ENGINEERING RESEARCH JOURNAL (ERJ)

Volume (52),Issue (3)

July 2023, pp:164-171
https://erjsh.journals.ekb.eg

Effect of Suction Slot Location and Width on the

Performance of Centrifugal Fan
S. A. Beskales 1,*, Ibrahim Shahin 1, Samir S. Ayad 1, Osama E. Abdullatif 1 and Tarek A. Mekhail 2
1
Department of Mechanical Engineering, Faculty of Engineering at Shoubra, Benha University, Shoubra, Cairo, Egypt.
2
Department of Mechanical Engineering, Faculty of Energy Engineering- Aswan University, Aswan, Egypt.
*Corresponding author

E-mail address: samer.aagaiby@feng.bu.edu.eg, ibrahim.shahin@feng.bu.edu.eg, samir_ayad@mail.com,

osama.abdellatif@feng.bu.edu.eg, mina_tarek@hotmail.com

Abstract: Blade slots are known to reduce the noise and improve flow behavior and steadiness in centrifugal fans. The present work
considers an unsteady Computational Fluid Dynamic (CFD) study to investigate the effectiveness of slot location and width on the
performance and flow field inside the centrifugal fan blade passages. The computer program, Fluent 19, is used and the 3D fan model
is validated by comparing its results with those of earlier researchers. The aim of the present study is to find the best slot location and
appropriate slot width at that location. The flow pattern is then analyzed and simulated with no slot and with a slot at six different
radial locations; namely S1, S2, S3, S4, S5, and S6 with fixed slot width W of 2.5 mm. The computationally generated characteristics
of unsteady flow demonstrated that although cut slots at all locations on the fan blade from the hub to the shroud has a negative
influence on the centrifugal fan efficiency as well as the static pressure rise compared with the fan without slot, the slots added at S3
i.e. near the midpoint of the blade span gives the highest fan efficiency and largest increase in static pressure. Slots at S3 suppress
secondary flow at the blade passages and push it to the blade tip. Additionally, streamlines and velocity distributions in impeller
passages and over the blade surfaces (pressure and suction sides) validated the slots' benefits. Computations width different slot
width; namely 1.5 mm, 3.5 mm, and 5 mm are carried out to find the best slot location and width at the best slot location S3. At the
optimal efficiency point, the calculated performance of the employed fan with slot location S3 and 2.5 mm width, showed a 2.7 %
increase of efficiency and 1.5% increase in static pressure rise as compared to fans with slot at S1 with same width of 2.5 mm.
However, it showed a positive impact on the boundary layer buildup and flow separation over the blade suction side for the impeller
studied .

Keywords: Suction Slots, Secondary flow, Centrifugal Fan Performance, CFD, Simulation.

NOMENCLATURE
Symbols Description SI units Symbols Description
u Fluctuated velocity m/s Abbreviations
Time averaged velocity m/s BEP Best Efficiency Point
Ui
Cm Moment Coefficient - CFD Computational Fluid
Dynamics
Ds Slot pitch Diameter mm MRF Moving Reference Frame
2 2
k Turbulent kinetic energy m /s SST Shear Stress Transport
Greek SIMPLE Semi-Implicit Method for
Pressure Linked Equations

3
Instantaneous value of density kg/m
2 3
ɛ Turbulent energy dissipation rate m /s
-1
ω Rotational Speed S
ᴓ Flow coefficient = Q/ωD³ -
η Static Efficiency -
ᴪ pressure coefficient = ΔP/ρω²D² -

the blade tip and the fluid particles in the boundary layer
1. Introduction experience a negative velocity and the flow may be
In centrifugal fans, the recirculation zones occurring reversed. At the point of the separation one streamline
between the blade passages contribute to some losses. The intersects the wall at a definite angle as shown in Fig. 1, and
moving fluid with low velocity in the boundary layer the point of separation itself is determined by the condition
doesnot remain in the thin layer which sticks to the that the velocity gradient normal to the blade surface
impeller’s blade surfaces. The boundary layer will increase vanishes there.
its thickness substantially within the downstream path near

164
 Engineering Research Journal (ERJ) S. A. Beskales et al Vol.52, No1 January 2023, pp:164-171

2. COMPUTAIONAL TECHNIQUE
2.1 Governing Equations
Researchers utilize CFD to collect information on fluid
flow issues and flow parameters such as pressure, velocity
in situations when experimental procedures cannot be
conducted. This can be achieved by solving the
fundamental equations such as conservation of mass and
Fig 1. Flow in boundary layer near separation [1]
momentum for incompressible turbulent flow [1]
 
