1 99'1. '14 v.,2 Ji
JUL ~-I 1.00 13.00
b. II 21.d5 2l.t-5 0.61 0.,+U 9'1'.94 0.231
JUL V-I 1.00 13.00
6.11 21.d~ 2.1.65 0.61 .1.40 999.94 0.231
JUlV-1 1.00 B.UO
boll n.d~ I.l. b5 0.61 0.41,) 9'1-1.r..;4 0.231
JUlY-l 1.00 13. vLl
7.&0 21.'10 lO.60 0.62 O.~9 ",445.00 U.L9"
J'JNF-2' 2.00 6.50
21.9() 20.60 0.61. u.39 i.445.06 0.,94
JU~JF-2' 2.00 0.~0 7.bO
I "4 .. ~.Oo
21.9U I.0.6J 0.62 0.39 Li.i.'1 ..
JU~f-2b 2.00 a.50 7.bO
0.62 U. j9 L441J.Oh U.;Q4
JU"F-7' 2.00 b.~O 7.60 21.90 '<J.bU
I
21.711 ,4.10 0.59 .1 ... 1 100 ... 0 O.oz;
JULY-Zq 1.00 !J.()O 4.59
4.59 21.7U 44.10 0.59 0.41 7bO.4U Ll."V
Jl'l V-zq I. 00 13.00
21.7v 1. ... 10 0.59 0.41 700.40 O.4,J
JULV-Z<l 1.00 lJ.OO 4.59
4.59 21.70 ,4. ILl 0.59 0.41 7bO.40 J.o./J
J')LY-Z'l 1.00 1J.0O
4.59 21.10 ,4.1.1 0.59 0.41 7dU.40 v.42J
JUlY-Z" 1.00 13.uO
Table 4.0.1. Basic experimental parameters and initial data of

the temperature measurements.
-90-
4.1 MEAN TEMPERATURE
The mean temperature T(x,y) in °c for a particular distance from
the exit x, was plotted versus the transverse distance y (in cm), and
a Gaussian curve was fitted of the form
T(x,y) (4.1.1)
where b is the half-width of the temperature profile, as in Figures

T
4.1.l.a to 4.1.l.d and 4.l.4.a to 4.l.4.d. The temperature half-widths,
bT(x), for a particular run were plotted versus the distance from the
exit as in Figure 4.1.2 and 4.1.5, and a straight line
(4.1.2)
was fitted to the data, where D is the width of the jet at the origin
(jet exit). The virtual origin of the mean temperature profile is then
given by
The coefficients K , K2T from all the runs are tabulated in Table
1T
4.1.1 from where it can be observed that there is no apparent variation
of K1T , K2T with the initial Richardson number Ro' But it is interesting
to note that the coefficients K , K2T for runs PLVT2 and June 7 are
1T
quite different, although they both correspond to jet Richardson
numbers appropriate to a pure jet. The explanation of this appears to
be that the true character of the jet half-width is non-self-preserving,
i.e., the assumption of linear growth of the half-width although
(text continues on page 99)

-91-
o
TRANSVERSE MEAN TEMPERATURE PROFILE
0r-----.------.-----.-----.------r-----.------r-----.
LI'I
o
RUN JUNE 7
o RICH. NUMB.=0.0009
::I'
X/D=IO.O
EXI'ER JI'ENT !!l
u
o
o
zo
....... ~
W
cr.
:::J
t-
cr O
cr.~
WN
CL
L
W
t-
o
o
~
0
-11.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0
Y AXIS IN CM
Figure 4.1.1.a. Mean temperature profile for a
turbulent jet at x/D = 10.0.
0 TRANSVERSE MEAN TEMPERATURE PR~FILE

""!
""
0
~ RUN JUNE 7
"" RICH. NUMB.=0.0009
0 X/0=20.0
""! EXPERJI'ENT !!l
UN
0
ZO
0
>-< •
N
W
cr.
:::Jo
t-LI'I
cr'
cr.-
W
CL
o
Lo
W •
t--
0
LI'I
Cl
~
0
-6.0 -4.5 -3.0 -1.5 0.0 1.5 3.0 4.5 6.0
Y AXIS IN CM
Figure 4.1.1.b. Mean temperature profile for a

-92-
o
0.-____.-____.-____-.____-.______.-____.-____.-____-,
(Tl
o
If) RUN JUNE 7
N RICH. NUMB.=O.OOOS
X/O=30.4
EXPER I f'£NT [!]
.,u8.
N
Z
WO
a:lI!
:::J_
\
f-
a:
a:
Wo
0..0
~.
w-
f-
o
If)
~
O~ ____ ~~~~ ____-L_____ ~ ______L __ _ _ _ ~~~~ _ _~~
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0

Y AXIS IN CM
Figure 4.l.l.c. Mean temperature profile for a
o
o
.,u
;[' RUN JUNE 7
RICH. NUMB.=O.OOOS
X/O=51.0
EXPERlf'£NT [!]
z
WO
a:~
:::J_
f-
a:
a:
W
0..
~
Wo
f-l/')
o
[!)
[!)
o~ L--_...,,-::=-____~____- ' _________ -1_. ____~____ _ L_ _ _ __=r:==__ _---'
-12.0 -S.O -6.0 -3.0 0.0 3.0 6.0 S.O 12.0

Y AXIS IN CM
Figure 4.l.l.d. Mean temperature profile for a
o TEMPERATURE HALF WIDTH
~~r----~----~------~----T-----~-----r----~----~------~--~
0')
0
0
<Xl RUN JUNE 7
00
RICHARDSON NUMB.=O.0009
,~ br IO=O.136(X/D +1.82)
.... ['-
.D
·0
IO
I--t[)
0
~
3:0
0
LLlJ")
.
--1
ITo I
Io
--1::1'
. \.0
W
I
IT
20
°0
........
(f)crJ
2
Wo
LO
.........
ON
20
00
2~
0
0
0.0 7.0 14.0 21.0 28.0 35.G 42.0 49.0 56.0 63.0 70.0
NON DIMENSIONRL DISTRNCE FROM EXIT. X/D
Figure 4.1.2. Non-dimensional temperature half-width of a turbulent jet
plotted against non-dimensional distance along the jet axis.
a
NDN OIMENSIDNRL MEAN TEMPERRTURE PR~FILE
N~, _ _~_ _~_ _ _ _~_ _~_ _ _ _~_ _~_ _ _ _~_ _~_ _ _ _~_ _- ,
a
RUN JUNE '7
o
R1CH. NUMB.=0.03
X'=X+3.62
EXP( -41 7(2 )
o
ro EXPERIM. RESULTS
o
X/O SYMBOL
10.0 [!)
20.0 ~
I~g 30.4 J!:.
"
II- 0 51.0
60.0
+
X
I
\0
..,...
I
o
::I'
o
a
N
0,
~ L...---<"rn
-0.35-0.28 -0.21 -0.14 -0.07 0.0 0.07 0.14 0.21 • ~
-
0.28
I
0.35
, 1(=-Y IX'
Figure 4.1.3. Non-dimensional mean temperature profile of a jet·

plotted against non-dimensional distance from jet axis.
-95-
a
TRRNSVERSE MERN TEMPERRTURE PROFILE
a . -____. -____- r____- .______. -____. -____- .____- .____- - .
N
a RUN JUNE 28
a
RICH. NUM8.=0.216
X/O=6.0
EXPERI~NT [!)
(
U25.
ex>
Z
wa
a::C:;
:::JtO
I-
er:
a::
Wa
CLa
L'
UJ:::1'
I-
a
a
N
I!I
-11.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 11.0
Y AXIS IN CM
Figure 4.1.4.a. Mean temperature profile for a
turbulent plume at x/D = 6.0.
a
TRRNSVERSE MEAN TEMPERRTURE PROFILE
a . -____. -____- r____- .______. -____. -____- ,____- . , -__- - ,
to
o RUN JUNE 28
a
lJ) RICH. NUMB.=0.216
X/O=111.0
EXPERI~NT [!)
a
u a
(, ,;.
Z
W25
a:: •
:::J'"
I-
er:
a::
wa
CLC:;
LN
W
~:l"J -8.0 -6.0 -11.0 -2.0 0.0

Y AXIS IN CM
I
2.0 11.0 6.0 B.O
Figure 4.1.4.b. Mean temperature profile for a

-96-
o
~r-----~-----r-----.------.-----.-----
M -r---
o
o
M
RUN JUNE 2B
RICH. NUHB.=0.216
o XI0=30.2
~
, UN
Zo
....... 0
N
w
a:
::J
~o
a:"!
a:-
w
(L
::Lo
W o
~...:
o
~
-16.0 -12.0 -B.O -4.0 0.0 4.0 B.O 12.0 16.0

Y RXIS IN CM
Figure 4.l.4.c. Mean temperature profile for a
o
0r-----.------r-----.------r-----.------.-----.------,
('t'I
0
"! RUN JUNE 28
N RICH. NUHB.=0.216
X/0=36.0
EXPERIMENT [!)
t
ug.
N I
I
Z
~
wo
a:"!
:::l_
~
ffio~
(Lo
::L'
w-
~ 1
~l ~ I
-18.0 -13.5 -9.0
I
-4.5 0.0
Y RXIS IN CM
4.5
I!l
9.0 13.5 lB.O
Figure 4.1.4.d. Mean temperature profile for a

o
TEMPERRTURE HRLF WIDTH
~~I----~----~----~----~----~----~----~~----r_'----r_--~
~
o
o
(0
.-- RUN JUNE 28 .
RICHRRDSON NUMB.=O.216
o
'- b r ID=O.122(XID +2.63)
1--0
.Do
::C ~ --
I-
o
1-40
3:0
/
lL.::::J'
.-1
a:
Io
I
.-1~ 1.0
a:('() "'-J
z I
E)
~
(nO
ZO
w'
~C\I
t-1
o
z8
E)'
z .....
o
OL I I I
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0
NON DIMENSIONAL DISTANCE FROM EXIT,' X/D
Figure 4:1.5. Temperature half-width of a plume as a function of
distance from plume orifice.
-9S-
Run R X/D KIT K2T CIT C

0 2T
from to
1 2 3 4 5 6 7 S
PLVT2 O.OOOOS 20.S0 93.75 0.167 -1.56 0.20 -2.44
June 3 0.0009l 10.0 50.00 0.130 3.90 0.27 -1.0
June 7 0.00094 10.0 60.0 0.136 1.S2 0.27 -1.0
PLVT3 0.002 33.75 93.75 0.141 -2.12 0.31 -4.70
July 3 0.005S 6.0 36.0 0.125 4.35 0.35 -0.1
July 5 0.0050 6.0 36.2 0.121 4.46 0.35 -0.1
July 25 0.0297 6.0 36.0 0.123 2.3 -- --

June 6 0.031 10.0 60.0 0.120 3.3 -- --
July 10 0.063 6.0 36.0 O.l1S 3.3 -- --
PLVTl 0.17 9.0 39.0 0.120 2.10 -- --
June 28 0.216 6.0 36.2 0.122 2.63 -- --
July 1 0.231 6.0 36.0 0.121 2.87 -- --
June 26 0.294 6.0 lS.0 0.113 4.0 -- --
July 29 0.424 6.0 36.0 0.112 4.42 -- --
Table 4.1.1. Linear approximation of the growth of the temperature

. half-width and the centerline dilution.
-99-
providing a convenient way of plotting and fitting data which is
compatible with the overall accuracy of a particular run is not
strictly valid. This will be clarified in Chapter 8.
The mean temperature T normalized by the mean temperature on the
jet axis TM has been plotted versus the non-dimensional distance y/x'
(where x'= x - x ) for two representative cases of Richardson number.

oT
Figure 4.1.3 gives T/TM for a jet and Figure 4.1.6 for a plume. The
Gaussian curve which is plotted on these figures is defined by
(4.1.3)
Similar plots for the other experimental runs are presented in Appendix
B.
The centerline dilution, defined as To/TM has been plotted in
Figure 4.1.7 for the case of a pure jet (R ~ 0.0009) and a line
a
(4.1.4)
fitted to the data. In Figure 4.1.8 it can be seen that for a pure
-1 -1/2
plume the centerline mean temperature T decays as x ,and not as x
as occurs for a jet (Figure 4.1.7). The centerline temperature for
many runs with initial Richardson number greater than 0.15 (case of a
plume) has been plotted in Figure 4.1.9 in a fashion suggested by the
non-dimensional number (see Eq. 2.2.14)

C)
NON DIMENSIONRL MERN TEMPERRT~RE PROFILE
N~_ _~_ _~~_ _~_ _~_ _ _ _~_ _~_ _ _ _~_ _~_____~_ _~
C)
RUN JUNE 213
o
R!CH. NUMB.=0.216
X'=X+2.S3
EXP( -47 Tl2 ) -
C)
to
EXPERIM. RESULTS
o
x/a SYt1BCJL
6.0 [!J
111.0 ~!
\~~l"- 1
22.0 .o!I
II- 30.2 + I
X I-'
36.0 o
o
I
C)
:;;j' I
01
C)
I
N
o
.'
o
oI ~~ I I I ~I 4~4d
-0.35-0.28 -0.21 -0.14 -0.07 0.0 0.07 0.14 0.21 0.28 0.35
, Ti=Y IX'
Figure 4.1.6. Non-dimensional mean tenperature profile for a p~ume.
C)
CENTERLINE DILUTI~N ~F R 20 JET
~~I------~----~~----~------~------~-----~
C)
N
Ro =0.0009
RUN SYMBOL
JUNE 3 [!]
~~
JUNE 7 (!)
[6/
C)
/
)(
0
C) ·
.....
---
if-~ I!J / '
I
I-'
"-
N 0
/(!) 0
I-'
I
If-
a
·
0
LIl
/ID
·
0
O' I I I I I I
0.0 10.0 20.0 30.0 ijO.O 50.0 60.0

N~N DIMENSI~NAL DISTANCE FR~M JET ~RIGIN,X/D
Figure 4.1.7. Temperature distribution along jet axis.

o
CENTERLINE DILUTION
o
~~I--~--~~--~--~--~'--~----~--~--~'--~
RUN JULY 29
Ro =0.1.123
o
o
N
o
o
0>
~~ /
~ f-'
11-
......
II-
0
-1 0
N
I
0
0
('I)
.
C)
o I!! I I I L
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 ijO.O ij5.0 50.0
NDN OIMENSIDNAL DISTANCE FR~M EXIT, X/O
Figure 4.1.8. Temperature distribution along plume axis.
o PLUME-. CENTERLINE OILUTI(jN
co
01
RUN SYMBOL
PLVTl (!]
JUNE 28 o
o JULY 1 A
<.0
o JUNE 26 +
JULY 29 X
><
!!J
:2:
Ir--
~ !!J ~ !!J~
><
t-
o
0
::i'
...._.,. .,+X +
6
X
1!1- - - -..
X
!!J <1.
X
!!J
I 0
>< I
I-'
0' o
W
tj I
"-
rr>
N o
><
C\I
o
.
.Q:l..
o
06 I I I I I
0.0 10.0 20.0 30.0 40.0 50.0

N~N DIMENS. DISTRNCE FR~M EXIT .X/O
Figure 4.1. 9. Temperature distribution along plume axis.
-104-
(4.1.5)
where the thermal expansion coefficient a(x) is calculated from Eq.
(2.2.5) at a temperature Ta + TM(x) and the kinematic buoyancy flux
S(x) is calculated from the relation
a(x)gH
o
S(x) (4.1.6)
pc
p
where H the heat flux at the origin of the jet. The mean value for
o
0T suggested by the results tabulated in Table 4.1.2 is 0T ~ 0.42 with a
standard deviation of 0.02. Since
where PM(x) is the mean density on the jet axis at a distance x from
the jet origin, then using Eq. 4.1.5 it is apparent that
-2/3
0.42 g (x - xoT)S(x) (4.1.7)
This value for 0T is higher than the value 0T ~ 0.38 suggested by Koh
and Brooks (60), or by Lee (55).
The mean centerline temperature TM(x) and the temperature half-
width bT(x) for each measured temperature profile are given in Table
4.1.2.
-105-
Run R D x
-T -
TM b crT
0 0 T
em em °c °c em
I
1 2 3 4 5 6 7 8
n.74ll <.OOC 3.300 1.7511 0.7 0 0 2.b~0

PlVT2-F C.CC0051
PlVT7-1 U.OOOfPo 11.240 q.ooo 5.4,,0 2.C20 1.4UO 1.722
Pl VT 2-1' n.,1t'0,1<Q 0.240 14.100 3.~00 1.160 2.040 1.5Cl
PLVT7-A n.rvOl23 C.2 4 O 22.~GO f.A50 I.BOO 3.300 o .AcB
J'J",': _'1 U.O()o·oq U.5(') 5.!) rll' 11.2')0 7. IOU 0.900 1.151
JIJNc-3 0.001') 9 00 u.~oo I (J. ,1(1(, 11.2UO 5.150 1.531J O.GH
C. < C') 4.0(,0 2.710 0.8B~
JII"c -, U.OCOao o 1 ' . lOC 1l.2JO
J'I"c -3 0.01')(1")0 ,). SuO 2~.JOO 11. <-.It,} 3.100 1.51. 0 0.734
JUNF-7
j!p.,j!: -7
(.COC94U
iJ.Of'()Q40
O. ~0'1
O.5CJ
~ .. 1)0(,
10.;)0r.
6.~JO
6. <'ivO
~.I>Ut,)
•• 250
3.200
2.,00
I 0.800
1.500
2.700
1.174
O.qll
C.818
JIINc-7 (I.OO094r 0.5011 15.7(1J
JUW'-7
JIf,,"c -7
Pl VT~-P
0.GOC<;4U
0.100940
O.~Cr:
0.50U
O. '41)
2'.500
3n.I1 n r
A.IOO
6.1>00
6.6UO
0.2uU
I 1.8LO
1.7411
3.050
3.620
4.200
1.160
0.714
O. t 35
O.~OO
'l.0l1I AC 4
PlVT3-A 0.01')2742 C.24L V.SOC 10.400 1.7BU 3.100 0.518
JlllY-, (').005821 l.roo f..OOO 5.lno 3. 6~C 1.2QO 0.705
J'lL V-, 0.Cl1<A21 l.luO 14. )01) 5. Iu a 2.430 7.300 U .675
J'll V-l 0.005~21 1.0(,0 n.ooo 5.1uO 1.760 3.200 0.'>12
Jill Y-3 0.005A21 1.1)00 30. 4 00 5.100 I. StU 4.2nO n. 5 27
JUL Y-, 0.O'1 c R21 1.000 3t.0('C 5.100 1.430 4. AOO 0.407
J Il V-5 o.OO~OCI 1.000 ~.ooe 4.500 3.000 1.270 0.838
14.1)00 4.500 2.150 2.200 0.~6q
JIll Y-~ (.'05eOI 1.000
JlJlV-5 0.0050CI 1.0GO n.fJCC 4.5JO 1.560 3.330 0.649
J'Jl V-5 (1.11115001 1.')01) 30.400 4.500 1.370 4.2 JO 0.562
'
4. 'laO o .~20
Jill V-5 0.OC5'101 1.000 3'.200 4.5uJ 1.250
Jill V-? < 0.0'G'41 1.000 f .<10G 18.3uO 12.UOO 1.13) U .607
0.020741 1.0'10 14.JOL IA.3'10 7.250 2.GOO O.~ 32
JUlY-~5
Jill V-2 ~ (I.02G741 1. lUI) 22.1)00 1~. 3uO 5.250 3.000 0.5C2
JUl V-2 5 0.029 7 41 1.000 ~U.Oor, 18. 3l () 4.150 4.050 0.4(13
JULY-7S
JU"I<-!>
(I.,nQ741
O.03CQ<;1
I. ,)",)
1).5 CC
U.501)
3~.OOO
".1)0('
10.00e
IA.301l
23.5uu
23.5UO
I 3.550
1l.7ulI
7. ECC
t..700
O.ROO
1.430
0.47'1
0.467
0.455
Jll"c -6 0.030 0 07
JI/NF-e Q.03CQG1 U.5(0 I' .'JUO 23.5'lO 5.5(;C 2.030 0.4t3
JIJNc -~ O.f)1cqn 0.500 25.0CC ?3.5(;0 3.750 3.270 0.434
JlJlojf-~ (I.030R Q 1 0.500 30. )00 2 J. 501l 3.400 3.770 0.405
JlJl V-I a n.C67~51 I.OCO ~. lOC 15.1')0 B.300 1. 100 0.509
JIJL Y-l 0 0.062R53 1.uoe 14.000 15.1CO 5.250 2.070 0.445
JULY-I!) 0.{)S2e,3 1.000 22 .')1)0 1<'.100 3.450 3.000 0.41l
JUlY-IO 0.(;~?o~3 I.oeo 30.4CO 15.1UO 1.000 3.A30 o .41()
J'JLY-10 0.Of2°<3 1.000 ?~.20e 15.1,,0 2.500 4.670 0.422
PLVT1-F 0.17150 1.COO <? lOG 11.500 3.870 1.7(10 U .50
"lVT 1-1' O. I 7 qc 71 1.roo 14. )0(1 11.550 3.250 2.340 0.416
PLVTl-l 0.16SA43 I .() ,1'1 10.1l00 1l.JOO 2.200 2.~50 0.458
PLVTI-C 0.HBQ7 l.u()J 23.5<)0 10.9VO 2.<l()0 3.250 0.414
Pl VT I-~ 0.\1/,Q45 l.rou 2A. QO C ' 1. 150 1.5~C 3.7'iU 0.452
PlV T 1-R (1.UoQ43 1.0,)0 33."'0( 1l.000 1.;50 4.400 0.446
PLVTI-~ O.lel~·1 1.0GO 3Q.IOC 11.400 1.210 4.950 0.4B
J'lPl:~-~B ('.71~~54 I.CO(l '.<lOn 20.900 <;.600 1.070 0.447
0.216254 1. ')01 II. .,JUU 20.~JO 5.6UO 1.970 0.~13
JUIoj~-~·
J1)"I~ -2 ~ 0.2lf2S4 I.OCL 27. ,JOO 20.900 3.500 '.000 0.451
J'iNc-2° 0.21f?~4 I.UOO 3C.20 r 20.900 2.700 4.130 0.449
JIJlliC -7 R (1.21625 4 1.00(' 36.2CU 20.9l10 2.300 4. BOO 0.44~
JUlY-l 0.2~1431 1.(GO 6.U()0 21.B50 1).200 1.070 0.41A

Jill V-I 0.231431 I.OOf) 14.'100 21.850 S.tSO 1.G70 0.414
Jill Y-l (,.231431 1. JOO 22.1)0(' 2l.8,U 3.700 2.810 0.438
JlllY-l 0.231431 1.0,)C ~O.40( 21. A5<) 2.80t 4.00<) 0.438
JlJl V-I fI.231431 1.00'J 3~.40" 21.850 2.400 4.700 0.435
JII"IE-lf 0.7C4~le .?OtC 17. ,)00 21.9UO Q.500 2.260 0.419
JUNO -"., 0.7q451~ 7.001) 70.100 21.'1)0 6.250 3.320 0.417
JIINE-2 f 0.2~·~IR 2. ,)(J(I 2B.')00 21.900 4.BOO 4.000 0.429
,IUN~ -2 ~ U.294~IR 2.(100 3~.'1uG 21.9JO 4.200 5.000 0.404
J'll V-20 0.42)977 I.U.JO 6.1)011 21.700 9.000 1. 170 O.31R
JUl Y-10 0.423077 1.000 14.00U 21.700 4.650 2.010 0.374
JUL Y-Z9 0.42H17 1.000 22.000 71.70u 3.12U 2.950 0.3'14
Jill Y-2o 1).4?]" 77 1.000 30.000 21.70U 2.300 3.850 0.414
Jill Y-Z9 0.423 0 17 l.tOO 36.000 21.700 2.06;) 4.000 0.3<;5
Table 4.1.2. Experimental values for temperature half-

width and temperature distribution along
jet axis for turbulent jets and plumes.
-106-
4.2 TURBULENCE INTENSITY
The intensity of turbulent temperature fluctuations, defined as
T,2 1
= Lim ""2 J t ' [T(x,y,t)
0 - -T(x,y)] 2 dt, (4.2.1)
t-+oo
or, when conditional sampling is performed, as
-2
T' = -NlNL -
[T(x,y,t.) - T(x,y)]
2
(4.2.2)
i=l 1 '
is plotted in Figure 4.2.1 (representative case for a plume) and in
Figure 4.2.2 (representative case for a jet). It is interesting to
notice the dramatic difference in the structure of the turbulence
intensity profile for the above two cases. The reason for this is
probably the production of turbulent energy by the buoyancy forces
(see Eq. 2.2.15). The turbulent energy equation (see Eq. 2.2.15) has
the form
dq'7 -- agu'T' + production by Reynolds

~
stresses + diffusion and dissipation terms, (4.2.3)
where q,2 u,2 + v,2 + w,2 and agu'T' is the production of turbulent
energy by the buoyancy forces. The production of turbulent energy on
the jet cross section is obtained by integrating Eq. (4.2.3) across
the jet:
B(x)
J a(x)g u'T'dy,
-B(x)
o
t:"'-
el
RUN ...IU£ 26
Ro aO.295
X·2X+4.0
·
01.....
XlO STt18O....
6.0 I!J
10.0 (!)
1
0 + 14.0 6
+
~
o
• + ~ 18.0 +
+ -+ (!)
"i!>
I T-
o(Y)
II-
E
0 · I
I--'
o
...... "I
S
0
N
0
.0 ·
0
01 (!).
-0.40 -0.32 -0.24 -0.16 -0.08 -0.00 0.08 0.16 0.24 0.32 0.40
Y/X'
Figure 4.2.1. Profile of the intensity of turbulent fluctuations of plume temperature
plotted against non-dimensional distance from jet axis.
o
Ln
orl------·.-------,------,-------.-------.------.-------,-------,-------.------,
RUN JlLY 3
<'II
:J' Ro=O'CXl)
o
X'=X+4.9
XlO SYHBCl..
6.0 [!)
~r
14.0 (!)
22.0 AI
30.2 +
36.0 X
Ln
<'II
I
o I-'
a
00
I
I~ r--
.
" 0
~ CD
0
0
o
01 I ~ I I I I I I I II!! I
-0.40 -0.32 -0.24 -0.16 -0.08 -0.00 0.08 0.16 0.24 0.32 0.40
Y/X'
Figure 4.2.2. Profile of the intensity of turbulent fluctuations of jet temperature
against non-dimensional distance from plume axis.
-109-
where the thermal expansion coefficient a(x) is calculated from Eq.
(2.2.5) at a temperature Ta + TM(x). However, the kinematic buoyancy
flux
l3(x) = JB(x)a(x)gu(x,y)T(x,y)dy
-B(x)
= JB(X)a(x)g~(x,y) T(x,y)dy + JB(x)a(x)gu'T'dY (4.2.4)

-B(x) -B(x)
so that the production of turbulent energy by the buoyancy forces is
equal to the turbulent contribution to the buoyancy flux. Assuming
that ~(x,y) and T(x,y) are approximated by the Gaussian curves Eq.
4.1.3 and Eq. 5.1.3 respectively, and considering that for a plume
(see Section 2.2) it is found that
B(x) a(x)g~(x,y)T(x,y) ag(x-xOT)TM(x)

J 13 dy ~ 2/3 (4.2.5)
-B(x) 13
i.e., the non-dimensional number Eq. (4.1.5) can be regarded as the
ratio of the kinematic buoyancy flux of the mean motion to the total
kinematic buoyancy flux.

-110-
Further, assuming that the distribution of the turbulent energy
flux is self-similar, it is possible to write
where (u'T') the value of u'T' on the jet axis, so that

M
For a plume, u(x,y,t) and T(x,y,t) are well correlated (see Section 6.2),
so that it is possible to write
Using this result, it then becomes possible to write
a(x)g(x-x )/Tf2
SB(X) a(x)gu' (x,y)T' (x,y) oT
dy .... 2/3 (4.2.6)
-B(x) S(x) S(x)
where S(x) is calculated by Eq. (4.1.6).
We can therefore conclude from Eqs. (4.2.5) and (4.2.6) that Eq.
4.2.4 can be rewritten as
+ 1 ,
for some constants A and B. Furthermore, since the first term is
constant in a turbulent plume (see Figure 4.1.9), it is apparent that
the second term should also be constant.

