Proceedings of the Sixth Triennial International Symposium on Fluid Control, Measurement and Visualization,
Sherbrooke, Canada, August 13-17, 2000.
PASSIVE CONTROL ON JET MIXING FLOWS
BY USING VORTEX GENERATORS
Hui HU, Tetsuo SAGA, Toshio KOBAYASHI and Nobuyuki TANIGUCHI
Institute of Industrial Science,University of Tokyo
Roppongi 7-22-1, Minato-Ku, Tokyo 106-8558, Japan
ABSTRACT
In an effort to increase mixing in jet flows, a passive
control method, using vortex generators in the form of
mechanical tabs or small protrusions at the exit of a nozzle
had been investigated experimentally in the present paper.
The techniques of Planar Laser Induced Fluorescence
(PLIF) and Particle Image Velocimetry (PIV) were used to
accomplish flow visualization and velocity field
measurements of the jet mixing flows simultaneously. The
vortical and turbulent structure changes in the near field of
a jet mixing flow caused by mechanical tabs were studied
based on the PLIF and PIV simultaneous measurement
results. The experimental results revealed the great
changes of vortical and turbulent structures in the near
field of jet flow due to the mechanical tabs intrusion.
Compared with a nature jet flow (a circular jet flow
without mechanical tabs intrusion), the tabbed jet flow was
found to have shorter laminar length, smaller size of the
spanwise Kelvin-Helmholtz vortices, earlier appearance of
small scale turbulent structures. The existence of the
streamwise vortices generated by mechanical tabs was
verified clearly from both PLIF flow visualization and
simultaneous PIV measurement results. Due to the engulf
of the streamwise vortices, an inward indentation of
ambient flow into core jet flow and the outward ejection of
core jet flow into the ambient flow was found in the cross
planes of the tabbed jet. As the streamwise distance
increasing, the cross section form of the tabbed jet flow
was found to change from “round” to “oval” gradually.
Keywords: tabbed jet flow, passive control of jet
mixing, mixing enhancement, PIV and LIF technique
INTRODUCTION
In an effort to increase mixing in jet flows, a passive
control method, using vortex generators in the form of
mechanical tabs or small protrusions at the exit of a nozzle
had been under investigation in the past several years.
Bradbury and Khadem [1] were the first to study the effect
of mechanical tabs on jet flows in detail. They reported
that, for the low speed jet, mechanical tabs or small
protrusions in the jet flows at the exit of a nozzle can
increase jet spread rate significantly, reduce the potential
core length and even bifurcate the jet flows. Ahuja et
al.[2,3] and Zaman et al.[4-8] began to investigate the
mixing enhancement performance of mechanical tabs
systematically. They found that mechanical tabs not only
can increase the jet mixing for the low speed jets, but also
have good mixing enhancement performance at high speed
and high temperature jet flows as well. The mechanical
tabs had been used to suppress the jet noise of the air
breathe engines by Ahuja et al.[2,3] and Zaman et al.[4-8].
More recently, tabbed nozzle was also found to be used as
a fuel injector nozzle (Glawe et al.[9]) in supersonic
combustion chamber to enhance the mixing of fuel with
supersonic air.
About the fundamental research of how and why
mechanical tabs can enhance jet mixing process, Zaman et
al.[7] reported that large scale streamwise vortices can be
generated by mechanical tabs in jet flows and two sources
of the streamwise vortices were postulated in their paper.
In the research of the molecular mixing in a jet mixing
flow with mechanical tabs, Zhang et al.[10] found that
mechanical tabs can reduce jet transitional Reynolds
number and increase the molecular mixing about 35% at
the downstream location of six diameter of the nozzle. The
work of the Reeder and Samimy [11] revealed more detail
about the vortices and turbulent structures in tabbed jet
flows by using flow visualization and Laser Doppler
Velocimeter (LDV) measurements. They confirmed the
existence of the large-scale streamwise vortices caused by
mechanical tabs and reported the higher Reynolds stress
levels in the tabbed jet flow.
Although many important results have been obtained
through those previous works, much work still need in
order to understand the fluid dynamic mechanism of the
enhancement of mixing caused by mechanical tabs more
clearly. Regarding to the vortical and turbulent structure
changes in the near field of a jet flow caused by the
intrusion of mechanical tabs, the quantitative data are far
more limited. In the meanwhile, most of the previous
experiments were conducted by using Pitot probe, LDV or
Hot Wire Anemometer, with which it is very hard to reveal
the vortical and turbulent structures in tabbed jet mixing
flows instantaneously and globally due to the limitations
of these experimental techniques. In the present study, the
optical whole-field diagnostic techniques of Particle
Imaging Velocity (PIV) and Planar Laser Induced
Fluorescence (PLIF) were used to conduct flow
visualization and instantaneous velocity field measurement
of the tabbed jet mixing flow simultaneously. By using the
directly perceived PLIF flow visualization images and the
simultaneous velocity and vorticity distributions of the
PIV measurement results, the characteristics of the
evolution of the various turbulent and vortical structures in
the near field of tabbed jet mixing flows were discussed in
detail.
