Experimental study of the pitching blade effect

with a PIV application

Bilel BEN AMIRA, Zied DRISS, Mohamed Saleh ABID

Laboratory of Electromechanical Systems (LASEM), National School of Engineers of

Sfax (ENIS), University of Sfax, B.P. 1173, Rte Soukra 3038 Sfax, Tunisia.
benamira.bilel@yahoo.fr, Zied.Driss@enis.rnu.tn,

Abstract. The aim of this paper is to study experimentally by particle image ve-
locimetry system (PIV) the effect of the pumping direction introduced with a 60
pitched blade turbine. In fact, an Up- and a down-pitched blades turbine were
used. The particle image velocimetry technique is equipped with a double pulsed
Nd:YAG laser, a CCD camera resolution and a mini-synchronizer. Therefore, sev-
eral results were carried out to investigate the hydrodynamic structure and the en-
ergetic parameters in a stirred vessel.

Keywords: pumping direction, 60 pitched blade turbine, PIV, hydrodynamic

structure

1 Introduction
Since the efficiency of an agitated vessel is dependent on some geometric pa-
rameters, type and dimensions of impeller, internals and physical properties of the
medium processed, it looks necessary to analyze the effect of one or more parame-
ter on the vessel performance. Several numerical process and experimental appa-
ratus have been used to investigate the hydrodynamic structure of the flow gener-
ated by different impellers geometries. For example Mohan et al. (1992) used
numerous empirical heat transfers correlations to have a fundamental understand-
ing of the flow fields in stirred vessels. La Fontaine and Shepherd (1996) investi-
gated the time-averaged velocity vectors and the standard deviation of the fluctua-
tions flow in a stirred vessel using a scanning particle image velocimetry (PIV)
system. Guillard et al. (1998) used Laser Induced Fluorescence (LIF) and Planar
Laser Induced Fluorescence (PLIF) techniques to determine mixing and entrain-
ment effects in many flow configurations. Guillard et al. (2000) used Laser In-
duced Fluorescence (PLIF) to investigate the turbulent mixing process in a model
2 Names of the authors

stirred tank equipped with two Rushton turbines of standard geometry. Paglianti
and Pintus (2001) developed an electrical impudence measurement to measure
hod-up and mixing time of a stirred tank equipped with a six blade Rushton tur-
bine. Aubin et al. (2004) used laser doppler velocimetry (LDV) data to validate re-
sults of the single phase turbulent flow in a stirred tank generated by a down- and
an up-pumping pitched blade turbine. Gabriele et al. (2009) used angle resolved
particle image velocimetry to compare the turbulent flow characteristics generated
with a 45 up- and down-pumping pitched blade. Driss et al. (2010) developed
specific computational fluid dynamics (CFD) code to analyze numerically the tur-
bulent flow field in a stirred tank equipped with the pitched blade turbines. Driss
et al. (2011) used specific computational fluid dynamics code to compare the lam-
inar hydrodynamic flow generated by mono and twin screw profile with simple
and modified profiles. Ben Amira et al. (2011) used particle image velocimetry to
analyze the Reynolds number effect in a cylindrical stirred tank generated with a
six flat blade turbine. Feng et al. (2012) used an explicit algebraic stress model
(EASM) to simulate anisotropic turbulent flows in baffled stirred tanks equipped
with a standard Rushton turbine.
This work is an attempt to enhance the studies of the pitched blades turbine to
improve the mixing in a stirred vessel instead the 45 turbine. PBTU 60 and the
PBTD 60 are proposed according to these previous studies and as a continuation
of a previous works.

