Aeronautical Journal

May 2013 Volume 177 No 1191 505

Experimental studies on
rectangular jets with
trapezoidal tabs
A. Arokkiaswamy
argy_swamy@hotmail.com
S. B.Verma
sbverma@ead.cmmacs.ernet.in
S. Venkateswaran
venkatesh_w@hotmail.com
Council of Scientific and Industrial Research
National Aerospace Laboratory
Bangalore, India

ABSTRACT
An experimental investigation was carried out to study the flow development of a jet issuing from
a 2:1 rectangular nozzle with mixing tabs using two-component hotwire anemometry. A pair of
tabs of trapezoidal configuration (with 2% total blockage area) is placed on the minor-axis side
of the rectangular nozzle and tested for two tab inclination angles of 135° and 45°, with respect to
the flow direction. Tests were conducted for a nominal jet exit velocity of 20m/sec corresponding
4
to a Reynolds number based on nozzle equivalent diameter of 5·013 × 10 . Relative to the plain
jet, the jet with tabs show significant reduction in jet-core length (by 67%) followed by a faster
decay in jet centreline velocity (U/Ue). This is also accompanied by a significant upstream shift
in peak centreline turbulence intensity (u’/Ue). The presence of tabs is observed to inhibit the jet
growth along the minor-axis plane thereby introducing large distortion in the jet cross-sectional
development that ultimately leads to jet-core bifurcation along its major-axis. While a mushroom-
like flow structure develops behind the tab with 135° inclination, the flow structure behind a 45°
inclined tab rather takes the shape of the tab itself. The former flow development is seen to enhance
the jet growth more along the minor-axis while the latter improves the jet growth more along the
major-axis plane. From application point of view, since both tab inclinations result in more or less
similar jet characteristics, a 135° inclined tab would be preferable over a 45° inclined tab from
the view of improved jet mixing.

Paper No. 3800 Manuscript received 7 March 2012, revised version received 18 August 2012, accepted 7 September 2012.

3800.indd 505 03/05/2013 11:28:29

 506 The Aeronautical Journal May 2013

NOMENCLATURE
De equivalent diameter of the rectangular nozzle, mm
h projected height of the tab to the oncoming flow, mm
ReDe Reynolds number based on equivalent diameter of the rectangular nozzle
u’/Ue non-dimensional streamwise velocity fluctuation
2
u’v’/Ue Reynolds shear stress
–1
Ue mean streamwise jet exit velocity, ms
–1
U local mean streamwise jet velocity, ms
–1
V local mean jet velocity along Y-direction, ms
X streamwise distance along the jet centreline, mm
Y cross-stream distance along minor-axis plane, mm
Z cross-stream distance along major-axis plane, mm
Y0·5 jet half-width growth along minor-axis plane, mm
Z0·5 jet half-width growth along major-axis plane, mm
w width of the tab base, mm
δmi jet exit shear-layer thickness along minor-axis plane, mm
δmj jet exit shear-layer thickness along major-axis plane, mm
φ tab angle relative to the flow direction, degrees

1.0 INTRODUCTION
Mixing enhancement in jet flows is of paramount importance in many engineering applications and
therefore, has been the subject of continuing research. Frequently the jet geometry is dictated by
the nature of application since jet characteristics are known to be closely related to the dynamics
of shear flow originating at the nozzle exit and hence, are strongly affected by the shape of the
(1)
nozzle from which they issue . As a result one of the most commonly used methods of shear
flow control in jets is the use of nozzles with non-circular exit cross-sections which significantly
changes the jet flow development as compared to a jet issuing from a circular nozzle. Jets from
non-circular nozzle geometries spread and mix faster thereby providing a unique capability to
control the jet development (both fine- and large-scale).
(2) (3)
Studies on jets issuing from non-circular nozzle geometries such as triangular , square ,
(4,5) (6-9)
rectangular and elliptic have been reported in the past. It has been found that the initial
instability mode in a jet issuing from an elliptic nozzle is strongly linked to the thinnest jet
(6,10) (11)
momentum thickness around the nozzle circumference . Rectangular jets , on the other hand,
combine the aspect ratio features of an elliptic jet with the corner (vertex) features of square jets.
(4-7)
Later studies on the coherent structures of non-circular jets revealed that the jet undergoes a
three-dimensional deformation process associated with the azimuthal distortion and bending of
the vortex ring wherein ambient mass is brought in towards the jet-centreline along the major-axis
side, and jet mass is ejected out along the minor-axis side. This non-uniform self-induction process
(4-8)
results in enhanced mixing between the jet and the ambient irrotational mass . As a result, the jet
undergoes an axis-switching phenomenon wherein it entrains more fluid and spreads faster in the
major-axis plane while it shrinks along the minor-axis side. The behaviour of coherent structures
was also found to be strongly effected by initial flow conditions such as the jet aspect-ratio, initial
(7)
momentum thickness, excited or unexcited and therefore could be manipulated .

