Experimental Thermal and Fluid Science 35 (2011) 15051513

Contents lists available at ScienceDirect

Experimental Thermal and Fluid Science

journal homepage: www.elsevier.com/locate/etfs

Turbulent boundary layer separation control and loss evaluation of low prole
vortex generators
Davide Lengani a,1, Daniele Simoni a,, Marina Ubaldi a, Pietro Zunino a, Francesco Bertini b
a
b

DIMSET, Universit di Genova, Via Montallegro 1, I-16145 Genova, Italy

Avio R.D., Via I Maggio, 99 I-10040 Rivalta (TO), Italy

a r t i c l e

i n f o

Article history:
Received 7 March 2011
Received in revised form 30 June 2011
Accepted 30 June 2011
Available online 8 July 2011
Keywords:
Turbulent separation
Vortex generator
Deformation work
Total pressure losses
Turbine intermediate duct
Dissipation mechanisms

a b s t r a c t
The present paper analyses the results of a detailed experimental study on low prole vortex generators
used to control the turbulent boundary layer separation on a large-scale at plate with a prescribed
adverse pressure gradient, typical of aggressive turbine intermediate ducts. This activity is part of a joint
European research program on Aggressive Intermediate Duct Aerodynamics (AIDA). Laser Doppler Velocimetry and a Kiel total pressure probe have been employed to perform measurements in the test section
symmetry plane and in cross-stream planes to investigate the turbulent boundary layer, with and without control device application.
Velocity elds, Reynolds stresses, and total pressure distributions are analysed and compared for the
controlled and non controlled ow conditions to characterize the mean ow behaviour. The detail and
the accuracy of the measurements allow the evaluation of the deformation works of the mean motion
in the test section symmetry plane. Normal and shear contributions of viscous and turbulent deformation
works have been analysed and employed to explain the distribution of the total pressure loss. For the
controlled ow the discussion of the ow eld is extended considering the effects of the vortex generated
in the cross-stream planes. The experimental data allow the evaluation of the global amount of losses,
considering a balance of total pressure uxes in the different measuring planes.
2011 Elsevier Inc. All rights reserved.

1. Introduction
In the last 30 years many efforts have been done to apply ow
control devices inside a real environment in a reliable and efcient
way. Even though the concept of boundary layer control was introduced by Prandtl [1] at the beginning of the 20th century, only
recently it has been thought to control the ow inside complex
machines such as aeroengines. For a modern turbomachine, one
of the most interesting applications of boundary layer control is
the prevention of ow separation. Boundary layer separation is in
fact one of the main causes of total pressure losses and, therefore,
the prevention of separation may have a positive impact on turbomachinery efciency, or the suppression/delay of separation may
allow the application of more aerodynamically loaded airfoils
and ducts, without decay of aerodynamic performances [2]. For
these reasons, the investigation of boundary layer separation
control methods applied to internal aeroengine ows becomes of
primary importance.
For external aerodynamics, different ow control devices have
been proposed and often employed to avoid boundary layer separa Corresponding author. Tel.: +39 010 353 2447; fax: +39 010 353 2566.
E-mail address: daniele.simoni@unige.it (D. Simoni).
Present address: Institute for Thermal Turbomachinery and Machine Dynamics,
Graz University of Technology, Austria.
1

0894-1777/$ - see front matter 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.expthermusci.2011.06.011

tion. Control devices can be passive, requiring no auxiliary power,

or active, requiring energy expenditure [3]. Up to now, one of the
most tested passive devices are the vortex generators (VGs), thanks
to their relatively easy applicability. They consist of little wings
embedded in the boundary layer, which generate a tip vortex able
to apply momentum transfer from the outer to the inner region of
the boundary layer. Until the1980s their height (VGs characteristic
dimension) was of the order of the boundary layer thickness [4], but
later, in order to reduce parasitic drag, different authors [57] introduced sub-boundary layer VGs, also called low prole VGs (height
approximately 20% of the boundary layer thickness). Recently,
VGs have been applied to internal ows, in particular to diffusing
ducts, with two different purposes: secondary ow control and
mixing enhancement. Reichert and Wendt [8,9] carried out tests
on circular S-ducts with tapered-n VGs. Sullerey et al. [10] experimentally investigated the effects of various vortex generator congurations in reducing the exit ow distortion and improving
pressure recovery in two-dimensional S-duct diffusers.
Flow simulation of vortex generators effects has been carried
out using two different numerical approaches: VGs may be included in the computational domain modelling their geometry,
or VGs may be replaced by volume forces that create vortices in
an analogous way VGs do. This last approach seems computationally more efcient than the rst one and it has been implemented
in several works (e.g. [1114]). Most of these works came out from

1506

D. Lengani et al. / Experimental Thermal and Fluid Science 35 (2011) 15051513

Nomenclature
Cf
Cpt

skin friction
total pressure coefcient =

(Dn)Tx

deformation work operated by the Reynolds normal

stresses along the x direction
deformation work operated by the Reynolds shear stresses along the x direction
deformation work operated by the Reynolds normal
stresses along the y direction
test section inlet height
boundary layer shape factor
VGs length
static pressure
total pressure
inlet Reynolds number = U0H0/m
momentum thickness Reynolds number = U0h/m
turbulence intensity