-111-
The centerline turbulence intensity for many runs was scaled
according to Eq. (4.2.6) and the results are plotted in Figure (4.2.3)
versus the non-dimensional distance from exit X/D. Two important
conclusions can be drawn. First the right-hand side of Eq. (4.2.6) is
constant for a plume and it is approximately equal to 1.0, i.e.
a(x)g(x-XOT)~
~ 1.0 (4.2.7)
8(x)2/3
Second, the turbulence intensity increases as the initial or local
Richardson number increases, and tends asymptotically to the turbulence
intensity of the plume. Combining (4.2.6) and (4.1.5) it is seen that
for a plume
~ 0.42 . (4.2.8)
The effect of the initial Reynolds number on the turbulence intensity
(see for example Heskestad (5)), which should be relevant in the cases
of R « R , was of second order in this investigation because the

o p
initial Reynolds number for all the experiments remained essentially
constant. The centerline turbulence intensity is plotted again in
Figure 4.2.4 where it is non-dimensionalized using the local centerline
mean temperature TM(x). Basic parameters and experimental results
regarding the centerline turbulence intensity are given in Table 4.2.1;
column 3 is the Grashof number, defined as 8x 3 /v 3 , and column 10 is the
non-dimensional ratio given by Eq. 4.2.7. The profile of turbulence
intensity, as defined by Eq. (4.2.1) gives the gross picture of the

100 -r r ,---,--,--,---,---.---,--,---.----~--~--,--,---.---,---. II-~'-'T-'-~"-
50
~~.:~23 ~~ -.:t:~._~.-~~~BO:L_
...
JUNE-7-
JULY 3 -
.• ..
IO-::4-~
- 000('-
- 0---
---0--
TuL:Tz5-- '003 - ---...-
JUNE-6-- -003-- - >--~--
""iV-, e)___ oj.§~___ ~~
.
PLVT I 0 17 •
~~n~~:llu~Gl
-~+ --+
~
~__-"""F==:=i!IT~--+'
>~> o~~=.:======-=--
__ • --u-
t;' .+
o tlO • ±
>-
_____ ;
~
CQ
"-
'"
,-J-.,
0.5
_0 I
f-'
f-'
tv
I
~ .... 01
----'.
0
'"
I
'"
;:' 005'-
:::
t:l
'--v-'
001 ~ ~--~~~~--~-~--~~
00 10.0 20.0 30.0 40.0 50.0 60.0
X/O
Figure 4.2.3. Intensity of turbulent fluctuations of temperature on jet axis as a

function of non-dimensional distance from jet orifice and initial
Richardson number.
0.50i~~~~-r-r~~.-.-~~r-~~~~~-r-r~~'-.-~.-.-~;~~~-.-r~~~
~N Ro SYMBOL
PLVT2 Cl.OOO09 ()
JUNE 3 0.0009
------
e
OAO JUNE 7 0.0009 0
JULY 3 0.0006 Q
JULY 5 0.0005 6.
JULY 25 0.029 A
JUNE 6 0.030 ~
0.30 PLVTI 017 0
JlJ1LY 1 0231 'V
~LJLY 29 0.423 •
I>-~ ~
~y ~
I
f-'
..... 0.20 f-'
W
~
I
0.10
0.01 1'1
o 10 20 30 40' 50 60 70
X/D
Figure 4.2.4.' Intensity of turbulent fluctuations of temperature on jet axis
normalized by the local mean temperature as a function of non-
dimensional distance from jet orifice.
-114-
II
Run R G x
-T -
TM
i
i
11,2
--
1T,2 1T,2
--- - L:
0 r 0 I -
x 10 10 cm °c °c °c TM T
0
1 2 3 4 5 6 7 8 9 10
1.750 0.130 u.074 U .Ll3'1 0.J28

PLVT?-E o.uvO()~1 C.IA4 ". CJO 3.3JU
001J48 0.075
PLy T 2-1 1.0('()lPQ 1.7 P 7 a.coo 5.400 2.02() r.2b,) 0.12"
0.053 0.115
l.. 04 7 1 4 .100 ,.guO I. 160 ('.200 0.172
DLVT2-" n.e( OO~R 0.220
PLV'7-A 0.COO123 41. no 72. '00 6.~5U 1. AOI) 0.340 O.IR" I 0.050
0.56U O.C 19 I 0.050 O.OM
JilW-, ( • a G('"'I) 9 1.3~7 ~. 0')0 11.2U<.I 1. I'JO
0.4AO 0.(191 e.043 (J.C~1
p.132 I <.I. O.lO II. ZlJv 5.1<0
JtI"lr _l.
JI I"J e - 3 I O.COO~OO
O.OJr)"Cq 2h.U13 I 1';.000 I!. h·J 4.0'10 0.4RO I 0.120 0.043 0.135
I
n. VO(POO IOH.o03 25.0011 11.2,jv I 3.IUO 0.490 0.158 v.U44
I
0.215
I
J' I'lf - ~
JiI"Jr -7
J'jr,iF _or I 0.CCC940
·1. 0()O~40
Ci.404
-; • R"1A II ~.O')O
1(" •.')J'J
h.6011
b. &OJ I 4. 2~U
).?OO
0.392
(;.320
I U.290
I 0.U90
0.100
0.116
u.U5B
0.048
0.044
I
O.U77
O.llQ
0.142
2. ~OJ
JIINc-7
JlI'JI=-7
JUN' -7
U.('U0040
1).'1I..)O Q4 O
:).01"'0"40
Q.17"-
4C.~6()
f5.~76
15.2'10
?~. ~OO
3'1. 0)0
".6)U
6.6U')
6.6UO
I. gOO
1.;40
0.315
0.30~
0.175
0.112
I U.U4B
(J.v~5
0.245
0.271
0.704
0.ql~ 0.2JJ 3. 0<0 0.0 0.0 0.0
PlVT1-P 1).00IQ~4 ". lao ,).v61 0.683
1~.C~'l n.50() I U. 4(,) 1.78'1 0.630 0.354
DlV T 3-A 0.007?47 u.0 9 4 0.185
(1. on f.UOO s .IUO 3.6A'J (,.480 O.ll()
JUt Y-l O.105P21 0.077 0.260
~.HC 14.'100 5.100 2.41J 0.395 0.163
JI1l V-l G.CC5·21 u.088 0.418
73.<;74 22.0'n 5.1( 1) 1.760 0.450 0.256
J'IlV-3 0.OC5P21 v.J15 0.4t2
1-1. e15 30.4'J() 5.1') ) 1.5W O.3~1) 0.244
JUlV-3 0.005°71 ll.<.I78 0.563
101.177 3'.COO 5.10'} 1.430 0.400 .).2·0
J'ILY-l 0.005'21 0.1 ()7 0.191
0.4A2 6.()0) 4.50J 3.000 0.480 \J .160
J'IL Y-" 0.005001 U.ll87 0.271
<.5f<) 14.0lJO 4.50J 2.150 0.39J O.lAI
JUl Y-~ O.oosrOI U.078 0.346
20.217 2? 00lJ 4.50U 1.5"0 0.350 0.224
JUlY-5 0.005UCI 0.076 0.441
4.50) 1.370 0.341) 0.24~
JUl V-5 o.on~()(11 52.177 30.40()
I. ?50 (1.330 0.264 u.1.I73 0.4<;9
J'IlV-5 I).!JC5LCI ~to •• "1) 3~.?00 ... 500
12.000 1.920 0.1"0 0.105 0.264
t.~ ~ 6. C)0 18.'v,",
J'ILY-?~ 0.0~Q741
"• v.079 0.376
JJJLv-l"
Jill Y-2",
IJ.0~"741
(1.02<;'4\
45.905
141.e C ?
1 ... UOI)
22.000
30.(01)
13.1JU
18.3 )()
18. 3J)
7.25)
5.2~O
4. 1 ~'l
1.450
1.260
1.100
I 0.200
0.240
I 0.265
I
I lJ.069
'J.(J6U
LJ.478
0.549
JIJLV-25
JUl V-2 ~
JIJ"IC-f>
0.02°"741
C.029HI
0.010·'17
33" .140
~43.ll1
1.30t
36.'lOI)
". CJn
I A • 3).1
22.5 JJ
3. ~5'J
12.700
1.120
I.QOO I U.31~
0.150
u.061
0.081
U.069
I 0.658
0.320
0.4b(J
().01na9l 1-.5P4 10.0')'1 2 j . 50 J 7.POO 1.63') 0.209
JIJNc-6 u.010 0.648
17. c IQ 15.LaO 23.5()O 5.500 I .f 50 0.300
JlJt-.I~ -,., 0.030A<;1 U.v54 0.787
0.031)A97 61 • <;R' 2~.(OO 23.5aO 3.750 l.2A!) 0.341
JU"I" -6 0.353 (J.u5t O. A72
JIJ'JF -, (J.OJlJAH Ill.2f7 30.00a 23.50() 3.40(J 1.200
15.IUJ d.300 2.2°0 0.P5 0.151 0.5100
J'JLY-IO 0.U2Q~3 I.el~ 6.<:00
15. lUO 5.25 1) 1.36U 0.25Q 0.090 0.582
J'JlY-IO fJ.0l,2P ~3 17.IU 14.C00
t .230 0.357 U.081 0.7~4
C.L62A~3 5".333 V.OOO 15.100 3.4~0
JUlY-I'1 0.060 0.740
110.17" 30.1. )0 15. lu J 3.000 0.910 0.303
JULY-IU 0.U2·~3
15. Iv) 2. SOO 0.910 0.364 v.06U 0.862
J'JlY-IO 0.062R53 22?6Re 3h.200
q .GQ·j 11. SUl) 3.~70 1.500 0.3.8 U. I 3u 0.171
PL \IT 1-" [).111"f~ 1.40?
14.(0') 11.55 J 3.:>50 1.100 O.31~ U .095 0.813
PLvTI-r n.17R"7~ " .U17
PLVT I-l
PI VTI-C
0.16 Q O"3
fl. 1"~8~7
11.12 0
20.412
19.u00
23."()O
II. JO)
1 U. '~UJ
I 2.7(JU
2.')00
0.890
().Q~O
0.405
0.425
0.U81
IJ.J76
0.A83
I.G21
11. 15) I. 53(' 0.680 0.444 0.061 0.984
PLYTI-K 0.176 0 45 41,.730 2 •• FOO
0.600 U.444 u.055 0.997
PL VTt-p 0.11,9 Q4 3 "6.004 33.tOO II.uOn 1.350
3~. 100 11.40·) 1.220 0.540 0.443 u.047 I. J22
PlVT I-H 0.IAI P 41 ItO.740
0.68A
6.000 2U.9Jv <;.600 2.951) 0.307 0.141
JIJ"Ic -2 ~ 0.2162% 1.752
I~.IIO 14.00iJ 2 O. 9)U 5.600 1.100 (J.30 4 0.081 0.735
JIJ"-JC_ZP n.216?54 0.925
3.5JO 1.480 0.423 I) .071
JIJ"'c-ZP O.21~254 47.C54 72.000 2 O. 9uJ
30.2 JQ 2).90,) 2.71)0 1.131) 0.419 U.uS't 0.932
JIJ~C-?P 0.216154 111.541,
JJI'J c -2P
J'IL Y-I
O.2lt2~4
v.?31t.~1
I 183.772
1.9 0 7
16. ZUO
fo. COO
20."00
21.95J
2.300
11.1.200
I.OOJ
3.0"0
O.b 35
0.300
0.(,4<1
U.140
O. ~71
0.717
14.UOJ1 21.850 5.650 I.AIO C.320 0.U83 0.714
J')L Y-I f).2314~1 111.450 I
Jut V-I 0.23l4~1
0.231431
5 Z. 175
124.06 4
n.coo
30. 4 00
21.35u
21. A5v
3.~OO
2.800
1.420
1.120
0.30,4
0.400
O.06~
0.051 I O.P 76
0.914
JULY-I
J·JL Y-I 0.2314~1 20~.324 36.400 n.1l5J I 2.400 l.02(J
2.890
0. 4 25
0.340
I 0.041
o .13Z
0.<;78
U .811
8.500
JIJ"I" -26
J'JNF-26
JU"J c -2 /,
i
Ci.2Q451P
0.294518
t.2G451"
30.837
11".249
26'>.62"
12.(00
20.0')0
2R.COO
I
21.901)
21.900
21.900
I ~.25o
't.800
I 2.250
i.800
0.360
Q~375
I 0.103
0.062 I 0.364
o. e 75
36.000 21.900 4.200 1.68J 0.400 0.071 0.991
JU"l~ -2 6 11.2 0 45·IP S37.4~1
6.000 2) .10J 9.000 3.200 0.35& 0.147 1.082
JlJlY-2° tJ.423'l77 1.664
JULY-2~ 0.423077 14. 1~5
47.241
14.000
22.000
21.70.;
21.7JO
4.&50
3.120
1.100
1.3aO
0.366
0.417
0.07&
0.060
I 0.979
1.051
JUlY-2Q 0.423911
'\0.01}0 21.700 2.300 CJ.980 0.421. 0.045 1.029
JlJlY-2~ 0.423<;11 110.170
JUlY-?9 0.423'<71 lf~.7C3 36.001) 21.700 2.060 0.850 0.413 0.039
I 1.045
I I I
Table 4.2.1. Tabulated experimental values for the intensity of
turbulent fluctuations of temperature along jet axis
for turbulent jets and plumes.
-l1S-
phenomenon but there is a lot of additional information which is
important and should be considered when intermittent flows are in-
vestigated. Specifically the turbulence intensity T'2 can be analyzed
as the sum of two terms as follows
-- . 1 t'
T'2 = L~m i' J [T(x,y,t) - T(x,y)]2dt,
t.4.00 0
1 t'
Lim"':-l J [T(x,y,t) - T(x,y)]2[1 - I(x,y,t)]dt ,
t.4.00 t 0
1
+ Lim i'
t.4.00
Jt' [T(x,y,t)
0
-
- T(x,y)]
2
I(x,y,t)dt, (4.2.9)
where
1 t'
T(x, y) Lim -:-t JoT (x, Y, t) d t , (4.2.10)
t4.<x, t
and
I( x,y,t ) ={ l i f T(x,y,t) > T

a
+ o.osoe,
o otherwise.
Hence T,2 = T,2 + T,2

cold hot'
where
1
Ti7
cold = Lim i' t.4.00
J t' [T(x,y,t)
0
-
- T(x,y)]2[1 - I(x,y,t)]dt , (4.2.11)
-116-
and
T,2
hot
Lim
t4<:o
i, Jot' [T(x,y,t) - T(x,y)]2 I(x,y,t)dt . (4.2.12)
Clearly, if I(x,y,t) = 1 for all the time t (which means y(x,y) = 1)
then T'2
cold
o and T'2 = T'2.
hot
If I(x,y,t) = 0 for all the time t
(which means the temperature probe is always in the cold ambient water)
then
T'2 T,2 =T,2=0.

hot cold
Figure 4.2.5 (where an intermittent temperature signal is sketched) can
be used to clarify the physical meaning of each of the two components
T,2 and ~ The turbulence intensity (see Eq. 4.2.1) is the moment
cold hot'
of the whole signal T(x,y,t)(the continuous and the dotted line) with
respect to the mean temperature T(x,y) (defined by Eq. 4.2.10). T,2 ex y)

hot '
is the moment of the continuous line with respect to the mean T(x,y) (again
defined by Eq. 4.2.10) and the T~;ld is the moment of the dotted line
with respect the mean T(x,y) (again defined by Eq. 4.2.10). Hence the
term T'2 can be considered as an apparent turbulence intensity

cold
generated by the intermittency, i.e., by cold bursts or puffs.
The presence of the term T~;ld does not necessarily imply turbulence
(it can even exist in the case of an unstable laminar jet). An intrinsic
look at the profile of turbulence intensity can be obtained from
Figures 4.2.6 (~pure jet) and 4.2.7 (~plume) where T ' 2, T'2 T'2
cold' hot
are plotted, and from which some interesting conclusions can be drawn.
The two peaks in the profile of the turbulence intensity T,2 of a pure
-117-
Ta -~--
T a, Ambient Temperature
~--------------------------------------------------------- Time
Figure 4.2.5. Sketch of an intermittent temperature signal.

"I
RUN JULY 5
Ro = 0.004
x/a = 36 ISYMBOL
T'2 / TI~ 0
NE -;z -2 0
Ir- T HOT / T m
.......
0
-J
- ,2 -2 t::.
0
u
T COLD / Tm ----- ---- ---
I:' 0.2
NE
Ir- I
f-'
....... f-'
~ 00
0 I
:r
I~r-
0.1
NE
Ir-
.......
11-
0.0 ------
-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3

Y IX'
!!:.sure 4.2.6. Profiles of turbulence intensities T ,2 , T'2 , T '2
hot co ld
for a turbulent jet.
r I I .I
RUN JULY 29
Ro = 0.423
XID = 30 SYMBOL
0.3 T,2 I :r~ 0
NE
II- -2 -2
...... 0
T'HOr/Tm
0
.J
0 -:z -: b.
"'u T COLD / T m
I~
0 0 0
0.2
0 0,;0-
'" E /'
.II- 1
...... cY I-'
l-
./' I-'
\0
0
J:
,,0 1
I~I- o /
/0
0.1 0/
N E ./'
II- ./
"
I~I-
0.0 I y I~ 9<"1 I.....>J

-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
. y/ X I
,2
Figure 4.2.7. Profiles of turbulence intensities T , Th~t' and T~~ld for a
turbulent plume.
-120-
jet (Figure 4.2.6) are mainly generated by the large cold bursts
(which may be irrotational), i.e., by T~;ld. The Reynolds equations
and the energy equation, where all phase information is lost by
averaging, cannot be used to predict T~~ld. It was found previously
(Section 2.2, or Figure 4.2.3) that the normalized turbulent intensity
~T'2/TM is greater for a plume than for a jet due to the action of
buoyancy forces. Since T,2 was analyzed into two components Th;t (hot
vortices) and T,2 (cold bursts) the question is which component

cold
in particular the buoyancy forces feed. The answer to this question can
be obtained from Figures 4.2.6 and 4.2.7 from where it can be observed
that the non-dimensional profile T~;ld/T~ (apparent intensity due to
cold bursts) is almost the same in magnitude and shape for both
jet and plume. However, the profile Th;t is much bigger for the
plume. It can be observed also that on the jet axis
T,2 ~ a
cold '
for both jet and plume and that
(112)
T~ jet
Hence it seems that the buoyancy forces intensify the T'2

hot term , 1·. e. ,
they feed mainly the hot vortices.
4.3 CROSS-SECTIONAL PROFILE OF MAXIMUM AND MINIMUM TEMPERATURE
The maximum and minimum temperature occurring at a given point (x,y)

-121-
during a sampling time, and made non-dimensional using the mean center-
line temperature TM(x) , are plotted in Figures 4.3.1 and 4.3.2 for the
two typical cases of R ~ 0.005 (jet) and R ~ 0.4 (plume). The

o 0
boundaries of the jet can be defined as the distance y/x' at which the
profile of maximum recorded temperature drops abruptly to zero, i.e.,
the equation
y = ±B(x) ~ ± 0.25 (x - x
oT
) ,
can be considered as defining the boundaries of the jet (Figure 4.3.1).
For the same experiment the equation of half-width of temperature
profile is given by y = ±0.12l (x - x oT ) and the equation of standard

deviation by y : ±0.103 (x - x ). Hence the boundary as defined by
oT
the maximum temperature profile is approximately at 2.4 standard
deviations from the axis. The centerline maximum and minimum tempera-
tures are plotted in the Figure 4.3.3 for many initial Richardson
numbers. The conclusion from Figure 4.3.3 is that the non-dimensional
centerline maximum temperature Max TM/1M and the non-dimensional center-
line minimum temperature Min TM/1M depend on the Richardson number, R.
Specifically Max TM/1M increases with R and tends to an asymptotic
value equal to 2.5 for R ~ Rp while Min TM/1M decreases with increasing
R and becomes zero for R ~ R. This is consistent with the results of

p
Section 4.2, i.e., that the turbulence intensity increases with R.
The interesting result Min TM/1M = 0 for a plume implies that the
intermittency of the temperature at the center of the plume is not one,
and cold ambient water can reach the axis of the plume.
1.9~1------~--------~--------~--------~----------~--------~--------~----__,
~
RUN JULY 5
Ro = 0.005
X/D Max T Min T
1.5
Tm Tm
14 0
•
22
30
0
X
•
+
-
36 '7 ~
0
~ / 0 I"\.
E
I~ 1.0
"-
~
c:
.-
~ I
I~
"-
E
0.5
t 1° )(
f-"
N
N
I
~
>C
0
~
~
•
o.oLo I Q 1 '7. ± + ~ +
• + I
"' . • •'" • 0.11+ e-t +1 j . "...0 0 0 0
-0.3 -0.2 -0.1 0.0 0.2 0.3
Y/ X'
F~gure 4.3.1. Maximum and minimum temperature for a turbulent jet as a function
of distance from jet axis.
2.5r-
RUN JULY 29
Ro = 0423
t::. \]
\]t::.\]~ X/D MoxT MinT
2.0 Tm Tm
6
14
t::.
0
••
22
30
0
X
•+
36 \] ~
E 1.5 --
I~
"'-
~
c:
~
I
f-'
1.0 N
Vol
E I
I~
"'-
~
><
o
~
0.5
O.OL-.~
•
- 0.3 -0.2 0.0 0.1 0.2 0.3
y / X'
Figure 4.3.2. Maximum and minimum temperature for a turbulent plume as a

function of distance from plume axis.
'TT---'-'-I T
• • . •
2.0 RUN Ro Max TM MinTM
JUNE 3 0.0009
JULY 3 0.005
•
.A.
0
b.
I
~ JULY 10 0.06 ~
JULY29 0.423 • 0
1.5
~
n- Ro e
A \ e
"l- ~ ~ eT
e
- e
I
~ I-'
1.0 N
+:'-
~
II-
-, I
" ~
1-. 0.5
)(
0
~
0.01 I IJ':o:-.t f1 1=r:1 0 B-- !
0.0 10 20 30 40 50 60 70
X/O
Figure 4.3.3: Distribution of maximum and minimum temperature along jet axis
as a function of initial Richardson number.
-125-
4.4 INTERMITTENCY
The intermittency y of the temperature was defined as
. 1
y(x,y) = Llm -;
t~ t
J. t' I(x,y,t)dt,
0
where
(I if T(x,y,t) - T > O.05°C

I (x, y, t) = {O a
, otherwise,
where T is the ambient temperature. The intermittency profile is

a
plotted in Figure 4.4.1 for an almos~ pure jet and in Figure 4.4.2 for
an almost pure plume. The intermittency at the center of the plume-
like buoyant jet was found to be of the order 0.985 to 0.999, which is
consistent with the profile of minimum recorded temperatures (Figure
4.3.2). An interesting conclusion for an almost pure jet (Figure 4.4.1)
is that the intermittency at a given y/x' location, where x' = x - x oT '

depends on the distance x from the exit. There is no previous experi-
mental investigation of the intermittency profile of temperature with
which to compare the above results, but there is an experimental in-
vestigation of the intermittency profile of velocity by Hanum [40].
By replotting his Figure 2 in new coordinates, (y versus y/x), it can
be seen that the same pattern is revealed, i.e., at a given location
y/x the intermittency decreases with increasing x, (see Figure 4.4.3).
Hanum [40, or 61] did not report the location of the virtual origin
x in his experiments so it has been arbitrarily chosen as zero. From

o
his replotted results (Figure 4.4.3) it can be obtained that y(x,y) ~ 0.6
I T-j
RUN JUNE 7
Ro=00009
X/D SYMBOL
1.0 0 0
1'0
20 t::.
30 0
50 0
0.8 60 \J
0.6
>--. I i fot ,\\ J I

~
N
, I I I '( \ \
(j\
I
0.4
0.2
0.0 I IV 1-,- I
-0 . 3 -0.2 -0.1 0.0 0.1 0.2 0.3

Y IX I
Figure 4.4.1. Temperature intermittency for a turbulent jet plotted against

non-dimensional distance from jet axis.
IT-TIT 1111\
RUN JULY 29
Ro :: 0.423
X/D SYMBOL
1.0r- A~'V 0
~ 0
14 6.
22 0
0
O.Sf- 'VJV \ ~36 \l
0.6 I
r ~
I \.
I
~
L \0 f-'
N
'-I
I
0.4
0.2
0.0 I Q IU I Y
- 0.3 - 0.2 - 0.1 0.0 0.1 0.2 0.3
y / X'
Figure 4.4.2. Temperature intermittency for a turbulent plume plotted against
non-dimensional distance from plume axis.
T-'l~ ~I
Ro = 0
DATA. FROM RANUM [40]
~Y~_~OL I
X/O ~
1.0.- ----?
/,
--......,
. ~ 6 ·
12 ·
, ·
/1/'l "", \ 18
, 0.81- I I, \\, \
I/
I,
\\
I \
\\
>-.. O.Gr- I I:1 I \ I
'\ \
r-'
I
N
1/
I I
\
I \
00
I
OAI-
I
r
I
I I \
II
I '
\\
0.2~ I
II \\
O.O~I ____________~~~~~~~~~~~~~~~~~________________~
-0.2 -0.1 0.0 0.1 0.2
y/x'
Figure 4.4.3. Profile of the velocity intermittency for a turbulent jet
(based on data by Hanum [40]).
I r-l~r~I--l-T 1- I -, '--'-T- I
RUN JULY I
Ro = 0.231
X/D SYMBOL
14 •
22 6-
30 x
36 V
>-
u
z 1.0
w
l-
I-
:::E
a::
w
/ ~
I
I-
Z
-r\x6- \ ~
N
~
I
)...
0.5
O. rr] y IP><J:II A
-0.20 -0.10 0.0 0.10 0.20
YiX'
Figure 4.4.4. Temperature intermittency for a turbulent plume plotted against

non-dimensional distance from plume axis.
-130-
at y/x = 0.10 at X/D = 18, which seems too low compared with the tempera-
ture intermittency results ot this investigation or with the results of
velocity intermittency of Heskestad [5],who reported only one inter-
mittency profile at x/D = 102 and who found y(x,y) ~ 0.92 at y/x' = 0.10.
The temperature (or concentration) intermittency Y can be measured more
H
precisely than the velocity intermittency y (where the vorticity should
u
be detected to distinguish the turbulent from the irrotational fluid).
4.5 FREQUENCY OF CROSSING OF THE HOT/COLD INTERFACE
The frequency of interface crossing fR in Hz is given by
M(x,y,t.)
1
i=l
where
if [I(x,y,t ) + I(x,y,t + )]
i i l
=0 or 2
M(x,y,t.)
1 = 1°1
if [I(x,y,t ) + I(x,y,t + )] 1
i i l
(see Figure 3.5.1) was plotted in Figure 4.5.1 for the case of an
almost pure plume, and in Figures 4.5.2, 4.5.3 for the case of an
~
RUN JULY I
Ro = 0.23
0.5 X/D SYMBOL
22 0
30 0
3,6 \J
0.4
o 0
0.3
if\ . .. . I
f-'
w
I
C
.-
N
/ °IO~ q /0:'/0\0\\ f-'

I
- a::
0.2
0.1
0.01 d ~ 0
-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
y / X'
Figure 4.5.1. Profile of frequency of hot/cold, cold/hot interface crossing

for a turbulent plume.
-132-
TfI-'1
rR,,--u-'-,N-JU--Cl--Cy-_ 3-1 I
Ro ' 0006
X/D SY MBOl
14 6
22 0
30 0
36 V
04
N
I
'= °l-
a::
~
0.2
0.1
020 0.30
Y I X'
Figure 4.5.2. Profile of frequency of hot/cold, cold/hot

interface crossing for a turbulent jet.
I
'I
o5 -j
-1
0.4
N
I
0.3
-'=a::
....
0.2
0.1
00
-0.30 -020 -0.10 00 0.10 020 030
YI x'
Figure 4.5.3. Profile of frequency of hot/cold, cold/hot
interface crossing for a turbulent jet.
-133-
almost pure jet. The same general structure was revealed in all
experiments performed.
It should be noted that the nature of these measurements relies
essentially on the formation of the test function l(x,y,t) (see Eq.
2.0.1) which depends on the threshold level T , i.e.,

L
{ ~ otherwise
if T(x,y,t) > TL '
l(x,y,t)
.
In this investigation TL ~ Ta + 0.05°e. The absolute value of the fR
depends on the value of the threshold level chosen (T ), but a systematic

L
investigation of this dependence was not made. The important point is
that keeping TL constant for a particular experiment resulted in the
same structure being revealed. A feeling of the reproducibility and
of the effect of the threshold TL can be obtained from Figures 4.5.2
and 4.5.3 which represent the results from two experiments with almost
identical initial Richardson and Reynolds number, but with slightly
different thresholds, i.e., TL = Ta + 0.04°e, for experiment "July 3"
(Figure 4.5.2), and TL = Ta + 0.07°e, for experiment "July 5" (Figure
4.5.3). As should be expected, the level TL T

a
+ 0.04°e gives
slightly lower values than the level TL = Ta + 0.07°e. The peak
frequency can be used to form a Strouha1 number based on the ha1f-
width and on the exit velocity and it is plotted in Figure 4.5.4 where
some results by Bradbury [2] are also plotted. From the intermittency
and the frequency of interface crossing the average duration of a hot
burst at a point (x,y) (time between two successive cold/hot, hot/cold
interface crossings) can be found.

---T-- -----1
0.8 SYMBOL RUN Ro I
0 JUNE 7 0.0009 I
0.7 0 JULY 3 0.006

PRESENT STUDY
"V JULY 5 0.005
0.6 ~ JULY I 0.23
X
-------
0.0- -
-----
BRADBURY 1965
0.5
0
I:::J 0.4 I
........ f-'
...- W
I- +:-
.0 I
0.3
......a:
II
(f) 0.2 I\. _ _ _ _- -
E ts D.
-"'I.. x--
0.1
0
_~~~---Ll--~------A-
...~---......JV-
0 10 20 30 40 50
X/ D
Figure 4.5.4. The Strouhal number, based on the peak frequency of hot/cold,
cold/hot interface crossing profile against distance from
jet orifice.
-135-
(in seconds)
and is plotted in Figure 4.5.5 and 4.5.6.
4.6 THE FREQUENCY DISTRIBUTION OF TEMPERATURE
The frequency distribution of the conditionally sampled ins tan-
taneous temperature is plotted in Figure 4.6.l.a-d for four typical
positions across the jet. It is clear that the turbulence is not
homogeneous (because the shape of the frequency distribution depends
on the non-dimensional position y/x'). The intermittency y(x,y) is
defined (see Section 4.4) as the fraction of the time for which the
temperature of the flow is above some level TL (where TL is close to
zero, i.e. TL ~ 0.05 - O.lOOC). Hence 1 - y(x,y) is the fraction of
the time for which the temperature of the flow is below the level T .
L
Now, the peak located near zero temperature (see Figures 4.6.l.b-d)
is due to the intermittency of the flow. Since the whole area under
the frequency distribution is unity, the area of the peak around zero
temperature is approximately 1 - y(x,y).
4.7 SKEWNESS AND FLATNESS FACTOR FOR THE TEMPERATURE DISTRIBUTION
The skewness factor, defined as
is plotted in Figure 4.7.1, and the flatness factor defined as
is plotted in Figure 4.7.2.