EXPERIMENTAL SYSTEM AND TECHNIQUES
Figure 1 shows the experimental set-up used in the
present research schematically. The test circular nozzle
was fixed in the middle of a water tank
(600mm*600mm*1000mm). Fluorescent dye (Rhodamine
B) for PLIF or PIV tracers (hollow glass particles, d=812µm) was premixed with water in a jet supply tank, and
jet flow was supplied by a pump. The flowrate of the jet
flow, which was used to calculate the representative
velocity and Reynolds numbers, was measured by a flow
meter. A cylindrical plenum chamber with honeycomb
structures was installed at the upstream of test nozzles to
insure the turbulent levels of the core jet flows at the exit
of test nozzles were less than 3%. An overflow system was
used to keep the water level in the test tank to be constant
during the experiment.
The pulsed illumination laser sheet was generated by a
double-pulsed Nd:YAG Laser system. After passing
through a second harmonic generator cell, the wavelength
of the light beams emitted from the double-pulsed
Nd:YAG Laser system is 532nm. By using a set of optics
(cylindrical lens and mirrors), the laser beam was bundled
in a planar laser sheet with thickness being about 1.5 mm.
The frequency of the double-pulsed illumination is 10 Hz.
The pulsed illumination duration is 4ns, and power is 200
mJ/pulse. The time interval between the two pulses is
adjustable, which was set as 3 ms in the present study.
In order to achieve the PLIF flow visualization and PIV
simultaneous measurement, a simultaneous image
recording system was designed by using at set of optics
and two high-resolution CCD cameras (TSI PIVCAM 1030, 1K by 1K resolution). The diagram of the
simultaneous image recording system was shown on the
right upper corner of the Figure 1.
Rhodamine B was used as fluorescent dye in the present
study. It was well known that the emission peak of
Rhodamine B is about 590nm, and the wavelength of the
illuminating laser light scattered by the PIV tracer
particles is 532nm. Two kinds of optical filters were used
in the present study to separate the LIF light from the
scattered laser light, and then recorded them separately to
obtain PLIF and PIV image simultaneously. As shown in
the left upper corner of the Figure 1, the combined light
including both LIF light (peak at 590nm) and scattered
laser light (532nm) were divided into two light beams by
using a beam splitter. Once light beam is go straight to
CCD camera #1 for PIV image recording. A bend pass
optical filter (532nm±5) was installed at the head of the
camera #1. Therefore, only the scattered laser light is
transmissible to generate PIV images in the camera #1, the
LIF light is blocked out. Another light beam from the
beam splitter is reflected by a mirror, then, goes into the
camera #2. A high pass filter (>580nm pass) was installed
in the head of the camera #2 to filter out the scattered laser
light (wavelength 523nm). The LIF light (peak at 590nm)
passed through the high-pass optical filter to generate LIF
image in camera #2. More detail about the simultaneous
PLIF and PIV image recording system can be got from Hu
et al.[14].
Mirror for cross plane measurement
Laser sheet
low pass optical filter
mixing region
Beam splitter
CCD camera #1
for PIV
Double-pulsed Nd:YAG Laser
CCD camera #2
for PLIF
High pass optical filter
mirror
water tank
Overflow system
A cylinderica
plenum chamber
jet supply tank
Test nozzle
pump
Flowmeter
Figure 1. Experimental set up
synchronizer
top view
8mm
90
a. test nozzle with mechanica tabs
1mm
Z
135
b. mechanical tab
Figure 2. Test nozzle and delta tabs
The double-pulsed Nd:YAG Laser and the simultaneous
image recording cameras were connected to a host
computer via a synchronizer, which controls the timing of
laser illumination and image acquisition. The host
computer is composed of two high-speed CPU (800MHz,
Pentium III, CPU), colossal image memory and Hard disk
(1GB RAM, Hard Disk 100GB). It can acquire the
continuous PIV and PLIF image pairs up to 250 frames
every time at the framing frequency of 10 Hz.