(Dt)Tx
(Dn)Ty
H0
H12
l
p
pt
Re
Reh
Tu

pt0 pt
0:5qU 20

the necessity to determine the optimal VGs geometry for different

ow conditions. In the last years the aforementioned CFD works,
theoretical models (e.g. Smith [15] and Velte et al. [16]), as well
as extensive experimental parametric analyses (such as Wendt
[17]) produced a large amount of data concerning global performance evaluation parameters (as reviewed by Lin [18]), which
may allow a relatively easy and efcient implementation of VGs
in common applications. Nevertheless, for the design and prediction of ow control in more complex environments more data
are required on turbulence and loss generation mechanisms.
In the present work, which is part of the European research
project AIDA (Aggressive Intermediate Duct Aerodynamic for
Competitive and Environmentally Friendly Jet Engines), loss generation mechanisms are analysed with and without VGs in a linear diffuser, the pressure gradient of which is typical for newly
designed aeroengine intermediate ducts. Since in aeroengine
intermediate ducts the boundary layer is fully turbulent and
prone to separation, this condition has been imposed at the inlet
of the diffuser test section. In previous works [19,20], carried out
in the same test rig, the effectiveness of low-prole VGs was demonstrated and the mechanism by which VGs modify the ow
structure was explained by means of complementary measurement techniques.
The results presented herein concern a measurement campaign carried out by means of LDV and a Kiel total pressure probe
aimed at understanding the loss production mechanisms in a separating boundary layer in both controlled and uncontrolled cases.
Measurements have been performed in the test section symmetry
plane with and without VGs, and in two cross-stream planes located downstream of the VGs. Reynolds stresses and velocity gradients in the meridional plane have been adopted to evaluate the
turbulent deformation works contributing to loss generation. The
results conrm some important features of the non-controlled,
fully separated boundary layer (for an extensive review of turbulent boundary layer separation see Simpson [21]). For the
controlled ow some important differences in loss generation
mechanism are revealed, although the deformation work terms
in the symmetry plane are only partially suitable to justify the total pressure distribution, because of the three-dimensional ow.
The distributions of the total pressure coefcient in the crossstream planes are explained by analyzing the effects of the
cross-stream vortex action. A quasi-3D method of the overall loss
evaluation from the velocities and the total pressure measured in
the meridional and the cross-stream planes is proposed and
adopted for the controlled case. In this method overall losses

u
u0 v 0
rms(u0 )
U0
v
x
y
z
d
dloss
d0loss
h

l
m
q

streamwise velocity
Reynolds shear stress in the (x, y) plane
streamwise velocity uctuations root mean square
inlet free-stream velocity
velocity normal to the wall
axial coordinate
coordinate normal to the wall
cross-stream coordinate
boundary layer thickness
loss coefcient
loss coefcient as a function of cross-stream coordinate
boundary layer momentum thickness
dynamic viscosity
kinematic viscosity
density

are evaluated from the balance of the total pressure uxes and
not, as classically, as a mass-weighted total pressure loss coefcient, since this last approach seems to be questionable for a separated boundary layer.
2. Experimental apparatus and methodology
2.1. Facility
A detailed description of the open-loop low-speed wind tunnel
used for this study is given in Canepa et al. [19].
The test section (Fig. 1) was designed to provide several adverse
pressure gradients typical for aeroengine diffusers. The boundary
layer develops on a large-scale at plate 1700 mm long and
400 mm wide with the leading edge located about 600 mm upstream of the test section inlet (x = 0). The inlet test section height
H0 is 196 mm. Before the test section inlet, the lateral and top wall
boundary layers are sucked to avoid inlet section blockage and to
obtain a two-dimensional ow inside the test section. Furthermore, on the inclined top wall, the boundary layer was controlled
applying suction, to avoid interaction with the lower plate boundary layer, where the experiments were performed.
For the experiment analysed in this paper the test section top
wall was inclined by 16, which corresponds to an ideal overall
acceleration factor K value of 3.16  107. This parameter represents the non dimensional velocity gradient of an ideal one-dimensional ow between the inlet and the outlet of the test section. The
mean velocity at the test section inlet was kept constant for all
experiments at the value of 28.1 m/s. The boundary layer parameters and free-stream turbulence intensity in the inlet section are
reported in Table 1.

Fig. 1. Experimental test section.

1507

D. Lengani et al. / Experimental Thermal and Fluid Science 35 (2011) 15051513

Table 1
Inlet section ow parameters.

Table 2
Vortex generators: geometrical parameters.

d (mm)

h (mm)

H12

Reh

Cf

Tu

VGs no.

h = c (mm)

l (mm)

s (mm)

a ()

b ()

80

56

1.24

11000

0.0032

0.01

16 (0.2d)

64 (4/5d)

80 (d)