-136-
RLJ N JULY I
R. ' 023
x/a SYMBOL
22 0
36 a
10
Ul
'0
Q
0
CJ
aJ
c.~
Q
-.-I
10
-
~
;:-
0
o
01
-0.30 -0.20 -0.10 00 0.10 0.20 0.30
VI x'
Figure 4.5.5. Average duration of hot bursts in a
turbulent plume.
RUN JULY 5
R.'OOO5
XI D I
SYMBOL
22 I 0
36 I 0
10
Ul
'0
Q
0
CJ
aJ
Ul 0
0
Q
-.-I 1.0
-
0
~
; :-
J
-0.30
0
-0.20 -0.10
VI x'
00 010
I I
020
I
0
J
0.30
Figure 4.5.6. Average duration of hot bursts in a

turbulent jet.
-137-
0.6 RUN PLVT I-H

y/x': 0.005
0.5
>-
~
(f)
z 0.4
w
0
>- 0.3
~
-.J
m
<:(
m 0.2
0
0:
a..
0.1
0.0
0.0 1.0 2.0 3.0
TEMPERATURE °C
Figure 4.6.l.a. Probability density of the temperature
of a turbulent plume at y/x' = 0.005.
0.6 RUN PLVT 1- H

Y/X': 0.055
0.5
>-
~
(f)
z 0.4
w
0
>- 0.3
~
-.J
co
<:( 0.2
m
0
0:
a..
0.1
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
TEMPERATURE °C
Figure 4.6.l.h. Probability density of the temperature
-l38-
2.5~----~~----~------~------~------~------~~
RUN PLVT I-H

Y/X': 0.098
2.0
>-
f-
(f)
z 1.5
w
0
>-
f- 1.0
-l
aJ
<t
aJ
0 0.5
0::
a.
O.O~ ____- L______- L______ ~ ______ ~ ______ ~~ ____ L-~
0.0 1.0 2.0

TEMPERATURE °c
Figure 4.6.l.c. Probability density of the temperature
0.6r----.~----------~------------~--------------~
RUN PLVT I-H

0.4 Y/X'= 0.148
>-
f-
(f)
Z 0.3
w
o
>-
f- 0.2
.J
aJ
<t
aJ
o 0.1
0::
a.
0.0 ...1.-1
, - - - - I
0.0
-~----'--""'-_~---=:::::::t:::_--L.-..J
1.0 2.0 3.0
TEMPERATURE °c
Figure 4.6.1. d. Probability density of the temperature
of a turbulent plume at y/x' ~ 0.148.
8.0
I J r I I
6.0
4.0
0.2
'X O
0.0 "'f'.:x_O ____t:oX I
-
N
,.,
-1.0
-~~ ......
W
1.0
I
11-
'" RUN JULY 25
11- Ro = 0.03
X/D SYMBOL
14 0
22 6- ,
30 X
36 0
--'----'-------'-----I---..IL--J'---'------'-------'--- _ L J I I I I I I I I
-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
Y I Xl
Figure 4.7.1. Skewness factor of temperature fluctuations of a buoyant: jet as a function
of distance from the jet axis.
I T
RUN JULY 25
Ro=0.029
\5.0 X/O \ SYMBOL
14 0
22 I 6.
30 1 0
36 1 0
(\J
\0.0
r--
I~.-
---
-.......
I
~
~
o
I!r- I
5.0
0.01~~1~~_~~~~~~~~~~~~~~~~~~~~_~~~~~~-J
-0.3 -0.2 -0.\ 0.0 0.\ 0.2 0.3
y/ XI
Figure 4.7.2. Flatness factor of temperature fluctuations of a buoyant jet as a function
of distance from jet axis.
-141-
For a Gaussian curve the skewness factor is zero and the flatness
factor is 3.0. The conclusion from these figures is consistent with
the results of the previous chapter, i.e., the turbulence is not
homogeneous.
-142-
5. VELOCITY MEASUREMENT RESULTS
5.0 BASIC EXPERIMENTAL PARAMETERS AND INITIAL DATA
Complete information regarding the basic parameters and the
initial data of the experimental runs is given in Table 5.0.1. An
explanation of the data columns is as follows:
Column 1 gives the name of the experimental run. The mean transverse
velocity profile was measured for all the runs except the runs
Center, Cent 1 and Cent 2 where only the centerline mean velocity
was measured.
Column 2 gives the width of the jet (in em) at the orifice.
Column 3 gives the aspect ratio of the jet at the orifice.
Column 4 gives the exit velocity (in em/sec) of the jet.
Column 5 gives the temperature of the ambient water in °c.

Column 6 gives the excess (above the ambient) temperature at the jet
orifice in °c.
Column 7 gives the kinematic viscosity in cm 2/sec at the jet orifice
and multiplied by 100.
Column 8 gives the thermal expansion coefficient (11°C) of the discharge
calculated at the jet orifice and multiplied by 1000.
Column 9 gives the Reynolds number of the jet at the orifice.
Column 10 gives the Richardson number of the jet calculated at the
orifice.
a. gT D
o 0
R =
o u2
o
-143-
-u
Run D
Aspect 0 -T T
a
\i a R
e
R
0
cm
-- 0 0 0
ratio sec °c °c x 10 2 x 10 3
1 2 3 4 5 6 7 8 9 10
1.1.60 0.90 u.l.~ lbd3.j~ 1l.Uuuu)1

PIVT}-f' 0.l4 '4.20 .,2. q 3 3 • .lO
bl.OO O.u '.l.UV 0.94 v.4t.,j 1,"4.1" O.U
PlV T 2- P 0.24 54. ,0
1117.0_ O.OIJOvti'1
J.24 "I..~J ~.4V .... l.dO 0.65 u.27
NVT2-1 '4."V U.O
0./4 54. <0 bl..bJ o.v .::J.OI,) 0.~4 L1."J IbLlLldi
"lVT2-C u .uJuu:>1j
b4.iLl .l.6v LI..Lv 0.67 v.l<> i1,l.70
°lVT2-H
PlVT7-'l
v.24
0./.
1)4. 4 0
'4.l0 bl.9l
b2.90
O.ll
".d5
.:.l.OU
Lj.'~
II 0.94
0.60
v.l J
vdv
lovl.ijij
Id, 7 • .lu
O.U
L1.uuuII.J
P(VT~-A 0.l4 '4.2v
O.vl 1.4.'i~ o.~o 0.£:> £,J'L.t..t!. o.UOJUvd
0-7ft 1.00 IJ.OO Id.IO
,I I ",." I O.UJld94
.14.9~ 0.90 v.I., ivl.<.l< u.uOv\JUtj
f'I_1P. 1.00 IJ.OO H.IO O.vl
PlVf3-FI. 0.24 51t • .:. ,J 1 'i.uo '4 • .:.1.1 t.2.,)J v.' , O.Jl
H.UO 10.4U "3.b) 0.74 U.:":>i 01". i1 O. L1u.!.l4l
Pl v n-A 0.2.4 ,4.<0
p-2Q
P-1A
P-3A
1.00
I. UO
1.00
IJ.Ov
13.UV
13.00
7.5b
7.56
15.11
I 14.10
Ib.90
20.00
Lit .vu
I.l • 00
21..uv
0.b7
0.66
0.6l
1,).";0
0 • .Jo
a.)d
llLl.~J
1I>9.1J
2409db
O.UtiS
v.IJl
O.L1JJ
l4.vU 0.73 v • ..,)j ,bJb4.1"1 lJ.v14
"-t.~ 1.00 H.uv 15.10 10.00
P-46 1.00 !.l.OO 15.10 II.ILI
10.VO
I.". JO
l4.vLl
0.75
0.73
Ud2
v.J)
1.0",.;5
2Ub4.19
v.ul;
V.U14
P-4( 1.00 13.(,0 15.10
11.00 £1..OLl 0.75 1.1.3£ oUI/.v:> U.U.d
PI vTt-F 1.00 I>.LV 4. 5~
4.55 11.l> a.uv 0.74 0."", oIV.~'i 0.112
Pl vTl-1'l 1.00 13.00
a.uv v.75 Udl. bU~.u5 U.lo~
PlVT!-1 1.00 13.UO 4.55 11.00
4.5, IU.~'! l" Jv 0.75 v.JL ou 7.76 0.1,,7
PlVTt-C 1.00 lJ.UO
4.;5 11.15 23.00 0.73 OdJ 01. J. 93 0.116
P'-VTI-" 1.00 IJ.GO
4.55 11.00 ;.£. LhJ 0.75 u • .>2 6v9.U5 O.lb'l
°lVT!-A 1.00 l.l. UU
4.,5 11.4U I.J.vu 0.7 J U. J3 621.11 O. I ~I
rlVTI-f' 1.00 lJ.UU
13.00 4.53 17.UU l4.U() 0.b4 ud8 lU'J.50 u.J1J
r-Ir 1.00
p_Ir lJ. UO 4.53 17.UU ,4.00 0.~4 Odo 1019. ,,, U. Jl 3
1.00
lJ.vO 4.53 11.UU I. ... uv 0.64 U.J~ lv'l.'" U..>13
r-l1 l.llO
4. ' j j 17.1.10 24.UU 0.64 Udd 7u~.~fJ u.JI j
~-I~ 1.00 1 J.UO
4.53 11.010 2 ... J0 0.64 U.Jd lv{~ .lju 0 • .>13
P-lA 1.00 lJ.UO
p_c.~
P-~A
1.00
1.00
l.l. UO
l.l. UO
b.04
b.04
b.04
30.4U
32o.lU
30.10
, ... 10
1..1.'U
"j.6u
I O. ,I
L1.51
0.52
u.4O
U.4t.j
U... ,
1115.~"
lido.HS
1157.20
u.4UV
v.4JJ
U.jtjtj
"-~n 1.00 13.01l
I)_C::C 1.00 LJ.OU 6.04 30.1.0 l4.UO 0.52 U.47 11"1.9~ U.J9"
4.,3 30.5U ,4.0U 0.52 O.'tu .j7-'.'17 0.7U7
P-fP 1.00 1>.00
l.l. OJ 4.53 30.40 1.4.70 0.51 O. 4~ ddl.'!" 0.712
P-~A 1.00
... 5J 30.10 24.IU 0.54:' U.4' l:ili.4:> U.b9't
f"J-Ar 1.00 13.00
C f" ~T ~p 5... l.v 62.01} ,1. Uv 1.00 hdd.UU U.V
0.24
5 ... l0 62.00 I.I.uu 1.00 14/jO.UU O.U
cr"lnO 0.24
ret-Teo 54. ,0 ti.UO "l.JU 1.00 1.4Mb.OIl u.O
0.24
Cr"·HFQ 0.24 54.2U b2.UO 1.1. VU 1.00 14dd.OU O.U
54.20 62.00 Ll.UU 1.0U 14dR.OU U.U
CFNTfo 0.24
(~~·n.:o 0.,4 5... 2L1 62.00 ,I. JU 1.00 14btl.OIJ 0.0
54.20 d I. 8 7 ,1.00 1.00 1~o4.~O u.O
C</,TI 0.24
~4.I.U 50.40 21. UU 1.00 1,09.60 U.U
UNTI 0.l4
54.20 50.40 L I. UU 1.00 llU~.bU U • .!
rF"'TI 0.24
54.20 81.67 21.UO 1.00 1~64.~d u.O
rr~Tl 0.24
r r. ...'T 1 81. B 7 <.1.00 1.00 19"4.RH U.V
0.24 54"~0
0.24 54.2U 02.90 21.00 1.00 I)U,.6v O.U
«"IT! v.U
0.1.4 54.20 50.40 21.00 1.00 IlU9.6v
U"Tl 1:> b. UU 0.0
erNTI 0.24 54.;':'U H.50 "I.UO 1.00
, ... 20 31.50 il.OO 1.00 1>".00 0.0
c." Tl 0.24
(C~ Tl 0.24 ')4.,0 ~I. d 7 21.0J 1.00 1~b4.tso u.u
u.24 54.20 I> I. d 1 I. I. ULI 1.00 1-104./jd 0.0
(F"ITI
0.24 54.20 B I.d 7 I. 1.00 1.00 Hb4.Bd o.u
eC~lT2
54.2u 6.1.96 Ll.OU 1.00 1511.51. u.u
re"Tz 0.l4
0.24 54.20 62.'18 <.1.00 1.00 1)11.5" 0.0
r '~I' 2 7,7. ,,~ v::-O
(F"IT7 0.24 ,4.20 H.S8 ll.UO 1.0v
cruZ 0.24 54.<0 JI.58 I. I.UO I.UO 7') 1. "ii. o.v
54.20 61.81 1.1.1l0 1.00 1"'04. tHI 0.0
UNT? 0.24
0.24 54.1.0 61.61 d.OO 1.00 1964.dd v.o
U"IT2
r. f 1Ij' 2 0.24 54.l0 dl.87 II.OU I.OU l'ih"'.hrt ll.O
Table 5.0.1. Basic experimental parameters and initial data of

the velocity measurements.
-144-
5.1 MEAN VELOCITY PROFILE
For every experireental run the mean velocity profile u(x,)7; in
ern/ sec was plotted versus t[~e tr,:Il SVen.l8 ~istXiC e y (in em), the half-
width b was calculated, and a Gaussian curve defined as

u
;-(x,y) ~(x) exp[-£n2 (y/b)2] (5.1.1)
was plotted as in Figures 5.l.l.a-f (representative case for a pure
jet), and 5.l.4.a-f (representative case of an almost pure plume). The
velocity half-widths b (x) for a particular run were plotted versus the
u
distance from the exit as in Figures 5.1.2 (jet) and 5.1.5 (plume) and
a straight line
(5.1.2)
was fitted to the data. The virtual origin of the mean velocity profile
is given by xou = -K2u D. The coefficients K ' K

lu 2u
for all the runs
are given in Table 5.1.1 from where it can be observed that there is no
correlation of the variation of these coefficients with the initial
Richardson number. However, it is interesting to notice that the
coefficients K ' K of the runs PLVT2 (~pure jet) and P-7 (~pure jet)
lu 2u
are quite different. An explanation for this is the apparent non-self-
preserving character of the pure two-dimensional jet and it will be
discussed in detail in Chapter 8.
The mean velocity ;-(x,y), normalized by the mean velocity ~(x) on
the jet axis, has been plotted against the non-dimensional distance
y/x' (where x' = x - x ) for the case of a jet, Figure 5.1.3 and the
au
-145-
o MERN VEUJC I TY PROF I LE
0r---_.----r_--~----~--~----_.----r_--~----,_--_,
o
::f'
o
o
If) fIUN f'LVT 2-E
en
Ro cO.OOOOS2
o
o X/O=20.B
o EXf'ERII£NT I!I
en
~
O~ __ ~ __ ~~ __ ~ ____ ~ __ ~~ __ ~ ____L -_ _ ~~ __ ~ __ ~
-2.0 -1.6 -1.2 -0.8 -O.ij -0.0 O.ij 0.8 1.2 1.6 2.0
Y AXIS IN CM
Figure S.l.l.a. Mean velocity profile for a
turbulent jet at X/D = 20.8
o MER~VELCJC I TY PROF I LE
0r----r----~--~----._--~r_--_r----~--~----._--~
o
en
o RlM'l PLVT2-B
o
If)
N Ro cO
X/D=33.3
o EXI'ERI t1f.NT I!I
o
u~
w
en
'0
::Eo
Ut',
·
z~
>-0
r-~
...... 0
u~
o
~
Wo
>~
If)
~
O~ __ ~ ____L -__ ~ ____ ~ __ ~L- __-L____L-__ ~ __ ~~~-J
-2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5
Y AXIS IN CM
Figure S.l.l.b. Mean velocity profile for a
turbulent jet at X/D = 33.3.
-146-
o
MERN VELOCITY PROFILE
0r----.----.----.----,---~r_--_r----~--_.----,_--_,
In
N
~ PlVT2-Z
g flo -0.00009
o
N )(10-37.0
Exrrnlt£NT I!l
U O
w~
(f)1n
,~
~
u
Zo
~o
>-~
I-
......
U
D
-lO
w~
>IJ')
~
oL-__~~=-L---~----J---~L----L----L---~~~~--~
-3.0 -2.!l -l.B -1.2 -0.6 -0.0 0.6 3.0
Y AXIS IN CM
Figure S.l.l.c. Mean velocity profile for a
MEAN VELOCITY PROFILE

g
~---r----r---~----._--_.----_r----._--_.----._--_,
In
N
RUN PLVT2-C
a
o Ho =0
o
N X/0=5Y.2
EXPER I t£NT I!l
U o
wC!
,r
(f)1J')
....
u
Zo
...... 0
>-~
I-
......
U
D
-lO
w~
>IJ')
a
O~~~~~L- ___ L_ _ _ _ ~ _ _~_ _ _ __ L_ _ _ _L __ _~_ _~~~~
-!l.O -3.2 -2.!l -1.6 -0.8 -0.0 0.8 1.6 2.ij 3.2 ij.O
Y AXIS IN CM
Figure S.l.l.d. Mean velocity profile for a
-147-
o
0r----.----.---~----~--~----_r----r_--_r----~--~
U"l
N
o I\UN PLVT2-H
o flo c{).OO006
o
N X/OzSB.7S
EXPERH1ENT [!J
U O
W~
,-
U"JU"l
:::E
U
Zo
~~o .
~
O~ ____ ~__L -_ _-LI____~__- J_ _ _ _~_ _ _ _L -__~~~~__~
-S.O -1,1.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 14.0 5.0
'( AXIS IN CM
Figure 5.1.1.e. Mean velocity profile for a
o
0r----.----.----.----.----.-----r----r----r----,----,
CD
o RUN PLVT2-0
o
U"l Ro =0
X/0=711.5
EXPERHENT [!J
u~
o
o .
W
U"J
'0
LO
U·
en
:z:
>-0
t-~
...... CD
U
CD
-1
Wo
>~
(I')
~
O~ __ ~~~L- __-L____ ~ __ ~ ____ ~ ____ ~ __ -L~~~ __ ~
-6.0 -14.8 -3.6 -2.4 -1.2 -0.0 1.2 2.4 3.6 14.8 6.0
'( AXIS IN CM
Figure S.l.l.f. Mean velocity profile for a
turbulent jet at X/D = 74.S0
o
VEL~CITY HALF WIDTH
~rl-----r----~----~----.-----'-----'-----'-----~----~--~
co
.-.
RUN PLVT2
o
Ro -0
-
o bu ID=0.109(X/D-2.S)
Ln
o
'"
~
::10
.o~
.(\/
I'-'
I-
o
.......
~I
3:0
o ./ I
LL~ f-'
.I>-
.....J 00
0: I
:c
.....Jo
0:0
z
OW
·
.......
(f)
z
Wo
~o
........
0('1")
Z
~
z: .
~I I L I
0.0 10.0 20.0 30.0 YO~O 50.0 60.0 70.0 80.0 90.0 100.0
N~N OIMENSI~NAL DISTANCE FR~M EXIT, X/O
Figure 5.1.2. Non-dimensional velocity half-width for a turbulent
jet as a function of distance from jet orifice.
,
o
NON DIMENSIONAL MEAN VELOCITY PROFILE
N
...:r- I
o
RUN PLVT2
o
-- Ro =0
X'=X-2.5
EXP( -58.37]2 ) -
o
co
o · EXPERIM. RESULTS
X/O SYMBOL
20.8 (!]
0 33.3 (!)
·
:!; (£)
I:;:) 37.5 A
0
........ Sl!.2 + 1
58.75 X I-'
I:;:) +:-
7l!.5 ~ \0
0 93.75 .... 1
::r
0 ·
o
N
o
+
o
0' · '=="W ""'I =::.....&.
-0.50 -0 .40 -0 .30 -0 .20 -0.10 -0 .00 0 .10 0 .20 0 .30 :0 .40 0 .50
7]=Y/X'
Figure 5.1.3. Non-dimensional mean velocity profile for a turbulent jet.
-150-
o MERN VELOC IT'!' PRClF I LE
o~---r----r---~----~---.-----r----r---~----.----'
LIl
fOol PL VTI-E
o
o Ra cO.l7
"::3'
XlO-9.0
EXPERI Io£NT (!J
U O
w~
(f)<'l
.........
:E
U
Zo
-0·
o
O~ __ ~ __ ~~ __ ~ ____ ~ __ ~~ __- L____ L-~~~ __ ~ __ ~
-6.0 -4.8 -3.6 -2.4 -1.2 -0.0 1.2 2.4 3.6 4.8 6.0
Y AXIS IN CM
Figure 5.1.4.a. Mean velocity profile for a
turbulent plume at X/D = 9.0
° MERN VELClC IT'!' PROF I LE

0r----r----r----.----.----.-----.----.----r----.----,
Lf)
1IUr~ PLVTl-O
o
o Ro =0.172
"::3'
X/O=llJ.O
fXr'ERlI"ENT
\\
o
Jl.-2---_-~O-:.0---1L.2----2L• 4----3.1-.- -..=:::::==4"'".-8---1•0
'-:6=-.-=0---:-:4.....8:==-----:3-:-.-=-6---:_2-:!-.-::-4----
0_
6 6
Y AXIS IN CM
Figure 5.1.4.b. Mean velocity profile for a
-151-
o
MEAN VELCCITY PR~rILE
o~---'----.----.----.-----r----r----~---.-----r---'
III
~ I'LVTI-Z
o
o
::J'
"0 aQ.I?
XlO-19.0
EXPEftHENT [!J
U O
w~
(/")(1')
.......
::E
U
Zo
~o
>-N
I-
......
U
o
...10
w~
>-
~
OL-__ ~~=-~ __ ____ ____L -__
~ ~ - L____L -__ ~~~ ____ ~
-6.0 -6.~ -~.8 -3.2 -1.6 -0.0 1.6 3.2 ~.8 6.~ 8.0
Y AXIS IN CI1
Figure 5.1.4.c. Mean velocity profile for a
o
MEAN VELCCITY PRDFILE
o~---.-----.----.----'----'r----r----~---.----.----'
III
RUN PLVTI-K
"0 =0.177
X/O=28.8
EXPERJIoENT [!J
>-N
I-
......
U
o
...10
W O
>..:
~
O~~~----~--~ ____J -_ _ ~~ _ _~_ _ _ _L -__~____J-~~
-9.0 -7.2 -5.~ -3.6 -1.8 -0.0 1.8 3.6 5.4 7.2 9.0
Y AXIS IN CM
Figure 5.1.4.d. Mean velocity profile for a
-152-
a
MERN VELOCITY PROFILE
0r-.--~-----r----r----.-----r----'----'-----r----.-~~
lfl
RUN PLYTI-5
o
°
::r
Ro :0.17
Xl0233.6
EXPERlrENT [!J
U O
U.!~
(Den
"-
~
u
Zo
-0
~
O~ __ ~~~~ ____L __ _ ~ ____ ~ _ _ _ _L -_ _ ~ _ _ _ _- L_ _ ~~ __ ~
-12.0-9.6 -7.2 -ll.B -2.11 -0.0 2.11 1l.B 7.2 9.6 12.0
Y AXIS IN CM
Figure 5.1.4.e. Mean velocity profile for a
o MERN VELOC I TY PROF 1LE

0r----,----.-----,----,-----r----,----,-----r----.----,
I/')
I\UN PL VTI-H
o
o Ro =0.18
::r
X/O .. 39.1
EXPERlrENT [!J
U O
w~
(Den
"-
~
U
Zo
_0
>-N
-
I-
U
o
-.l0
W O
>-=
o
O~~~~ __J __ _ _ _L __ _-L____J __ _ _ _L __ _ ~ _ _ _ _ _ L_ _ ~~~~
-J2.0-9.6 -7.2 -ll.B -2.11 -0.0 2.11 1l.B 7.2 9.6 12.0
Y AXIS IN CM
Figure S.1.4.f. Mean velocity profile for a
o VEL~CITY HALF WIDTH.
C)
~~I----~--~----~--~----~--~----~--~~--~--~
RUN PLVTl
o Ro =0.17
~r- -
<D bu IO=0.093(X/O +5.87)
o
'" ::::10
.o~1-
''If)
--1
I
r
o
1--10
3:~r
m~~
LL:l'
.-l
a: I
IC) ~
Ln
.-l~1- - W
a:<n ,~., . I
Z -,/" " f!f
El
1-10
(f) 0 , _
~'
Z 'r-
WN
-
::E:
1--1
o
______ m~
o
Z~I- ---1
8 ....
Z
C)
oL I I I I
0.0 ~.O 8.0 12.0 16.0 20.0 2~.0 28.0 32.0 36.0 ~O.O
N~N DIMENSI~NAL DISTANCE FR~M EXIT, X/D
Figure 5.1.5. Non-dimensional velocity half-width of a turbulent
plume plotted against non-dimensional distance from
jet orifice.
-154-
--
Run R X/D K K
0 1u 2u
from to
1 2 3 4 5 6
P-7 0.0 14.0 37.0 0.087 6.7
PLVT-2 ,..,0.0 20.8 93.7 0.109 -2.5
PLVT-3 0.002 33.7 94.0 0.106 -3.1
P-4 0.015 14.0 37.0 0.087 5.6
P-2 0.10 14.0 24.0 0.09 3.8
PLVT-1 0.17 9.0 39.1 0.093 5.87
P-1 0.31 12.5 43.0 0.095 4.85
P-5 0.40 14.0 37.2 0.104 1. 78
P-6 0.70 14.0 37.2 0.093 4.20
Table 5.1.1. Linear approximation of the growth of the

velocity half-width.
-155-
case of a plume, Figure 5.1.6. The Gaussian curves in Figures 5.1.3
and 5.1.6 are defined by
~(x,y)
(5.1.3)
~(x)
where Xl = x - x ou
The kinematic mass flux ~(x) and the kinematic momentum flux m(x)
were calculated for each profile using the Gaussian curve given by
Eq. 5.1.1 as a reasonable approximation to the measurements so that
(5.1.4)
m(x)
_rrr-
"2Tn2 -2
b u (x) l1}1(x) • (5.1.5)
The kinematic buoyancy flux Sex) was calculated from the relation
sex) = agHo fpc p (5.1.6)
where the thermal expansion coefficient a is calculated from Eq. 2.2.5
for a temperature T +
a
Tc where H fpc ~(x),
Tc=o p
i.e., Tc is the mean
cross sectional temperature. The two non-dimensional numbers, the jet
angle
c = ~(x)f[m(x)(x-xou )]1/2
and the Richardson number
were calculated for each cross section. The basic experimental results
o NON DIMENSIONAL
. MEAN VELOCITY PROFILE
...
~~i- -_ _----~--~----~--~----~--~--~----"~--~
~~ ~ +
RUN PLVT-l
Ro =0.17
X'=X+S.9
:2
~
0
to
0
. EXP( -80.1 7] )
I:::J EXPERIM. RESULTS
.........
X;O SYMBOL
I:::J 0
<.D
9.0 [!]
0 f!) 1
1'l.0 i-'
19.0 A VI
Q'\
23.5 + 1
~~ 1" \ 28.8
33.6
39.1
X
~
..,.
o
C'\I
;L -O.SO-O.ijO -0.30
=
-0.20 -0.10 -0.00
7] = YIX'
0.10
'e~
0.20 0.30 O.ijO 0.50
Figure 5.1.6. Non-dimensional mean velocity p,ofile for a turbulent plume.