Figure 2 shows the test nozzle and mechanical tabs used
in the present research. The diameter of the circular nozzle
at exit was D=30mm, and the mechanical tabs are the
“delta tabs” suggested by Zaman et al.[7]. Each
mechanical tab had about 1.5% blockage area after
mounted on the nozzle exit. During the experiment, the
core jet velocity was about U0=0.14m/s and 0.27m/s. The
Reynolds Number of the jet flows, based on the circular
nozzle exit diameter and core jet velocity, were about
4,000 and 8,000.
For the PIV image processing, rather than tracking
individual particle, cross correlation method [13] was used
in the present study to obtain the average displacement of
the ensemble particles. The images were divided into 32
by 32 pixel interrogation windows, and 50% overlap grids
were employed. The resolution the PIV images for the
present research is about 120µm/pixel. The postprocessing procedures which including sub-pixel
interpolation[14] and spurious velocity deletion [15] were
used to improve the accuracy of the PIV result.
RESULTS AND DISCUSSIONS
1). In the axial sections
PLIF flow visualization and simultaneous PIV
measurement results for a natural jet flow (circular jet flow
without mechanical tab intrusion) were shown in the
Figure 3. It can be seen that, for a natural jet flow, three
different regions, which are laminar, transition and
turbulent regions, can be identified clearly in the figures.
A laminar region exists at the downstream of the circular
nozzle trailing edge. At the end of laminar region,
spanwise Kelvin-Helmholtz vortices were found to roll up.
side
view
Y
c. the two veiws
The pairing and combining of the spanwise KelvinHelmholtz vortices conducted in the transition region. In
much downstream, the big Kelvin-Helmholtz roller was
found to break down, many small-scale turbulent and
vortical structures were found in the flow field, and jet
flow transited to turbulence. All these process can be seen
clearly from both the PLIF visualization image(Fig.3(a))
and simultaneous PIV measurement result (Fig. 3(b)).
400 frames of instantaneous PIV results were used to
calculated the ensemble-averaged values of the flow
parameters, which include mean velocity, turbulent
intensity distribution and Reynolds stress field. The mean
velocity field of the nature jet flow was given on the
Figure 3(c). It can be seen that the nature jet flow
expanded linearly along the downstream.
Figure 4 gives the PLIF flow visualization image and
simultaneous PIV measurement results in the side view
slice (X-Z plane, Fig 2(c)) of the tabbed jet mixing flow.
Compared with the natural jet flow, the laminar region of
the tabbed jet flow in this slice was not straight. It became
a convergent region instead. This can be explained by that
there was a local contraction at the exit of the nozzle due
to the existence of mechanical tabs. The length of the
laminar region was much less than that in the natural jet
flow. It also can be found that the scale of spanwise
vortices rolled up by the Kelvin-Helmholtz instability was
much smaller, and the transition region also became much
shorter. The small-scale vortices appeared earlier in the
tabbed jet flow and the flow field is much more zigzag
instead of symmetrical in the natural jet flow. From the
ensemble-averaged PIV results, the expend angle of the
tabbed jet in the side view section is found to be smaller
than that of the nature jet.
Figure 5 shows the PLIF flow visualization image and
simultaneous PIV measurement results for the tabbed jet
flow in the top view section (X-Y plane, Fig.2(c)). It can
be seen that, similar to the natural jet flow, a straight
laminar region can be found in the tabbed jet flow at this
section, but it was a bit shorter than that in the nature jet.
Small scale turbulence and vortical structures appeared
earlier in the jet flow. From the ensemble averaged results,
the trailing edge of the mechanical tabs. The existence of
the streamwise vortex pairs generated by mechanical tabs
can be seen clearly from the simultaneous PIV
measurement result and ensemble-averaged velocity
distributions. Due to the engulf effect of the streamwise
vortices generated by the mechanical tabs, the ambient
flow was pumped into the core jet in the mechanical tab
intrusion plane and core jet flow was extracted out in the
plane normal to the mechanical tab intrusion. So, an
inward indentation of ambient flow into core jet flow and
the outward ejection of core jet flow into the ambient flow
can be observed in the ensemble averaged PIV result.
it can also be seen that the spread angle of the tabbed jet in
this axial section was much bigger that in the side view
slice and that in the nature jet flow.