23

25

2.2. Control device conguration

In order to control boundary layer separation, low-prole vortex
generators are employed. A brief description of their geometrical
parameters is reported here, while a more comprehensive explanation of their choice may be found in Canepa et al. [19]. The VGs are
arranged in a co-rotating pattern at an angle a = 23 to the incoming
ow and the tunnel centreline provides a plane of symmetry for the
conguration, as shown in Fig. 2. The VGs geometrical dimensions,
as dened in this picture, are reported in Table 2. The VGs axial location which was most effective in delaying separation was established in the previous work by Canepa et al. [19]. It resulted to be
95 mm (x/H0 = 0.48) upstream of the detachment point, that was
detected without VGs at x = 385 mm (x/H0 = 1.93).
2.3. Measuring techniques
Total pressure measurements were performed by means of a
miniaturized Kiel total pressure probe. The transducer output
was sampled at a frequency of 1 kHz by means of a Metrabyte A/
D converter board with a 12 bit resolution. The total number of
samples collected for each position was 10,000.
A four-beams two-colour Laser Doppler Velocimeter (Dantec Fiber Flow), in backward scatter conguration, was employed for the
velocity measurements [20]. The probe consists of an optical transducer head of 60 mm diameter, with a focal length of 300 mm and a
beam separation of 38 mm, connected to the emitting optics and to
the photomultipliers by means of optic bres. The probe volume
dimensions are 0.09 mm  0.09 mm  1.4 mm. The ow was
seeded with mineral oil droplets with a mean diameter of 1.5 lm.
To process the bursts, two Dantec Enhanced Burst Spectrum Analysers were employed. For each measurement point 30,000 samples
were collected with a maximum record length in time of 120 s.
2.3.1. LDV experimental uncertainty
A specic evaluation of errors for LDV frequency domain processors is given by Modarress et al. [22]. For the present experiment
the uncertainty of the instantaneous velocity was evaluated to be

less than 1%. Statistical moments were weight-averaged with transit time to avoid statistical bias. Statistical uncertainty in mean and
rms velocities depends on the number of independent samples, the
turbulence intensity based on the local velocity and the condence
interval. Thanks to the large number of samples (30,000) the statistical uncertainty on the mean velocity was estimated lower than 4%
for a probability of 95% and a local turbulence intensity of 100%,
which may occur in the near wall region. In order to obtain high
accuracy also in the boundary layer separating region, where the
integral time scale of the ow increase, the data rate has been reduced acting on the photomultiplier voltage, while the acquired
number of samples has been kept constant. Thus, according to
George [23] and Satta et al. [24] the sampling period has been always kept larger than 1000 times the integral time scale of the ow.
2.3.2. Experiment organization
The boundary layer developing over the at plate was surveyed
along traverses normal to the wall. For the case without VGs, measurements were performed only in the symmetry plane, while for
the case with VGs measurements were performed in the symmetry
plane and in two cross-stream planes, as depicted in Fig. 3, which
shows the Kiel miniaturized total pressure probe measurement
grid.
For the LDV analysis each normal to the wall traverse was constituted of 103 measurement points along the normal to the wall
direction. The rst point was located at 50 lm from the wall (corresponding to y+ 2 at ow separation position) and the distance
between adjacent points was progressively increased in the outer
ow region.
The probe volume was oriented with the larger dimension along
the plate spanwise direction z in order to have better spatial resolution in the x and y directions. The LDV probe was moved using a
three-axis computer controlled traversing mechanism with a minimum linear translation step of 8 lm. The measurements of the
two velocity components were performed in coincidence mode.
Typical data rate was 2000 Hz. As mentioned before, to take into
account the increase of the ow integral time scale, the data rate
was reduced to about 50 Hz in the separating ow region.

Fig. 2. Vortex generators conguration.

1508

D. Lengani et al. / Experimental Thermal and Fluid Science 35 (2011) 15051513

y [mm]

150
600
100

500
400

50

300
200

0
0

z [m50 100
m] 150

100

m]
x [m

Fig. 3. Streamwise and cross-stream measurement grid.

For the total pressure measurements, each normal to the wall

traverse was constituted of 71 measurement points along the normal direction. The rst point was located at 1 mm from the wall
with a distance between adjacent points of 1 mm in the region of
the boundary layer close to the wall, that was progressively increased in the outer part.
Results from total pressure measurements are presented in
terms of a non dimensional total pressure loss coefcient:

C pt

pt0  pt
0:5qU 20

where pt0 is the total pressure in the undisturbed ow at x = 0 mm,

pt is the local total pressure, and the pressure difference is normalized by the inlet dynamic pressure.
3. Results and discussion
3.1. Flow analysis in the meridional plane

was identied at x = 300 mm, where the maximum of rms(u0 ) starts

to depart from the wall. Moving downstream along the duct the
separated ow area extends in the direction normal to the wall
and the maximum of the velocity standard deviation moves away
from the wall. This maximum may be identied within the separated shear layer just above the zero mean velocity line.
With VGs application, the boundary layer separation is delayed
(Fig. 4, bottom). The most relevant differences with respect to the
baseline case can be observed dowstream of the VGs position
(x = 350). Here, the ow is accelerated close to the wall since
streamwise momentum is transferred from the outer to the inner
part of the boundary layer. This streamwise momentum migration
toward the wall prevents separation, as analyzed in detail by Satta
et al. [20]. This ow transfer is one of the main effects of the vortices generated by the control device [19] and represents the main
mechanism through which the ow overcomes separation. At the
same time these vortices induce coherent velocity uctuations
[13] and, as a consequence, high Reynolds shear stresses u0 v 0 which
are much larger than for the baseline case, as shown in Fig. 5.
The distributions of the total pressure loss coefcient Cpt are
shown in Fig. 6. For the baseline ow the total pressure loss increases rapidly in the y direction around the detachment position
(x = 385 mm), showing a behaviour similar to the velocity standard
deviation. On the contrary, near to the wall the value of Cpt is lower
after the boundary layer has separated. As discussed in Satta et al.
[20], in this region the wall shear stress is reduced due to low
velocity and low turbulence occurring within the separated region
and thus, the total pressure loss is reduced as well.
When VGs are applied the total pressure loss growth is reduced
(Fig. 6, bottom). The acceleration of the ow induced by the VGs
leads to a signicant reduction of the total pressure loss coefcient
near to the wall with respect to the baseline case, but even with respect to the boundary layer development upstream of the VGs
location. Losses start to increase again further downstream, but
at the exit plane the coefcient Cpt is about 10% lower compared
to the uncontrolled case in the boundary layer outer region, and
it is further reduced up to 20% in the near wall region.