3800.indd 506 03/05/2013 11:28:29

 Arokkiaswamy et al Experimental studies on rectangular jets with trapezoidal tabs 507

Means have also been explored to introduce secondary instabilities in the form of streamwise
vortices generated from mixing devices such as tabs to alter the flow-field development signifi-
(12-20)
cantly . The mechanism of their formation, evolution, and interaction with Kelvin-Helmholtz
type vortices is of great importance for practical applications. Generally tabs were observed to
introduce circumferential variations in the jet flow development by splitting the jet into two high-
(13)
velocity regions on either side of the diameter joining the two tabs . Various tab configurations such
(16) (22)
as triangular and rectangular have been studied in the past. The tab inclination , with respect to
the jet axis, has also been found to strongly govern the initial vortex-dynamics. Individual studies
(23,24)
on the flow past a trapezoidal tab mounted on a flat plate have also been carried out. However,
the flow development downstream of a trapezoidal tab mounted on a flat plate (no-slip condition)
changes significantly when such a tab configuration is in a jet flow (slip condition) at the nozzle
exit. The focus of the present work was, therefore, to study the flow structure development behind
a tab of trapezoidal configuration in a jet flow and its effect on the overall jet flow development,
thereafter. Two such tabs are placed on the minor-axis sides (and tested for tab inclinations of
45° and 135° to the oncoming flow) of a 2:1 rectangular nozzle. Results are compared with the
jet issuing from a 2:1 rectangular nozzle without tabs (hereafter referred to as the ‘plain jet’). A
two-component hotwire probe (X-wire) was used for detailed grid measurements in the Y-Z plane
of the jet in (i) the tab-wake region and, (ii) in the jet half-plane at various axial locations to study
the flow structure development behind the tab and its effect on the overall jet flow development,
respectively in the absence and presence of tabs.

2.0 EXPERIMENTAL SET UP AND PROCEDURE

2.1 Test facility and models
Experiments were carried out to investigate the jet flow development from a 2:1 rectangular nozzle,
with and without tabs, Fig. 1(a). The nozzle has a circular section of 203 ± 0·1mm diameter,
smoothly contoured to a rectangular section (2a = 47·0 ± 0·1mm and 2b = 23·5 ± 0·1mm) over
a length of 300 ± 0·1mm, where 2(a) and 2(b) are major-and minor-axis lengths, respectively.
The equivalent diameter (De) of the rectangular nozzle is 37·5 ± 0·1mm. The measurements are
–1
carried out at a nominal jet exit velocity (Ue) of 20 ± 0·5ms and the Reynolds number based on
4
the equivalent diameter of the jet (RDe) is 5·02 ± 0·13 × 10 . A pair of trapezoidal tabs (2% total
blockage) is placed in the minor-axis plane of a 2:1 rectangular nozzle to modify the initial jet
development in this plane, Fig. 1(b).

4.0

(a) (b) (c)

Figure 1. Schematic of (a) 2:1 rectangular nozzle,

(b) tab placement (φ=135°), tab dimensions and grid measurement plane for overallo jet development study
Figure 1: Schematic of near-wake
and, (c) (a) 2:1 rectangular nozzle, (b)plane;
grid measurement tab placement (φ=135
all dimensions are),intab
mm. dimensions
and grid measurement plane for overall jet development study and, (c) near-wake grid
measurement plane; all dimensions are in mm

3800.indd 507 03/05/2013 11:28:29

 508 The Aeronautical Journal May 2013

2.2 Instrumentation
Hot-wire measurements were carried out in the tab-wake region and at several axial locations in the
Y-Z plane by means of a 99N10 DANTEC anemometry system using a Dantec 55P11 2-component
probe. The probe has platinum plated tungsten wires (1·25mm long and 5μm diameter) and can
be used for air applications with turbulent intensities up to 5-10%. The X-wire was positioned in
the flow in such a way that it allowed to measure the fluctuations of streamwise velocity (u’) and
transverse velocity (v’). The positioning of the sensor was performed by using a PC controlled
DANTEC 3-dimensional precision traverse (Model # 41T33). The probe was calibrated using Dantec
2 –1
9054H01 calibrator with 120mm nozzle in the velocity range between 0-25ms . The signals from
the probe were acquired at a sampling rate of 3kHz with 10,000 samples. The anemometer analog
output was acquired by using a differential mode National Instruments PCI -6036E having 16-Bit
resolution, operating range of ± 10V and maximum scan rate of 200Ks/samples. The linearisation
and processing of the hot wire signal was then carried out digitally. The actual streamwise velocity
U and perpendicular velocity V were calculated from the hotwire anemometer output according
(25)
to King’s law equation and equation procedure by Jorgenson . The uncertainty in the jet exit
–1
velocity Ue and in the positioning of the hotwire X-probe is ± 0·5ms (2·5%) and ± 0·5mm (2%), 4.0
respectively. The projected dimension of the sensor elements to the oncoming flow is 0·8mm
and the uncertainty in measurements with regards to the probe dimensions is approximately 2%.