2). In cross planes
Figure 6 shows the PLIF flow visualization and the
simultaneous PIV measurement results in the X/D=0.5
cross plane of the tabbed jet mixing flow. An inward
indentation at the downstream of the mechanical tab
intrusion locations can be seen clearly from the PLIF flow
visualization image. Small-scale turbulent structures were
found to appear in the inward indentations, which is due to
the earlier shedding of the Kelvin-Helmholtz vortices from
4.5
4.5
4
4
0.2 m/s
3.5
3.5
3
3
2.5
2.5
2
2
1.5
1.5
0.3m/s
1
1
0.5
0.5
0
0
-2
-1
0
1
2
-2
3
-1
0
1
2
3
a. PLIF flow visualization
b. simultaneous PIV velocity vectors
d. ensemble averaged velocity
Figure 3. PLIF and PIV simultaneous measurement results in the axial section of the nature jet (Re=8,0000)
4.5
4.5
4
4
0.2 m/s
3.5
3.5
3
3
2.5
2.5
2
2
1.5
1.5
0.3m/s
1
1
0.5
0.5
0
0
-2
-1
0
1
2
-2
3
-1
0
1
2
3
a. PLIF flow visualization
b. simultaneous PIV velocity vectors
d. ensemble averaged velocity
Figure 4. PLIF and PIV simultaneous measurement results in the side view section of the tabbed jet (Re=8,0000)
4.5
4.5
4
4
0.2 m/s
3.5
3.5
3
3
2.5
2.5
2
2
1.5
1.5
0.3m/s
1
1
0.5
0.5
0
0
-2
-1
0
1
2
3
-2
-1
0
1
2
3
a. PLIF flow visualization
b. simultaneous PIV velocity vectors
d. ensemble averaged velocity
Figure 5. PLIF and PIV simultaneous measurement results in the top view section of the tabbed jet (Re=8,0000)
1
1
0.05 m/s
0.5
0.5
0.05m/s
0
0
-0.5
-0.5
-1
-1
-1
0
1
-1
-0.5
0
0.5
1
a. PLIF flow visualization
b. simultaneous PIV measurement result c. ensemble-averaged PIV result
Figure 6. PLIF and PIV simultaneous measurement results in the X/D=0.5 cross section of the tabbed jet (Re=4,000)
1
1
0.05 m/s
0.5
0.5
0.05m/s
0
0
-0.5
-0.5
-1
-1
-1
0
1
-1
-0.5
0
0.5
1
a. PLIF flow visualization
b. simultaneous PIV measurement result c. ensemble-averaged PIV result
Figure 7. PLIF and PIV simultaneous measurement results in the X/D=1.0 cross section of the tabbed jet (Re=4,000)
1
1
0.05 m/s
0.5
0.5
0.05m/s
0
0
-0.5
-0.5
-1
-1
-1
0
1
-1
-0.5
0
0.5
1
a. PLIF flow visualization
b. simultaneous PIV measurement result c. ensemble-averaged PIV result
Figure 8. PLIF and PIV simultaneous measurement results in the X/D=2.0 cross section of the tabbed (Re=4,000)
1
1
0.05 m/s
0.5
0.5
0.05m/s
0
0
-0.5
-0.5
-1
-1
-1
0
1
-1
-0.5
0
0.5
1
a. PLIF flow visualization
b. simultaneous PIV measurement result c. ensemble-averaged PIV result
Figure 9. PLIF and PIV simultaneous measurement results in the X/D=3.0 cross section of the tabbed jet (Re=4,000)
At X/D=1.0 cross plane(Figure 7), the inward
indentations in to the core jet flow due to the intrusion of
the mechanical tabs were found to become bigger. Much
more small-scale turbulent and vortical structures at the
inward indentations can be found from the PLIF flow
visualization image. The mushroom structures can also be
seen in the flow field due to the growth up of the azimuthal
instability[16]. The simultaneous PIV measurement result
shows that the jet flow will be more turbulent than that in
the X/D=0.5 cross plane. Due to the "engulf effect" of the
streamwise vortices, the jet flow was found to expend
rapidly in the plane normal to the mechanical intrusion
section. This also verified the measurement results of that
the tabbed jet flow was found to spread more rapidly in the
top view plane than that in the side view plane, which were
given in Fig.4 and Fig.5.
As the streamwise distance increased to X/D=2.0 (Figure
8) and X/D=3.0(Figure 9), small-scale turbulent and
vortical structures was found to almost fill the whole flow
field from the PLIF visualization results. The instantaneous
velocity vector fields of the simultaneous PIV measurement
also show that the flow fields become much more turbulent
as the downstream distance increasing, and many
streamwise vortices can be identified in the instantaneous
velocity vector fields, which may be due to the growth of
the azimuthal instability. Form the ensemble-averaged
results of the PIV measurement, it can be seen that the
cross section form of the tabbed jet flow was found to
change from “round” to “oval” gradually due to the "engulf
effect" of the streamwise vortices generated by the
mechanical tabs.