The baseline ow evolution is reported on top of Fig. 4; the

decelerating ow detaches at x = 385 mm, as shown by the time
mean velocity vectors. The boundary layer prole at x = 385 mm
(x/H0 = 1.93) is in fact characterized by a zero mean velocity gradient in y direction at the wall. From the PIV measurements analysed
in Canepa et al. [19] it was found that the detachment position is
characterized by back-ow for a fraction of time equal to 50%, in
agreement with Simpson [21]. The beginning of backow events

3.1.1. Deformation works without boundary layer control

The distribution of the total pressure loss coefcient may be
better explained taking into account the viscous and turbulent contributions to the overall deformation work of the mean motion.
The present LDV results are suitable for evaluating the deformation
works operated by the normal and shear Reynolds stresses with a
reasonable accuracy which is of the order of 10%.

Fig. 4. Mean velocity vector plots and rms velocity colour plots: baseline case (top)
and controlled case (bottom).

Fig. 5. Reynolds shear stress colour plots: baseline case (top) and controlled case
(bottom).

D. Lengani et al. / Experimental Thermal and Fluid Science 35 (2011) 15051513

Fig. 6. Total pressure loss coefcient: baseline ow (top), controlled ow (bottom).


u


u

 2 =2
 2 =2 u
 @p
@u
@u
v

@x
@y
q @x
 






1 @
@u
1 @u
@u
 l


u
 qu0 2
l  qu0 2
@x
@x
q @x
q @x
 






1 @
@u
1
@
u
@u
 l

u
 qu0 v 0
l  qu0 v 0

@y
q @y
q @y @y
@v 2 =2
@v 2 =2 v @p
v

@x
@y
q @y
 



1 @
@ v
1 @ v
@ v

v l  qv 0 2 
l  qv 0 2
q @y
@y
q @y
@y
 



1 @
@ v
1 @ v
@ v
0
0
v l  qu v 
l  qu0 v 0

q @x
@x
q @x
@x









@u
1 @u
@u
~  rpt  1 @ u
l  qu0 2 
l  qu0 v 0
U
@x
q
q @x
q @y @y




1 @ v
@ v
1 @ v
@ v
l  qv 0 2 
l  qu0 v 0 TD VW

q @y
@y
q @x
@x
4
Considering a two-dimensional incompressible ow, the mean
ow energy equations along the x and y directions may be written
as Eqs. 2 and 3, respectively [25]. The sum of these two equations
(Eq. 4) leads, on the left hand side, to the convective derivative of
the total pressure. The dissipation terms, appearing on the right
hand side of Eq. 4, have been computed from the LDV velocity results, while the remaining terms, which represent the turbulent
diffusion (TD) and the work done by the viscous forces (VW) on
the control volume, have not been evaluated, because of their negligible contribution according to Moore et al. [26].
 =@x2 ; l@ u
 =@y2 ; l@ v
 =@y2 and l@ v
 =@x2
The terms l@ u
represent the deformation work of the mean motion due to the viscous stresses. All these terms appear with a negative sign and
consequently they reduce the global amount of the mean ow
mechanical energy. These contributions are not negligible only in
the region close the wall upstream of the separation onset, where
high velocity gradients normal to the wall are present. Downstream of the detachment point the viscous terms are negligible
due to the moderate strain rate of the ow.
The major contributions to the dissipation of the mean ow total pressure are due to the deformation work operated by the Rey

nolds stresses. These terms qu0i u0j @ ui =@xj , appearing on the right
side of Eq. 4, are also involved in the turbulent kinetic energy
transport equation [25], but with opposite sign. They act exchang-