-157-
from each cross section, i. e. , Q.(x), b (x), ~(x)/~ , m(x)/m , S(x)/S , C,

M U 0 0 0
It can be observed from Table 5.1.2 that tne t;.i.C:~2.Tdsol1 numte'-
R(x), which is identically zero for a pure jet, tends to a constant
value R for an almost pure plume. The mean value, suggested for runs
p
with R > 0.15 is R 0.63 with a standard deviation 0.05. Hence it
o p
seems that the Richardson number of a pure plume is a constant, not
only in the case of conserved kinematic buoyancy flux (see Eq. 2.2.12),
but also in the case of a variable thermal expansion coefficient (not
conserved kinematic buoyancy flux). It can be observed also from
Table 5.1.2 that the non-dimensional number C (see Eq. 5.1.7) is a
constant for a wide range of initial Richardson numbers, from R =0

(pure jet) to R~ 0.63 (pure plume), with a mean value C = 0.54 and a
standard deviation 0.02.
5.2 ON THE CONSERVATION OF MOMENTUM AND ON THE DECAY OF THE
CENTERLINE VELOCITY OF A PURE JET
The kinematic momentum flux m(x) at a distance x from the exit,
normalized by the initial kinematic momentum flux m =

o
u02D (see Table
5.l.2),has been plotted versus the distance from the exit x/D in
Figure 5.2.1 for the run PLVT2 and P-7 (case of a pure jet). The
interesting conclusion is that the kinematic momentum flux is not con-
served in the case of a pure jet out of a wall, but it decreases as the
distance from the exit increases, in qualitative agreement with the
theoretical prediction made in Chapter 2.1 and given by the continuous
line of Figure 5.2.1. The same conclusion can be drawn from Figure
-158-
x - b \lex) m(x) sex)

uM u C R(x)
Run \lo m So
cm cm cm 0
sec
1 2 3 4 5 6 7 8 9
PLIIT2-E 5.00 .;>3 .00 C.S1S 2.390 0.ab8 0.943 0.580 0.001
PLVT2-tI 8.10 i5.:Ju C.6u0 ,0.919 0.tl49 1.000 0.55<; 0.000
PlVTZ-l d.9u ,,4.50 ~.d5U 2.<;35 u.oub 0.009 0.549 0.1l04
PlVf2-C 13.0J 19.00 1.300 3.4 :'0 u.7~O 1.000 0.557 0.000
i>lVT2-ri 14.10 10.00 1.430 3.b75 0.753 0.908 0.5bl 0.i.JOb
PLVT2-i) 17.80 ;t.20 1.700 3.5b2 O.t"b 1.0011 0.542 O.OUl
PlVTZ-A 22.50 1..> .00 2.400 4.4vll U.6,+3 0.655 0.572 0.OJ4
P-7A 14.0U 1".:'0 1.800 2.2Zj U.9.L 1.0JO a.512 U.UOO
P-78 37.bU b.9a 3.7bO 3.052 u. ",,2 1.000 0.~06 u.OOO
Pl \iD-a b.l0 o.la 0.&40 3.177 0.9~7 o .8i5 0.586 0.U58
PlVf3-A ,,2.50 5." 5 Z.5CO 6.595 1.J8b 0.7dl O.SOb u.189
P-2B 14.00 6.30 1.0011 L.d39 i. 672 O. b05 0.52u U.347
P-ZA 24.00 " .50 2.~00 3. b 72 .... ':11 0.732 0.5211 0.517
P-3A 24.00 7.90 2.900 3.2L8 1.1<;3 0.731 0.603 0.488
P-48 14.00 .0.30 i.700 2.46" 1.1<;1l 1l.8'>6 a.51L 0.11U
P-4A 24.00 7.40 2.620 2.733 1l.9,+7 0.021 0.517 0.308
P-'.c ~7.00 7.:50 3.700 3.912 1..>74 0.817 0.512 0.271
PlVTl-E 9.00 4.25 1.500 2.9tU 1. ·no 0.017 0.55b O. '082
PL\iTl-D 14.00 3.0:' 2.000 3.b03 ".1~5 0.793 0.550 0.6'>0
PLVTl-l 19.UO 3.90 2.30c) 4.197 2.54.3 0.770 0.532 0.581>
Pl VTl-C 23.50 3.dO 2.100 4.801 2.0.34 0.16b 0.5.30 0.625
PLvTl-K 28.80 3.80 3.200 5.690 3.3~9 0.75'0 0.53(; 0.646
PlVTl-8 3J.bv J.110 3.b50 b.490 .j.d~~ 0.744 0.530 0.614
PLVTl-H 39.10 3.75 3.900 6.84j 3.987 O.l'+b 0.513 0.686
P-lO 1.2.5U 4.b5 1. b50 3.000 2.617 0.756 0.53~ 0.019
P-IE 24.50 4.05 2.550 5.573 4.04" 0.719 0.512 0.5b9
P-ll 32.00 4.55 3.500 1.484 5 • .314 0.701 0.535 0.613
P-IB 36.0U '0.55 3.950 8.44b 5.997 0.695 C.527 0.608
P-IA 43.00 4.50 4.550 9.622 0.7,7 0.6tl9 0.535 0.623
P-~B 1 ... 00 0.00 1.050 3.4;;9 2.4~ '1 0.671 O,5bl v.777
P-5A 24.00 6.LC 2.800 6.119 '0.44') U.50d 0.572 0.6b3
P-'iD 32.00 b.40 3.bOO 8.i21 b.Jd3 0.584 0.561 0.5'01
P-5C 37.20 6.15 'o.20U 9.104 0.553 0.578 0.57V 0.b08
P-6B 14.0U 5.bO 1.bOO 4.211 ,j.bOU 0.043 0.515 0.682
P-tA 2'0.00 5.80 2.500 b.814 b.l08 0.b02 0.517 0.~16
p-tC 37.20 5.50 3.850 S.'751 1>.541 0.575 0.529 0.032
CEil TER
CEf\TER
10.50
10.50
19.60
19.52 . 0
0
CENTER 11.80 19.52 a
CEr.TER 11.80 16.61 0
CEIITER 14.50 11.46 0
CEr.TER 14.50 16.98 I,)
CEt-Tl 11.50 ,:,.'00 0

C EH 1 11.50 15.28 0
(EIITl 8.00 11>.93 0
CENT! 8.00 30.'09 a
CENT! 14.40 22.91 0
CEl\Tl 14.40 11.68 0
CENTl 14.41,) 14.'08 o·
CHIT! 14.40 a.78 0
CENU 14.40 8.11> 0
CEH1 11.90 25.02 0
CENTl 12.00 25.40 0
CEH2 16.70 21.10 0
0
C=NT2
UIIT2
16.70
10.10
16.05
1b.22 I 0
CENT2 16.70 7.92 0
CHT2 19.00 7.'00 0
CENTZ 19.0U 19.59 0
CENT2 19.00 19.'01 0
C ENT2 17. au 20.51 0
Table 5.1.2. Tabulated experimental values for volume,

kinematic momentum and buoyancy fluxes for
turbulent jets and plumes.
a
M~MENTUM ~F A 20 JET ~UT ~F A WALL
o
Nr I
Ro =0
RUN SYHBClL
PLVT2 .D
aLf) o
P-7
.....
a
o
o
....
.
Section 2.1
E @
I
I-'
'- I!ll!l
\.Jl
~
)(
I!l I
I!l
E a
u: '--
o
~ll 0.0 20.0

I·
ijO.O 60.0 80.0 100.0 120.0
1 ,
140.0 160.0
N~N DIMENSI~NAL DISTANCE FR~MJET ~RIGIN,X/D
.Figure 5.2.1. Experimental values for the kinematic momentum flux m(x)
of a turbulent jet (R = 0) as a function of distance
from jet orifice. 0
-160-
5.2.2, where the centerline velocity is plotted versus the distance
from the exit x/D. If the kinematic momentum flux is really conserved,
then the decay of the centerline velocity should be given by the
relation
u \ 2
(- ~ , I (5.2.1)
\ '\I~x) /
which is represented by the dotted line plotted in Figure 5.2.2. It can
be seen that all the experimental points are well above the dotted line.
This means that the centerline velocity decays faster than is predicted
from the conservation of the kinematic momentum flux, i.e., the
kinematic momentum flux is not conserved. The straight line which fits
the experimental data (Figure 5.2.2) has the form
(5.2.2)
where C ' C are the coefficients given in Table 5.2.1. A discussion

lu 2u
of Eq. 5.2.2 will be given in Chapter 8.
Using the relation Eq. 5.1.5, it can be shown that
m(x) = J_TT_ (5.2.3)

m 12,Q,n2
o
or for x/D much greater than K ' C

2u 2u
(5.2.4)
a
CENTERLINE VELOCITY DECAY
a~~F------~------~~--~------~------~--'--~
~
RUN PLVT2
a RICHAROS~N NUMB.=O.OOOOS
a
Ln
N
· (UO/UM)2/0=0 .21ij( X/O-S.O)
I!]
G)
a
a
N
·
/'
",.
a ",.
0
Ln · ",.
..".
I
N
.---.
><
-- ",.
,/'
,/' I--'
(j\
I--'
I
0 ,/
0
:E
· /'
I::J
"-
I::J
~
0 -
0
,/'
",./' /' ' \ For Conservation of

0 /
',/' Momentum (Eq. 5.2.1)
0
Ln · ....;:.... , /
,/'
,/
0
a ·
0.0 20.0 40.0 60.0 80.0 100.0 120.0
N~N OIMENSI~NAL DISTANCE FR~M EXIT, x/a
Figure 5.2.2. Velocity distribution along jet axis (R 0).
o
-162-
The results obtained by several previous investigators are given in
Table 5.2.1. An explanation of the data of the columns is as follows:
Column 1 gives the name of the investigator while the number in
parentheses refers to the reference number.
Column 2 gives the geometrical configuration of the jet, i.e., out of
a wall or without a wall (see Figure 5.2.4).
Column 3 gives the distance of the closest measurement to the exit,
9.,. (see Figure 5.2.4).

m~n
Column 4 gives the distance of the most distant (from the exit)
measurement, 9., (see Figure 5.2.4).

max
Columns 5 and 6 give the coefficients K ' K2u (see Eq. 5.1.2).
lu
Columns 7 and 8 give the coefficients C ' C2u (see Eq. 5.2.2).
lu
Column 9 gives the ratio Eq. 5.2.3 for x = 9.,may. •
(5.2.5)
Column 10 gives the ratio m(x)/m for ~ » 1, i.e., Eq. 5.2.4.

o
A careful examination of the ratio m(x)/m (Table 5.2.1) for the
o
experimental results of many experimenters who investigated a pure
two-dimensional jet out of a wall confirms the failure of the conserva-
tion of momentum argument. It is believed that the reason why the
coefficients C ' C ' K ' K and the ratio m(x)/m varies consider-
lu 2u lu 2u o
ably from investigator to investigator, is the non-self-preserving
character of the two-dimensional pure jet, as will be clarified in

£ . £ m(£ ) I
Experimenter Geometry m1n max K K C C max m(x)
of the Jet 1u 2u 1u 2u
D D m m
0 0
1 2 3 4 5 6 7 8 9 10
Miller et a1 Out of a 1 40 0.096 1.57 0.181 -0.62 0.83 0.79 i
[12] wall 0.163 1.05 0.89 0.88
Heskestad » 47 155 0.110 -4.2 0.27 -10.0 0.63 0.61
[5]
Go1dshmidt » 15 67 0.099 0.66 0.207 -5.8 0.80 0.72
[10]
Jenkins-
Go1dshmidt » 10 60 0.088 4.5 0.16 4.0 0.83 0.83
[8]
Zijnen » 10 40 0.096 0.6 0.162 0.6 0.89 0.89 ,
I
[38] 10 35 0.099 1.7 0.205 1.7 0.72 0.72 I
f-'
0'>
Forthmann » 0 25 0.096 0.50 0.18 0.11 0.88 0.80 W
I
[3]
Nakaguchi » 10 100 0.106 -0.3 0.189 -0.3 0.84 0.84
[37],[59]
Knystautas » 50 295 0.106 -1.5 0.18 -1.4 0.88 0.88
[7]
Householder » ? ? 0.091 1.46 0.193 -6.98 0.70(?) 0.70
[44]
Bicknell No wall 15 126 0.115 -2.70 0.112 6.70 1.45 1.53
[57]
Present Study Out of a 14 37 0.087 6.7 0.17 1.72 0.87 0.77
Run P-7 wall
Present Study » 20 94 0.109 -2.5 0.214 -5.0 0.77 0.76
Run PLVT2
Table 5.2.1. The kinematic momentum flux of a plane pure jet.
0
=:
C\J
1----1 1'
c
C\J
f- INVESTIGATOR
Pf'£:3(HT STUll
GEOMETRY SYMBOL
!!l
~ Pft:5ErIT STUll (!)
.D.\(INS .6
t£Sl(ESTffi +
g~ FI'lf\THfoR.I OUT OF A WALL. X
c-: GO..D50t11OT ¢
HILli1I ...
HIU£R x
gL •
Z!HH
ZIJI£H
Z
y
1fi\RG(Q11 )(
610<hUl. I() I4U. II!
0
~
- S
~S
o
,......
~
til
o
::1'
0
• I
I--'
0\
~
N
'-' I
S
o
o
o
co
o
~l(el
y
...
..
. f!)
I!' Section 2.\
o
w +
o
o
::J'
c:iL 1 1 1 1 \ \ 1 \
0.0 20.0 ~O.O 60.0 ~O.O 100.0 120.0 l~O.O 160.0 180.0 200.0
~ maxI 0
Figure 5.2.3. Kinematic momentum flux m(x) of a two-dimensional
turbulent jet (R = 0) as a function of distance
t /D and jet g~ometry.
max
-165-
Chapter 8. The Figure 5.2.3 demonstrates essentially that the momentum
ratio m(£ )/m from different investigators is correlated with £ /D

max 0 max
(where £ is the most distant measurement of the centerline velocity
max
from the jet exit). In fact, it can be seen that m(x) /m decreases as x
o
increases.
It is interesting to notice that Bicknell (see Table 5.2.1), who
investigated a pure two-dimensional jet without a wall, obtained the
result that the ratio m(x)/m is greater than one, i.e., the momentum
o
flux in the jet increased. This agrees with the theory proposed in
Section 2.1 because the momentum of the induced flow is now in the same
direction with the input momentum, see Figure 5.2.4.
JET OUT OF A WALL JET SPACE

m(x)<mo
m(x»mo
Figure 5.2.4 Schematic of the induced flow by a two-dimensional

pure jet.
It can be observed from Figures 5.2.1 and 5.2.3 that the solution
of the mathematical model developed in Chapter 2.1 underestimates the
decay of the kinematic momentum flux. A definite explanation for this
is not presently available but one possible cause is that the

-166-
theoretical prediction is based on a first approximation with external
streamlines meeting the axis of the jet at ¢ = rr/4 for all x. A
second order approximation should find that the angle ¢ should increase
as x increases. This can be observed in the interesting picture of
Lippisch ([29], his Figure 22). When ¢ increases the rate of decrease
of the kinematic momentum flux increases, as is obvious from Eq. 2.1.16.
5.3 CENTERLINE VELOCITY OF A PLUME
It was shown in Chapter 2.2 that for a pure plume the centerline
mean velocity ~ is related to the kinematic buoyancy flux S(x) by
= (] S(x)1/3 (5.3.1)
u
where (] is an experimental constant. The kinematic buoyancy flux S(x)

u
is given by
(5.3.2)
where the thermal expansion coefficient a(x) is calculated from the
algebraic relation Eq. 2.2.5 at a temperature
e T
a
+ Tc
and where
H
o
T pc 11 (x) •
c
o p
T can be considered as a cross sectional mean temperature.

c
The basic experimental results for the centerline mean velocity
-167-
~ (cm/sec), and the kinematic buoyancy flux S(x)/So (where
a gH /p c ) are given in Table 5.1.2. The value of 0 calculated

o 0 0 p u
from the values of ~ and Sex) given in Table 5.1.2 is plotted in
m
Figure 5.3.1. o was found to be constant with the value
u
o 0.04 (5.3.3)
u
5.4 THE FREQUENCY DISTRIBUTION OF THE VELOCITY
The frequency distribution of the velocity was generated by
taking many sweeps of the spectrum analyser using the process described
in Chapter 3.2, part III. Two representative cases of the frequency
distribution of the velocity are plotted in Figure 5.4.1 (plume) and
Figure 5.4.2 (jet).
The occurrence of "negative" velocities (i. e., velocities with
direction opposite to the jet flow) was observed on the oscilloscope
of the spectrum analyser while an estimation of their probability can
be obtained from the figures. There are two possible mechanisms for
the generation of such negative velocities. First, large vortices
produced at the edge of the jet near the exit could contain velocities
larger than the advection velocity of the vortex. Second, the external
induced flow has a reverse component as can be seen in the flow
visualization experiments by Lippisch [29].
5.5 TURBULENCE INTENSITY
An estimation of the turbulent intensity of the velocity fluctua-
tions u ' 2 can be obtained from the frequency distribution of the

PLUME -CENTERLINE VEL~CITY
~I-r~~--r-T---r-~
o
LJ)
~I
OJ
A
[!)
r!Jt.C!J C!>-t!J-i(!)
(!)~ t!l (!)
.....
0 I!l
LJ)
r<'l
RUN SYMBOL
rLVT-l (!J
>< r-l (!)
.....
0
0
CQ r-5 6
I
.......... r-6 + t-'
0\
00
>< I
E
I;::) 0
LJ)
.1-
0
~I I I I' I
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 1.,10.0 l,I5.CI 50.0 55.0 60.0
NDN DIMENSIDNAL DISTANCE FRQM EXIT ,;(/0
Figure 5. 3 .1. Velocity distribution along plume axis.
0.2 r, -,----.-----,.---,~r__,_-_,__-I-I·-I--
RUN PLVT IZ
RUN PLVT IZ
Ro' 0.17
0.20 YI X', 0.091
Y/X', 0.01
>- >-
~ Ro' 017
~
(/)
(/)
z z
w 015
0 w
0
>- 0.1
~
>-
>-
--' --' 010
m m
<t <I
m m
0 0
0: 0:
D.- D.- 005
0.0 0.0
-I 0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 -1.0 0.0 1.0 2.0 3.0 40 50 60 70
VELOCITY emlsec VELOCITY emlsee
I
0.5 07 f-'
cr-
\0
RUN PLVTIZ PLVT IZ I
0.6
04 Ro' 0.17
>-
YI X', 0.133 >-
>- (/) 05
~
(/)
z
z 0.3- w
0
w 04
0
>-
>- >-
~
--' 0.3
J 0.2 m
ill <I
<t m
m 0 0.2
0 0:
0: D.-
D.-
0.1
0.1
OD 00
-1.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0
VELOCITY em/sec VELOCITY em I sec
Figure 5.4.1. Probability density of the velocity for a turbulent plume as' a
function of distance from jet axis.
0.10 . . 0.06rl~----,---~r---,-----'---~----~----~
RUN PLVT 2-B