3.0 RESULTS AND DISCUSSIONS

Figure 1: Schematic of (a) 2:1 rectangular nozzle, (b) tab placement (φ=135 ), tab dimensions o

and grid measurement plane for overall jet development study and, (c) near-wake grid
measurement plane; all dimensions are in mm
3.1 Centreline velocity decay and jet half-width growth
In general, as the jet fluid moves away from its origin, it slows down due to the process of
mixing initiated with slower moving ambient 4.0 fluid. This interaction between the jet and the

ambient fluid forms the mixing layer, or shear layer. Due to Kelvin-Helmholtz instability, the
primary jet structures begin to roll-up which grow in size as they move downstream, due to
entrainment of slower moving ambient fluid. As a result, jet decay is proportional to the velocity
gradient across the shear layer and is a strong function of the distance downstream of the jet exit
normalised by equivalent diameter of the nozzle. Figures 2(a) and (b) show the comparison of
Figure 1: Schematic of (a) 2:1 rectangular nozzle, (b) tab placement (φ=135o), tab dimensions
centreline
and grid measurement velocity
plane decayjetand
for overall their corresponding
development study and, (c)turbulent
near-wakeintensity
grid plots, respectively. Relative
to theallplain
measurement plane; dimensions
jet, are
theinpotential
mm core length of jets with tabs is significantly reduced from 3·0De

Figure 2. (a) Centreline velocity (U/Ue) decay for jets issuing from a 2:1 rectangular
nozzle with and without tabs and (b) effect of tabs on the u’/Ue centreline distribution.
Figure 2: (a) Centerline velocity (U /Ue) decay for jets issuing from a 2:1 rectangular
nozzle with and without tabs (b) effect of tabs on the u’/Ue centerline distribution and, (c)
Jet half-width growth for plain and tabbed jets along X-Y and X-Z planes

11

3800.indd 508 03/05/2013 11:28:29

 Arokkiaswamy et al Experimental studies on rectangular jets with trapezoidal tabs 509

3 3
2.5 Y 0.5/D e
Y0.5/De , Z0.5/De
2.5

Y0.5/De , Z0.5/De
Z 0 .5/D e
2 2
1.5 axis-switching 1.5
location
1 1
0.5 0.5
(b) φ=45o
(a) Plain 0
0 0 2 4 6 8 10 12 14 16
0 2 4 6 8 10 12 14 16
X /D e X/De
3
2.5
Y0.5/De , Z0.5/De

2
1.5
1
0.5
(c) φ = 135°
0
0 2 4 6 8 10 12 14 16
X/De

Figure 3. Half-width plots showing the variation in jet growth along Z- and Y-planes for
the two tab inclination angles tested (a) plain jet, (b) jet with tab inclined at 135°,
Figure 3:with
(c) jet Half-width plots
tab inclined showing
at 45° and, (d) the variation
equivalent in jetforgrowth
jet width along
all the test Z- and Y-
cases.
planes for the two tab inclination angles tested (a) plain jet, (b) jet with tab inclined at
o o
to (for plain , (c)tojet1·0D
135jet) withe tab
(byinclined at 45 and,by(d)
67%) followed equivalent
a faster jet of
decay width
the for
jet all the test cases
centreline velocity (U/
Ue), Fig. 2(a). For plain jet, the centreline turbulent intensity (u’/Ue) shows a peak in value at
X/De = 7·8 whereas for jets with tabs this location shifts considerably upstream to X/De = 2·67
for 45° tab followed by 135° tab at X/De = 2·0, Fig. 2(b). The above trends show enhancement
of both large-scale and small-scale activity for jets with tabs relative to the plain jet and also
with variation in tab inclination.
Figure 3 shows a comparison of the jet half-width growth for the test cases investigated. It
can be seen and is well known, the plain jet from rectangular nozzle grows along its minor-axis
plane while it shrinks along the major-axis plane. At approximately X/De = 4·0, the jet half-
width plots along the two planes cross each other indicating the axis-switching location of the
jet, Fig. 3(a). Tabs are initially seen to inhibit the jet growth along the minor-axis plane due to
the blockage effect of the tabs. Thereafter for X/De > 4·0, the jet growth increases significantly.
While the plain jet is observed to switch its axis at 4·0De, the jet with tabs switch their axis at
4·2-4·5De, Fig. 3(b)-(c). Further it can be observed that when the tab is inclined at 45° to the
oncoming flow, the jet half-width growth increases significantly along the major-axis side but
Figure
for the 135° 4: Contours
inclined of mean
tab, the velocity atgrowth
jet half-width differentincreases
axial locations in Y-Z plane
significantly alongforthea plain jet
minor-axis
a) X/D
side. This e =0.5, (b)in
behaviour X/D
jete =1.0,
growth X/D
(c) is e =2.0 (d)toX/D
contrary thee =results
4.0 observed for cylindrical tabs (26)
(placed 90° to the oncoming flow) in an elliptic jet where axis-switching did not occur due
to reduced jet growth along minor-axis and enhanced jet growth along major-axis plane. This
variation in the behaviour of jet growth with tab inclination angle becomes clear when the flow
development in the tab wake region is discussed in detail in the later sections. The equivalent
jet spread which signifies the overall mixing and jet spread along the axial location for a given
exit condition is shown in Fig. 3(d). It can be seen that the overall jet growth is more for the jet
with tabs and shows higher overall jet growth for 12 tabs inclined at 45° followed by the jet with
tabs inclined at 135°.