1509

ing energy between mean ow and turbulence and, if positive, they

dissipate the mean ow total pressure and increase the turbulent
kinetic energy.
The terms more relevant in the total pressure loss production
rate have been found to be the deformation work operated by
the Reynolds normal and shear stresses along the x direction
 =@x and Dt Tx qu0 v 0 @ u
 =@y, respectively) and
(Dn Tx qu02 @ u
the deformation work operated by Reynolds normal stress along
 =@y. Their distributions are rethe y direction Dn Ty qv 02 @ v
ported for the uncontrolled condition in Fig. 7.
Up to x = 300 mm, the near wall losses are increased mainly by
the term (Dt)Tx. After this position, both near wall velocity gradi =@y and Reynolds shear stress u0 v 0 decrease in the separatents @ u
ing boundary layer (Fig. 5), thus reducing the term (Dt)Tx, which
vanishes at the separation onset location. On the contrary, the
deformation work operated by the Reynolds normal stress along
the x direction (Dn)Tx starts to generate signicant losses at
x = 200 mm. After the back ow begins (x = 300 mm) the term
(Dn)Tx shows a steep increase due to the large ow oscillations,
and it becomes the main responsible of loss generation.
Consequently, the distributions of the terms (Dt)Tx and (Dn)Tx
clearly show that the classical assumption that the normal component of the deformation work is negligible cannot be applied for a
separating turbulent boundary layer. On the contrary, in this condition it is the shear contribution to be negligible. The large values
of the term (Dn)Tx depend on the velocity uctuations generated by
the low frequency phenomena, which occur on the edge of the separated ow region. The contour plot shown on the middle of Fig. 7
is, in fact, very similar to the contour plot of rms(u0 ) (Fig. 4, top).
Above the separated ow region the energy subtracted to the main
ow by (Dn)Tx is converted in turbulent kinetic energy and this fact
justies the large increase of Cpt observed in Fig. 6.
Inside the region of the reverse ow, both mechanisms of loss
generation are almost negligible (top and middle of Fig. 7). Thus,
as it came out from total pressure measurements, the total pressure losses do not increase in magnitude (Fig. 6).
The only non-negligible contribution to the total pressure dissipation rate along the y direction is given by the deformation work
operated by the Reynolds normal stress (Dn)Ty (Fig. 7 on bottom).
This term is due to the channel divergence that forces uid to move
away from the wall, and to the normal velocity uctuations. The
distribution of this term resembles that one of the normal dissipation term acting along the x direction and starts to be signicant as
backow appears, but shows negative values. It means that the
term (Dn)Ty performs a mean ow re-energization due to the positive gradient of the normal velocity along the y direction. Hence, it
tends to increase the mean ow energy subtracting it from the
turbulence.
3.1.2. Deformation works with boundary layer control
The contributions to loss generation due to (Dt)Tx, (Dn)Tx and
(Dn)Ty, evaluated with VGs installed in the test section, are reported
in Fig. 8. These terms give the greater contributions to the total
pressure loss also with boundary layer control. In this case, the viscous terms are not negligible downstream of the VGs due to the
strong normal velocity gradients induced by the momentum transferred towards the wall. However, their effects are limited to the
very near wall region and therefore they are not shown.
The deformation work operated by the Reynolds shear stress
along the x direction (Dt)Tx (Fig. 8, top), downstream of x =
350 mm, is larger than for the baseline ow. As previously described, VGs modify signicantly the boundary layer shape introducing distortions along the y direction, as well as larger
Reynolds shear stress u0 v 0 compared to the uncontrolled case, as
it has been discussed in Section 3.1. As a consequence, also (Dt)Tx
is larger for the controlled ow.

1510

D. Lengani et al. / Experimental Thermal and Fluid Science 35 (2011) 15051513

Fig. 7. Deformation works operated by Reynolds shear stress (top) and Reynolds normal stresses (middle and bottom) without boundary layer control.

Fig. 8. Deformation works operated by Reynolds shear stress (top) and Reynolds normal stresses (middle and bottom) with boundary layer control.

Hoverer, the Reynolds normal stress contribution (Dn)Tx (middle

of Fig. 8) is characterized by the largest values and consequently
provides the greater contribution to the increase of total pressure
losses observed in Fig. 6. Fluctuations of the streamwise velocity,
that are responsible for the growth of (Dn)Tx, are mainly generated
by the instationary vortices induced by VGs. Anyway, velocity uctuations are smaller as compared with the uncontrolled condition
where backow occurs (Fig. 4, top), thus (Dn)Tx is reduced as well.
This justies the considerable reduction of Cpt observed at the test
section exit plane with ow control as compared with the uncontrolled case (Fig. 6).

The presence of VGs modies also the shape of (Dn)Ty with respect to the uncontrolled condition (Fig. 8, bottom). In fact the
VGs, as previously discussed about Fig. 4, transfer uid toward
the wall producing a negative time-averaged normal velocity.
The term (Dn)Ty downstream of VGs is positive near the wall,
 =@y is negawhere the downward motion decrease and hence @ v
tive. That contributes to increase the total pressure dissipation
rate close to the wall. On the contrary, above y = 18 mm the term
(Dn)Ty returns negative (as observed for the uncontrolled condi =@y imposed by the channel
tion) due to the positive @ v
divergence.

1511

D. Lengani et al. / Experimental Thermal and Fluid Science 35 (2011) 15051513

Fig. 9. Overall deformation work distributions without (top) and with (bottom) boundary layer control.

The distributions of the overall dissipation work due to viscous

and turbulent terms for both uncontrolled and controlled conditions are reported in Fig. 9. The overall deformation work and
hence the induced dissipation rate appear, for the controlled case,
evidently smaller at the duct exit plane. Whereas, downstream of
VGs the overall dissipation rate appears slightly larger than for
the uncontrolled case close to the wall. This fact seems apparently
in contrast with the total pressure results which show a reduction
of losses in the near wall region (Fig. 6, y < 18 mm) just dowstream
of the VGs. This discrepancy cannot be explained by analysing the
ow only in the meridional symmetry plane, because the presence
of VGs induces a 3D ow.

3.2. Three-dimensional effects of VGs

Fig. 10. Cross-stream velocity vectors, streamwise velocity and total pressure loss
coefcient contour plots in the cross-stream plane at x = 350 mm.