RUN PLVT 2-B
R. ' 0
0.08 R.' 0
>- 0045 Y/X',0046
>- YI X': 0.021
l- l-
v; v;
Z Z
~ 006 w
o
>- >- 003
I- I-
.oJ
:::! 0.04
CD
<l
CD
'"<l
o '"oa: 0.015
g: 0.02,- CL
OOL-
0.0
-~~~~~~~
"'-1
10.0
~~~'>~;;~
20.0
__
30.0 400
____-L_______________ 0.0 I
0.0 10.0 200 300 400

,.,.
500 600
VELOCITY emlsee VELOCITY em/sec
I
0.08'-1 Ii 020, r-- f-'
"-l
a
RUN PLVT 2-B RUN PLVT2-B I
R.' 0 R. ' 0
::: 0.06 YI X', 0.088 >-
I- Y/X"0.154
(f)
z (f)
Z
w
o W
o
>-
I- 004 >- 01
I-
.oJ
.oJ
CD
<l '"<l
'"og: 0.02
'"oa:
CL
OO~,~~~~~__~__1-_______ 0.0 I ----.....

)(1 .....
0.0 10.0 200 300 40.0 -----.J
~,O 0
00 100 200
VELOCITY em Isee VELOCITY em/sec
Figure 5.4.2. Probability density of the velocity of a tur1bulent jet as

a function of distance from jet axis.
-171-
velocity. The turbulence intensity u l 2, normalized by the mean velocity
on the jet axis, is plotted against ylb (where b the velocity half-
u u
width) in Figure 5.5.1 for the case of a pure jet, and in Figure 5.5.2
for the case of a plume. It can be observed that the effect of the
buoyancy is-an increase in the relative turbulent intensity. Similar
results concerning the turbulent intensity of the temperature fluctua-
tions were described in Chapter 4.2.

I-----~~-
RUN P-78
Ro = 0
X/D = 37.0
0.4
I
......
E .......
N
I~ I
t; 0.2
0.0\'--__---1..-_ _ _ _...l-_ _ _---1-_ _ _--1

-2.0 -1.0 0.0 1.0 2.0
Y / bu
Figure 5.5.1. Profile of the intensity of turbulent fluctuations of the velocity
in a turbulent jet.
r ------- I .- r-~ --.---~ ---1
RUN P-IZ
Ro = .0.20
0.5
X/O=32.0
OAI- 0 0 0 0
0
0
0
I '1
0.3
E
oJ
p I
"'- / '\ 1 f-'

.......
LV
I
t;
0.1,-
O.OLI----,--~________L __ _ _ _ _ _~_ _ _ _ _ _ _ L _ _ _ _ _ _ _ ~_ _ _ _ _ _ _ _
-3.0 -2.0 -1.0 0.0 1.0 '2.0 3.0
Y / bu
Figure 5.5.2. Profile of the intensity of turbulent fluctuations of the
velocity in a turbulent plume.
-174-
6. VELOCITY-TEMPERATURE EXPERIMENTAL RESULTS
AND DISCuSSION
6.0 BASIC EXPERIMENTAL PARAMETERS-INITIAL DATA
The velocity and the temperature were measured simultaneously
using the instrumental set-up described in Chapter 3.4 (Figures
3.4.4.a-c). Complete information concerning the basic experimental
parameters is given in Table 6.0.1.
TABLE 6.0.1 Basic parameters of velocity-

temperature measurement
-u -
T T
Run D Aspect 0 0 a R R
e 0
cm ratio em/sec °c °c
PLVT2-E 0.24 54.2 62.93 3.45 21. 6 1680 0.00005
PLVT2-Z 0.24 54.2 62.0 5.40 21. 8 1780 0.00009
PLVT2-H 0.24 54.2 62.93 3.80 22.2 1760 0.00006
PLVT2-A 0.24 54.2 62.63 6.85 23.25 1900 0.0001

I
PLVT3-B 0.24 54.2 64.20 9.20 23.2 593 0.0019
PLVT3-A 0.24 54.2 62.93 10.40 24.0 620 0.0022
PLVTl-E 1.0 13.0 19.0 11.0 22.0 610 0.17
PLVTl-D 1.0 13.0 19.0 11.15 22.0 610 0.17
PLVTl-Z 1.0 13.0 4.55 11.0 22.0 610 0.17

I
PLVTl-C 1.0 13.0 4.55 10.90 22.0 610 0.17
PLVTl-K 1.0 13.0 4.55 11.15 23.0 620 0.17
PLVTl-B 1.0 13.0 4.55 11.0 22.0 610 0.17
PLVTl-H 1.0 13.0 4.55 11.40 23.0 630 I 0.17

1
-175-
6.1 MEAN VELOCITY AND MEAN TEMPERATURE PROFILE
The non-dimensional velocity and temperature profiles and the half-
widths b and b are plotted in Figures 6.1.1 and 6.1.2 for a pure jet,
u T
and in Figures 6.1.3 and 6.1.4 for a plume. The basic experimental
results from each temperature or velocity profile were given previously
in Tables 4.1.1, 4.1.2 and Tables 5.1.1, 5.1.2. In Table 6.1.1 the
ratio of the temperature half-width b

T
to the velocity half-width b
u'
and the ratio KIT/K ' are given for each profile. The interesting
lu
conclusion that can be drawn from Figures 6.1.1, 6.1.3 and Table 6.1.1,
is that the temperature profile is wider than the velocity profile for
a wide range of Richardson numbers (R,-oJ O,-oJ 0.6). Thus the suggest ion
of Rouse et al [21], that the velocity profile of a pure plume is wider
than the temperature profile, is not supported by this investigation.
At the present moment, it is not clear what is the mechanism which makes
the mean temperature profile wider than the mean velocity profile. The
statement which is used by some investigators (see Hinze [34], p.426)
that "the spread of heat is much greater than the momentum" should be
avoided because the terms "spread of heat" and "spread of momentum" are
not well defined for a free turbulent flow. As a matter of fact, a
particular mass of cold irrotational fluid without any axial momentum
that is engulfed by the jet acquires axial momentum much faster than it
is heated. The large vortices which are produced at the edge of the jet
contain velocities larger and of opposite direction than the advection
velocity of the vortex and may be partially responsible for narrowing the
velocity profile.
The heat transport by the mean flow, ~, can be calculated

-176-
0
NON DIMENSIONRL MERN VELOCITY PROFILE
'"
~i
~~
RUN PLVT2
.. R. =0 i
i?~ X'=X-2.5 i
0 r ~
EXP( -58.37) 2 ) _Ji
~ EXPERIM. RESULTS
CJ
I!
I
X/O SYMBOL
Lt
l::l 0
~
20.6
33.3
[!]
(!)
~
CJ
37.5
S4.2
"+
56.75 X I
J 74.5
93.75
(>
• I
"':
CJ
l
CJ
"!
0
.
I!J
~
~
OL-__ ~ ____ ~=WL-~ ____- L____ ____- L____- L__
~ ~~ __ ~~~~
-0.50-0.40 -0.30 -0.20 -0.10 -0.00 0.10 0.20 0.50

7)=Y/X'
Figure 6.l.l.a. Non-dimensional mean velocity profile

in a turbulent jet (R = 0).
o
o
NON DIMENSIONRL MEAN TEMPERRTURE PROFILE
'" , . . - - - - - - - , r - - - - - - - , - - - - - , - - - - - - - r - - - i - - --.-----,------~Ir-_____r--I
o
RUN PLVT2 I
o
RICH. NUMB.=O.OOOO-1
X'=X-I.56 Ii
.
+ EXP( -257)2)
o
<Xl
CJ
EXPERIM. RESUL~ ~II
I
1
X/O SYMBOL
+
20.8 [!]
37.5
58.75 (!)
...
93.75 +
+
~r +
---l
~lJ·,
I!J
~
,
-j
I
-0.50-0.40 -0.30 -0.20 -0.10

I
-0.00 0.10 0.20 0.30 0.40 0.50
I
7)=Y/X'
Figure 6.l.l.b. Non-dimensional mean temperature

profile in a turbulent jet (R 0).
o
-177-
o
YEL~CITY HRLF ~IDTH
0 , -_ _- ._ _ _ _. -_ _ _ _. -_ _- ._ _ _ _- ._ _ _ _. -_ _- ._ _ _ _- ._ _ _ _. -_ _- .
~
RUN PLVT2
R" e!0
b u 10=0. lO9( X/0-2 .5)
o
'-
,,0
.o~
oN
I-
f-
o
3:0
o
LL";
-1
a:
I
-10
a:O
Z·
0'"
......
Cf)
Z
Wo
~~
0""
Z
o
Zo
O~ _ __ L_ _ _ _J -_ _ _ _L -_ _- J_ _ _ _- L_ _ _ _ ~ __ ~ -L____
____ ~ __ ~
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 60.0 90.0 100.0
N~N DIMENSIQNAL DISTANCE FRQM EXIT. X/D
Figure 6.1. 2 . a . Velocity half-width of a jet (R ~ 0)
as a function of distance alongOthe jet.
o
TEMPERRTURE HRLF ~IOTH
0r----.----.-----.----.-----,----.----,-----r----.---~
~
RUN PLVT2
RICHRROSON NUMB.=O.OOOOS
b T 10=0.1670/0 -1.56)
>-
.0
·0
IO
f-'
o~
~O~
-10
cr:
I'"
.
-1
~~I
({)",r- -i
I
Z
W
I
,
I
:L
~
!
I I
D~L
/
/ i
~I~i ~~ ~ ~I
oz"'l,
Z ,
____ L -_ _- J_ _ _ _- L1 ____ __ ____-LI____ ____IL___-LI__

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 60.0 90.0 100.0
NON DIMENSIONRL DISTRNCE FROM EXIT. XID
Figure 6.1. 2. b • Temperature half-width of a jet (R ~ 0)
as a function of distance along thg jet.
-178-
o NClN 0 I MENS I ClNRL MERN VELLlC I TY PRClF I LE

N.-----,-----,------r____- .____- .____- , r -____. -____. -____. -____~
0 RUN PLVT-I
c:::
Ro =0.17
X·=Xt5.9
0
co EXP( -BO.I TJ 2 )
0
~ EXPERIM. RESULTS
1<
1;:1
X/O SYMBOL
0
~ 9.0 [!J
0 I!)
14.0
19.0 AI>
0
23.S +
26.6 X
~
33.6 (>
0
39.1 ...
o
N
c:::
O~ ____ ~ ____ ~ __ ~-L ____- L____ ~ ____ ~L_ _ _ _ _~~_ _~----~----~
-0.50-0.40 -0.30 -0.20 -0.10 -0.00 0.10 0.20 0.30 0.40 0.50
TJ = YIX •
Figure 6.l.3.a. Non-dimensional mean velocity profile
in a turbulent plume.
OIMENSILlNRL MERN TEMPERRTURE PRLlFILE
If-
0
ex:
0
~o
u:
x\ EXPERIM. RESULTS
X/O
9.0
SYMBOL
[!J
'-0 14.0 I!)
If-
x.
19.0
23.S ""
+
:,
0
•
..
26.6
33.6
39.1
x
(>
...
I \
~ "~" ~
0
(!)
~
0
•I!l
~~ ~ ~. ~ ~~ ~~ ~I
-
OL-__- L__ __ __- J_ _ _ _ __ ___ __ _ __LI__
-0.50-0.40 -0.30 -0.20 -0.10 -0.00 0.10 0.20 0.30 0.40 0.50
TJ=YIX'
Figure 6.l.3.b. Non-dimensional mean temperature profile

in a turbulent plume.
-179-
a
VEL~CJTY HRLF ~JDTH
a
.... r - - - ' - - - ' - - ' - - - ' - - - , - - - ' - - T
RUN PLVTl
a Ro =0.17
a
<D b u /O=0.093(X/O +5.87)
a
"~8
I~
f-
a
~a
3:,,?
LL:::l'
.....J III
a:
Ia
.....J
a•
a:'"
Z
o
a
a
za
o
z~
·
a
aL-_~ _ _-'---_ _-L-_ _L-_---'-_ _-'---~_~~~~_~~_~
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 40.0
NON DIMENSIONAL DISTANCE FROM EXIT.
Figure 6.1.4.a. Velocity half-width of a plume as
a function of distance along the plume.
a
TEMPERRTURE HRLF ~JDTH
....ar---.---r--.---.---.---r---~-~~-~--~
a
a RUN PLVTl
a
(!)
RJCHAROS~N NUMB.=0.17
" br IO=0.12(X/O +2.1)
"-0
.co
I~
f-
a
~o
3:0
LL:::l'
.....J
a:
Ia
.....J"? ~
a:'"
z
o !!l
§2~
WN
~gl
0""::
Z I
i
aa L I_ _ --LI_ _-'---_ _-L-_ _L-_---'-I_ _-'---_ _-:'-I_ _ _LI_ _-:,-1_ _-,
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0
Figure 6.l.4.b. Temperature half-width of a plume as
a function of distance along the plume.
-180-
b KIT HT
X T - -H
Run R(X)
D b K
u 1u 0
PLVT2-E 0.001 20.8 1.54 1. 53 0.06
PLVT2-Z 0.004 37.0 1.65 1. 53 0.06
PLVT2-H 0.006 58.7 1.43 1.53 0.08
PLVT2-A 0.034 93.7 1. 38 1.53 0.06
PLVT3-B 0.058 33.7 1. 38 1. 33 0.10
PLVT3-A 0.189 93.7 1.24 1. 33 0.12
PLVTl-E 0.482 9.0 1.13 1.29 0.21
PLVT1-D 0.64 14.0 1.17 1. 29 0.24
PLVT1-Z 0.59 19.0 1.10 1.29 0.38
PLVT1-C 0.62 23.5 1. 20 1. 29 0.32
PLVT1-K 0.64 28.8 1.17 1. 29 0.44
PLVT1-B 0.61 33.6 1.20 1.29 0.41
PLVT1-H 0.68 39.1 1.27 1. 29 0.42
Table 6.1.1. Ve1ocity·temperature basic results.

-181-
for each measured velocity and temperature profile by using the
Gaussian curves Eq. (5.1.1) and Eq. (4.1.1) as a reasonable fit to the
experimental results. In this case
~ = pc IB(X)~(x,Y)T(x,y)dY
P -B(x)
- - 1/2
pc Q.(x) TM(x) b (x)(IT)
p .M. u
(6.1.1)
Hence the cross sectional heat transport by the turbulence is
pc IB(x) u'T'dy (6.1.2)

p -B(x)
where H the conserved input heat flux. The local Richardson number
o
R(x), see section 5.1, and the ratio HT(x)/H for each experiment are
o
given in Table 6.1.1. The interesting point brought out by these
results is that the fraction of heat transported by the turbulence ~,
increases as the local Richardson number increases. For a low local
Richardson number R(x) < 0.01 (jet)
IB(x) u'T'dy ~ 0.06 IB(x) uT dy (6.1.3)

-B(x) -B(x)
i.e., the turbulent heat transport is a negligible fraction of the
longitudinal heat transport for a jet. But for a local Richardson
number R(x) '" 0.6 (plume)

-182-
IB(x) u'T' ;:;;; 0.38 IB(x) uT

dy ,
-B(x) -B(x)
or
IB(x) u'T' dy;:;;; 0.6 IB(x) u T dy (6.1.4)
-B(x) -B(x)
Now, given significant distance to develop, any slightly heated jet
will eventually become a plume because the initial momentum will be a
small fraction of the momentum which will be gained by the continuous
action of the buoyancy forces. It was found in Section 2.3 that the
importance of the initial kinematic momentum flux relative to the
kinematic buoyancy flux is given by the length scale,
m /8 2 / 3
00'
which is infinite for a pure jet and zero for a pure plume. A heated jet
with low Richardson number at the jet orifice (R «0.6), becomes a

o
plume at a distance
x » m /S2/3
o 0
On the basis of this investigation transition to a plume is assured at
any point for which
2 3
xS o / /m0::: 10 •
The turbulent heat transport will increase with the distance x, or
the local Richardson number R(x) , and eventually for
xS2/3/ m » l
o 0
-183-
will be comparable with the heat transport by the mean motion.
Hence, the assumption made by previous investigators «21), (27),
(24), (20» concerning a pure plume, that
u'T' « u T
has no physical basis, and the turbulent heat transfer (and conse-
quently the turbulent kinematic buoyancy flux) should not be neglected
when a plume is investigated. It should be noted that for a pure plume
the transport of heat by the mean flow ~ can be calculated using the
expressions
1.66 S(x)1/3,
for the mean velocity, (see Eq. 5.3.3),
for the mean temperature, (see Eq. 4.1.5), to give
~ = pc JB(x) u(x,y) T(x,y)dy

p -B(x)
pc
p
~TM(x - x oT ) Joo exp [-£n2 (1 +
_00
K~T)( K1Tx
Klu
Y ,)2] d(~)
(where x' = x - x oT )' then using K1T/K ~ 1. 3, K1T 0.12 (see

lu
Tables 6.1.1 and 4.1.1)
~ = 0.62 pCp S(x)/a(x)g = 0.62 Ho • (6.1.5)

-184-
The ratio HT/Ho' given in Table 6.1.1, is plotted in Figure 6.1.5.
Assuming HT = ~(R)Ho' expanding ~(R) = ~o + ~l R + 0(R 2), neglecting

the terms 0(R 2), and considering that for a pure jet ~(O) = 0.06, and
for a pure plume ~(R ) ~ 0.38, the turbulent heat transfer as a function
p
of the local Richardson number R(x) is obtained as
(0.06 + "U • .:J.l.

,.., R(x» ,. c. , c. \
H \U • .l.. V)
o
Hence, the turbulent heat transfer HT normalized by the initial heat
flux H can be represented by the following empirical function of the

o
local Richardson number
H /H = 0.06 + 0.51 R(x) • (6.1. 7)

T 0
which is plotted in Figure 6.1.6 together with experimental results.
It follows also from Eq. 6.1.5 that for a plume
SB(X) u'T' dy ~ 0.6 SB(X) u T dy •

-B(x) -B(x)
6.2 PROFILES OF uT AND ~
The purpose of these measurements was to find the profile of the
advective heat flux
u(x,y,t) T(x,y,t)
and then to derive the profile of the turbulent heat flux from the
relation
RUN Ro SYMBOL
PLVT 2 10- 5 0
1.0 PLVT 3 10- 3 A
-
o PLVT I 0.17 0
I~
_._-------_.-
-
I::J
.......
..---.
>-
0.8
'U
I~ 0.6
~
L..----.I I
II t-'
ex:>
o VI
I o (value for Plume (Eq. 6.1.5) I
....... 0.4 o
o 0
t-
I
"
,,"" 0
0.2
0 ....... ,-0'
------;;~~I~--:J:-~
~
.....
0.0 I b l':. o
0.0 01 0
20.0 40.0 60.0 80.0
x/a
Figure 6.1.5. Turbulent heat flux in a buoyant jet normalized by thE~ heat flux at
the jet orifice, as a function of distance along the jet.
RUN SYMBOL
o PLVT 2 0
I~o PLVT 3 6.
,:J PLVT I 0
1.0
........
..----.
>-
"'0
r~
~ ,
t-'
L...-....J (Xl
"o ,
(J'\
0.1 6.
I
........ o
~ r-------·----~O 0 -
I Eq.6.1.7
0.01, ""' 11111 11111 11111 11111

10- 5 10- 4 10- 3 10- 2 10- 1 10°
R( X )
Figure 6.1.6. Turbulent heat flux in a buoyant jet, normalized by the heat flux
at the jet orifice, as a function of the local Richardson number.
-187-
u' (x,y,t)T' (x,y,t) u(x,y,t)T(x,y,t) - ~(x,y)T(x,y) (6.2.1)
Ideally one would like to measure
R(O,O) Lim 1:.. Jt u(x,y,t!)T(x,y,t)dt , (6.2.2)

t-+<x> t 0
without disturbing the flow (without inserting probes), but this has
not yet become possible. In this investigation the velocity was
measured at the focal volume (x,y) of two focused beams of a laser
Doppler velocimeter, while the temperature was measured with a small
(0.4 mm diameter) fast response thermistor located at a point (x+~x,y).
The distance ~x was chosen to be 1-2 mm, the criterion being that the
thermistor bead should be close to the focal volume of the laser beams,
but not so close as to block the laser light or to disturb the flow.
Hence the correlation
R(~,O ) u(x, y, t)T(x+~x, y, t)
Lim t1 Jt u(x,y, t~T(x+~x,y, t~dt' , (6.2.3)

t-+<x> 0
was measured.
The interesting question is how the correlations R(O,O) and
R(~x,O) are related. To the best knowledge of this writer, the
correlation R(~x,O) as a function of ~x has not been examined in a
buoyant jet. For a plume the instantaneous velocity u(x,y,t) at a
point (x,y) should depend upon the instantaneous buoyancy forces at
(x,y) and since the instantaneous buoyancy forces at (x,y) depend upon
-188-
the instantaneous temperature T(x,y,t), it follows that the velocity
u(x,y,t) and the temperature T(x,y,t) should be very well correlated.
Hence the correlation
R(O,O) Lim l Jt u(x,y,t)T(x,y,t)dt

t-+<:o t 0
in a pure plume should be larger than the correlation
R(~x,O) = Lim l Jt u(x,y,t)T(x+~x,y,t)dt ,

t-+o t 0
because the signals u(x,y,t) and T(x + ~x,y,t) are not necessarily
well correlated, and
T(x + ~x,y) ~ T(x,y)
(strictly T(x + ~x,y) < T(x,y».
For a pure jet, where the heat behaves as a passive tracer, the
fluctuations of the signals of the velocity u(x,y,t) and of the
temperature T(x,y,t) should not exhibit any strong correlation, as the
experimental results of Corrsin and Uberoi [52] have demonstrated. The
schematic of Figure 6.2.I.a-b is expected for the correlation of the
fluctuation u', T' for a plume and a jet respectively.

u'(x,y,t) u'(x,y,t)
----r.-~~--~ TI (X, y. t) ----f-'-:-:-+---'-~-+-. T 1(X, y, t )
Figure Schematic of the fluctuations of the temperature

T'(x,y,t) and vertical velocity u'(x,y,t); (a) plume
(b) jet.
-189-
Hence it is expected that for a plume R(~x,O) < R(O,O), but for a jet
R(llx , 0) "" R(0 , 0) •
A direct measurement of R(O,O) and comparison with the R(llx,O) was
not possible in this investigation. However, a reasonable understanding
can be obtained by calculating the integral of R(llx,O) across the jet
rB(x)
H - pc J R(l:.x,O)dy
c p -B(x)
and comparing it with the input heat flux H. The conservation of the
o
heat flux gives
H pc IB(x) R(O,O)dy , (6.2.5)

o p -B(x)
and any deviation from this relation should be attributed to experi-
mental errors (if radiation effects are absent). Since in general
R(llx,O) + R(O,O) it is expected that
H =1= H
c 0
It seems reasonable then to assume that
IB(x)
R(0 ,0) ~ y
_-B_(.,-x..,...)_R_(_O_,O_)_d_
(6.2.6)
R(l:.x,O) "" IB(x)
R(llx,O)dy
-B(x)
Keeping these preliminary results in mind, the experimental
results as they were obtained will be presented, some observations will
be made, and then the profile of u'T' will be derived. The profile of
-190-
the correlation pc R(~x,O) in ca1/sec is plotted in Figures 6.2.2.a-d

p
(jet) and 6.2.3 (plume). The half-width bUT was calculated for each
profile and a Gaussian curve given by
(6.2.7)
fitted to the data, where ~ is the value of R(~x,O) on the jet axis.
The half-widths buT(x) of the R(~x,O) profiles of a jet were plotted
in Figure 6.2.4 and a straight line
b -/D = K -(~ + K -)
uT luT D 2uT
was fitted. The values of K

1uT
= 0.105 and K2uT = -5.0 were found.
The profile of R(~x,O), normalized by ~ was plotted against y/x' in
Figure 6.2.5 and a Gaussian curve given by
R(~x,O)
(where x' = x + K D) was fitted. The integral of the correlation

2uT
R(~x,O) across the jet or plume can be calculated for each profile by
H pc B(x)
J-B(x) R(~x,O)dy ~ pc Joo
c p p -00
(6.2.8)
The ratio
H
c
H
o
-191-
o
R{6x,O) .HEAT FLUX PRClF ILE
0r----r----~--_,----_r----._--_.----._----r_--_r--__,
o
CD
o /IUH f'lVT2-E
o
o fl. ·O.CXXlOS
LIl
XlO-2O.S
(Xf'(RltENT I!l
H
c
H~ 0.95
o
o
O L -__- L____~~-J-----L----L---~----.~--~~--~--~
-2.0 -1.6 -1.2 -0.8 -O.lI -0.0 O.lI 0.8 1.2 1.6 2.0
Y AXIS IN CM
Figure 6.2.2.a. Profile of the correlation pCpR(~x,O)
in a turbulent jet at x/D = 20.8.
o
R{6x,O).HEAT FLUX PROFILE
0 r -__- r____~---,----,,----._--_.----._----._--_.--__,
o
CD
oo R~ PLVT2-Z
o flo =O.0DJ9
LIl
X/0=37.0
EXf'ER I tENT I!l
H
c
H~ 0.98
o
~
OL-__ _ L_ _ _ _ ~ __ &d~ __ ~ ____L __ __ L_ _ _ _ ~ __ ~ _ _ _ __ L_ _ ~
-S.O -lI.O -3.0 -2.0 -}.O 0.0 1.0 2.0 3.0 lI.O 5.0
Y AXIS IN CM
Figure 6.2.2.b. Profile of the correlation pC p R(8X,0)
-192-
o
R(llx,O),HEAT FLUX PReJFILE
o.----r----~---,-----r----._--_.----._--_;r --,----,
':3'
N
o fUl f'L VT2-H

o
o
N
flo -0. (XXXl;
~JQa56.7S
o EXPfRHENT I!l
uo
w' H
U)~ C
'-
~
HI'::; 0.96
a: o o
Uo
ZN
.......... 1
x
::::>0
~~
lL.a>
I-
a:
Wo
I~
':3'
~
O~ __ ~~~~ __ ~ ____- L_ _ _ _ ~ __ ~ ____ ~ ____ ~~~ __ ~
-S.O -ILO -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 1l.0 S.O
Y AXIS IN CM
Figure 6.2.2.c. Profile of the correlation pCpR(Ax,O)
o
R(llx,Ol,HEAT FLUX PRClFILE
o
do .------.-----.----r-
~ PL",_:-l1
N
o
o
o Ro =0.0001
N
Xl0=93.75
D<l'ERlI'ENT I!l ~
H
C
HI'::; 0.94
o
x
3~t-
LLco
I-
a:
Wo
I~
~
o
OL--__---'-____-"---=='--I.____-L
-!2.Q-9.6 -7.2 -4.8 -2.11 -0.0 2.4

\--=---1__~I_
4.8 7.2 9.6 12.C
'( RXIS IN CM
Figure 6.2.2.d. Profile of the correlation pCpR(Ax,O)
a
R( b. x . ,0) PRCJF ILE
a
C\I ·
.....
}- I
a RUN PLVTl-Z
0
a
· .....
.... Ro =0.11'
XlO=19
a EXPERlt1E:NT [!]
0
(Xl · H
u
Q)
IIc 0,,83
<.Il
a o I
"-
0
u
0
· .....
CD
I-'
\0
W
I
c
0-
a
o a · .....
x ::!'
<J
0:::
0. a
U
o
C\I
· .....
~
o
OL '''''''''''-= -, I ==-r-I
-10 .0-8 .0 -6 .0 -ll.O - 2 .0 0 .0 2 .0 ij .0 6 .0 8 .0 10 .0
Y AXIS IN eM
Figure 6.2.3. Profile of the correlation ~cpR(6x,0) in a turbulent
plume at x/D = 19.0.
C)
HERT FLUX HRLF WIDTH
~rl- - - - - - - , r - - - - - - - , - - - - - - - - , - - - - - - - - , - - - - - - - - , - - - - - - - - ,
U1
RUN PLVT2
o
Ro =0.00005
'0
I I-
::l
0
• b· uT 10=0 .10S( X/O-S.O)
..0 N
....
:r:
I-
o
1-4 C)
3:~
/
0}
I..L.
-.J
IT I
I I-'
C) \0
_J~ ~
I
IT to
Z
D
........
(f)
Zo
Wo
:L .
........ ('t')
o
Z
Do
:ZOI!3
0.0 20.0 ijO.O 60.0 80.0 100.0 120.0
N~N OIMENSI~NAL DISTANCE FR~M EXIT. x/a
Figure 6.2.4. Heat flux· half-width for a turbulent jet as: a
function of distance along the jet axis.
o NON DIMENSIONAL HEAT FLUX PROFILE
-
~II
~ . RUN PLVT2
-= r ~ Ro =0
: ~+~ X'=X-5.0
EXP( -62.9 7J 2 )
o
co
o
. EXPERIM. RESULTS
x/a SYMBOL.
20.8 [!)
o 37.0 (!)
CD
:2: 58.75 ..t.
I
o
0::: 9;.75 + f-'
\0
VI
"o I
o
><
<l
::r
o
.
0:::
o
N
o
o
o I........ ===-w==
-0.59-0 •ijO -0 .30 -0 .20 -0 • 10 -0 .00 O. 10 0 .20 0 .30 01 •ijO 0 .50
'Tl=-Y/X'
Figure 6.2.5. Non-dimensional heat flux profile in a turbulent jet.
-196-
is indicated in Figures 6.2.2.a-d and 6.2.3. It can be observed that
for a jet the ratio H /H was found to be between 0.9 and 0.98, i.e.,
c 0
H is approximately equal to the input heat flux H. This implies that

c 0
for this particular experimect for a pure jet
R(6x,0) ~ R(O,O) for 6x ~ 1-2 mm .
The experimental result H ~ H for a pure jet can be considered, also as

c 0
an overall check of the velocity and temperature measurements.
For the case of a plume, it was found that (see Figure 6.2.3)
J B(x) R(6x, 0) dy
Hc _ -B(x)
0.83. (6.2.9)
Ho - JB(x)R(O,O)dy
-B(x)
The reason for this probably is that the correlation R(6x,0) is less than
R(O,O) for a plume.
The turbulence heat flux u'T' ideally must be calculated from the
relation
u'(x,y,t)T'(x,y,t) = R(O,O) - ~(x,y)r(x,y). (6.2.10)
During this investigation the quantities R(6x,0), ~(x,y) and r(x + 6x,y)
were measured. For 6x/x « 1 it can be assumed r(x + 6x,y) = r(x,y).

The correlation producL R(6x,0) can substitute R(O,O) only when
R(6x,O) ~ R(O,O) or presumably only when
J B(X)R(6X,0)dy = JB(x)R(O,O)dy ,
-B(x) -B(x)
which was appropriate for a jet. The profile of u'T' for a jet
-197-
calculated from the relation
u'T' = R(~,O) - u T
and normalized by the local mean velocity ~(x) and temperature TM(X)
on the jet axis, is plotted in Figure 6.2.6. For a plume R(6x,O) <
R(O,O), hence an adjustment is necessary in order to obtain a meaningful
profile of u'T'. It is assumed, according to the discussion in the
beginning of this section, that
R(O,O) = R(llx,O) H /H . (6.2.11)

o c
Hence the product u'T' is calculated from the relation
H
u'T' = HO R(6x,O) - u(x,y)T(x,y) (6.2.12)
c
The adjustment given by the Eq. 6.2.11 gives the correct balance of the
Eq. 6.2.12 integrated across the plume
B(x) H
pc I o
u'T' dy = -
H
pc IB(x) R(6x,O)dy - pc I uT dy =
p - B(x) c p -B(x) p
B(x)
H
o
- pc I
p -B(x)
u T dy •
The profile of u'T' for a plume calculated using the Eq. 6.2.12 and
normalized using the local mean velocity ~(x) and temperature TM(X) on
the jet axis is plotted in Figure 6.2.7. It is interesting to notice
the dramatic difference both in magnitude and shape of the normalized
profile of u'T' for a jet and for a plume. This is consistent with the
n-I IT I II
RUN PLVT 2-Z

R (x )= 0.004
X/Dr 37
0.08
0.06
~ 0/ \0 \ ~
I
I-'
I~
E
I t-
E
0.04
I \0
co
I
I~
0.02
0.0 r
-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
y / X'
Figure 6.2.6. Profile of turbulent heat flux in a jet.

~----.--.-----.- 1-- I --,
RUN PLVT~-Z
R(x);; 0.6
0.3
XID = 19
0.2
::e
II-
I I
I:J
::e
....... t / 1 j f-'
\0
\0
I
l~
0.1
oL 11" , -0.2
L_~~ ___ L_
-0.1 0.0
-'---_1_.....1
0.1
_L
0.2
7] = Y/ Xl
Figure 6.2.7. Profile of turbulent heat flux in a plume.

-200-
previous discussions concerning the turbulence intensity (see Section
4.2) and the turbulent heat flux (see Section 6.1). Although jets and
plumes share some common characteristics (for example, the jet angle
expansion given by the non-dimensional number C, see Section 5.1, is
almost the same for both jets and plumes) they are radically different
with respect to the forces which drive or maintain their growth, the plume
is driven by gravitational forces and the jet by the input inertia
forces; as a result their turbulence structure is entirely different •
.Finally, the correlation coefficient
R-
uT
is plotted in Figure 6.2.8 for a jet and in Figure 6.2.9 for a plume.
III rT r TT T I I I
~
UN PLVT2-Z
R{x) = 0.004
1.0 X/D = 37
0.8
~
~
,
0\ TI \ II \ J
I
N
0
I-'
I
I~ 0.4
0.2
0.0 I -0.2
-0.1 0.0 0.\ 0.2
7J=Y/X·
Figure 6.2.8. Jet correlation coefficient of velocity and temperature as a