2. Experimental setup

The schematic of the experimental setup is shown in figure 1. Obviously, the

experiments were carried out in a cylindrical stirred vessel. The diameter of the
stirred tank is equal to T = 0.3 m. Four baffles with width of 0.03 m are equipped
equally-spaced at the wall. The liquid height for all experiments is set at H=T.
The off-bottom clearance (from the middle of the impeller to the vessel bottom) is
T/3 in experiments. Two different types of stirrers are used, namely, 60 pitched
blade turbine down flow (PBTD) and 60 pitched blade turbine up flow (PBTU).
The experimental data were taken from a particle image velocimetry (PIV) system
which equipped with a 532 nm green light Nd:YAG laser, a 1600 x 1200 pixel of
CCD camera resolution and a mini-synchronizer.
Title of your paper 3

Fig. 1 Schematic of the experimental setup

In this study, we are interested in the analysis of the hydrodynamic structure of

turbulent flow. Up- and down-pumping pitched six-blade turbines with a blade in-
clination equal to 60 were used (figure 2). In fact, we present results associated to
the vertical measurement plan which located midway between two blades. The di-
ameter of the impeller is equal to 102 mm and the tangential velocity at the tip of
the blade is equal to 0.8 m.s-1.

(a) PBTU 60 (b) PBTD 60

Fig. 2 Geometrical arrangement

3. Experimental results

3.1 Velocity field

4 Names of the authors

Figure 3 shows the velocity field for both the up- and down- pumping pitched
six-blade turbines. Two different zones are presented in the vessel. In fact, one of
these zones is characterized by the vortex attached to the blade and the passage of
the vortex from the previous blade, but the other zone is characterized with the ab-
sence of vortices. In the up-pumping configuration, the volume upper the impeller
is more extended than in the down-pumping configuration. However, in the down-
pumping configuration, the volume lower the impeller is more extended than in
the up-pumping configuration. Two main recirculation loops are presented in the
up-pumping configuration. Thus, one is localized in the third upper zone of the
vessel and attached to the free surface. Otherwise, the second one is attached the
wall of the vessel, which it is larger than the first recirculation loop. However, one
main recirculation loop is presented in the down-pumping configuration which lo-
calized in the bottom of the vessel and near the vessel wall. In addition, there are
no vortices presented in the volume upper the impeller which characterizes by the
lowest velocity.
Close to the impeller, the stream discharged from the blade as the fluid acceler-
ates from a downward to an upward direction in the upper reaches of the vessel. In
fact, the sloping stream turns vertically close the impeller, then, included velocity
direction appears between the main recirculation loops. Otherwise, the stream dis-
charged from the blade causing a local velocity in the down-pumping configura-
tion. Whereas, the velocity increases as the fluid accelerates from an upward to a
downward direction in the inferior reaches of the vessel. The sloping stream con-
fined by the bottom wall and the side wall.
V=49.9 mm/s V=51.2 mm/s

(a) PBTU 60 (b) PBTD 60

Fig. 3 Velocity field

3.2 Mean instantaneous velocity

Title of your paper 5

Figure 4 shows the variation of mean instantaneous velocity with time for both
the up- and down- pumping pitched six-blade turbines. Subsequently, for both the
two applications, the x-axis presents the time axis in which the unit of time re-
flects to one shaft rotation. Thereby, it is clearly evident that the flow oscillates in
a sinusoidal form. In addition, it has been seen that for the up pumping configura-
tion, the velocity oscillation is the same along the experience. However, for the
down pumping configuration, the velocity oscillation decreases and increase along
the experience.
Velocity (mm.s-

Velocity (mm.s-

(a) PBTU 60 (b) PBTD 60

Fig. 4 Variation of mean instantaneous velocity with time

3.3 Vortices

Figure 5 shows the distribution of the vortices for both up- and down- pumping
pitched six-blade turbines. The maximum values are observed in two locations
corresponding to the location of highest value and the lowest values for both up-
and down- pumping configuration. Consequently, the highest and the lowest local
vortex values were found at the impeller level and become medium moving both
upward and downward from this level. For the up-pumping configuration, the
trailing vortices cross the plane to the back side at the impeller tip and to the front
side above the blade. However, for the down-pumping configuration, the trailing
vortices cross the plane to the back side above the blade and to the front side at the
impeller tip. Indeed, it has been observed that the high vortex value of the down-
pumping configuration is greater than the high vortex value of the up-pumping
configuration. However, the values of the lowest vortex of the down-pumping
configuration and the up-pumping configuration are almost the same.
6 Names of the authors