160
VGs z=0mm
VGs z=25mm
VGs z=60mm

120

No VGs

y [mm]

The three-dimensional effects of the vortices generated by the

VGs on velocity and total pressure losses are shown in Fig. 10.
The comparison between the velocity proles at selected transversal positions and the prole for the uncontrolled ow is reported in
Fig. 11.
The VGs generate vortices that rotate counter-clockwise, as
indicated by the velocity vectors superimposed to the colour plot
of the streamwise velocity component on the left of Fig. 10. The
vortex centre, which may be identied at y = 18 mm and
z = 25 mm, is characterized by low velocity magnitude. However,
at the wall, the controlled boundary layer prole is fuller than
the baseline one, as depicted in Fig. 11 (z = 25 mm). In the upward
motion (on the right side with respect to the vortex centre) the
vortex concentrates and lifts low momentum uid from the wall
region. This vortex action leads to an almost separated boundary
layer in a small and conned region around z = 40 mm. Anyway,
at the right boundary of the vortex (z = 60 mm) the VGs effect is
positive, and the boundary layer has higher momentum compared
to the uncontrolled ow (Fig. 11). In the region of downwash, near
the test section symmetry axis, the boundary layer appears
strongly attached, and the largest difference with respect to the
uncontrolled case may be observed (Fig. 11, z = 0 mm).
The region in the cross-stream plane characterized by high total
pressure loss coefcient Cpt (Fig. 10 on the right) is the result of two
distinct phenomena associated with the cross-stream vortex action.
A loss core, correlated to the low velocities of the vortex, may be
identied in the proximity of the vortex centre. Furthermore, a local
loss increase, due to the upwash of the vortex, may be observed
close to the wall on the right with respect to the vortex centre.
Far downstream of the VGs trailing edge, in the cross-stream
plane at x = 600 mm, the ow distortions induced by the vortices
are almost completely disappeared and the ow is almost twodimensional, as it was shown in [19]. Consequently, due to the mix-

80

40

0
0

10

15

20

25

u [m/s]
Fig. 11. Streamwise velocity proles for the baseline and controlled cases at
x = 350 mm.

ing process, the total pressure distribution in this downstream

plane (not shown in the paper) presents an almost uniform pattern.
3.3. Total pressure losses
3.3.1. Without boundary layer control
In this case the global losses between inlet and outlet sections
may be evaluated on the meridional plane (x, y) where the

1512

D. Lengani et al. / Experimental Thermal and Fluid Science 35 (2011) 15051513

measurements of total pressure and velocity have been performed.

The presence of a boundary layer in the inlet section and reverse
ow in the outlet section makes the classical approach based on
the integration of the mass-weighted total pressure loss coefcient
not suitable to evaluate overall losses. With the 2D-ow assumption, since no work is done inside the test section and no massow
passes through the bottom wall, the total pressure loss in the control volume may be evaluated from the following balance of total
pressure uxes across the boundaries of the 2D measuring domain:

pt losses

Z
0

~
y

u0 ypt0 ydy 
Z

0.025

VGs

loss(x=350)

0.015

0.01

~
y

u~x ypt~x ydy

0.005

~x

v y~ xpty~ xdx

However, because of the turbulent ow separation, the trans is not negligible within the reverse
versal velocity component w
ow region [27]. The massow exiting from the meridional
plane in the transverse direction was computed via a massow balance in the same rectangular domain considered for the balance of
total pressure uxes. The massow exiting at x = 450 mm is less
then the 1% of the inlet massow, while at x = 600 mm this contribution is around the 8% of the inlet massow. Since the total pressure is low inside the separated region (Fig. 6), the global amount
of the total pressure ux exiting in z direction is relatively small
(less then 4%).
The loss of total pressure has been computed for the baseline
ow at different axial coordinates considering a rectangular do~ 150 mm and a variable position ~
main with a xed height y
x of
the exit section. The loss coefcient dloss is dened as follows:

dloss

Baseline

0.02

pt losses

1=2qU 30 H0

where the denominator is representative of the inlet inviscid ow

power.
The evolution of dloss along the axial coordinate is shown in
Fig. 12. An abrupt loss increase may be observed after
x = 300 mm where backow begins. In particular, from this position to the outlet of the test section the loss coefcient increases
of almost 6 times.
3.3.2. With boundary layer control
The presence of VGs induces an evident three-dimensional ow,
as previously mentioned in Section 3.2. Therefore, losses may be
evaluated only through a proper balance of the total pressure
uxes in a 3-D domain. This balance is written in Eq. (7) where
the double overbar indicates the integral average along the z direc-

VGs
Baseline

0.02

loss

0.015

0.01

0.005

200

400

600

x [mm]
Fig. 12. Evolution of the loss coefcient along the axial coordinate for the baseline
and controlled ows.

20

40

60

80

z [mm]
Fig. 13. Loss of total pressure with vortex generators at x = 350 mm.

tion. The rst two terms on the right hand side represent the total
pressure ux entering and leaving the control volume along the
streamwise direction, respectively. The third term represents the
total pressure ux through the top side, while the two latter terms
account for the net total pressure ux through the boundaries of
the measuring domain normal to the z direction. Thus, if the mean
velocity component w is zero on the lateral planes of the control
volume (z = 0 and z ~z), the two latter terms in Eq. 7 do not contribute to the balance. On the basis of previous results reported in
Canepa et al. [19], the lateral domain boundaries (z = 0 and z ~z)
have been chosen in order to satisfy this assumption.