This study's objective is to enhance the flow on the    U i   0 (1)
suction side of the blade’s impeller by decreasing the size t x 
i

of the recirculation zones. This can be done by cutting slot
in the blades of the impeller to control the flow passively in
order to delay flow separations. slot will move the flow (2)
from the pressure side of the blade to the suction side, they
are installed near the point of separation to control the
development of the boundary layer over the blade surface. 2.2 Shear Stress Transport (SST) K- ω Model
Many researchers have utilized CFD to investigate In the present study, the Shear Stress Transport (SST) k- ω
centrifugal fan performance. [2] studied numerically the model is implemented. The k- ω model consists of two
effect of the operating condition, such as the impeller speed, equations. The Wilcox k- ω model [12] is more stable than
on the centrifugal fan performance and flow characteristics. the k- ε model in the viscous sublayer close to the wall and
[3] analyzed both transient simulations and steady moving is one of the most well-known forms of k- ω. However, its
reference frame (MRF) simulations. [4] studied the effect of results are particularly sensitive to the free stream value of
blade angles on the velocity and pressure distribution inside ω in the free shear layer and adverse pressure gradient
the impeller passages using the CFD techniques. boundary layer flows. Consequently, the k- ε model is not
Furthermore, detailed information on the flow field through the most appropriate model for wake area applications. In
the fan may be collected, providing guidance for fan the wake areas, however, the k- ε model performs better,
impeller modification by cutting slots in the impeller according to Menter [13], the combination of the two
blades. [5] investigated the effect of a cut slot in the blower models is the optimum way to model the flow
blades, the results showed The magnitude of pressure characteristics along the wall and in the boundary layer's
variations decreases when slots are added to the impeller wake. Accordingly, the SST k-ω turbulence is believed to
blades and the fluid poured through the slots into the be the best acceptable model for predicting the flow
impeller passages re-energizes the separated flow. [6] characteristics. The first equation in the model defines the
Studied the effect of constant-diameter slots in a turbulent kinetic energy (k), while the second describes the
perpendicular direction to the blade surface on radial fan turbulent dissipation rate (ω).
impeller. The results revealed that the slots enhanced the
internal efficiency and overall pressure of the fan impeller
without volute casing. [7] analyzed unsteady numerical (3)
simulations the slot cuts near the impeller shroud and the
results showed that both the performance and flow field are
well improved by applying the blade surface slots with 3.6
% for the total pressure and of 2.6 % for the efficiency at (4)
the design mass flow rate. Single and twin slot
configuration were investigated by [8] at different locations 3. CFD PROGRAM VALIDATION
on the chord length from the blade root. The results showed 3.1 Geometric Model
that the optimum single slot size is 5 mm with 5.5 %
increase in the static pressure while the twin slot located at The aim of this section is to verify the reliability of the CFD
50% and 75% from the blade root with 3 mm slot size model such that it can be used with confidence for
enhance the static pressure by 7% compared with the base numerical simulation and decision making in the design and
blade model. [9] studied the effect of the slot on an airfoil, the modification of centrifugal impeller blades with
results confirmed the slot benefits in reducing the boundary different slot locations and widths. The predictions of the
layer separation and stall control over different incident utilized computer software, FLUENT 19, are validated and
angles. [10] used a passive control using an opening in the compared to the results produced by Schönwald et al. [14]
airfoil thougth which the flow passed and converted the using HP workstation (Intel Xeon E5-2680 V3 -2.5 GHz
spearation region into small vorticies. [11] used a slotted and 128 GRAM). The modeled centrifugal fan in the
blade on a centrifugal pump, which improved hydraulic present validation is with ten backward impeller blades. The
performance. whole fan’s design specifications and settings are provided
in Table 1 and Fig. 2.

165
 Engineering Research Journal (ERJ) S. A. Beskales et al Vol.52, No1 January 2023, pp:164-171

TABLE 1 The design parameters for the centrifugal fan.[14] in greater numerical model stability [15]. Figure 4 shows
Item Dimension the mesh used for inlet pipe (A) , impeller (B) and volute
Angle of cover plate 11 ° casing(C) using polyhedral elements.The leakage flow
Blade inlet angle 31 ° through the clearance between the impeller and inlet pipe
Blade outlet angle 41 ° (fig 4. D) has a great impact on the performance of the
Curvature of the cover plate 45 mm centrifugal fan. consequently, in the present validation, the
Curvature of the inlet nozzle 45 mm
Clearance 3 mm
gap dimensions mentioned in Table.1 is considered in
Clearance width 13 mm present CFD simulations.
Casing width 315 mm
Curvature of the blade 383 mm
Blade number 10
Diameter of the inlet duct 400 mm
Diameter of impeller outlet 722 mm
Diameter of impeller inlet 318 mm
Blade inlet width 33 mm
Blade outlet width 93 mm