function of distance from the jet axis.
I. ~I I J I 1--,-,
----------1
-RuN PLVTI-Z I
1.0 R(x)= 0.6
XI D = 19
0.8
~ O.6r
~ ~~ 1
I
~
N
0
N
I
"-
I~ 0.4
0.2
0.0 .~~~~~~~~LJ.~~~~~~~~~~~~~~~
-0.3 -0.2 '-0.1 0.0 0.1 '0.2 0.3
l
'rj=y/x
Figure 6.2.9. Plume correlation coefficient of velocity and tempE~rature as a
function of distance from the jet axis.
-203-
7. COMPARISON OF MODEL PREDICTION WITH
THE EXPERIMENTAL RESULTS
7.0 INTRODUCTION
The mathematical model developed in Chapter 2.3 predicts the
evolution of the kinematic mass flux ~(x), kinematic momentum flux m(x) ,
and the kinematic buoyancy flux 8(x). Hence the experimental results
of Chapter 5, where the above quantities were measured, can be compared
wi th the model. However, the experimental results of Chapter 4, where
only temperatures were measured, cannot be used for a direct comparison
with the model, but the centerline dilution TofTM can be used to pre-
dict the mean dilution p(x) (see Section 7.2).

]10
7.1 COMPARISON OF THE MODEL WITH THE EXPERIMENTAL RESULTS
The basic experimental results obtained from the velocity measure-
ments are given in Table 5.1.2, where ~

o
= u D is the initial volume
0
flux, m = u 2 D is the initial kinematic momentum flux, and H =

0 0 0
- a gH
pc u T D is the initial heat flux. 13 = - - the initial kinematic
0
o p 0 0 o pc
op
buoyancy flux and the kinematic buoyancy flux at any location along the
jet is calculated by the relation
agH
o
13 (x)
pc
p
where the thermal expansion coefficient is calculated at a temperature
T + T. H fpc ~(x)
Tc=o is the cross-sectional mean temperature.
a c p
Hence the apparently unjustified approximation that the buoyancy flux
is conserved, i.e., Sex) = So is avoided.

-204-
The experimental results for the kinematic mass, momentum and
buoyancy fluxes are plotted in Figures 7.1.1, 7.1.2, 7.1.3 respectively.
The theoretical prediction, \-lhich is the solution of the system of the
differential equations given in Chapter 2.3, is also plotted on these
Figures for purposes of comparison.
7.2 CO}WARISON OF THE MODEL WITH EXPERI~mNTAL RESULTS OF TEMPEP~TURE
The experimental results for temperature give the centerline
dilution To/1M' but the model (Section 2.3) predicts the mean dilution
~(x)/~.
0
Hence a relation should be found between
0 0
T /1M and ~(x)/~
in order to extend the application of the model to cases where the
centerline dilution is desired. Using the Gaussian fit to the experi-
mental data, the heat flux by the mean motion is given by
pc IB(x) u.T dy (7.2.1)

p -B(x)
and the kinematic mass flux is given by
~(x) (7.2.2)
so that
(7.2.3)
r··--'
RUN Ro SYMBOL
12.0 P-5 0.4 0
11.0
P-I
Pl\/T- I
0.31
0.17
•"Y
10.0
PLVT - 3
PLVT-2
0.002
0.0
•
A
P-7 0.0 ~
9.0 MODEL
Section 2.3
8.0
0
7.0
::l.. I
......
6.0 N
o
>C lJl
~
I
::l..
5.0
4.0 A
3.0
2.0
1.0 l 1.._1. I
0 10 20 30 40 50 60 70 80 90
X/D
Figure 7.1.1. Comparison of the model with the experimental results for the
kinematic mass flux.
10.01:
I~I
RUN Ro SYMBOL
P--5 0.4 6.
P-I 0.3 0
PLVT 2 0.0 0
P-7 0.0-- •
MODEL
0
1.0
E .""1)
~
I
.......
- )(
F 001 o 0 0 0
N
0
0\
I
O.I~I__~. __~__~__~__~__~~~~__~__~_______._________
o 10 20 30 40 50 60 70 80 90 100
X/O
Figure 7.1.2. Comparison of the model with the experimental results of the
kinematic momentum flux.
T
RUN Ro SYMBOL
P-5 0.40 0
P- I 0.31 0
1.0
PLVT-I 0.17 ~
MODEL
0.8
0.6
0 I
en 0.5
N
o
....... -...J
,-... I
><
0.4
en
0.3
0.2
0.1
0.0
0 10 20 30 40 50 60 70 80 90
X/O
Figure 7.1.3. Comparison of the model with the experimental results for the
kinematic buoyancy flux.
-208-
Combining Eqs. (7.2.3) and (6.1.7) it is found that
(7.2.4)
and for Klu/K ~ 0.8, it is found that

1T
T
o [0.83 + 0.66 R(x)] ~(x) (7.2.5)
~o
Then the mean dilution ~(x) can be calculated from the centerline
~o
dilution To/TM using Eq. (7.2.5) (see Table 4.1.2). Both the experi-
mental results and the prediction based on the model (Section 2.3)
are plotted in Figure 7.2.1.

~1l1 I I
RUN Ro SY MBOL
PLVT 2
JUNE 7
000005
0.0009
•
~
10.0 JULY 29 0423 0

JULY 3 0005 0
r-:::::' .JULY I 0.23 ~
-
II-
)(
E
9.0
MODEL --~~~
"-0
8.0
II-
L---'
-
r;::::'
)(
a::
7.0
I.D
6.0
~ I
0 N
+ 5.0 o
.~
\0
to I
CD
0
.......
4.0 •
L---'
3.0
"0
::l
....... 2.0
- )(
::l 1.0
0 LL I 1. LL~ 1 ~ J I I I I I I I I I I
0 10 20 30 40 50 60
X/O
Figure 7.2.1. Comparison of the model with the experimental results for the mean
temperature distribution on the jet axis.
-210-
8. ON THE SELF-PRESERVATION OF A PURE JET
The self-preservation of the velocity or temperature profile of
a pure two-dimensional jet has been assumed by previous investigators
(Townsend (18), Reichardt (13), Miller et al (12)). A consequence of
this assumption is that velocity b and temperature b half-widths were

u T
thought to depend linearly on the distance x from exit,and equations
of the form
b
u
(8.1)
D
(8.2)
were used to describe the half-widths. The coefficients Klu ' K2u '
KIT' K2T were thought to be unique for all the experimental results
concerning two-dimensional pure jets. However, the variations of these
coefficients from investigator to investigator were not negligible
(see Table 1.3.1). Some investigators found the location of the
virtual origin Xo -K D to be positive, others negative.

2u
In the past, these variations were attributed to experimental
errors. The opinion of this writer is that although the linear assump-
tion is probably valid locally, these variations are most likely due
to the fact that the growth of the jet is not linear on a larger scale.
The self-preservation argument which leads to the linear approximation
for jet growth should therefore be reexamined. Keeping in mind the
results in Chapter 2, a hypothesis was therefore made that the true

-211-
growth of the velocity half-width has the shape as in Figure 8.1. The
Experimenters A and B would fit straight lines to their experimental
results thus indicating linear growth. Possibly Experimenter A could
not extend his region of experiments to a greater X/D,perhaps because
he was restricted by the aspect ratio of the slot and the requirement
of two-dimensionality. Experimenter B could possibly not extend his
region of experiments closer to origin because the physical dimensions
of the jet profile (say in mm) would be too small to ensure accurate
experimental results. Obviously, Experimenter A finds smaller slope
(i.e., smaller coefficient K ) and a negative virtual origin whereas

lu
Experimenter B would find a positive virtual origin.
The preceding discussion indicates that if the hypothesis of a non-
self-preserving two-dimensional jet is correct, a correlation should
be found between the experimental results K ' K of different experi-

lu 2u
menters and t /D, where t is the distance of the more distant
max max
cross section from the origin for each investigator. The coefficients
K from each jet investigator are plotted versus the coefficient K2u
lu
in Figure 8.2. The coefficient K is plotted versus t /D in
lu max
Figure 8.3, and the coefficient K is plotted versus £max/ D in Figure
2u
8.4. It is interesting to notice in Figure 8.2 that the coefficients
K ' K do not vary randomly from investigator to investigator but

lu 2u
vary as predicted by the hypothesis, i.e., when K increases, K
lu 2u
decreases and becomes negative, giving a positive virtual origin
Also the coefficient K increases as £ /D increases

lu max
and K decreases as £max/D increases, as predicted. The coefficients
2u
~
Investigator Symbo I
A 0
Investigator B
B 0 I
(l)
o
u
U)
I
N
>-
~
I-'
N
o I
.a
~
I nvestigator A
c::(
o
"-
/~tan-I
~
..0
./ K
..,."
/ t IU
~ K2UO~ X/O 1
Arbitrary Scale
Figure 8.1. Hypothesis concerning the nonlinear spread of the mean

velocity profile of a turbulent jet.
TWO DIMENSI~NAL JET
--
('t)
C;;r-
INVESTIGATOR SYMBOL
PRESENT STUOY CJ
N PRESENT STUOY (')
JENKINS A
o JENKINS +
IiESKESrRO X
fmT:iHRNN ~
GClLDSCHHIOT ...
KNYSTAlITFIS ~
--
o
•
HILLER
ZIJNEN
Z
y
~~ X
ZU':EN
fLORA
)(
•
~. NAKRGUCHI
BICKNELL
X
.L I
0 N
f--'
(:) W
::I I
:x:::
+'.
en
'''" "",,"
"
0 """
... '''',~ '-..(!]
0
.....
00
o
oL- I I I I I j
-7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0 7.0
K 2U
Figure 8.2. Correlation of spreading rate Kl and non-dimensional

virtual origin K . u
2u
~
TWD DIMENSIDNRL JET
o r-~·· ·---.--1-- --]---,
INVESTIGATOR SYMBOL
::3' P?oE~NT SlUIY m
o
P~.5ENT
S'l(INS
JEt.'t(INS
STUOr
...,r
(!)
HESKESTP.O X
en fOATHMRNN ~
o GOUJS:::tt1IOT +
HILLER ~
ZI..JI\'EN Z
ZIJI£N V
N )(
fl.Of\l1
o NflKRGLCHI liE
BICKNtll. X
lI:
)( J
0 N
I-'
::J ~
~
I
0
•
en
q
0 I!l ...
(Xl
o
o1 1 1 I I I I I I ----I
0.0 20.0 IJO.O 60.0 80.0 100.0 120.0 11!0.0 160.0 180.01 200.0
-Rmox /D
Figure 8.3. Spreading rate K1 as a function of non-dimensional

distance £ /D. u
max
0
TW~ OIMENSI~NAL JET
0
.
.....
C\I
~~
INVESTIGATOR SYMBOL
PRESENT SlUlY [!)
PRESENT STUOY (!)
JENKINS 4>.
HESI'iESTAO +
~~
I!l
FORTHHANN X
.Ii. GtUJSCHHIOT ~
~
HILLER +
~~
ZIJNf.N ~
~
"
C\J (Y) ZJJNEN Z
;J FLreR Y
X NflKRGOCHI )(
~
I
X N
BICKNELL 31( I-'
lJ1
~I l( I
0
01
. (!)
(Y')
.
~I0.0
~+
. ~
I I I I I
20.0 LJO.O SO.O 80.0 100.0 120.0 lLJO.O lS0.0 180.0 200.0
~ maxI 0
Figure 8.4. Non-dimensional virtual origin K2 as? function of
non-dimensional distance ~ /D. u
max
-216-
K ' K given by any experimenter, should therefore, apparently be

lu 2u
viewed as the best linear approximation, for each experimenter, of the
9, • 9,
. mln x mDax (where 9,. is the
growth of the half-width in the reglon ---- < - <
D D mln
closest to origin measurement). Equation 8.1, with coefficients so
determined, should not be extrapolated outside this region.
Using the above results the jet half-width b /D was plotted versus
u
the distance x/D from different investigators (Figure 8.5) from which
it is easy to observe that b /D is not a linear function of x.

u
The interesting conclusion is that the velocity profile of a two-
dimensional pure jet is not self-preserving. Similar arguments show
that the temperature profile is not self-preserving either, as the
Figures 8.6, 8.7, 8.8, 8.9 demonstrate.
The self-preservation of the pure two-dimensional jet, and the
conservation of the kinematic momentum flux, require that
lI£i ,.., x -1
-2
and previous investigators (see Table 1.3.1) and this writer (see
Table 5.2.1), give the decay of the centerline velocity in the form
(8.3)
The above relation should be viewed as a linear approximation to some

:;;,
o
o
N
oC::
o
t:l
.....
N
o (!)
/
......... ~
~
..0
.0 INVESTlGATelR SYMBOL
IC:: fORTHtt=lNN [!J
f-lJ)
0 .... KNYSTAUTRS (!)
t-1
~
HILLER AI.
o HESKESTAO +
LLC:: JENKINS X
---1N GOLDSCl1t1IDT
CI ....
I ZIJNEN .,.
~
N
I
X f-'
---10
Present {RUN P-7 ......
study RUN PLVT2 Z I
CIo
Z .
DOl
t-1
en
Zo
Wo
L .
t-1 (£)
o
Z
00
Zc:: ('t')
o
o ~ ______~________I~ I
0.0 30.0 60.0 90.0 120.0 150.0 180.0 210.0 2L10.0
N~N OIMENS. DISTAN(E FR~M EXIT ,X/O
_Figure 8.5. Nonlinear spread of half-y,Tidth b of mean velocity profile of a
two-dimensional turbulent pure j~t.
en
TW~ DIMENSIONAL JET
.-;
o~1----~-----~--~----~----~----~----~--~-----,
--0 1 1 1 1 1 1 1 1 1 1
-3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 li.O 5.0 6.0
K 2T
Figure 8.6. Correlation of spreading rate K1T and non-dimensional virtual
origin K .
2T
TWO OIMENSI~NAL JET
-
co
0
•
I
INVESTIGATOR SYMBOL
RUN JLl.Y 3 Present (!)
RUN PLYT2 (!J
study
RUN JUNE 3 A
RUN JlJt£ 7 +
.£NKINS X
~INS ~
l/ll
...... Zl.A£N +
0
r
-
I
I
~ / N
f-'
\0
I
+
-01
N
I I I 1__ ..... __ I __ . ~_L ____ . J_ . . __

0.0 12.0 2ij.0 36.0 ija.o 60.0 72.0 8ij.0 96.0 108.0 120.0
~ max /0
Figure 8.7. Spreading rate KIT of mean temperature profile as a function.of
distance R,
max
ID.
o TW~ DIMENSI~NAL JET
C)
·~i------~----~------~----~-------~----~
<.0
INVESTIGATOR SYMBDL
Present 1RUN JUtY 3
I!l
~~ ~~ RUN PLVT2 (!)

.& study RUN JUNE 3 A
RUN JUNE 7 +
)( JENKINS X
ZIJNEN ~
I
0
C)
•
'"
C\I
~
I ~
N I
N
~ N
0
I
¢
C)
.
L
C)
C)
C)
~I I I I
f!)" ' "
0.0 20.0 YO.O 60.0 80.0 100.0 120.0
~ max /0
Figure 8.8. Non-dimensional virtual origin K2T as a function of distance- ~ /D.
max
020 PURE JET
~Ir----r----~--~-----r----~---'----II----r----'----II----''----'----'---~
: I
<D
.r RUN
PLVT2
JUNE 3
sYH8I1
!!I
(!)
~
JUNE 7 '"
JULY 3 +
o
o
::I'
o
o
N
00
,~
.... 0
.o~
I
N
:c N
I-
00 t-'
./0./
...... 0 I
3 00
LL
-I
a:
:coo
<D
o
o
::I'
~+/
o
/fJ/
o ",/
N
/:l
/,+
o
~Jr- I I I I I I I
0.0 10.0 20.0 30.0 [D.O 50.0 60.0 70.0 80.0 90.0100.0110.0120.0 130.0 1~0.0
NON OIMENS. OISTRN(E FROM EXIT .X/O
Figure 8.9. Nonlinear spread of temperature half-width b T of a two-
dimensional pure jet.
-222-
nonlinear decay of the centerline velocity. This becomes clear if the

- 2
non-dimensional centerline velocity (;:) is plotted versus X/D from
many experiments by this writer (Figure 8.10) or from the results of
numerous other experimenters (Figure 8.11).
The coefficient C versus C is plotted in Figure 8.12; the co-

lu 2u
efficient G versus £max/D in Figure 8.13; and the coefficient C2u
lu
versus £ /D in Figure 8.14. The results all indicate a similar
max
behavior to the rate of decay of centerline velocity.
The above results concerning the decay of the centerline velocity
and temperature for a pure two-dimensional jet out of a wall, are
consistent with the analysis of Sections 2.1 and 5.2, where it was
found that the kinematic momentum flux of a pure jet out of a wall is
not conserved.
C)
PURE JET- CENTERLINE VELOCIT~
o~ __~____~__~____~__~~__~__- ,____~__"___. -__~
0
(T)
0
0
RUN SYMBOl
If)
C\I
P-7 (!)
PlVT-2
/
(!)
CENTER ~
0
0
CENT-l +
0
CfNT-2 X
C\I
1/
0 X/
N
0
IJ")
./*/0
-
l::l
><
~
0
0
I
N
N
W
......... I
NO
'::l
....
0
0
0
LJ")
o
oL-- I I I
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
NON DIMENSIONAL DISTANCE FROM EXIT ,X/D
Figure 8.10. Velocity ratio u2~(x) as a nonlinear function of non-
dimensional distgnce x/D.
o
o
Or
Ln I
b
~t- ...
~ A
A
o INVESTIGATOR SYMBOL
~t- A
PRESENT :5TUDY<P-7) I!l
-
-
N:E
1::3
)(
In
C\I
A f!)
PRESENT :UUDY<PLVT2)
HESKESTRI)
(!)
~
I
N
N
+0-
I
"'-
(!) MILLER +
NO o A
(!) GOLDSCHMIOT X
1::3 ~t (!)( fORTHMANI~ ~
C\I
..... ·x
(!)(
~[!J+
.J!l«
~~tJf-t I 1 L
0.0 25.0 50.0 75.0 100.0 125.0 150.0 175.0 200.0
N~N OIMENSI~NAL DISTANCE FR~M EXIT, X/O
Figure 8.11. Velocity ratio u~!UM(x) as a nonlinear function of non-dimensional
distance x/D. Data from several investigators.
=r
TW~ DIMENSI~NAL JET
~~i----~----~----~----~--~----~----~----_r-----~--~
Cl
I NVESTI GATOI~ ~YMBOL

PRESENT SllJOY (!}
~~
fRESENT STIlOY (!)
JENKINS A
HESKESTAO +
FffiTHtlANN X
cn.oSC\iMIOT ~
<D
C\t KNYSTRUTAS +
Cl · ZIJNEN ~
::J
ZlJNEN Z
NflKAGOOiI y
U
HIl.LER )i(
C\t
HIl.LER )(
C\t
·
Cl
. (!) '--
'-.""
'-''- Z
I
N
N
V1
I
'.'~
',v.
~~ + )(' X,
~ .III.
-·
::j'
0
-12.9-10.0 -8.0 -S.O -1l.0 -2.0 0.0 2 .. 0 1l .. 0 6.0 8.0
C 2.u
Figure 8.12. Correlation of coefficients C and C .
lu 2u
to
TW~ 01MENSI~NAL JET
(1')
or-
INVESTIGATOR SYMBOL
PRESENT STUDY !!J
PRESENT STUOY (!)
Jf.N?(INS .A
~ ~ HESi{ESTAD +
• fOflTHHRNN X
o GCl..DSCHHlOT ~
ZIJNEN +
ZIJNEN ~
~RGUCHI Z
MIll.fR Y
tllu...ER )(
:::r
N
01 /' I I
N
N
0\
:::l
-
U
/(!) I
/ z
~~ X
y. I>
(\1
o
. _-.L__-L_._l...-,.._..__l __ -1...... I I I
0.0 20.0 ijO.O 60.0' 80.0 100.0 120.0 lijO.O 160.0 1.80.0 200.0
Xmox / D
Figure 8.. 13. Coefficient C1.

u
as a function of non-dimensional distance.R,
max
ID.
TWO DIMENSIONAL JET
o
wl~--~----~----r---~-----r----~--~~--~-----~--~
INVESTIGATOR SYMBOL
o PRESENT STUDY (!J
• A
::t' PRESENT STUDY (!)
JENKINS A
HESKESTAD +
f(~RTHHANN X
Q GelLOSCHMIOT ~
:J
a
•
ZIJNEN ...
(\J ZIJNEN X
U HILLER Z
J
I
Q N
• HlUER y N
~
::1' I
i
I •
Q
•
CD
I
C)
•
C\I
.....
I
0.0 20.0 tJO.O 60.0 80.0 100.0 120.0 lijO.O 160.0 180.0 200.0
.Qmox /O
Figure 8.14. Coefficient C as a function of non-dimensional distance ~ /D.
2u max
-228-
9. SUMMARY OF MAIN RESULTS AND CONCLUSIONS
9.0 INTRODUCTORY NOTE
The present study contributes to the understanding of the mechanics
of a vertical two-dimensional turbulent buoyant jet in its transition-
state from a pure jet to a pure plume. The structure of the mean flow,
the turbulence structure, and the structure of the large scale motion of
a buoyant jet as a function of the Richardson number are examined.
Several interesting results have been brought out and numerous con-
clusions can be drawn.
9.1 SUMMARY OF THE MAIN RESULTS
I. The mean flow structure
The integrated equations of motion (i.e., the equations for the
conservation of the kinematic mass, momentum, and buoyancy fluxes) were
examined in a rigorous way and the basic assumptions made by previous
investigators were critically reviewed.
The concept of the jet boundary was clarified and its usefulness
was demonstrated in Section 2.1.
The conservation of the kinematic momentum flux of a pure jet in a
uniform non-flowing environment has been accepted as a valid assumption
by almost all previous investigators. This study has shown, however,
that the kinematic momentum flux is not conserved, but it decreases or
increases (depending on the external geometry of the flow) due to the
flow induced by the jet (see Sections 2.1, 5.2).
The conservation of the kinematic buoyancy flux of a buoyant (heated)
jet or pure plume in a homogeneous environment was thought to be an exact

-229-
statement by previous investigators. This study has demonstrated,
however, that the dependence of thermal expansion coefficient upon the
temperature (see Sections 2.2 and 7.1) leads to the kinematic buoyancy
flux being not conserved.
Previous investigators had assumed that the kinematic buoyancy
flux sex) of a plume could be written as
sex) = IB(x) a(x)gu(x,y) T(x,y) dy ,

-B(x)
but the experimental results of this study have shown that this is not
justified because the turbulent buoyancy flux term for a plume (see
Section 6.1)
I B(x) u' T' dy ~ 0.6 J B(x) u T dy

-B(x) -B(x)
It was found that the angle expansion of a buoyant jet, given by the
non-dimensional number C (see Section 2.3), or by the growth of the half-
widths, b and b , of the velocity and temperature profiles respectively,

u T
(see Sections 5.1 and 4.1), is essentially independent of the Richardson
number, i.e., it is the same for both jets and plumes. For an accurate
description of the growth of a turbulent jet the non-self-preserving
character of velocity profile should be taken into account (see Chapter
8). Suggested values for the linear part of the growth of the profile of
a buoyant jet are

-230-
)leX)
c 0.55 ,
lm(x)(x-x )
ou
db
u
0.10 ,
dx
0.14 .
The entrainment coefficient can be calculated for a pure jet by using
Equation (1.2.11) to give
( ~·)1/2
Q,n2 ~ 0.055 ,
and for a plume, using Equation (1.2.18),
K,
( Q,n2
.LU ~\)1/2
a ~ 0.11 •
p 2
It was found that the time-averaged mean temperature profile is
wider than the mean velocity profile for both jets and plumes (see
Section 6.1).
The centerline dilution of a pure plume was found to be (see
Section 4.1)
-2/3
0.42 gxS (x).
The centerline velocity of a pure plume was found to be

-231-
1.66 S(x)1/3 .
It was an open question in the literature as to why the virtual
origin of the mean velocity profile, defined as x

au
= -DK
2u
' was sometimes
found upstream and sometimes downstream of the jet orifice. Furthermore,

-1
the jet angle expansion (given by tan K ) has varied from investigator
lu
to investigator. This study has proved that this is mainly because the
mean flow for a two-dimensional pure jet out of a wall is not self-
preserving (see Chapter 8), The nonlinear growth of the jet velocity
half-width can be empirically approximated by the algebraic equation
b 2
u
= 0.0937 + 0.097 (~) + 0.000052 (~)
D
where 6 < x/D < 200 .
II. The turbulence structure of jets and plumes
Turbulent jets and plumes share some common characteristics; their
expansion angle (given by the growth of the half-width of velocity profile)
is approximately the same. Time exposure pictures of jets and plumes are
amazingly alike. An extrapolation of these similarities of jets and
plumes in their turbulent structure is, however, not justified, because
the forces which drive or maintain their motion are radically different.
A pure jet is maintained by the input inertia forces whereas a pure plume
is driven by the buoyancy forces. For a pure plume the velocity u(x,y,t)
at any point (x,y) depends upon the buoyancy at time t, which depends on
the temperature T(x,y,t) which implies that u(x,y,t) and T(x,y,t) are
well correlated. Combining this observation with the turbulence energy

-232-
equation (see Section 2.2) we concluded that the turbulence intensity and
the turbulent heat transfer should increase with the local Richardson
number. These conclusions were verified by the experimental results of
this investigation (see Sections 4.2, 6.2).
It was found that the profiles of the turbulence intensity of a
plume and of a jet are quite different and that the normalized turbulence
intensity on the jet axis increased with the Richardson number and became
constant for a plume (see Section 4.2). Similarly, it was found that the
turbulent heat transfer u'T' increases with the local Richardson number
(see Section 6.1). Hence the turbulence structure of jets and plumes is
entirely different.
To the best knowledge of this writer there is no previously published
investigation of the turbulence structure of a buoyant jet as a function
of the Richardson number. It is believed that similar results should be
obtained for a round buoyant jet.
III. The structure of the large eddies of jets and plumes
The mechanics of the buoyant jets can be described adequately by the
mean flow and the turbulence structure, but the characteristics of the
phase information of the intermittent velocity and temperature signals
are of vital significance for a better understanding and explanation of
jet behavior. In this investigation the large eddy structure of the
temperature field was investigated as a function of the Richardson
number. The intermittency was studied and it was found that the inter-
mittency is not unity on the axis of a plume (i.e., ambient unmixed water
does reach the axis of the plume). For an almost pure jet the
-233-
intermittency was found to be unity on the jet axis, but the inter-
mittency profile was found to narrow with increasing distance from the
jet exit (see Section 4.4).
The intensity of the turbulent fluctuations of the temperature,
T'2, was separated into two distinct components, one associated with
the hot vortices of the buoyant jet and the other associated with the
cold bursts. It was found that the normalized profile of the
turbulence intensity associated with the cold bursts is essentially
the same for both jets and plumes and that the cold bursts are mainly
responsible for the two maxima in the distribution of the intensity
of turbulent fluctuations T,2 across a pure jet (see Section 4.2).
The frequency with which the hot/cold and cold/hot interface passed
through any given point was studied and it was found that the occurrence
of the hot vortices has an orderly structure. The profiles of the
average duration of a hot burst (defined as the time between two
successive cold/hot and hot/cold crossings) was obtained (see Section 4.5).
The profiles of the maximum and minimum temperatures were studied
as a function of the Richardson number. The profile of the maximum
temperature provides an accurate way to determine the boundaries of the
jet. It was found that there is a finite probability of finding jet
fluid at the jet boundaries with temperature approximately equal to the
mean temperature on the jet axis. This suggests that for some applica-
tions the buoyant jet should be viewed from a probabilistic rather than
a deterministic point of view (see Section 4.3).
The frequency distribution of the turbulent fluctuations of

-234-
temperature and velocity are in general skewed (see Section 4.6 and 5.4).
The skewness factor is approximately zero (appropriate for symmetric
distribution) only on the jet axis and the flatness factor is approxi-
mately 3.0 (appropriate for Gaussian distribution) again only on the jet
axis (see Section 4.7).
The main conclusion from the study is that the nature of turbulent
jets and plumes are quite different because the nature of the forces which
act upon them is quite different. A pure jet is maintained by the input