3800.indd 509 03/05/2013 11:28:29

 X/De

Figure 3: Half-width plots showing the variation in jet growth along Z- and Y-
planes for the two tab inclination angles tested (a) plain jet, (b) jet with tab inclined at
135o, (c) jet with tab inclinedhe
510 T at A o
45eronautical
and, (d) equivalent
Journaljet
width for all the test casesMay 2013

Figure 4. Contours of mean velocity at different axial locations in Y-Z plane

Figure 4: Contours
for a plainofjetmean velocity
a) X/De
at different
= 0·5, (b) axial
X/De = 1·0, locations
(c) X/De
(d)Y-Z
= 2·0 in X/Dplane
e
= 4·0.for a plain jet
a) X/De =0.5, (b) X/De =1.0, (c) X/De =2.0 (d) X/De = 4.0
3.2 Overall jet development mean velocity distribution contours
Grid measurements were carried out at various axial locations to study the process of overall jet
development. Extensive grid measurements were made in one-half Y-Z plane of the jet, Fig. 1
(b), with a step size of ∆y = ∆z = 1·0 ± 0·1mm or 0·026De for X/De = 0·5, 2·0 (approx 1,600 and
2,400 points, respectively) while a step size of ∆y = ∆z = 1·5 ± 0·1mm or 0·039De was kept for
X/De = 3·0, 5 (approx. 1,250 and 1,600 points, respectively). The variation in each successive
contour level is 0·1Ue. Contours of the normalised
12 streamwise mean velocity (U/Ue) in the tab
wake region were also acquired at two streamwise locations X/w = 3·0 and 4·0 (where w = 4·5mm),
respectively. These contours are obtained from grid measurements on both sides of the tab in the
Y-Z plane (with Δy = Δz = 0·5 ± 0·1mm or 0·013De and resulting in 1,073 points), as shown in
Fig. 1(c). Such detailed measurement grids were required because of the highly three-dimensional
evolution of the flow behind the tab. The jet-exit shear-layer thickness (U = 0·99Ue) along the
minor-axis side, δmi , was about 1·25mm at the tab axis and that along the major-axis side of the
nozzle , δmj, was 1·75mm, respectively. The corresponding displacement thickness along each
nozzle axis is 0·381mm and 0·534mm, respectively. With respect to δmi the tabs protrude well
into the mean flow. The longitudinal turbulent intensity (u’/Ue) at the jet exit was about 0·3% at
–1
20ms . Figures 4, 5 and 7 show contours of normalised streamwise mean velocity (U/Ue ) for

1.5
jet-core
1 bifurcation

indentation
0.5
Y/De

-0.5

-1
(a) (b) (c)
-1.5
(d)
0 0.3 0.6 0.9 1.2 1.5
Z/De

Figure5.5:Contours
Figure Contoursofof mean
mean velocity
velocity at different
at different axialaxial locations
locations Y-Z plane
in Y-Zinplane for a
for a trapezoidal
trapezoidal
tab withtab with inclination
inclination of 135
of 135°. (a) X/Dedeg.
= 0·5, X/X/D
(a)(b) De =0.5, X/X/D
(b)(c)
= 2·0, De =2.0, X/X/D
(c)(d)
= 3·0 De =3.0
= 5.
e e e
(d) X/De =5

0.1 U/Ue 0.1 u' /Ue

(a) (b)
1 0.16
0.9 0.14
0.2 0.2
0.8 0.12
0.7 0.1
Y/De

Y/De

3800.indd 510 0.3 0.6 0.3 0.08 03/05/2013 11:28:31

0.5 0.06
 Z/De

Figure 5: Contours of mean velocity at different axial locations in Y-Z plane for a
trapezoidal tab with inclination of 135 deg. (a) X/De =0.5, (b) X/De =2.0, (c) X/De =3.0
(d) X/De =5
Arokkiaswamy et al Experimental studies on rectangular jets with trapezoidal tabs 511

0.1 U/Ue 0.1 u' /Ue

(a) (b)
1 0.16
0.9 0.14
0.2 0.2
0.8 0.12
0.7 0.1
Y/De

Y/De
0.3 0.6 0.3 0.08
0.5 0.06
0.4 0.04
0.4 0.4
0.3 0.02
0.2
0.5 0.1 0.5
0.2 0.1 0 -0.1 -0.2 0.2 0.1 0 -0.1 -0.2
Z/De Z/De
0.1 U/Ue 0.1 u' /Ue
mushroom (c)
(c) (d)
0.16
structure 0.9
0.2 0.8 0.2 0.14
0.7 0.12
0.6 0.1
Y/De

Y/De
0.3 0.3 0.08
0.5
0.06
0.4
0.04
0.4 0.3 0.4
0.02
0.2
upwash 0.1
0.5 0.5
0.2 0.1 0 -0.1 -0.2 0.2 0.1 0 -0.1 -0.2

Z/De Z/De

Figure 6. Contours of mean velocity and streamwise turbulence behind the tab at different
Figure axial locations in Y-Z
6: Contours of plane
meanfor velocity
tab inclination
andof 135°. (a)-(b) X/wturbulence
streamwise = 3·0 and, (c)-(d) X/w = 4·0
behind the tab at
different axial locations in Y-Z plane for tab inclination of 135 deg. (a)-(b) X/w=3.0 and,
each test
(c)-(d) case while Figs 6 and 8 show the flow development in the region of the tab-wake for
X/w=4.0
various axial locations in the Y-Z plane.