Fig 4. Polyhedral mesh at mid-section by FLUENT mesher program

for: [A] Inlet pipe, [B] Impeller, [C] Volute and [D] A close view to the
Impeller Volute

3.3 BOUNDARY CONDITIONS AND SOLUTION

CONTROLS
The centrifugal fan's boundary condition consists of a
velocity inlet boundary condition applied at the pipe inlet
with a hydraulic diameter of 0.4 meters and a turbulence
intensity of 5% in all tested fan performance points. The
outflow boundary condition is used at the volute outlet. The
hub, shroud, and fan blades are described as moving walls
that revolve at zero velocity relative to the neighboring cell
Fig 2. Centrifugal fan dimensions. [14] zone. The inlet pipe and volute casing are similarly defined
as moving walls, but with zero absolute velocity. The
The complete fan model (inlet duct, impeller, volute and impeller fluid domain is defined as frozen rotor (MRF) at
outlet duct) is constructed in SPACE-CLAIM program as 1500 RPM at the first numerical simulation with a fixed
shown in fig. 3. In order to capture the flow filed in the position relative to the volute casing, then the sliding mesh
fully developed region, a suitable length for the (inlet and is applied to the impeller fluid domain for the unsteady
outlet) duct must be given in order to properly achieve the simulation afterwards.
uniform velocity profile. In the present validation the length The solution is obtained by using SIMPLE algorithm
of the inlet and outlet duct are five times the duct diameter. between the pressure and velocity, three dimensions in
The rear distance in fig.2, between the hub and volute wall, space and steady in time (frozen rotor) for the first 2000
is assumed to be 31.5 mm. iterations, after that the mesh motion is applied to the
unsteady calculations for one complete revolution and the
flow statistics are initialized at the end. The iterations are
then carried out again by rotating the impeller for a second
revolution. During the second round, the data sampling for
each time step is used to get a picture of how the flow
changes over time.
3.4 CONVERGENCE HISTORY
The solutions for the centrifugal fan are obtained with
Fig 3. Complete 3D model created in Space-claim program, present
validation.
this criterion. During the solutions, iteration history is also
checked for the stability and the trend of the residual
3.2 Mesh Generation curves. Figure 5 shows a typical convergence history graphs
The present computational solution utilizes a polyhedral for the steady and unsteady solutions obtained for the
mesh, where each individual cell has many neighbors and centrifugal fan. In Fig 5.A It is seen that for almost the first
gradients can be approximated much more accurately than 2000 iterations the residuals for the variables are decreased
with tetrahedral. Additionally, polyhedral structures are less to at least three orders of magnitude for steady solution.
sensitive to stretching than tetrahedral structures, resulting From 2000 to 5500 iterations the unsteady convergence
shows a stable trend till the end. In Fig 5. (B, C and D) the

166
 Engineering Research Journal (ERJ) S. A. Beskales et al Vol.52, No1 January 2023, pp:164-171

convergence histories of other solution variables like field across the critical areas of the centrifugal fan, it is
pressure outlet, pressure inlet and coefficient of moment necessary to construct the smallest elements around the
(Cm) are also monitored to decide whether a reasonable blades, such as the leading and trailing edges, etc. In this
solution is reached or not. unsteady graphs are divided into section a mesh convergence test is carried out to assure that
three convergence stages; each stage represents a complete the CFD validation results is in a good agreement with the
revolution of the impeller. Stage 1 shows the transition experimental results by Schönwald et al. [14]. In the present
from the steady to the unsteady calculations represented in work, we used six different points starting from the coarse
dramatically change in the output value followed by some mesh (point 1) to the fine mesh (point 6). Figure 6 shows
fluctuation to flow time at 0.04 seconds. Prior to the second the effect of the number of cells on the Static pressure rise
stage, the flow statistic is initialized and calculations are between the inlet and outlet of the fan (fig.6 A) and
done for the second impeller revolution to time 0.08 impeller torque (fig.6 B) for one of the cases of the
seconds. The final stage from 0.08 to 0.12 is carried out to validation study at flow coefficient of 0.027. The figures
insure that the output variables are unchanged. show that the results are almost independent of mesh
elements number for all values of number of elements equal
3.5 MESH SENSITIVITY STUDY or above 3.4 million.
Mesh size is one of the most influential elements in the
outcomes of numerical simulations. To explore the flow

Fig 4. Convergence history graphs at [ N=1500 RPM and Φ =0.027]: (A) Residuals graph for mass flow rate, velocity components, turbulence kinetic
energy (K) and dissipation rate (ω), (B) outlet pressure, (C) Coefficient of moment CM and (D) Inlet pressure

Fig 5. Eeffect of mesh elements number on the static pressure rise coefficient (A) and impeller torque (B) at flow coefficient 0.027