inertia forces and a pure plume is driven by buoyancy forces. This
difference explains why the evolution of the volume flux, the centerline
mean temperature and mean velocity, the intensity of turbulent fluctua-
tions of temperature and velocity, and the turbulent heat transfer, are
entirely different for a plume and for a jet. However, the shape and the
spreading rate of the mean velocity profiles are approximately the same
for both jets and plumes. Hence we can conclude that the shape of the
mean velocity profile and its spreading rate are independent of the nature
of the driving force. This conclusion is also supported by the fact that
similar results were found for a round jet (Corrsin and Uberoi [52]) and
a round plume (Rouse et al [21]).
Previous mathematical models (see Section 1.2) used to predict the
evolution of mean quantities in a turbulent buoyant jet, were found to be
lacking in two basic respects. Future models of the evolution of mean
quantities in a buoyant jet should consider that the turbulent kinematic
buoyancy flux is not negligible for a plume, and that the kinematic
-235-
momentum flux of a pure jet in a uniform nonflowing environment is not
conserved. The assumption of a universal entrainment coefficient a

e
for both jets and plumes is not correct (see Section 1.2). Previous
mathematical models based on this concept and used for design should be
replaced by more reasonable models that recognize the variation of the
entrainment coefficient.
The increase of the intensity of turbulent fluctuations and of the
turbulent heat transfer with the Richardson number leads to the conclusion
that the turbulent mixing is much more effective for plumes than for
jets. In particular, for a given ratio of the volume flux ~(x) to the
initial volume flux ~o' the centerline dilution T /T}1(x) is much bigger
o
for a plume than for a jet (see Section 7.2).
Finally, the large scale motion of buoyant jets is not chaotic but
has a structure and an order which is essential to the understanding and
explanation of some of the basic jet features.

-236-
APPENDIX A
A few photographs are presented here to illustrate that the short
time exposure photographs of turbulent jets and plumes are entirely
different. Figure A.l and A.2 are short exposure pictures (1/100 of a
second) and Figure A.3 is a medium exposure picture (~2 seconds) of a
turbulent two=dimensional plume in a uniform nonflowing (except for the
external induced flow) environment. Figure A.4 is a short exposure
picture (1/100 of a second) and picture A.S is a medium exposure (~2
seconds) of a pure two-dimensional turbulent jet in nonflowing (except
for the external induced flow) environment. A feeling for the flow
induced by a jet or a plume can be obtained from the Figures A.3, A.4,
and A.S.
It is interesting to observe the large scale waving motion of the
plume. A detailed stability analysis will not be presented here, but the
basic mechanism which produces these large scale waves can be explained
as follows.
Assume that at some time t a plume achieves the configuration of

o
Figure A.6 and consider the volume V which is included between the
AB
cross sections AA and BB. Now, the temperature on the plume axis
decreases and the density increases as the distance x from the plume
orifice increases. Hence the center of the mass of the volume V will
AB
be above the center of the pressure forces. Two external forces are
acting on the mass m of the volume V , (i) the gravitational force,

AB AB
with absolute value
G
of pressure the
the forces
is the mechanism is
are observed a
-243-
A-
--- Archimedian Force
-- --- --A
8-_
-- _-#-_ Gravitational
-- -
Force
--8
Figure A.7. Schematic to explain the production

of large waves in a plume.
-244-
APPENDIX B
The experimental results concerning the mean temperature profile
for all the runs are presented here. For every experimental run the
mean temperature profile T(x,y) in °c was plotted versus the distance
from the jet axis in cm, the temperature half-width b were calculated
T
and a Gaussian curve defined by Equation 4.1.1 was fitted. The ha1f-
widths bT(x) for each particular run were plotted versus the distance
from the exit and the spreading rate KIT was calculated according to
Equation 4.1.2. The mean temperature T(x,y) normalized by the mean
temperature on the jet axis, TM(x) , was plotted against the non-dimen-
sional distance from the jet axis y/x' (where x = x - x OT ) for each
particular run and a Gaussian curve defined by Equation 4.1.3 was fitted.
The experimental values for the half-width b and for the temperature
T
on the jet axis TM(X) for each run were presented previously in Table
4.1.2.
Since all the necessary information concerning each experiment is
indicated in each figure, no titles will be given in the figures of
Appendix B.
-245-
o
TRRNSVERSE MERN TEMPERRTURE PRDFILE
N~_ _-.---_
0 RUN PLYT2-H
~
RICH. NUI1B.=O.OOOO6
X/O=58.7S
0
EXFERII1::NT (!J
o
u(O •
0
Z
-.
wg
a:.. !
:::)0
I-
a:
a: O
W
o...=':
:Eo
w
l-
e
"!
0
~
0 _---L
-8.0 -6.0 -1l.0 -2.0 0.0 2.0 ll.O 6.0 8.0
Y AXIS IN CM
e
o~ __.____. -______- .______- , . -______ ~ ______- r_ _ _ _ _ _ ~
RUN PLYT2-H
.....
o
CD RICH. NUMB.=O.OOOI3
X/O=93.7S
EXFERIl'fNT (!J
o
=t'
o
~
a 1--_ _ _-=":::l..-_ _ _ _ _--1.._ _ _ _ _ _ _ L _ _ _ L -__....L.::"'-_ _- '
-12.0 -8.0 -4.0 0.0 4.0 8.0 12.0
Y AXIS IN CM
-246-
o
OJ
o
lJ) RcIN flVT2-E
RICH. NCM8.=O.COOOS
X/O·20.8
o EXPEfi WENT (!J
()
UN .
z
UJ~
cr: •
~O~I
W O
tL~
::<:0
W
r-o!
~ LL'__~__
-3.2 -2.4 -1.6 -0.8
.-1 ____ '----_ _'-------'==_'--_---'
0.0
Y RXIS IN CM
0.8 1.6 2.4 3.2
o
TRRNSVERSE MEAN TEMPERRTUR~ PROFILE
-r--------r-------~r_------
N
o
(0
.- RUN FLVT2-Z
RICH. NU~8. =0 •c0009.
a X/O=37.S
lJ)
EXf'ERINCNf (!J
U ....
<>
Zo
>-1 N
o(I)
o
o~ '--_ _ _ _~_______ L_ _____ 1_____ ~ __=L_ _____1

-6.0 -4.0 -2.0 0.0 2.0 4.0 6.0
Y RXIS IN Ci1
-247-
o
TEMPERRTURE HRLF wrOTH
o
:I.UN PLV~2
~ RICHRRD5~N NUMB.=O.OOOOS
E2~ bT IO=O.167(X/O -1.56)
l-
n
·0
IO
f-'
o~
.......
3:
...J
cr:
z 0
0
~--'I ~
(f)<D
Z
W
.:E
00
o
z ·
El(Y")
Z
~
O~ __ ~ ____- L_ _ _ _ ~ ____ ~ ____ ~ ____ ~ __ ~ _ _ _ _- L_ _ _ _ ~ __ ~
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
NON DIMENSIONRL DISTANCE FROM EXIT, X/D
I\CJN DIMENSIONRL MEAN TEMPERATURE PROFILE

...:ir-----r
0
N
0
RUN PLVT2
~ RlCH. NUM8.=0.OOOO
X·=X-l.S6
EXP(-251/1
0
~
0
- EXPERIM. RESULTS
XlO SYM80L
:;;0
If-~
.. 20.S
37.S
[!J
(!)
56.75 6
-"""0
II- 93.75 +
0
~
0
I!)
0
N
0
~
0 _L ---L I
-0.50-0.40 -0.30 -0.20 -0.10 -0.00 0.10 0.20 0.30 O.lIO 0.50
7]=YIX'
o TRRNSVERSE MERN TEMPERRTURE PR~FILE
C)
'~r----~----~----~----~--~----~-----~--~
N
RUN JUNE 7
o R1CH. NUMB.=O.0009
..........
If)
X/O=60.0
EXPERIMENT [!)
u
o
Z
t-1
I
N
wo .J>.
o:~ 00
I
:::J ....
I-
cr:
a:
w
0....
L
Wo
I-lf)
o
.
o
oL ,.----- I I I __ ,
-16 .0 -12.0 -8.0 -I.L 0 0 .0 ij .0 8.0 12 .0 16 .0
Y RXIS IN eM
-249-
o
TRANSVERSE MEAN TEMPERRTURE PROFILE
0r_----~-----r----_.-------r_----,_----_r----_.-----_.
CD
RUN JUNE 3
RlCH. NUMS.=O.0009
o
o X/O=lO.O
w [XPERII'fNr (!J
U
o
o
O~~~~~~~ ___ ~ ____~____~____-L_=~~____~
-q.O -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 q.O
'( AXIS IN CM
o TRANSVERSE MERN TEMPERRTURE PROFILE

0r-----.-----_r-----.-----,.-----,-----,,-----.----~
:
~
"UNJ~3
RICH. NUMS.=O.0009
]
X/O,,20.0
o [XPfRlHENr (!J
o
ou .
:3'
z
......
w8.
a:
:::l(Y)
~
a:
a: o
W
o....<=!
LN
w
~
o
o
o
o~~--~~~~----~I-_----L-----~-----L.-2~-L----~
-6.0 -Q.5 -1.5 0.0 1.5 3.0 Q.5 6.0
'( AXIS IN CM
-250-
0
TRRNSVERSE MERN TEMPERRTURE PR~FILE
II"!
::3' I
0 RUN JU~ 3
(D
('I")
RICH. NUMB.=0.OOO9
X/O=30.0
EXPfRHENT [!)
U
0
0
zt-
..-.~
w
a:
:J
i-
a:o
a:~
w-
a....
L
w
t-
o
~
0
~
0
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Y RXIS IN CM
0
II"!
J
('I")
0 I!J
~ RUN JUNE 3
('I")
RICH. NUMB.=0.0009
0 X/O=50.0
LJ)
EXPER"'ENT [!)
UN
0
Zo
......, 0
N
W
a:
:J O
t-LJ)
IT •
a: ....
w
a....
LO
WO
I-~
°
II"!
0
~
0
-12.0 -9.0 -6.0 -3.0 0.0 3.0 6.0 9.0 12.0
Y RXIS IN CM
-251-
0
TRmJ5VERSE MERN TEMPERRTURE PROFILE
~
co
0
~ RUN JULY 5
co
R1CH. NUMB.=O.OOS
0
X/0=6.0
~ EXPERII'fNT !!l
UN
0
Zo
>-10
WN
a:
-.
~o
r- ln
0: •
a:-
W
0-
:Eo
WO
1--=
0
~
0
~
a
-4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0
Y RXIS IN CM
0
TRRNSVERSE MEAN TEMPERATURE PROFILE
In
NI
RUN JULY 5
0
~ R1CH. NU~1B. =0 .005
N
X/O=14.0
EXPERIMENT !!l
U
0
0
Z~
.......
W
-
a:
:::J
1-'0
CIo
Ej-=
0-
l
:E
W
I-
a
~
0
~
a I ----L_
-B.O -6.0 -2.0 0.0 2.0 4.0 6.0 B.O
Y AXIS IN CM
-252-
If)
TRRNSVERSE ME~N TEMPERRTURE PROFILE
r-
T
RUN JULY 5
RICH. NUMB.=0.005
X/O=22.0
EXFERll'ENT C!J
u
o
If)
z~
W
a:
::J
I-
cr: a
a:":
We
(L
L
W
I-
If)
rt)
o~ '--_ _-==:::L
[!)
_ _ _ _' - -_ _ _-L._ _ _--l_ _ _ _ _'-_::=,_, _ _---'
-12.0 -B.O -4.0 0.0 4.0 B.O 12.0

Y RXIS IN CM
11)
r - , -_ _. -____- .______. -____- .___- .______. -___. -___- ,
RUN JULY 5
RICH. NUM8.=0.005
X/O=30.2
EXPEA ll'ENT C!J
u
o
11)
Z~
W
a:
::J
I-
CIa
0:":
We
(L
L
W
I-
11)
rt)
:o.L~_-, _ _ I ____ ---L ___ ~ ____ L ___

-16.0 -12.0 -8.0 -4.0 0.0 4.0
(!)
qC'J___=_1__.
8.0 12.0 16.0
Y RXIS IN CM
° TRRNSVERSE MERN TEMPERRTURE
U')
PR~FI~E
.....·
~~ I!l RUN JULY 5

~ RICH. NU!MB .=0 .005
X/O=36.0
UO I 1- -, EXPERIMfNl' (!J
o ~
....
Z
t--1
wU')
a:~1
1
I
::Jo
I-
IT
a:
\ 1 N
Ln
w
I
Wo
O-U')
L .
O
W
I-
U')
C\I
0 ·
(:)
Q ·
-lB.O -13.5 -9.0 .-L!.5 0.0 L!.5 9.0 13.5 18.0
Y RXIS IN eM
-254-
o
TEMPERRTURE HALF WIDTH
o~__~__---.----~-----.-----.----
....
a
o 'RUN JULY 5
o
<D RICHRRDSON NUMB.=O.OOS
'-.
....
b r /D=O.121<X/O +4.46)
.Dg
.~
::r:
~
o
~o
3:0
u..::I'
...J
a:
Io
...J~
a:(Y)
z
o
~
<nO
z~
WN
L
~
o
o
zo
0"':
:z
~
o~ __~~ __ ____ ____ ____ ____ ____ ____ ____ ____
~ ~ ~ ~ ~ ~ ~ ~ ~
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0
aNON DIMENSIONRL MEAN TEMPERATURE PROFILE

Nr-----r---~r_---~----_.----.~----~----_r----_r----_r----~
o RUN JULY .5
a
RICH. NUM8.=0.00S
X'=X+4.46
a(X) EXP( -47.0 Tl. 2 )
o EXPERIM. RESULTS
X/O SYMBOL
6.0 (!]
111.0 (!)
22.0 ~
30.2 +
35.0 X
a
::I'
a
a
N
o
I!I
+
a
c:i .. 0 -..l ___--LI_ -'------'----L-----'--___--'--~~'"___"=='.." _ J
-0.35-0.28 -0.21 -0.14 -0.07 0.0 0.07 0.14 .0.21 0.28 0.35
Tl.=y IX'
-255-
TRRNSVERSE MERN TEMPERRTURE

o0 , - -_ _....-_ _--..,-_ _---.-_ _
PR~FILE
":1' -r
RUN JULY 3
R1CH. NUMB.=0.006
o
o X/O=6.0
(Y)
EXf'EA JI'ENT [!J
U
o
z
C!
o~-=~~_ _- L_ _~__ ._~L-_ _~_ _~_~L£~~~
-l!.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 l!.O

Y RXIS IN CM
o
TRRNSVERSE MERN TEMPERRTURE FR~FILE
0 . - -_ _. -_ _- . - - - . - _ - , , - _ - . - -_ _. -_ _ _ _- ._ _ _ _- ,
(Y)
o RUN JLLY 3
lJ'l
N R1CH. NUMB.=O.006
X/O=14.0
EXfEfl JI'ENT [!J
o
o
ou •
N
w~
a: .
::J ....
r-
cr:
a:
llJO
o...C!
:::E ....
w
r-
o
lJ'l
o
C!
C) L-..---ln--dm::=--.-Id..-L_ _ _-.L_ _ _- - l L -_ _ _.L____ -L.J:l_-e-"_'t---'
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Y RXIS IN CM
-256-
lJ)
~r-----.------r----~------.-----.------'-----'-----'
o
lJ)
RUN Jl1.. Y 3
RICH. NUMB.=0.006
U1 X/O=30.2
N
EX PEA II£NT [!]
o
u-
Zo
....... 0
w-
a:
::JU1
I-~
a: •
a:o
w
CL
LO
wU1
1-
o'
U1
N
o
~ I~~ I I ___~~-2~~____~
-16.0 -12.0 -8.0 -4.0 0.0 4.0 8.0 12.0 16.0
Y RXIS IN CM
0
TRRNSVERSE MEAN TEMPERATURE PROFILE
Ul
~I
I
U1
~
RUN JUlY 3
RICH. NUMB.=O.006
X/O=36.0
EXPERlrENT [!]
U O
o ~
Z
t-1
WU1
a:~
::Jo
I-
a:
a::
Wo
CLU1
::E •
O
W
I-
U1
~
0
!!I
jg I -.l [!ll
-9.0 -4.5 0.0 4.5 9.0 13.5 18.0

Y AXIS IN CM
-257-
o
O~---T----~-----r-----r----'-----,-----.-----r----.-----'
r--
o
o RUN JULY 3
a
<D RICHRROSON NUMB.=O.006
'- b r /D=O.125(X/O +4.35)
....
.D8
.~
:c
f-
a
...... 0
3:0
u..::l'
~
a:
:Co
_1 C!
a: co
z
D
.......
(flo
z~
WN
::L
a
o
zo
D"":
z
~O~ __ ~. ____ ~ ____- L_ _ _ _- L_ _ _ _ ~ ____ ~ ____ ~ ____ ~ __ ~ ____ ~
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 40.0
N~N DIMENSI~NRL DISTRNCE FR~M EXIT. X/D
o
NON DIMENSIONRL MERN TEMPERRTURE PROFILE
N r - - - - , - - - - - , - - - - - , - - - - - , - - - - I r - - - - , - - - - - , - - - - - , -__- ,____-,
C!
C!
.,;..,
. . RUN JULY 3
RICH. NUrlB.=0.OO6
X'=X+4.9
EXP( -49 71 2 )
0
~ EXPERIM. RESULTS
0
X/D SYMBOL
6.0 [!)
:EO llJ.O (!)
If-~ 22.0
'-0
If- 30.2
'"
+
36.0 X
0
=':
C)
a
"!
0
. '"0
x+
I
-0.14 -0.07 0.0 0.07 0.14 0.35
71=Y/X'
o
o
~-r----~----~-----~--~----~----~-----~r
.-1
o
o
•
C\J RUN .,JUNE 6
---
o RICH. NUMB.=O.031
ou~ X/O=10.0
o
z--- EXPERltoENT (!]
........ 0
o
woo•
a: I
N
,::)0 V1
1-0 00
cr·
a:<D
I
Wo
0...... 0
L
W:::!'
•
I-
o
o
•
C\J
o
o lc [!J (!II ttl mJ...Zr I ...!2t:::::. '" Y I ~ ~ I
-ij .0 - 3 .0 - 2 .0 -1 .0 0 .0 1 •a 2 .0 3 .0 lL 0
Y AXIS IN eM
-259-
~ ____~____~____~____~r-____~__- - '
o•
CD
..-4
en
0
0
. EI
Lf)
•
lO CO
II
.
UJ
Z
:)
L
:::>0 z
Z
0
. ~....
-, ..... o
J:
• ('\J
II a: •
Z U Cl UJ en
:::> ...... "- 0-
><
a:: a:: x UJ
-
Lf)
W •
-
...J
IJ....
o
a::
(L
w
a::
:::>
I-
a:
a::
w
(L
~
W El
I-
Z
a:
w
~
w
(f) Lf)
0::: :jf
•
W I
>
(f)
z 0
a: •
a:: CD
I
I-
0~6 OS·L OO·S OS·n OO-E OS-! 0·0
Jo NI 38nl~83dW31
-260-
o
~----~----~----~----~------r-----' CO ·
.-I
('(')
0
0
II
· El
o
•
CD CD
~
·
UJ:::JO
ZZ .z
a ~
I-
:::!
--,
J: /I
· ('(')
....a:
LLJ
o
::::f4
•
ZUO 0...
:::J-'
a::: a::: x
x
LLJ
o
I.LJ •
N
-.J
t-1 ::E
I.L U
0
a:: z:
CL -.t
o
I.LJ
a:::
o(f) -.t
::J X
I-
a: a:
a:: o
I.LJ N• >-
D- !
~
I.LJ
!-
Z
o
•
a: ::::f4
I
UJ
~
I.LJ
~
(f) o
a:
I.LJ CD
I
·
>
(f)
z I
a:
a::
I- I~----~I----~I----~I----~~----~--~I~
00·9 00· SOD· fi 00· € 00· (; 00· T O· 0
Jo NI 38nl~83dW31
CJ
CJ
~-r----~-----~-----~--~----~----~-----~--~~
~~ (!]
RUN JUNE 01
m~ RICH. NUMB.=0.031
aU CJ
z~
I
. - . ('I')
r ~
X/D=50.0
EXPERlt£NT (!]
/ \
I
W N
a: (J'\
.....
I
::::>0
~CJ
a:.
a: N
W
CL
LO
WCJ
~
....... .
CJ
•
0
-12.0 -9.0 -S.O -3.0 0.0 3.0 6.0 9.0 12.0
Y AXIS IN eM
CJ
TRANSVERSE MEAN TEMPERATURE PR~FILE
°:::r: r- 1
o
RUN JUNE 6
o RICH. NUMB.=0.031
ou~ X/O=60.0
z EXPERIMENT (!)
~
WCJ
cr: O I
:::J" tv
0"
I-C\I N
I
a:
cr:
w
a....
~o
CJ
W " ...-
~--
-16.0 -12.0 -8.0 -ij.O 0.0 ij.O 8.0 12.0 16.0

Y AXIS IN eM
-263-
a
__ --~----~----~----~----~--~r_--~----_r----~O
r--
·
b
(Y)
CD
·
~o
CD,
lJ>x
a • I--•
a> .......
::J'x
W
OL
· 0
Nee
::J'u..
W
au
'Z
~a:
I--
(f)
~
~O
en " co
N.-1
O('r)
• • a:
a en Z
:r:
..-. II 00
Cl
.......
• +
III Cl
~,
~
-(f)
·
,::)x NZ
~
Z '-J W
lJ..
(£)ZN
o L
-1 0"'-'
a: o- .0
:c W
Z Cl 0
(f) • ::J'
-Z
:::>0::11 o
w -')a:Cl
cr:: :z:, Z
:::> Z u....... ;-. o
I-
a:
a::
,::)
0::0::..0 ·
w
a....
L o
W-
I ~ __
~ ____ ~ ____~ ____
~ ____ ~ ____
~ __
~ ____
~ __ ~~O
·
00' 6 00' BOO' L. 00' 9 00· 5 00' f1 00· £ 00· ~ 00· t O· a
O/lq 'HIOIM ~l~H l~NQISN3WIO NQN
o N~N OIMENSI~NAL MEAN TEMPERATURE PR~FILE
~
.-.·~1--~~--~--~----~--~----~--~--~----~--~4
o
o RUN JUNE 6
·
.....
RICH. NUMB.=0.03
o
X'=X+3.33
CO
o · EXP( -497]2)
EXPERIM. RESULTS
::EO
11- (£)
X/O SYMI30L
I
o
"
U_ 10.0
20.0
(!J
(!)
N
cr-
-i'-
I
30.0 ~
o
50.0 +
=:I'
o · 60.0 X
o
~
·
~
o
o
oL I ... !!W::@ (!)
-0.5&0.40 -0.30 -0.20 -0.10 -0.00 0.10 0.30 0.40 0.50

'iJ = YIX'
-265-
TRRNSVERSE t1EFifj ITt~PERR fURE PROF I LE

o
: ,---- . - - - - 1 - - - . - -
RUN JULY 25
RICH. NUMB. =0.03
o
o X/D=1l1.0
CD
[x~E.HH1ENr [!)
u
o
z
W~I
a:~
:::J~
I-
/
a:
a:
W
(L
:L
WO
I-~
N
~
OL-_ _~~~~_ _ _l -_ _-L_ _ _~I_ _ _~~~-L~_~
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Y RXIS IN CM
o
o
:1.-----.----.-----r----r--RI-UN-.-J-UL-Y-r2-5-~
~ RICH. NUMB. =0.03

X/D=22.0
E.XPEA I ~IENr [!)
z
WO
a: o
:::Jc0
I-
a:
a:
Wo
0....0
L .
N
W
I-
o
o
(!)
~ (!)
o '-e---_ _.....=._ _ _ ----L_____ L- ____ L_._--'-m~~alll_____-'
-12.0 -8.0 -4.0 0.0 4.0 8.0 12.0
Y RXIS IN CM
-266-
,~,~:. _JL y 25
R!~H. NUMB. =0.03
110=30.2
fXPfRlMENT [!J
~
o-'-------'-eF='----'----~___.L. _ _ __._L __
· _--L_-=~J...-._ _.....
-16.0 -12.0 -8.0 -4.0 0.0 4.0 8.0 12.0 16.0
Y RXIS IN eM
o
o
~r
RUN JULY 25 l
~~
RICH. NUMB. =0.03
X/D=36.0
EXPEA I MENT [!J
u
o
z
-267-
o
o~---'r---~----~-----r-----r-----r-----.---
c-
o RUN JULY 25
o
o
ID RICHRRDSON NUMB.=O.03
....... b T ID=O.123( X/D +2.3)
....
.D8
-vi
::c
I--
o
...... 0
3:0
u...::l'
--.J
(I
~
-'-0
--.J~
(Ie<)
Z
~
.......
(f)0
zC?
WN
~
o
o
zo
~....:
z
C?
o~ ____~__~'____-Jt____~____~____~ ____~____~____~____~
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0
NON DIMENSIONAL DISTRNCE FROM EXIT, X/O
o
NDN DIMENSIONRL MERN TEMPERRTURE PROFILE
N~----.----'r----.-----.-----.-----.----~-----r-----.----,
RUN JULY 25
o
a
R1CH. NUMS.=0.03
X·=X+2.3
EXP( -46 17 2 )
0
~ EXPERIM. RESULTS
a
X/O SYMBOL
6.0 (!J
:;:0 11/.0 (!)
If-- ~ 22.0 A>
'-0
II-- 30.2 +
36.0 X
0
=:r:
0
0
"! C'l
0
..
+
i!l
~
o \..-=~t-=.._---l____.-L I I ___ L ____ ._l_x_(!)~._~_..
-0.35-0.28 -0.21 -0.14 -0.07 0.0 0.07 0.14 0.21 0.28 0.35
Ti=Y /X'
-268-
0
TRRNSVERSE MERN TEMPERRTURE PROfILE
~
0
RUN JULY 10
0
Cl RICH. NUMB.=O.07
IX>
X/D=6.0
~
U
0
0
Z~
....... (D
~J
a:o.
a:
/
w::1'
CL
L
W
~
0
~
N
~
0
-5.0 -3.8 -2.5 -1.3 0.0 1.3 5.0
'( AXIS IN CM
o TRANSVERSE MEAN TEMPERATURE PROFILE

o
~I
~r RUN JULY 10
RICH. NUMb.=O.07
X/D=j4.0
[Xf'ER!HENf ~
U8f'
o
Z
~
.......
w8
a: .,L
::JCTl
~J
cr~1
LN
w
~
~J~ -8.0 -6.0 -4.0 -2.0

I
0.0
Y AXIS IN CM
2.0 4.0 6.0 8.0
C)
~~i------~----~------~------~------~----_.
=t'
RUN JULY 10
RICH. NUMB.=0.07
.
C)
C)
X/O=22.0
('I')
EXFERIMENT [!]
u
o
z
>--j
I
N
wC)
'"
\.0
I
a:~
::IN
I-
eI
a:
w
D-
L:
WC)
I-C)
......
oC) ..........- E9 L.:J !:!l~

-12.0 -8.0 -4.0 0.0 4.0 8.0 12.0
Y RXIS IN eM
-270-
o
TRRNSVERSE MERN TEMPERRTURE PROFlLE
0 r -_ _ _ _. -_ _ _ _~------._----~--
"" --r
RUN JULY 10
RICH. NUMB.=0.07
o
o X/D=30.2
(Y)
EXf'ERlMENf I!J
I!l
u
o
wg
a: .
::IN
f-
a:
a:
w
CL
:::;:
wo
I-~
~
O~ ____ ~~~~ ______L __ _ _ __ L_ _ _ _ ~ ____ ~~~~_L ____ ~
-16.0 -12.0 -8.0 -4.0 0.0 4.0 8.0 12.0 16.0

Y RXIS IN CM
o
o
00 I..-------r-----.---.--------r----.---------,
TRRNSVERSE MEAN TEMPERATURE PROFlLE
RUN JULY 10
o
=t' RICH. NUMB.=0.07
N
X/D=36.0
w
a:
::J
1-0
a:C'J.
a:
w-
CL
:::;:
W
I-
o
<D
o
I!l
~ I!l
o~______~______~. ----~I________~~--.--~.~------J
-18.0 -12.0 -6.0 0.0 6.0 12.0 18.0
Y AXIS IN CM
-271-
o
TEMPERRTURE HRLF WIOTH
o~ __- ,____ ~ ____ ~ ____ _____
~ ~ ____ ~ __ ~~ __ ~ ____ ~ ____ ~
r--
8 RUN JULY 10
w RICHRROSON NUMB.=0.063
~ b T /O",0.118( x/a +3.3)
~
..t:i8
-u:
::x:
...-
o
...... 0
~C?
1.1.-:::1'
..J
ex:
:Co
..JC?
a:(Y')
z
D
.....
(/)0
zC?
WN
~
o
o
zo
DW-:
Z
C?
O~ __ ~ ____ ~ ____- L_ _ _ _ ~ ____ ~ _ _ _ _L -__ ~L- __ ~ ____- L____ ~
0.0 11.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0
NON DIMENSIONAL OISTANCE FROM EXIT. X/O
~N DIMENSI~NRL MEAN TEMPERATURE PR~FILE

0~__~____~____- r____, -____~____r -__ -.r-__-.____-r____1
:r :~CH~~~~.O.O' ~
X'=X+3.3
EXP( -lj9. 76 )
o EXPERIH. RESULTS
~
o
x/a SYMBOL
6.0 !!l
111.0 (!)
22.0 ....
30.2 +
36.0 X
o
N
o
.
o
o~~~. __ L-~l- __~__~__~__~~~~~~~
-0,35-0.26 -0.21 -0.14 -0.07 0.0 I 0.07 0.35
'fIX
-272-
0
TRR~JSVERSE MERN TEM~ERRTURE PR~FILE
~ ,----
=1' I
M
~
co
\ RUN PLVTI-E
RICH.
XlD"9.0
t~.mB."0.17
\
EXPERIMENT [!)
t-
w
ou •
llJg
N
i:r:--;
::IN
I-
eI:
1
a:
WCO
o...~
::E-
W
l-
t-
~
0
0
0 I
-6.0 -!405 -3.0 -1.5 0.0 1.5 3.0 4.5 6.0
'( RXIS IN eM
o
~ r - - I----r---.------r----.----,---,--,
N RUN PLVTl-O
Ol
N RICH. NUHB.=O.173
XlO=l'!.O
EXPERIMENT [!)
u("/")
o ~
N
Z
WlJ)
a:c-;-
::J_
I-
a:
a:
Wt-
0... .~
::E •
w-
I-
to
lJ)
o~ L--_~=- __-,-- __L _ _L_------L-------1. _ _-=:lO=-_----l

-8.0 -6.0 -4.0 -2.0 0.0 2.-0 ll.O 6.0 8.0
'( RXIS IN eM
-273-
o
TRRNSVERSE MERN TEMPERRTURE PRnFILE
~r-----.------r-----.------r-----.------r----~----~
('\j
RUN PLVTI-Z
o
o RICH. NUH6.~.17
('\j
X/O=18.0
EXfERII'fNT I!J
o
u
o
z~
...........
o
Lf)
~
o '--___ ~= __-J...._.
-10.0 -7.5 -s.O -2.5 0.0 2.5 5.0 7.S 10.0

Y AXIS IN CM
o
TRR~SVERSE MEAN TEMPERATURE PR~FILE
Lf)
NI~~~~~~~~~
RUN PLVfI-C
o
o RICH. NUHB.;Q.16B
('\j
X/O=23.S
EXPERII'fNT I!l
u
o
o
z~
..........
o
Lf)
o
_ 0 __ .L ___- L_ _ _ _ _ _.L...-_ _ _ ~ _ _ _ ___'__ ___":='O_ ____'
-12.0 -9.0 -6.0 -3.0 0.0 3.0 6.0 9.0 12.0
Y AXIS IN CM
-274-
TRRNSVERSE MERN TEMPErlRTURE PROFILE
0
=':
C!l
-,-
0
~ RUN PLYTI-B
RICH. NUHB.;O.17
0 X/O;33.6
0
EXf'ERII'£NT C!J
u-
0
20
>-1 (0
O
W
cr:.
::::>0
r-w
IT •
cr:.o
w
CL
LO
W'::I
f -0'
0
~
0
0
0
-18.0 -13.5 -9.0 -4.5 0.0 4.5 9.0 13.5 18.0
Y AXIS IN CM
o
'::I ______~----_r-----~------r_---
r- --r-
o
N
RUN PLYTI-H
RICH. NUH8.=O.182
o X/O;39.0
o
-EXf'ERII1:NT [!J
u'"
o
ZO
~-4~
o
W
a:
::::>0
r-W
cr'
o
a:
W
§:o
LLl =':
f-O
o
(\I
oC? ____L- '--___ -L. ___L::::=-'-____---'

-20.0 -15.0 --10.0 -5.0 0.0 5.0 10.0 15.0 20.0
Y AXIS IN CM
-275-
a
0 r - - - - .____~-- __-r----~----~----~--~1---
c-
g RUN PLVT1
OW RICHRROSON NUMB.=0.17
" bT 10=0.120/0 +2.1)
"'0
.Do
.~
I
J-
o
....... 0
3: C! _
l1...~
.-J
CI
:r:o
o
.-J •
CI(I')
Z
o
.......
(1)0
zC!
WN
::E
o
o
zo
0"':
z
<:)
<:) ' - -_ _- - ' -_ _ _ _--1-_ _ _ _- ' -_ _ _- - ' -_ _ _J -_ _ I I
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0
NON DIMENSIONRL DISTRNCE FROM EXIT, x/a
a
NON DIMENSIONRL MEAN TEMPERRTURE PROFILE
::1"'---------.-----.-----.---------.----.-----.-,--,---,-----,
RUN PLVTl
ao
RICH. MJMB. =0 • 17
X'=X1"S.S
EXP( -57 T/ 2 )
EXPERIM. RESULTS
X/O SYMBOL
9.0 (!]
IIJ.O (9
19.0 A>
23.S +
28.8 X
33.S ~
o 39.1 +
:I'
o.
a
N
o
o I!I
a I..-_==--_--l __-'-_~ __ L ____ _'_I_ _ __l I
-0,35-0.28 -0.21 -O.lli -0.070.0 0.07 0.114 0.21 0.28 0.35
T/=-Y/X'
o
TRANSVERSE MERN TEMPERATURE PROFILE
o
~~I------~------~----~------~------~-----~
RUN JUNE ;~8

o
RICH. NUMB.=O.216
o X/D=22.0
('/')
EXPERIMENT I!l
oU
z
..........
wo
a:~ I
=>N IV
-...J
I- 0'\
CI I
a:
W
CL
~
Wo
1-0
o
o I aJ.L.::!7 - - ;;t=:t+""
-12 .0 -8 .0 -1.1. 0 . 0 .0 l4 .0 8 .0 12 .0
Y AXIS IN eM
-277-
0
TRR~JSIJERSE l1ERN TEMPERRTURF.: PROFILE
c: I--
~
0
c: RUN JULY
RICH. NUMB.=0.22
° X/o=6.0
EXFEAll"ENf I!l
o<-J 0 °
0)
Z
I-i
WO
r.- 0
::Jw
.-
cr
CC
Wo
D-o
~.
w::l'
.-
0
N
c:
c:
0
-4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0
Y AXIS IN eM

c:°
<0
°c: RUN JULY 1

In RlCH. NUMB.=0.22
X/o=14.0
EXFEAIMENf I!l
oU °O.
::l'
Z
I-i
UJ8
CC •
::JCY'l
~
cc O
W
D-
~('J
0
r
0
c:
- c: 0
(!J
I --.l
-B.o -6.0 -4.0 -~.O 0.0 2.0 4.0 6.0 B.O
Y AXIS IN eM
-278-
o
0 , -_ _ _ _ _ _- ,_ _ _ _ _ _ _ _, -_ _ _ _ _ _- ,_ _ _ _ _ _ _ _, -_ _ _ _ _ _- ,_ _ _ _ _ _- - ,
;1 RUN JULY 1
o
RICH. NUMB.=0.22
o X/0=22.0
C')
EXFERIMENT [!J
U
o
z
<=!
0~ _____~~~------L--------L--------L-----~~~~----~
-12.0 -8.0 -4.0 0.0 4.0 8.0 12.0
Y RXIS IN CM
o
U"l
~~I--~--~--~--~--~--~---~~
o
o
C')
RUN JULY
~ RICH. NUMB.=0.22
I!l
o X/0=30.2
U"l
EXPERIMENT [!J
UN
o
o
U"l
'?
t!l
o
aL 1 -==--=BLI____- i_ _ _ _ _ _..L1 _____--<-I______'---~=_-'-_ _ _ ___'
-16.0 -12.0 -8.0 -4.0 0.0 4.0 8.0 12.0 16.0
Y RXIS IN CM
-279-
0
co
..-<
·
N
N
. If)
,...... 0
.
II
EJ ·
('I')
.......
ro
>- L
-1 ::J 0
::J
....., Z .z I-
W
(D
....
~
0
:r: /I a:
z u CJ Q...
"
w
::J ....... "'- x
(Y')
en ·
a: a:x w E1
E1
If)
W
-1
::1'·
...... ::E
LL U
0
a: z
CL >--1
0
W O(f)
a:
:::)
>--1
X
I- a:
a: If)
a: · >-
w ::1'
CL I
::E:
W
I-
0
Z
a:
w
en
I
·
L
W
en If)
a:
--
('I')
w
> I
en 0
,co.....·
Z
a:
a: I I I
l- I
00"£ OS·~ OO·~ OS·1 00·1 OS·O 0·0
J O NI 3tlnltltEldW31
-280-
o
fEMPERRTURE HRLF WIDTH
o
~I I
g RUN JULY
-w RI CHHROSOr J NUr113. = :J .;;:
o
'-. bT /O=O.12ieX/O +2.87)
.00
...
a
.~
I
r-
o
~o
3:0
LL ~I
~
Io
I
-1~
0:«)
Z
o
.-.
(flo
Z~
W'"
:E
o
o
zo
0"':
z
~
o '--_--'-_ _---'-_ _--'-_ _ _-'--
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0
NON DIMENSIONRL DISTRNCE FROM EXIT, X/D
MERN TEMPERRTURE PROFILE
RUN JULY 1
o
a
ff\+~~
,"1 CH. NUMB. "'0.22
X·=X+2.87
EXPC -47.071 2 )
o x + +
~I
EXPERIM. RESULTS
X/O
'\
SHIBClL
x 6.0
14.0
If-::Oaf
~ _
22.0
'-.0
30.2 +
i~-
.!x
+I T\ 35.0 x
t "\
a
f +\
a
'"a
~
a
~/
/
..
=:.::dr~----L ___ J
J
~
A
____ ~ ___ ~---1 _____ L -__ L

"~ .~~
",(!) ~
__ ~}l~="
-0.35-0.28 -:J.2i -O.ill -0.07 0.0 0.07 0.14 0.21 0.28 0.35
71'::YIX'
-281-
Q
0r--_--.-_
C1)
o RUN JUNE 26
1Il
I' RICH. NUMB.=0.295
X/D=6.0
ExrERl~NT (!J
o
ug.
to
wo
a:~
::)::.1"
I-
a:
a:
Wo
0...0
::iC: •
W('<)
1-
o
1Il
....
o
0~ __~~~__L -_____ L-____~____- L____-L~~~~~~
-8.0 -6.0 -~.O -2.0 0.0 2.0 ~.O 6.0 8.0
Y AXIS IN CM
o TRANSVERSE MERN TEMPERATURE PR~FILE

C!
I'
0
~ RUN JUNE 26
<D
RICH. NU~IB .=0 .295
0 X/O=lO .0
~ lXI't:RI MENT (!J
Olll
0
Zo
.-.-..~
::l'
tU
a:
:::>0
1--0
a:'
a::(T)
w
0... 0
~o
w·
I-- C\I
0
~
c:! _ _ ~I
0 I
-12.0 -9.0 -6.0 -3.0 0.0 3.0 9.0 12.0
Y AXIS IN CM
-282-
0
TRRNSVERSE MERN TEMPERRTURE PROfILE
C? --
Lf) ----r I
0
RUN JUt...'E 26
C? R1CH. NUMB.=0.29S
::J"
X/O=14.0
[XPERlrENT Q1
U
0
0
zo
t--I~-
W
a:
=>
~
(L0
a:C?
WN
Q..
::L
W
~
0
C?
~
~
0 I
-16.0 -12.0 -8.0 -4.0 0.0 4.0 12.0 16.0
Y AXIS IN CM
o
o
:l' RUN JUNE 26

R1CH. NUMB.=O.295
~~~
X/O=18.0
EXf'ER I i1ENT Q1
llJ
a:
=>
~
a: 0
a:C?-
WN
Q..
::L
W
~
o
o
o
O~ ____ ~-C~-L ____- L____ ~~ ____L -____ ~~~~ ____ ~
-20.0 -1S.0 -10.0 -S.O 0.0 5.0 10.0 1S.0 20.0