3.2.1 Plain jet

Immediately downstream of the nozzle exit, the plain jet is seen to retain its original rectangular
(27)
shape and shows a thin mixing-layer initially 13 , as is evident from the closely spaced contours,
Fig. 4(a). As the mixing between the jet and the ambient mass is initiated and the shear-layer grows,
(4-8)
the jet begins to gradually deform due to non-uniform induction of velocity along the nozzle
azimuth, while vortices are generated from the nozzle corners (seen as a outward bump in contours),
Fig. 4(b). Further downstream, the mixing layer begins to thicken and the jet cross section begins
to change its shape as is indicated by a higher growth along minor-axis side (Fig. 4(a) – 4(d)). The
spacing between the contour levels is seen to increase with increase in downstream distance. As
a result, the jet undergoes a three-dimensional deformation process associated with the azimuthal
distortion and bending of the rectangular vortex ring wherein ambient mass is brought in towards
(7)
the jet centreline along the major-axis side, and jet mass is pushed out along the minor-axis side .

3.2.2 Jet with tabs

Introduction of tabs in minor-axis side significantly modifies the local as well as the overall jet
development, Fig 5 and 6. Tabs are seen to cause an inward indentation in the jet flow development
due to the blockage effect that prevents the jet flow from reaching the tab-wake region immediately.

3800.indd 511 03/05/2013 11:28:32

 512 The Aeronautical Journal May 2013

Such indentations help increase the contact area of the mixing layer with the slower moving ambient
fluid resulting in enhanced mixing relative to the plain jet. However, the inward penetration of the
tab wake initially inhibits the jet growth along the minor-axis side, as was observed for cylindrical
(26)
tabs . In fact, in the present tests, tabs not only bifurcate the jet core at some downstream distance
but are also observed to enhance the jet growth along both planes, Fig. 3(b)-(c). This flow modifi-
cation delays the overall axis-switching phenomena for both tab inclinations.
Variations in the overall jet flow development for the two tab inclinations tested can also be
observed. The jet is observed to get bifurcated much earlier for tab inclination of 135° than for
tab inclination of 45°. This modifies the jet development along each axis as observed in the jet
half-width plots. Further details of the flow physics involved with tab inclination angle is studied
using very fine grid measurements in the tab-wake region.
For tab inclination of 135°, the flow immediately behind the tab develops into a very well
defined ‘mushroom structure’, Fig. 6(a) and (c). The mean velocity is seen to decrease sharply as
the tab-wake region is approached, decreasing to approximately 0·3Ue indicating strong cross-
stream gradients. In the region of the tab-base, a sharp incursion of contours can be seen on each
side of the tab which indicates the presence of a streamwise vortex (of clockwise rotation on right
side and vice-versa). Such streamwise vortices are known to induce intense local mixing thereby
(26)
promoting fluid and momentum exchange . Further downstream the thickness of the mixing
layer increases both along the tab axis and along the axis normal to the tab which results in the
mushroom structure to grow in size. Figure 6(b) and (d) show the corresponding distribution
of turbulence intensity in the tab wake region. It can be seen that the turbulence intensity is the
highest on both sides of the tab base region and in the region of the streamwise vortices. The pair
of streamwise vortices shed from each tab interacts with the azimuthal vortices embedded in the
mixing layer and increase the entrainment of the mixing layer with the ambient fluid. The vortex
system so generated is seen to split the jet-core at some downstream distance with high-velocity
cores on either side of the minor-axis plane, Fig. 5(b)-(d). The observed jet bifurcation explains
the significant reduction in jet core-length and the rapid centreline velocity decay associated with
increased turbulent mixing activity thereafter for jets with tabs, as observed in Fig. 2.
Figure 7 shows contours of normalised streamwise mean velocity (U/Ue) for a trapezoidal tab
inclination of 45°. Some similar modifications in the overall jet flow development phenomena are
observed in Fig. 6(a). However, the most notable feature is the much sharper inward indentation
caused by this tab inclination, Figs 6(a) and 7(a). It can be seen that the flow while passing around

1.5

0.5
Y/De

-0.5

-1
(a) (b) (c) (d)
-1.5
0 0.3 0.6 0.9 1.2 1.5
Z/De

Figure
Figure7.7:
Contours
Contoursof mean velocity
of mean at different
velocity axial locations
at different in Y-Z plane
axial locations in for
Y-Za trapezoidal
plane for a
tab with inclination of 45°. (a) X/De = 0·5, (b) X/De = 2·0, (c) X/De = 3·0 (d) X/De = 5.
trapezoidal tab with inclination of 45 deg. (a) X/De =0.5, (b) X/De =2.0, (c) X/De =3.0 (d)
X/De = 5

0.1 U/Ue 0.1

(a) (b) u' /Ue
1
0.2 0.9 0.2 0.16
0.8 0.145
0.7 0.135
Y/De
Y/De

0.3 0.6 0.3 0.12

3800.indd 512 0.5 0.1 03/05/2013 11:28:33
0.4 0.08
0.4 0.3 0.4 0.06
 (a) (b) (c) (d)
-1.5
0 0.3 0.6 0.9 1.2 1.5
Z/De

Figure 7: Contours of mean velocity at different axial locations in Y-Z plane for a
Arokkiaswamy
trapezoidal tab et
with
al Experimental
inclination of studies rectangular jets with trapezoidal tabs
45 deg.on(a) X/De =0.5, (b) X/De =2.0, (c) X/De =3.0513
(d)
X/De = 5