30.9 s-1 55 s-1

1.8 s-1 12.7 s-1

-27.3 s-1 -29.7 s-1

(a) PBTU 60 (b) PBTD 60

Fig. 5 Vortices

3.4 Turbulent kinetic energy

Figure 6 shows the distribution of the turbulent kinetic energy for both the up-
and down- pumping pitched six-blade turbines. The turbulent kinetic energy in the
up-pumping configuration has two maximum regions. The faraway region is local-
ized in the same direction of the velocity acceleration between the two main recir-
culation loops. Moreover, it has been noted that the impeller defuse the energy to
the vessel at the same direction of the discharge flow. Otherwise, it is clear that
the down-pumping turbine transfers the kinetic energy to the region near the bot-
tom of the tank. Therefore, in this case it has been observed that the turbulent ki-
netic energy region which generated with one main circulation loop is different
from that in the double main circulation loops which transferred to the upper-right
region of the turbine.

0.16 0.35

0.14
0.3

0.12
0.25

0.1
0.2
0.08

0.15
0.06

0.1
0.04

0.02 0.05

(a) PBTU 60 (b) PBTD 60

Fig. 6 Distribution of the turbulent kinetic energy

3.5 Dissipation rate of the turbulent kinetic energy

Title of your paper 7

Figure 7 shows the dissipation rate of the turbulent kinetic energy for both the
up- and down- pumping pitched six-blade turbines. The dissipation turbulent ki-
netic energy rate in the up-pumping configuration is localized close to the impel-
ler, which has the same direction of the velocity acceleration between the two
main recirculation loops. However, the down-pumping turbine transfers the ener-
gy to the region near the bottom of the tank.
0.045 0.11

0.04 0.1

0.09
0.035
0.08
0.03
0.07
0.025
0.06

0.02 0.05

0.015 0.04

0.03
0.01
0.02
0.005
0.01
0

(a) PBTU 60 (b) PBTD 60

Fig. 7 Distribution of the dissipation rate of the turbulent kinetic energy

3.6 Profiles

In this section we are interested to the radial revolution velocity profile, the ra-
dial velocity component profile, the radial rms velocity profile, the turbulent kinet-
ic energy profile, the dissipation rate of the turbulent kinetic energy profile and the
turbulent viscosity profile in several positions in the bulk tank for both the up- and
down- pumping pitched six-blade turbines. Figure 8 shows the radial revolution of
the profiles at the axial position equal to z=1/3 H from the free surface. Therefore,
for all configurations, it has been seen that the hydrodynamic structure of the ves-
sel has the same distribution along the vessel. The up- and the down-pumping con-
figurations have the similar effect at this region. However, it has been seen that the
up-pumping turbine is greeter that the down-pumping turbine at the upper region
of the vessel. Then, the highest value of the up-pumping configuration were pre-
sented at the half of the vessel near the vessel wall which described by the main
recirculation loop. A quick decrease of the velocity near the vessel wall has been
noted. However, the first half region besides the shaft is described by the absence
of the recirculation loop for both configurations. Consequently, at this region, the
up-and the down-pumping configuration were looked confounded.
8 Names of the authors

(a) Velocity (b) rms (c) vorticity

(d) Turbulent kinetic energy (e) Dissipation rate of the (f) Turbulent viscosity
turbulent kinetic energy

Fig. 8 High third position (z=2/3 H)

Figure 9 shows the radial revolution of the profiles at the axial position equal to
z=1/2 H. It has been seen that the up-pumping turbine is greeter that the down-
pumping turbine at the medium of the vessel. Moreover, the down-pumping con-
figuration has sinusoidal reparation along the vessel. Then, the highest value of the
up-pumping configuration were presented at the medium region of the vessel
which described by the discharge flow region. A progressive decrease of the ve-
locity has been noted.