Y AXIS IN CM
-283-
TENPERflTURE tiRLf W10TH
g~--~----~----r----.----~-- --~
o
IJ)
en
RUN JUNE 26
R1CIiRROSON NUr·lB. '" ~ .295
b r /O=O.113( X/O +4.0)
::c
...-0
01J)
>-<N
3::
~g
~ C'~
-.J
a:o
zLi"!
0_
.-.
(f)
C'Jg.
~
.-. .....
o
zo
0Li"!
20
o
O~ __ ~ ____- L_ _ _ _ ~ ____ ~ _ _ _ _L -_ _ ~L- __ ~ _ _ _ _- L_ _ _ _ ~ __ ~
0.0 3.0 6.0 9.0 12.0 15.0 IB.O 21.0 2~.0 27.0 30.0
N~N DIMENS1~NRL DISTRNCE FROM EXIT. x/a
o N~N 01 MENS I ~NAL MEAN TEMPERATURE P[10FILE

N
';1
0
RUN JUNE 26
~ RICH. NUMB.=0.295
X'=X+Y.O
EXP( -51! 7(2) - -
0
~ EXPERIM. RESULTS
0
x/a SYMBOL
6.0 [!)
::!;o 10.0 ~
I"'-~ 111.0 A>
,...-
'-0
16.0 +
0
::2'
0
0
~ +
0
0
. +
+
0 .L ! I
-0.35-0.2B -0.21 -0.14 -0.07 0.0 0.07 0.14 0.21 0.28 0.35
T/='(/X'
-284-
a
TRRNSVERSE MfRN TE~PERRTURE PROFILE
~ --
~
I
a
RUN JULY 29
~ RICH. NUMB.=0.424
(0
X/O=6.0
EXPER I I'£NT (!J
U
0
a
z~
1--<<.0
W
a:
::)
~
CIa
a:~
W::l'
CL
L
W
~
a
~
""
~
.a
-4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0
Y AXIS IN eM
a
TRANSVERSE MEAN TEMPERATURE PROFfLE
0r-----~----_r-----.----_,r_----._----~-----_.----_,
~f RUN JULY 29
RICH. NUMB.=0.424
X/O=14.0
EXPEfll~lENr (!]
O~~r
~(1')
w
§ \
~~~
w""l
CL
L
W
~
a
a I
a
o
-8.0 -6.0
-e--L
-4.0 -2.0 0.0
Y RXIS IN CM
2.0
I 6_-G~
4.0 6.0 8.0
-285-
o
o~ ______ ~ ______ ~ ______ ~ ______- r______- '______- '
""
RUN JULY 29
RICH. NUMB.=O.424
o
a X/O=22.0
M
EXf'ERII'ENT !!l
U
o
z
a
a~ ______ ~~ _____ ~ ______- L_ _ _ _ _ _- L_ _ ~~~ ______ ~
-12.0 -8.0 -4.0 0.0 4.0 8.0 12.0

Y AXIS IN eM
0
TRRNSVERSE MERN TEMPERATURE PROFILE
~
M
0
III RUN JULY 29
N RICH. NUM8.=O.424
X/O=30.2
O (XPERI~ENT !!l
U
o ~
Z
,"
......
wo
a:L1!
::J ...
I-
a:
a=-
Wo
(La
L'
w ....
I-
0
III
0
a
0
-16.0 -12.0 -8.0 -4.0 0.0 4.0 8.0 12.0 16.0
Y AXIS IN eM
o
TRRNSVERSE MERN TEMPERATURE PR~FILE
If)
N 'r--I
o
RUN JULY 29
o RICH. NUMB.=O.42Y
N
X/O=36.0
EXPERIMENT r:!:l
aU
o
Z~
1--1_
I
w N
a: (Xl
::> '"I
I-
CIa
0
0:
W...:
CL
L
W
I-
oIf)
o
o
o ,I I -=== ____ I
-18.0 -13.5 -9.0 -4.5 0.0 4.5 9.0 13.5 18.0

Y RXIS IN eM
-287-
o
o~---.-----.-----r----.-----.-----r----.-----.-----r----,
.....
RUN JULY 29
RICHARDSON NUMB.=O.424
b T /0=0. 1 12( x/a +4.42)
o
OL-__~____~____- L_ _ _ _- L_ _ _ _~_ _ _ _L -_ _~~_~' ____~_ _ _ _~
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 36.0 40.0
NON DIMENSIONAL DISTRNCE FROM EXIT, X/O
o NON DIMENSIONRL MERN TEMPERRTURE PROFILE

Nr----.-----.-----r-----.----.-----r----,r----.-----r----~
.....
o RUN JULY 29
o
... RICH. NUM8.=0.424
X·=X+4.42
EXP( - 56 77 2 )
o
00
o
EXPERIN. RESULTS
x/a SYMBOL
6.0 (!J
14.0 (!)
22.0 <I>
30.2 +
36.0 X
o
::1'
o
o
N
o
o
o I t!l
-0.35-0.28 -0.21 -0.i4 -0.070.0 0.28 0.35
YIX'
-288-
APPENDIX C
TEST OF THE TWO-DIMENSIONALITY OF THE FLOW
In this study, the two-dimensional jet (see Section 3.4) emerged
out of a slot 13 cm long and D cm wide (D varied from 0.24 to 2.0 cm).
The two-dimensionality of the emerging jet at x = 0 was determined by

measuring the velocity ua at the jet orifice, which was found to be in
good agreement (variations of the order 1-2%) with the mean exit velocity
calculated by the known volume flux ~o' i.e.
where W is the distance between the two confining walls. In the
following discussion, the Cartesian system of coordinates (x,y,z) will
be considered with an origin in the plane midway between the two confining
walls (see Figures 2.1.1 and 3.4.l.a).
It was considered that the two confining walls, located at
z = ± 6.5 cm, would serve two purposes. First, growth of the jet in the
z-direction would be prevented, second, the two-dimensionality of the
induced flow would be enhanced. Using dye tracers, it was found that
the induced flow was approximately two-dimensional. The flow at the
entrance of the two confining walls was found to be smooth and no
separation effects were noticed.
The important parameter in testing the two-dimensionality of the
jet is the ratio of the distance from the jet orifice X over the distance
W between the two confining walls. The jet loses its two-dimensionality
-289-
for large values of the ratio X/W. For the run PLVT2, the ratio X/W
was
O. 38 < X/W < 1. 7,
and for the run July I
0.46 < X/W< 2.76.
Two experiments were run to test the two-dimensionality of the buoyant
jet by taking measurements at different positions Y in the Z direction
(parallel to the slot). The mean temperature T(x,y,z) (see Figures
C.I to C.4), the intensity of turbulent fluctuations of temperature
T'2(x,y,z) (Figures C.S and C.6), the intermittency y(x,y,z) (see Figure
C.7), the maximum and minimum temperature (see Figure C.8) and the
frequency of the hot/cold interface crossing (see Figure C.9) were
measured in a buoyant jet. We can observe that even for the largest
value of X/W used in this study (X/W = 3) the flow can be considered
two-dimensional over the central one-half of the distance between the
two confining walls. The only exception appears to be in the frequency
fR of the hot/cold interface crossings, but we should note that the
experimental determination of fR using several probes is very delicate.
On the other hand, the two-dimensionality of the large scale motions of
the jet can be concluded from the Figures A.I and A.2 of the Appendix A.
The profile of the induced flow is shown in Figure C.IO. It can be
seen that boundary layers have developed. This suggests why the jet
temperature close to the wall is higher than the temperature over the
central region between the two confining walls (see Figure C.2).
-. ·---r~---l
RUN JUNE 21
Ro = 0.6
X/W = 2.35
Y/X SYMBOL'
0.016 0
0.049 8. -
0.116 r::J
0.143 V
10 . 21 -
X _.
4.0
I
.0 I
u
0
I N
I '"oI
~: ~~
c 3.0
I
11-
I
2.0 I
I
I
1.0
I
..-g
I
~ oK a I
I
WALL~ ~_WALL
01.- I 11 U I
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Z. in em
Figure C.l. Distribution of mean temperature in a direction parallel
to the slot.
r--- .. - - r - - - 1---- 1··'-
RUN JUNE 22"""""

Ro=0.33
X/W=3.0
Y/X SYMBOL
0.0 (;)
0.08 l!!.
0.14 []
I
7.0 N
u \0
0 I-'
I
.: 6.0
...
CI)
-:::l
:2
CI)
a.
5.0
(;) (;)
E 4.0
CI)
~
A
-.:;:;.."", A /!., 4li ~
.
~ 3.0
I~
2.0 -e- fJ D D 0-
WALL WALL
~/
1, __. .
1.0
0 I __ L.
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Z. in em
Figure C.2. Distribution of mean temperature in a direction parallel
to the slot.
I-----r-~
RUN JUNE 2.2

Ro = 0.33
Y/X = 0.0
X/W SYMBOL
1.6 0
2.0 0
2.4 8.
3.0__ \-l
---
--
7.0
u
0
~ 0 .t:>--- 1
N
.= 6.0
O 0 0 '1 s-
\.0
N
...
dI
1
-
::l
...0
dI
5.0
~
A A
A
.
6o-b--
'0/
0-
E 4.0 - V 17 :a ~
dI
I--
o 3.0
II--
2.0 I
WALL I WALL
1/ V'
1.0 I n
~ ,
0 I I I
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Z, in em

to the slot.
r---r--
RUN JUNE 22
Y IW = 2.25
Y/X SYMBOL
0.0 . 0
0.005 l::.
0.08 I:l
7.0 I
u
0
N
~
~ 6.0 w
I
~
:::l
0
~
Q)
5.0 &- til g g B
a.
E 4.0
Q)
I- c:r- 8 8 B
I:l
o 3.0
II-
2.0
1.0
0 . L ___ L _____ .1
-s.O -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Z, in em
to the slot.
RUN JUNE 21
Ro = 0.6
X/W=;~.35
Y/X SYMBOL
0.016 0
--
0.049 6.
0.116 0
- -
'0.143 V
0.21 X
1.4
:=:;
I
1.2
1.0
x 0 0
i. e-
x.-/
N
'!)
-i'"-
I
u
o
c: 0.8
I " - -x X 1
~ 06
I I
0.4
I I
I
__
I
0.21- I~ALL WI~LL
I1..,/'
OL
_
L-~I~__L-_~__.~__=-~~~~r_~I~ ~
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Z. in em
Figure C.S. Distribution of turbulence intensity in a direction

parallel to the slot.
T--T---
RUN JUNE~
Ro = 0.33
Y/X = 0
XI W SYMI30L
1.6 0
2.4 8.
[;)
3.0.
I
N
\D
Ln
3.0 I
---8 A A A b--
~ 2.0 o 0 0 0 [;)
c:
. t.) 0 0 0 0
II~
~ 1.0
OL-.~ ______ ~ ____ ~ ____ ~ ______ ~ ____ ~ ______ ~ ____ ~ ____ ~ ________ ~
'-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Z, in em
Figure C.6. Distribution of turbulence intensity in a direction parallel

to the slot.
.- T r---- ,---
RUN JUNE :~2
Ro = 0.33
X/W = 3.0
y/ X SYMBOL
0.0 0
1.0 0.08 A
o 0 0 (;) 0--- 0.14
-------
0
-_._----
0.9
A A 8: /J:J)--~--
A
0.8
0.7
~ I
[J N
>-
u
0.6 ------------~[J~.----~[J D 0----- '"
0'\
I
z
W 0.5
~
~
::?: 0.4
0::
W
~
z 0.3
0.2
0.1 ~ALL j,-~ALL
0 I__L __ I _ 1
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Z. in em
Figure C.7. Distribution of intermittency in a direction parallel

to the slot.
r- ----r --I --- (
12.0
RUN JUNE~
• s· • Ro = 0.33
Y/X=O !
• x/w
10.0
•
•
•
1.6
2.0
M"T"T
•
a
0
0
V
8.0
• • • 2.4
3.0
"Y
A- I 6.
u
o
V •
c
- 6.0
---
••
y
•• N
I
\0
-...J
I
I-
Z
:::!!
• 4.0
I-
x
<{
:::!!
2.0
~ALL
aI III
R /Sl
.
8( R
I
r
~ ALL
(
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Z, in em
Figure C.8. Distribution of maximum and mlnlmum temperature (see Section

4.3) in a direction parallel to slot.
RU~ JUS
Ro - 0.3.3
X/W = 3.0
Y/X SYMBOL
0.0 0
0.08 t:.
0.14 [)
0.4
- 1::1 I
tv
0.3~ 1::1 \0
(1:j
0 1::1 I
,.... t:.
N
:c
c: 0.2
--
a:
0.1
o
OLI__ ~ ________-L________ ~ ______ ~~ ______ ~ ________ ~ ________L -______ ~ ________ ~ __________ ~
-8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0
Z. in em
Figure C.9. Distribution of fR (see Section 4.5) in a direction

parallel to the srot.
-299-
z
4
Slot
Figure C.10. Schematic of the induced flow profile between

the two confining walls.
-300-
REFERENCES
1. Albertson, M. L., Dai, Y. B., Jensen, R. A., and Rouse, H. 1950.
Diffusion of Submerged Jets. Transactions, ASCE, Vol. 115,
pp. 639-664.
2. Bradbury, L. 1965. The Structure of Self-Preserving Turbulent

I')~ __ ')11:..1.
Plane J. Fluid t"1ech., Vol. L.J, pp. JJ..-U't.
3. Ftlrthmann, E. 1934. Uber Turbulente Strahlausbreitung. Ingenier
Archiv., Vol. 5, pp. 42-54 (or translated in English (1936)
NACA, TM-No. 789).
4. Flora, J., and Goldschmidt, V. 1969. Virtual Origins of a Free
Plane Turbulent Jet. AIAA Journal, Vol. 7, pp. 2344-2346.
5. Heskestad, G. 1965. Hot Wire Measurements in a Plane Turbulent
Jet. J. Appl. Mech., pp. 721-734, erratum, Sept. 1966, p. 710.
6. Hestroni, G., Hall, C. W., and Dhanak, A. M. 1965. Momentum
Transfer in Thermally Asymmetric Turbulent Jets. Transactions
ASME, Journal of Heat Transfer, Vol. 87, pp. 429-435.
7. Knystautas, R. 1964. The Turbulent Jet from a Series of Holes in
Line. Aeronaut. Quart., Vol. XV, pp. 1-28.
8. Jenkins, P. E. and Goldschmidt, V. 1973. Mean Temperature and
Velocity in a Plane Turbulent Jet. Transactions ASME, Journal
of Fluids Eng., pp. 581-584.
9. Gartshore, I. 1966. An Experimental Investigation of the Large Eddy
Equilibrium Hypothesis. J. Fluid Mech., Vol. 24, pp. 89-98.
10. Goldschmidt, V., and Eskimazi, S. 1966. Two Phase Turbulent Flow in
a Plane Jet. J. Appl. Mech., pp. 735-747.

-301-
11. Mih, W. C., and Hoopes, J. A. 1972. Mean and Turbulent Velocities
for Plane Jet. J. Hyd. Div., ASCE, Vol. 98, pp. 1274-1294.
12. Miller, D., and Comings, E. 1957. Static Pressure Distribution in
the Free Turbulent Jet. J. Fluid Mech., Vol. 3, pp. 1-16.
13. Reichardt, H. 1942. Gesetzmassikeiten der Freien Turbulenz. VDI
Forschung, Vol. 13, pp. 1-22.
14. Vulis, L. A., Kelmanson, I. A., and Platnik, I. B. 1970. Investiga-
tion of the Flow Structure in a Plane Turbulent Jet. Fluid
Dynamics (Translated from Russian), Vol. 5, No.4, pp. 684-687.
15. Vinogradov, Y. V., Gruzdev, V. N. and Talantov, A. V. 1972.
Influence of Initial Turbulence on the Mixing of Plane Iso-
thermal Jets. Soviet Aero., Vol. 15, No.2, pp. 37-40
(Translated from Russian).
16. Van der Hegge Zijnen, B. G. 1958. Measurements of the Distribution
of Heat and Matter in a Plane Turbulent Jet of Air. Appl. Sci.
Res., Vol. 7, pp. 277-292.
17. Van der Hegge Zijnen, B. G. 1958. Measurements of Turbulence in a
Plane Jet of Air by the Diffusion Method and by the Hot-Wire
Method. Appl. Sci. Res., Vol. 7, pp. 292-313.
18. Townsend, A. A. 1956. The structure of Turbulent Shear Flow.
Cambridge University Press.
19. Abraham, G. 1963. Jet Diffusion in Stagnant Ambient Fluid.
Delft Hydraul. Lab. Publ. 29.
20. Morton, B., Taylor, G. I., and Turner, J. S. 1956. Turbulent
gravitational convection from maintained and instantaneous
sources. Proc. Royal Society, Vol. A234, 1056, pp. 1-23.

-302-
21. Rouse, H., Yih, C., and Humphreys, H. 1952. Gravitational Con-
vection from a Boundary Source. Tellus, Vol. 4, pp. 200-210.
22. Fan, L. N., and Brooks, N. H. 1969. Numerical Solutions of Turbu-
lent Buoyant Jet Problems. W. M. Keck Laboratory of Hydraulics
and Water Resources, Report No. KH-R-1S, California Institute
of Technology.
23. List, E. J., and Imberger, J. 1973. Turbulent Entrainment in
Buoyant Jets and Plumes. J. Hyd. Div., ASCE, Vol. 99, pp.
1461-1474.
24. Morton, B. R. 1959. Forced Plumes. J. Fluid Mech., Vol. 5,
pp. 151-163.
25. Kotsovinos, N. E., and List, E. J. The Two-Dimensional Pure Plume.
To be submitted for publication.
26. Ellison, T. H., and Turner, J. S. 1959. Turbulent Entrainment in
Stratified Flows. J. Fluid Mech., Vol. 6, pp. 423-44S.
27. Lee, S. L., and Emmons, H. W. 1961. A Study of Natural Convection
above a Line Fire. J. Fluid Mech., Vol. 11, pp. 353-36S.
28. Taylor, G. I. 1955. Flow Induced by Jets. J. Aero. Sci., Vol. 25,
pp. 464-465.
29. Lippisch, A. 1955. Flow Visualization. Aeronau. Engrg. Rev.,
Vol. 36, pp. 24-32.
30. Sato, H. 1960. The Stability and Transition of a Two-Dimensional
Jet. J. Fluid Mech., Vol. 7, pp. 53-SO.
31. Forstrom, R. J., and Sparrow, E. M. 1967. Experiments on the
Buoyant Plume above a Heated Horizontal Wire. Intern. J. Heat
Mass Transfer, Vol. 10, pp. 321-331.

-303-
32. List, E. J. 1968. A Two-Dimensional Sink in a Density-Stratified
Porous Medium. J. Fluid Mech., Vol. 33, pp. 529-543.
33. Batchelor, G. 1970. An Introduction to Fluid Dynamics. Cambridge
University Press.
34. Hinze, J. 1959. Turbulence. McGraw-Hill Book Company.
35. Tennekes, H., and Lumley, J. 1972. A First Course in Turbulence.
The M.I.T. Press.
36. Werle, H. 1973. Hydrodynamic Flow Visualization. Ann. Rev. Fluid
Mech., Vol. 5, pp. 361-382.
37. Newman, B. 1965. Turbulent Jets and Wakes in a Pressure Gradient.
Symposium on the Fluid Mechanics of Internal Flow, Warren,
Michigan, (Editor G. Sovran).
38. Van der Hegge Zijnen, B. G. 1957. Measurements of the Velocity
Distribution in a Plane Turbulent Jet of Air. Appl. Sci. Res.,
Vol. 7, pp. 256-276.
39. Miller, D. 1957. Static Pressure Gradient in Turbulent Jet Mixing.
Ph.D. Thesis, Purdue University.
40. Hannum, W., and Griffith, W. 1955. On the Intermittency of a Two-
Dimensional Jet. J. Aeronau. Sci., Vol. 22, pp. 202-203.
41. Schorr, A. N., and Gebhart, B. 1970. An Experimental Investigation
of Natural Convection Wakes above a Line Heat Source. Intern.
J. Heat Mass Transfer, Vol. 13, pp. 557-571.
42. Tailland, A., Sunyach, M., and Mathiew, J. 1967. Etude d'un jet
plan. Comptes Rendus Academie des Sciences, Paris, t.264,
Series A, pp. 527-530.

-304-
43. Coles, D. 1962. Interfaces and Intermittency in Turbulent Shear
Flow. Mechanique de la Turbulence, Colloque International,
Marseille, pp. 229-250.
44. Householder, M. K., and Goldschmidt, V. W. 1968. Turbulent
Diffusion of Small Particles in a Two-Dimensional Free Jet.
TR FMTR 68-3, National Science Foundation, (reference in Flora
and Goldschmidt [4]).
45. Tollmien, W. 1926. Die Berechung turbulenter Ausbreitungsvorgange.
ZAMM 6, pp. 468-478, (NACA TM 1085, 1945).
46. Taylor, G. I. 1932. The Transport of Vorticity and Heat Through
Fluids in Turbulent Motion. Proc. Roy. Soc. London, A135,
pp. 685-702.
47. Schlichting, H. 1960. Boundary Layer Theory. McGraw-Hill, Fourth
edition.
48. Saffman, P. G. 1970. A Model of Inhomogeneous Turbulent Flow.
Proc. Roy. Soc. London, A3l7, pp. 417-433.
49. Laufer, J. 1975. New Trends in Experimental Turbulence Research.
Ann. Rev. Fluid Mech., Vol. 7, pp. 307-326.
50. Coles, D. 1975. Private Communication.
51. Batchelor, G. 1950. Note on Free Turbulent Flows, with Special
Reference to the Two-Dimensional Wake. J. Aeronau. Sci.,
pp. 441-445.
52. Corrsin, S., and Uberoi, M. 1950. Further Experiments on the Flow
and Heat Transfer in a Heated Turbulent Air Jet. NACA Report
998.
-305-
53. Brooks, N. and Koh, R.C.Y. 1965. Discharge of Sewage Effluent
from a Line Source into a Stratified Ocean. XI Congr. Int.
Assoc. Hydraul. Res., Paper No. 2.19.
54. Batchelor, G. K. 1954. Heat Convection and Buoyancy Effects in
Fluids. Quart. J. Roy. Meteor. Soc., Vol. 80, pp. 339-358.
55. Lee Shao-Lin. 1961. Natural Convection above a Line Fire. Ph.D.
Thesis, Harvard University.
56. Magrab, E. B., and Blomquist, D. S. 1971. Measurement of Time
Varying Phenomena: Fundamentals and Applications. New York,
Wiley Interscience.
57. Bicknell, J. 1937. A Study of Turbulent Mixing between a Plane Jet
of Fluid of Various Densities and Still Fluid. Master of Science
Thesis, Massachusetts Institute of Technology.
58. Fan, L. N. 1967. Turbulent Buoyant Jets into Stratified or Flowing
Ambient Fluids. W. M. Keck Laboratory of Hydraulics and Water
Resources, Report No. KH-R-15, California Institute of Technology.
59. Nakaguchi, H. 1961. Jet along a Curved Wall. University of Tokyo
Research Memo 4, (reference in Newman [37]).
60. Koh, R.C.Y., and Brooks, N. H. 1975. Fluid Mechanics of Waste-
Water Disposal in the Ocean. Ann. Rev. Fluid Mech., Vol. 7,
pp. 187-211.
61. Hanum, W. H. 1954. Intermittency of a two-dimensional jet as
investigated with a Hot Wire Anemometer. Master Thesis,
Princeton University.
-306-
62. Anwar, H. 1969. Experiment e1ll an Effluent Dischdrgillg from a Slot:
into Stationary or Slowly Moving Fluid of Greater Densitv.
J. Hyd. Res.. 7, pp. 411-43l.
63. Landau, L., and Lifshitz, E. 1959. Fluid Mechanics. Pergamon
Press.
64. Stevenson, W. 1970. Optical Frequency Shifting by Menns of a
Rotating Diffraction Grating. ~pl. Optics_, Vol. 9, pp. 649-h52.
65. Turner, J. S. 1973. Buoyancy Effects in Fluids. Cambridge
University Press.
66. Sato, H., and Sakao, F. 1964. An Experimental Investigation of the
Stability of a Two-Dimensional Jet at Low Reynolds Numbers.
J. Fluid Mech., Vol. 20, pp. 337-352.
67. Brooks, N. H. 1972. Dispersion in Hydrologic and Coastal Environ-
ments. W. M. Keck Laboretory of Hydraulics and Water Resources,
Report No. KH-R-29, California Institute of Technology.
68. Benjamin, T. B. 1968. Gravity Cllrn:nts and Relnted Phenomena.
]. Fluid Mec~., Vol. 31, pp. 209-248.
69. Monin, A. S. a.nd Yaglom, A. M. 197 1 • Statistical Fluid MechAnics,
Volume 1, The MIT Press.

A Study of The Entrainment and Turb Ulence in A Plane Buoyant Jet

Uploaded by

Copyright:

Available Formats

You might also like

A Study of The Entrainment and Turb Ulence in A Plane Buoyant Jet

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Study of The Entrainment and Turb Ulence in A Plane Buoyant Jet

Uploaded by

Copyright:

Available Formats

A STUDY OF

THE ENTRAINMENT AND TURB ULENCE

W. M. Keck Laboratory of Hydraulics and Water Resources

Report No. KH-R-32 August 1975

IN A PLANE BUOYfu~T JET

W. M. Keck Laboratory of Hydraulics and Water Resources

Report No. KH-R-32 August 1975

I wish to express my deepest appreciation and gratitude to my

advisor, Professor Ericson John List, whose inspired guidance, insight,

perspicacious criticism and enthusiasm were invaluable to this research.

I am greatly indebted to Professors Norman H. Brooks and Donald

Coles, whose advice and interest in this research continuously stimulated

and inspired me. I am grateful to Dr. Robert C. Y. Koh for numerous

ment and friendship during the incipient period of my graduate studies.

I wish also to express my deepest respect and appreciation to Professor

Vito Vanoni for his kind interest and encouragement throughout my

I gratefully acknowledge that the enthusiastic teaching of Professor

ThamrsXanthopoulos provided me with the necessary stimulation to pursue

I thank the California Institute of Technology for providing me with

financial support through Graduate Teaching and Research Assistantships,

and for its superb research environment facilities. Financial assistance

provided by the National Science Foundation, under Grant ENG75-02985, is

invaluable assistance in the construction of the laboratory equipment,

Joan Mathews for her patience in typing this manuscript.

Finally, I thank my wife, Kathy, whose love, support and understanding

were essential during the course of this work.

This report was submitted by the writer in May 1975 as a thesis

with the same title to the California Institute of Technology in

partial fulfillment of the requirements for the degree of Doctor of

The entrainment and mixing processes in a two-dimensional vertical

to a pure plume have been studied. The ambient fluid is of uniform

variations are assumed small.

general conserved. It has been proven that the flow in a two-dimensional

pure jet is not self-preserving.

A systematic set of experiments was carried out to examine turbulent

densimetric Froude numbers). Values of the Richardson number, which

describes the relative importance of buoyancy in a jet, extended from

plume (approximately 0.6). The buoyant jet temperature and velocity

fields were measured using calibrated fast response thermistors and a

laser Doppler velocimeter respectively. The velocity and temperature

data obtained were recorded magnetically in digital form and subsequently

processed to extract both mean and fluctuating values of temperature and

mean velocity and temperature profiles, velocity and temperature distribu-

(including the profile of turbulence intensity and turbulent heat

transfer, probability density distribution of temperature and velocity,

skewness and flatness factor of temperature fluctuations) and the large

scale motions (intermittency, profile of maximum and minimum temperature,

frequency of crossing of hot/cold, cold/hot interface) of a buoyant jet

were investigated as a function of the jet Richardson number. It was

increase with increasing the Richardson number. The spreading rate of

the transverse mean velocity and temperature profiles were found to be

independent of the Richardson number. The turbulent buoyancy flux in a

fully developed buoyant jet has been found to be a significant fraction

(38%) of the axial buoyancy flux.

1.1 Introductory note 7

2. THE MECHANICS OF JETS 25

3. EXPERIMENTAL PROBLEMS AND OBJECTIVES 53

4. TEMPERATURE MEASUREMENT RESULTS 88