0.1 U/Ue 0.1

(a) (b) u' /Ue
1
0.2 0.9 0.2 0.16
0.8 0.145
0.7 0.135

Y/De
Y/De

0.3 0.6 0.3 0.12

0.5 0.1
0.4 0.08
0.4 0.3 0.4 0.06
0.2 0.04
0.1 0.02
0.5 0.5
0.3 0.2 0.1 0 -0.1 -0.2 -0.3 0.3 0.2 0.1 0 -0.1 -0.2 -0.3
Z/De Z/De
0.1 0.1
(c) U/U e (d) u' /U e
1 0.16
0.2 0.9 0.2 0.145
0.8 0.135
0.7 0.12
Y/De
Y/De

0.3 0.6 0.3 0.1

0.5 0.08
0.4 0.06
0.4 0.3 0.4 0.04
dead-air region 0.2 0.02
0.1
0.5 0.5
0.3 0.2 0.1 0 -0.1 -0.2 -0.3 0.3 0.2 0.1 0 -0.1 -0.2 -0.3
Z/D e Z/D e

Figure 8. Contours of mean velocity and streamwise turbulence behind the tab at different axial
locations in Y-Z plane for trapezoidal tab with inclination of 45°. (a)-(b) X/w = 2·0 and, (c)-(d) X/w = 3·0.

the tab is unable to immediately reach behind the tab resulting in a huge portion of low speed or a
Figure 8: Contours of mean velocity and streamwise turbulence behind the tab at
dead air region, Fig. 8(a) and (c). Such a flow constraint imposed by this tab inclination prevents
different
the axialthe
flow behind locations in Y-Z into
tab to develop plane for trapezoidal
a mushroom tabaswith
structure inclination
observed of inclined
for 135° 45 deg. tab.
(a)-(b)
X/w=2.0 and, (c)-(d) X/w=3.0
Instead the flow in the tab wake region takes the shape that resembles the tab configuration and
with relatively more inward penetration, Fig. 6(c) and 8 (c). This change in flow development
relatively inhibits the jet growth in the minor-axis plane while an increase in the jet growth along
the minor-axis side is observed, Fig. 3(b) and (c). Figures 8(b) and (d) show the corresponding
distribution of turbulence intensity in the tab wake region. It can be seen that the turbulence
intensity is the highest at the top edges of the tab and in the region of the tab base. These varia-
tions suggest significant changes in flow structure development for tabs with different inclination
angles to the oncoming flow.
14
3.2.3 Profiles of mean velocity and turbulence intensity

Along tab axis

Figures 9(a)-(c) show the mean velocity (U/Ue) and turbulence intensity (u’/Ue) profiles measured
at various streamwise locations along the tab-axis (Z = 0), respectively. For tab inclined at 135°,
the mean velocity profiles reveal two distinct regions of velocity minimum and maximum, as seen
in Fig. 9(a). Close to the base region of the tab (i.e., at Y/De = 0·32) a maxima in U/Ue occurs (due
to the upwash from the base vortex) that is followed by a velocity deficit or a minima between
the tab mid-height and free-end (due to the downwash from the tab free-end). As the top-edge
of the tab is approached, the U/Ue value is seen to increase to its value in the jet-core causing

3800.indd 513 03/05/2013 11:28:34

 514 The Aeronautical Journal May 2013

1.2 0.3
1.2 0.3
X/De=0.2
1 (a) 0.25 (b) X/DX/D e=0.2
e=0.3
1 (a) 0.25 (b) X/DX/D e=0.3
e=0.5
X/DX/D e=0.5
e=0.7
X/De=0.7
0.8 0.2
0.8 0.2
U /Ue

u' /Ue
0.6 0.15
U /Ue

u' /Ue
0.6 maxima 0.15
maxima
0.4 0.1
0.4 0.1

0.2 0.05
0.2 0.05
maxima
0 maxima 0
00.1 0.2 0.3 0.4 0.5 00.1 0.2 0.3 0.4 0.5
0.1 0.2
Y/De 0.3 0.4 0.5 0.1 0.2
Y/De 0.3 0.4 0.5
Y/De Y/De
0.3
0.3
X/De=0.1
X/DX/D
=0.2e=0.1
(c) 0.25 (d) e
X/DX/D =0.2
(c) 0.25 (d) e=0.3e
X/DX/D
e
=0.5e=0.3
X/DX/D
e=0.7e
=0.5
0.2 X/De=0.7
0.2
u' /Ue

0.15
u' /Ue

0.15

0.1
0.1

0.05
0.05

0
00.1 0.2 0.3 0.4 0.5
0.1 0.2
Y/De 0.3 0.4 0.5
Y/De
Figure 9: Profiles of mean velocity and turbulence intensity in the tab
Figure
wake 9: Profiles of mean velocityparallel
and turbulence tabintensity in the tab
Figure 9.region
wake Profileswith measurement
of mean velocity andplane
turbulence to the
intensity inthe
the axiswake
tab for region
(a)-(b)with
o region with measurement
o plane parallel to tab axis for (a)-(b)
φ=135 plane
measurement parallelφ=45
and (c)-(d) , respectively.
to the tab axis for (a)-(b) φ = 135° and (c)-(d) φ = 45°, respectively.
φ=135o and (c)-(d) φ=45o , respectively.

relaxed flow flow constraint

relaxed flow
conditions flow constraint nozzle exit
conditions nozzle exit nozzle exit
nozzle exit
tab-wake region
tab-wake region

(a) φ = 135o o (b) φ = 45o o

(a) φ = 135 (b) φ = 45

Figure 10: Schematic of the flow development from a tab at different

Figure
Figure 10. Schematic of the10:
inclination flowSchematic
angles of from
development the flow
to the oncoming a tabdevelopment from a angles
at different inclination
flow. tab at to
different
the oncoming flow.
inclination angles to the oncoming flow.