CONCLUSIONS
The PLIF and simultaneous PIV measurement results of
the present investigation revealed the great changes of the
turbulent and vortical structures in the near field of a jet
flow by the intrusion of the mechanical tabs. Compared
with a natural jet flow without the mechanical tab intrusion,
the tabbed jet flow was found to have shorter laminar
region length, smaller scale of the spanwise KelvinHelmholtz vortices and earlier appearance of small-scale
turbulent and vortical structures in the axial planes. Jet
spread angle in the mechanical tab intrusion plane was
found to be smaller than that in the plane normal to the
mechanical tab intrusion.
The existence of the streamwise vortices generated by the
mechanical tabs can be seen clearly from both PLIF flow
visualization and simultaneous PIV measurement results in
the cross planes of the tabbed jet flow. Due to the engulf of
the streamwise vortices, an inward indentation of ambient
flow into core jet flow and the outward ejection of core jet
flow into the ambient flow occurred in the tabbed jet. As
the streamwise distance increasing, the cross section form
of the tabbed jet flow was found to change from “round” to
“oval” gradually.
REFERENCES
[1].Bradbury L.J.S. and Khadem, A.H. The Distortion of a
Jet by Tabs, J. of Fluid Mech., Vol.70, 1975,80l-8l3.
[2]. Ahuja K.K., Manes J.P. and Massey K. C., An
Evaluation of Various Concepts of Reduction
Supersonic Jet Noise, AIAA paper90-3982, 1990.
[3]. Ahhja K. K. Mixing Enhancement and Jet Noise
Reduction Through Tabs Plus Ejector, AIAA paper934347, 1993.
[4]. Zaman K.B.M.Q., Reeder, M.F.and Samimy, M. Effect
of Tabs on the Evaluation of an Axisymmetrical Jet,
NASA-TM104472, 1991.
[5]. Zaman K.B.M.Q., Reeder, M.F. and Samimy, M.
Supersonic Jet Mixing Enhancement by Delta Tabs,
AIAA paper92-3548, 1992.
[6]. Zaman K.B.M.Q, Streamwise Vorticity Generation and
Mixing Enhancement in Free Jet by "Delta
Tabs" AIAA paper93-3253, 1993.
[7]. Zaman K. B. M. Q., Reeder, M.F. and Samimy, M.
Control of an Axisymmetric Jet Using Vortex
Generators, Phys. Fluid, Vol.6 No2., 1994, 778-792.
[8]. Zaman K. B. M. Q. Axis Switching and Spreading of
an Asymmetrical Jet: the Role of Coherent Structure
Dynamics, J. Fluid Mech. Vol.316, 1996, 1-27.
[9]. Glawe, D. D. Samimy, M., Najad A. S. and Chen, T.
H., Effects of Nozzle geometry on Parallel Injection
into a Supersonic Flow, Journal of Propulsion and
Power Vol.12 No.6, 1996, 1159-1168.
[10]. Zhang S. and Scheider S.P. Molecular-Mixing
Measurements and Turbulent Structure Visualizations
in a Round Jet with Tabs, AIAA paper94-3082, 1994
[11]. Reeder M. F. and Samimy M., The Evolution of a Jet
with Vortex generating tabs: real-time visualization and
quantitative measurement, J. Fluid Mech. Vol.31l,
1996, 733-748.
[12]. Hu H., Saga T. Kobayashi T., and Taniguchi N.,
Simultaneous
Velocity
and
Concentration
Measurements in a Turbulent Jet Flow by Using PIVPLIF combined System”, Proceedings of the 4th JSMEJSME Joint Conference, 2000, Sept. Kobe, Japan
[13]. Willert, C.E. and Gharib, M., Digital Particle Image
Velocimetry. Exp. in Fluids Vol.l0, 1991, l8l-l93.
[14]. Hu, H., Saga, T., Kobayashi, T., Okamoto, K. and
Taniguchi, N., Evaluation of the Cross Correlation
Method by Using PIV Standard Images, Journal of
Visualization, Vol.1, No.1, 1998, 87-94.
[15]. Westerweel, J. Efficient Detection of Spurious
Vectors in Particle Image Velocimetry Data, Exp. In
Fluids, Vol.16, 1994, 87-94.
[16]. Liepmann D. and Gharib, M. The Role of
Streamwise Vorticity in the Near Field Entrainment of
Round Jet, J. Fluid Mech. Vol.254, 643-668, 1992.