(a) Velocity (b) rms (c) vorticity

Title of your paper 9

(d) Turbulent kinetic ener- (e) Dissipation rate of the (f) Turbulent viscosity
gy turbulent kinetic energy

Fig. 9 Half position (z=1/2 H)

Figure 10. shows the radial revolution of the profiles at the axial position equal
to z=1/3 H from the vessel bottom. Therefore, for all configurations, it has been
seen that the hydrodynamic structure of the vessel has the same distribution along
the vessel unless both velocity and vortex change direction near the vessel wall
which described by the second recirculation loop.

(a) Velocity (b) rms (c) vorticity

(d) Turbulent kinetic energy (e) Dissipation rate of the (f) Turbulent viscosity
turbulent kinetic energy

Fig. 10 Lowest third position (z=1/3 H)

6. Conclusion

In this work, we have studied the effect of the pumping direction on the hydro-
dynamic structure of a stirred vessel. In fact, it has been noted that the vessel
which equipped by the PBTU configuration is more stirred than the PBTD config-
uration. In addition, the velocity change direction only for the up-pumping config-
uration unless at the upper region of the vessel. Moreover, the trailing vortices and
the highest values were found in the same direction of the flow discharge.
In the future, we propose to develop a numerical simulation to study a new ge-
ometrical parameters.
10 Names of the authors

References

Aubin J, Fletcher DF, Xuereb C, Modeling turbulent flow in stirred tanks with CFD: the in-
fluence of the modeling approach, turbulence model and numerical scheme, Experi-
mental Thermal and Fluid Science 28 (2004) 431445.
Ben Amira B, DRISS Z, KAFFEL A, KARRAY S, ABID MS, Experimental characteriza-
tion of a stirred vessel equipped with eight flat blades turbine using a PIV system: Ef-
fect of the Reynolds number, Rcents Progrs en Gnie des Procds, Numro 101
2011 ISSN 1775-335X - ISBN 2-910239-75-6, Ed. SFGP, Paris, France.
Driss Z, Bouzgarrou G, Kchaou H, Abid MS, Simulation numrique de lcoulement lami-
naire induit dans des cuves agites gnres par les mobiles de proximit de type mono
et double vis profils simple et modifi, Mcanique & Industries 12 (2011)109121.
Driss Z, Bouzgarrou G, Chtourou W, Kchaou H, Abid MS, Computational studies of the
pitched blade turbines design effect on the stirred tank flow characteristics, European
Journal of Mechanics B/Fluids 29 (2010) 236245.
Feng X, Cheng J, Li X, Yang C, Mao ZS, Numerical simulation of turbulent flow in a baf-
fled stirred tank with an explicit algebraic stress model, Chemical Engineering Science
69 (2012) 3044.
Gabriele A, Nienow AW, Simmons MJH, Use of angle resolved PIV to estimate local spe-
cific energy dissipation rates for up-and down-pumping pitched blade agitators in a
stirred tank, Chemical Engineering Science 64 (2009)126143.
Guillard F, Fritzon R,, Revstedt J, TraKgasrdh C, AldeHn M, Fuchs L, Mixing in a con-
fined turbulent impinging jet using planar laser induced fluorescence. Exp Fluids
25(1998) 143150.
Guillard F, TraKgasrdh C, Fuchs L, A study of turbulent mixing in a turbine-agitated tank
using a fluorescence Technique, Experiments in Fluids 28 (2000) 225235.
La Fontaine RF, Shepherd IC, Particle Image Velocimetry Applied to a Stirred Vessel, Ex-
perimental Thermal and Fluid Science; 12 (1996) 256264.
Mohan P, Emery AN, AI-Hassan T, Review Heat Transfer to Newtonian Fluids in Mechan-
ically Agitated Vessels, Experimental Thermal and Fluid Science 5 (1992) 861883.
Paglianti A, Pintus S An impedance probe for the measurements of liquid hold-up and mix-
ing time in two/three-phase stirred tank reactors, Experiments in Fluids 31 (2001) 417
427.