15
15

3800.indd 514 03/05/2013 11:28:35

 Arokkiaswamy et al Experimental studies on rectangular jets with trapezoidal tabs 515

(26)
a shear layer to develop that gradually begins to roll-up into small-scale K-H vortices . The
corresponding turbulent intensity (u’/Ue) profiles, Fig. 9(b), show intense local mixing in the
region of the tab-base and the top edge of the tab. Further downstream, the velocity deficit at
tab mid-span reduces (due to vortex interaction and viscous diffusion) and is seen to gradually
move outwards while the shear layer from the tab top edge moves gradually inwards, between
X/De = 0·5 to 0·7. On the other hand, for tab inclined at 45°, the mean velocity profiles do not
show the presence of a velocity maximum and velocity minima in the region behind the tab, Fig.
9(c), as was seen in Fig. 9(a). This indicates that the tab inclination of 45° imposes a significant
constraint on the oncoming jet flow so that the flow is unable to completely wrap around the tab
and reach immediately behind it as seen for 135° inclined tab and also seen in Figs 6 and 8. The
corresponding turbulence intensity plots also show a similar effect, Fig. 9(d).
Figure 10 shows a schematic of the flow development from each tab inclination developed based
on the present results. It can be seen that as the flow approaches the tab, it has to move out of the
way from the sides and over the top of the tab. For 135° tab inclination, the flow while passing
over the tab is more ‘relieved’ in comparison with the flow passing over a 45° inclined tab, which
is more constrained, Fig. 10(a) and (b). The flow before passing over the top of a 45° inclined
tab first stagnates at the base of the tab and then speeds up over the top and sides to get out of
the way of the tab causing larger changes in velocity, Fig. 10(b). On the other hand, for 135° tab
inclination, the streamwise inclination eases the constraint on the flow (due to absence of flow
stagnation) and so the flow freely moves over the top and sides in a more relaxed fashion resulting
(28)
in lesser changes in velocity and pressure , Fig. 10(a). These changes experienced by the flow
due to variation in tab inclination result in different flow structure development in the wake of
each tab inclination tested, as seen in Figs 6 and 8. From application point of view, it is preferable
therefore to use a tab with 135° inclination because of its advantage of (i) lesser aerodynamic
drag (due to significantly less constraint on the flow) and, (ii) less heat transfer, especially at the
base of the tab in hot flows.

4.0 CONCLUSIONS
An experimental investigation was carried out to study the flow development of a jet issuing from
a 2:1 rectangular nozzle, with a pair of trapezoidal tabs placed on its minor-axis side. Further,
the effect of tab inclination (135° and 45° to the oncoming flow) was also examined. Detailed
two-component hotwire measurements were carried out to study the jet flow development.
Relative to the plain jet, the potential-core length of jets with tabs is significantly reduced (by
67%) followed by a faster decay of the centreline mean velocity (U/Ue). This is accompanied by a
significant upstream shift in the peak in centreline turbulence intensity (U’/Ue). The above trends
show enhancement of both large-scale and small-scale activity in jets with tabs. Close to the base of
the tab (135°), a maxima in U/Ue occurs (due to the upwash from the base vortex) that is followed
by a velocity deficit or a minima between the tab mid-height and free-end (due to the downwash
from the tab free-end). Detailed grid study of the tab wake region shows that the flow immediately
behind this tab develops into a very well defined mushroom structure with sharp incursion of
contours on each side of the tab indicating the presence of a streamwise vortex that entrains the
high-speed fluid from the jet-core towards the tab base and inwards along the tab axis resulting in
a strong upwash. On the other hand with 45° tab inclination, the flow on passing around the tab
is unable to immediately reach behind the tab resulting in a large portion of dead-air region near
the tab base. Such a flow constraint imposed by this tab inclination prevents the flow behind the

3800.indd 515 03/05/2013 11:28:35

 516 The Aeronautical Journal May 2013

tab to develop into a mushroom structure and takes a shape that resembles the tab configuration
itself. However, the flow structure generated by the tabs (irrespective of its angle of inclination)
is seen to split the jet-core at some downstream distance with high-velocity cores on either side of
the minor-axis plane. The jet growth in each case is significantly enhanced both along minor-and
major-axis planes which slightly delays the axis-switching location, relative to plain jet.
The results indicate that for 135° tab inclination (i.e., tab inclined in the direction the flow), the
flow while passing over the tab is more relaxed in comparison with the flow passing over a 45°
inclined tab (i.e., tab inclined against the flow), which is more constrained and results in different
flow structure development in the wake of each tab. Since both tabs result in more or less similar
enhanced mixing, from application point of view it would be preferable to use a tab with 135°
inclination because of its potential from the view of improved jet mixing.