Z y~
Z ~x



 pt dy 
 pt dy 
3D pt losses
u
u
v pt dx
0
0
0



~x
~
y
0

!
Z ~x Z y~
Z ~x Z y~


1

 t dx dy 
 t dx dy

wp
wp
~z


0
0
0
0
Z

~
y

~z

7
With these considerations the quasi 3D balance of the crossstream averaged total pressure uxes along x and y may give a representative estimation of the losses for the controlled case.
The coefcient d0loss , dened in Eq. 8, represents the contribution
to the overall losses for a given z transversal position. The crossstream averaged value of d0loss is equal to dloss and it is reported
for comparison in Fig. 12 (red2 symbols). The distribution of d0loss
in the cross-stream plane at x = 350 mm is shown in Fig. 13.

R y~
d0loss

0.025




R~
R~
 pt dy  0y u
 pt dy 0x v pt dx
u
~x

1=2qU 30 H0

~
y

In this plane, just downstream of VGs, a local maximum of

losses is located around the vortex centre. Here d0loss for the controlled case appears slightly larger than dloss for the uncontrolled
condition. The larger benecial effects induced by VGs may be observed near the symmetry axis, where the local total pressure loss
is extremely small. In fact high momentum uid has been transported from the outer region of the boundary layer in this position
by the cross-stream vortex, as it has been shown in Fig. 10. Also on
the right side of the vortex centre (z > 40 mm) losses are smaller
than the ones measured for the uncontrolled ow. The superposition of the upwash and downwash motions produced by adjacent
vortex generators, installed in a co-rotating pattern, locally reduces
losses with respect to the uncontrolled case.
2
For interpretation of colour in Figs. 213, the reader is referred to the web version
of this article.

D. Lengani et al. / Experimental Thermal and Fluid Science 35 (2011) 15051513

Moving downstream the mixing out process of the VGs vortices

induces a signicant loss increase. From x = 350 mm to x = 600 dloss
increases from 0.0048 to 0.0120 (Fig. 12). However, global losses at
x = 600 mm remain 50% lower as compared with the uncontrolled
case, since the separation is considerably reduced.

1513

For the baseline uncontrolled condition the total pressure losses

show a steep increase where the ow detaches. A strong reduction
of the aerodynamic losses has been found with the boundary layer
control applied. Losses for the controlled condition are in fact, at
the test section exit plane, reduced by 50% with respect to the
uncontrolled separated case.

4. Conclusions
References
A detailed experimental study of loss mechanism in a separating turbulent boundary layer controlled by low prole vortex generators has been carried out on a large-scale at plate with a
prescribed adverse pressure gradient. The boundary layer with
and without VGs has been investigated in the meridional and
cross-stream planes: velocity elds have been surveyed by means
of LDV, while the distribution of the total pressure has been measured by means of a Kiel probe.
In the baseline uncontrolled conguration the ow is affected
by a turbulent boundary layer separation. The momentum transfer
induced by VGs suppresses the separation and the total pressure
loss in the symmetry plane is highly reduced with respect to the
baseline case.
The dissipation mechanisms for both the separated uncontrolled condition and the controlled case have been in depth investigated through the analysis of the viscous and turbulent
contributions to the overall deformation work. Losses are mainly
produced by three terms: the deformation work of the Reynolds
shear stress acting in the streamwise direction ((Dt)Tx) and the
deformation works of the Reynolds normal stresses acting along
both the streamwise ((Dn)Tx) and the normal ((Dn)Ty) directions.
The largest contribution to the overall dissipation rate of the
mean ow mechanical energy may be identied within the separated shear layer and it is due to the term (Dn)Tx. The energy subtracted from the main ow by (Dn)Tx is converted in turbulent
kinetic energy and justies the large increase of both Cpt and
streamwise velocity uctuations observed in this region. It is
important to note that the classical assumption of negligible normal component of the deformation work cannot be applied for a
separating turbulent boundary layer. On the contrary, in this condition it is the contribution to the overall deformation work due to
the Reynolds shear stress to be negligible: it vanishes downstream
of the detachment position since the strain rate of the mean ow is
almost disappeared.
The only non-negligible contribution to the total pressure dissipation rate along the y direction is given by the deformation work
operated by the Reynolds normal stress ((Dn)Ty). Due to the channel
divergence it has been found negative for the uncontrolled case.
Hence the term (Dn)Ty tends to increase the mean ow mechanical
energy subtracting kinetic energy to the turbulence.
For the controlled ow the overall deformation work and hence
the induced dissipation rate appear smaller than for the baseline
case. In particular, the dissipation term (Dn)Tx is substantially reduced by VGs.
The presence of VGs induces cross-stream vortices able to suppress the separation and consequently the large ow oscillations,
responsible for the loss increase, that characterize the separated
shear layer. These cross-stream vortices induce a non uniform total
pressure distribution in the cross-stream plane: a total pressure
loss core is located at the vortex centre and another one close to
the plate surface where the vortex induces upward motion.
Due to the intrinsically three-dimensional structure of the ow
eld in the controlled case, the overall total pressure losses have
been computed through a balance of the total pressure uxes in
a 3D domain. The boundaries of the domain have been properly
chosen in order to evaluate a total pressure balance from the 2D
velocity eld measured in the cross-stream planes.