ACKNOWLEDGEMENTS
The technical work reported here is carried out in the experimental Aerodynamics Division (EAD)
at National Aerospace laboratories (CSIR), Bangalore. The first author is thankful to Dr A.R.
Upadhya, Director NAL and Dr. Sajeer Ahmed, former Head EAD, for granting permission to carry
out the experiments as a part of his PhD work. The authors also express their sincere thanks to Mr
Sudhakar and Mr Manisankar (scientists, EAD) for their assistance in conducting the experiments.

REFERENCES
1. Gutmark, E., Schadow, K.C., Koshigoe, S. and Wilson, K.J. Combustion related shear flow dynamics
in elliptic jets, AIAA J, 1989, 27, (10), pp 1347-1353.
2. Gutmark, E., Schadow, K.C. and Wilson, K.J. Subsonic and supersonic combustion related non-circular
injectors, J Propulsion and Power, 1991, 7, (2), pp 240-249.
3. Quinn, W.R. Experimental and numerical study of a turbulent free square jet, Physics of Fluids, 1988,
31, (5), pp 1017-1025.
4. Krothapalli, A., Bagdanoff, D. and Karamcheti, K. On the mixing of a rectangular jet, J Fluid
Mechanics, 1981, (107), pp 201-220.
5. Marsters, G.F. Spanwise velocity distributions in jets from rectangular slots, AIAA J, 1981, 19, (2), pp
148-152.
6. Ho, C.M. and Gutmark, E. Vortex Induction and mass entrainment in a small-aspect-ratio jet, J Fluid
Mechanics, 1987, (179), pp 383-405.
7. Hussain, F. and Husain, H.S. Elliptic jets, Part 1: Characteristic of unexcited and excited jets, J Fluid
Mechanics, 1989, (208), pp 257-320.
8. Quinn, W. R. On mixing in an elliptic turbulent free Jet, Physics of Fluids, 1989, 1, (10), pp 1716-1722.
9. Verma, S.B. and Rathakrishnan, E. Flow and acoustic properties of underexpanded elliptic-slot jets,
17, (1), AIAA J Propulsion and Power, January – February 2001.
10. Crighton, D.G. Instability of an elliptic jet, J Fluid Mechanics, 1973, 59, pp 665-672.
11. Gutmark, E.J. and Grinstein, F.F. Flow control with non-circular jets, Annual Review of Fluid Mechanics,
1999, 31, pp 239-272.
12. Ahuja, K.K. and Brown, W.H. Shear Flow Control by Mechanical Tabs, AIAA paper 89-0094.
13. Bradbury, L.J.S. and Khadem, A.H. The distortion of a jet by tabs, J Fluid Mechanics, 1975, 70, pp
801-813.
14. Zaman K.B.M.Q. Axis switching and spreading of an asymmetric jet: The role of coherent structure
dynamics, J Fluid Mechanics, 1996, (316), pp 1-27.
15. Chua, L.P., Yu, S.C.M. and Wang, X.K. Flow visualization and measurements of a square jet with
mixing tabs, Experimental Thermal and Fluid Sciences, 2003, 27, pp 731-744.
16. Verma, S.B., Venkatkrishnan, L. and Ramesh, G. 2-D PIV Study of Near-Field Flow Development
from a 2:1 Elliptic Jet with Tabs AIAA Paper No-2007-4498.

3800.indd 516 03/05/2013 11:28:35

 Arokkiaswamy et al Experimental studies on rectangular jets with trapezoidal tabs 517

17. Tanna, H.K. An experimental study of jet noise, part ii: shock associated noise, J Sound and Vibration,
1977, (50), pp 429-444.
18. Norum, T.D. and Seiner, J.M. Broadband shock associated noise from supersonic jets, AIAA J, 1982,
20, (1), pp 68-73.
19. Krothapalli, A., Wishart, D.P. and Mungal, M.G. Supersonic jet control via point disturbances inside
the nozzle, AIAA J, 1993, 31, (7), pp 1340-1341.
20. Zaman K.B.M.Q., Reeder, M.F. and Samimy, M. Control of an axisymmetric jet using vortex Generators,
Physics of Fluids, 1994, 6, (2), pp 778-793.
21. Behrouzi, P. and McGuirk, J.J. Effect of tab parameters on near-field jet plume development, AIAA J
Propulsion and Power, 2006, 22, (3), pp 576-585.
22. Reeder, M.F. and Samimy, M. The evolution of a jet with vortex real-time visualization and quantitative
measurements, J Fluid Mechanics, 1996, 311, pp 73-118.
23. Gretta, W.J. and Smith, C.R. The flow structure and statistics of a passive mixing tab, J Fluids
Engineering, June 1993, 115, pp 255-263.
24. Dong, S. and Meng, H. Flow past a trapezoidal tab, J Fluid Mechanics, 2004, 510, pp 219-242.
25. Jorgensen, F.E. Directional sensitivity of wire and fibre-film probes, DISA Inf, 1971, 11, pp 31–37.
26. Verma, S.B., Sudhakar, S. and Venkatkrishnan, L. Studies on an elliptic jet with cylindrical tabs, J
Turbulence, 2010, 11, (20), pp 1–24.
27. Arun Kumar, P., Verma S.B. and Elangovan. S. Study of jets from rectangular nozzles with square
grooves, Aeronaut J, March 2011, 15, (1165), pp 187-196.
28. Fundamentals of Aerodynamics, John Anderson, 4th ed, McGraw Hill, SIE, pp 474-475.

3800.indd 517 03/05/2013 11:28:35