[1] L. Prandtl, On the motion of a uid with very small viscosity, in: Verh. Int.
Math. Kongr., third ed., Heidelberg, 1904, pp. 484491.
[2] W.K. Lord, D.G. MacMartin, T.G. Tillman, Flow control opportunities in gas
turbine engines, in: Fluids 2000, Denver, CO, June 1922, 2000, AIAA paper n
2000-2234.
[3] M. Gad-el-Hak, Interactive control of turbulent boundary layers: a futuristic
overview, AIAA Journal 32 (9) (1994) 17531765.
[4] M. Gad-el-Hak, D.M. Bushnell, Separation control: review, Journal of Fluids
Engineering 113 (1991) 529.
[5] D.M. Rao, T.T. Kariya, Boundary layer submerged vortex generators for
separation control-an exploratory study, in: National Fluid Dynamics
Congress, 1st, Cincinnati, OH, July 2528, 1988, AIAA Paper n 88-3546-CP.
[6] J.C. Lin, F.G. Howard, G.V. Selby, Turbulent ow separation control through
passive techniques, in: Shear Flow Conference, 2nd, Tempe, AZ, March 1316,
1989, AIAA Paper n 89-0976.
[7] J.C. Lin, F.G. Howard, D.M. Bushnell, G.V. Selby, Investigation of several passive
and active methods for turbulent ow separation control, AIAA Paper n 901598, 1990.
[8] B.A. Reichert, B.J. Wendt, Improving diffusing s-duct performance by secondary
ow control, NASA TM-106492, 1994.
[9] B.A. Reichert, B.J. Wendt, Improving curved subsonic diffuser with vortex
generators, AIAA Journal 34 (1) (1996) 6572.
[10] R.K. Sullerey, S. Mishra, A.M. Pradeep, Application of boundary layer fences and
vortex generators in improving performance of s-duct diffusers, ASME Journal
of Fluids Engineering 124 (2002) 136142.
[11] K.A. Waithe, Source term model for vortex generator vanes in a NavierStokes
computer code, in: The 42nd AIAA Aerospace Sciences Meeting and Exhibit,
2004, AIAA paper n 2004-1236.
[12] J.C. Dudek, An empirical model for vane type vortex generators in a Navier
Stokes code, NASA TM-2005-213429, 2005.
[13] O. Trnblom, A.V. Johansson, A reynolds stress closure description of
separation control with vortex generators in a plane asymmetric diffuser,
Physics of Fluids 19 (11) (2007) 115. 115108.
[14] F. von Stillfried, Computational studies of passive vortex generators for ow
control, Technical Reports From Royal Institute of Technology Stockholm,
2009.
[15] F.T. Smith, Theoretical prediction and design for vortex generators in turbulent
boundary layers, Journal of Fluid Mechanics 270 (1994) 91131.
[16] C.M. Velte, M.O.L. Hansen, V.L. Okulov, Helical structure of longitudinal
vortices embedded in turbulent wall-bounded ow, Journal of Fluid Mechanics
619.
[17] B. J. Wendt, Initial circulation and peak vorticity behavior of vortices shed from
airfoil vortex generators, NASA CR-2001-211144, 2001.
[18] J.C. Lin, Review of research on low-prole vortex generators to control
boundary-layer separation, Progress in Aerospace Science 38 (2002)
389420.
[19] E. Canepa, D. Lengani, F. Satta, E. Spano, M. Ubaldi, P. Zunino, Boundary layer
separation on a at plate with adverse pressure gradients using vortex
generators, in: ASME Turbo Expo: Power for Land, Sea and Air, May 811,
Barcelona, Spain, 2006, ASME Paper n GT-2006-90809.
[20] F. Satta, D. Simoni, M. Ubaldi, P. Zunino, F. Bertini, E. Spano, Velocity and
turbulence measurements in a separating boundary layer with and without
passive ow control, Proceedings of the IMechE, Part A: Journal of Power and
Energy 221 (6) (2007) 815823.
[21] R.L. Simpson, Aspects of turbulent boundary-layer separation, Progress in
Aerospace Science 32 (5) (1996) 457521.
[22] D. Modarress, H. Tan, A. Nakayama, Evaluation of signal processing techniques
in laser anemometry, in: Fourth International Symposium on Application of
Laser Anemometry to Fluid Dynamics, Lisbon, Portugal, 1988.
[23] W.K. George, Processing of random signals, in: Proceedings of Dynamic Flow
Conference, 1978, pp. 757800.
[24] F. Satta, D. Simoni, M. Ubaldi, P. Zunino, Experimental difculties in measuring
separating boundary layers with the LDV technique, in: The XVIII Symposium
on Measuring Techniques in Turbomachinery Transonic and Supersonic Flow
in Cascades and Turbomachines, Thessaloniki, Greece, 2006.
[25] H. Tennekes, J. Lumley, A First Course in Turbulence, The MIT press, 1972.
[26] J. Moore, D.M. Shaffer, J.G. Moore, Reynolds stresses and dissipation
mechanisms downstream of a turbine cascade, Journal of Turbomachinery
109 (2) (1987) 258267.
[27] R.L. Simpson, Y.T. Chew, B.G. Shivaprasad, The structure of a separating
turbulent boundary layer. Part 1: Mean ow and reynolds stresses, Journal of
Fluid Mechanics 113 (1981) 2351.