167
 Engineering Research Journal (ERJ) S. A. Beskales et al Vol.52, No1 January 2023, pp:164-171

By comparing the efficiency curves in Fig.7 (B), the

4. CFD VALIDATIONS RESULTS steady and unsteady curves have the same trend along the
The iterations for the performance points are made by total flow coefficient values as the experimental results by
assigning different inlet velocity values at the inlet pipe Schönwald et al [14]. It is clear from the graph that both
boundary of the domain, resulting in corresponding flow the steady and unsteady efficiency curves are still below
rates for the pressure differentials across the fan. The the experimental curve by Schönwald et al. However, the
outlet boundary is outflow for all performance points. At unsteady efficiency curve shows good agreement with the
the end the overall solution the performance curves for the experimental curve by Schönwald et al. among the steady
centrifugal fan are given. Comparisons of the pressure calculations after flow coefficient 0.0321, with a
contours and velocity distributions are presented and difference in efficiency with the experimental of less than
discussed. 3.5% for Ø= 0.032 to the end of the curve, which is
accepted. At low flow coefficients, the steady calculations
4.1 Performance Curves Comparisons show good agreement with the unsteady calculations from
The aerodynamic performance of the designed fan by flow coefficient 0.0067 to the maximum efficiency point
Schönwald et al.[14] is simulated for the design rotational at flow coefficient 0. 0321.Finally, the comparisons serve
speed 1500 rpm. Present results are compared with thier as good verification of the used program, and it is possible
experimental measurement. comparisons of the to predict flow fields using the present CFD model.
experimental results of Schönwald et al with the present
computational results are shown in fig. 7. Figure. 7 (A) 4.2 CFD Post Processing Validation
shows the variation of the static pressure rise coefficient Flow visualization is required to create a surface at the
and the flow coefficient for the centrifugal fan. The mid-section of the impeller to capture the flow separation
unsteady pressure coefficient agrees well with the inside the blade passages. Figures 8 [a] and [b] show a
experimental results of Schönwald et al. across the entire comparison of the streamlines colored by the magnitude of
flow coefficient range. The unsteady pressure curve gives the velocity between Schönwald et al. and the present
more accurate values than those given by the steady unsteady CFD on a plane at the mid-section of the impeller
calculations. This result came from the solution method of at the maximum efficiency point. Both graphs show that
the unsteady simulation; the static pressure rise is the flow separation occurs first after the midpoint of the
measured under unsteady statistics by calculating the mean blade in all impeller passages, and they show good general
static pressure rise at each time step by rotating the agreement between them.
impeller (3°) with 120 time steps during a complete
rotation of the impeller. However, the steady pressure
curve could be considered an initial guess for the fan
pressure curve when computational time is limited.

Fig 8. Velocity vectors and contours at a cross section in the

centrifugal fan domain, [a] CFD by [14] and [b] CFD –present model

5. THE EFFECT OF THE SLOT’S LOCATION

A cut slot perpendicular on the blade span from the
hub to the shroud with a constant width as shown in fig.10
is investigated numerically using the present validation
model at various six configurations at diametric position
ratios (DS/D) equal to 0.5, 0.6, 0.65, 0.7, 0.8, and 0.9,
where (D) is the impeller diameter and (Ds) is the mean
diameter at the cut slot, at a rotational speed of 1500 RPM,
and with the other parameters remaining constant. Table 3
Fig 7. Comparisons between present results and experimental results shows the parameters used to construct the six cases at
of Schönwald et al. [14] at (N=1500 RPM): static pressure rise curved
(Fig.A) and efficiency curves (Fig.B) different slot locations.

168
 Engineering Research Journal (ERJ) S. A. Beskales et al Vol.52, No1 January 2023, pp:164-171

of the blade slot has a considerable effect on the fan

performance.The pressure decreased dramatically with
configurations S1 and S6 near the leading and trailing
edges, however the cut slot after the mid span of the blade
increased the pressure. Slot located at Ds/D of 0.65
namely S3 gives the best performcane compared with all
tested locations.
Figure 12 and 13 display the streamlines colored by
the velocity magnitude for the datum blade (S0) and
Fig 10. Slotted blade spesifications in present work
slotted blades at different slot locations S1, S2, S3, S4, S5
and S6. For the datum blade, the streamlines at the suction
side separated in all impeller passages. The point of
separation appeared at one-third of the blade span from the
blade’s trailing edge. For the slotted blades in case S1(near
the leading edge), S5 and S6 (near the trailing edge), it is
clear that an early separation occurs near the trailing edge
of the blades compared with S0, these slot locations have a
negative effect on the flow attachment near the blade
surface. For cases S2, S3 and S4, have a general
qualitative effect on the flow separation, these
configurations showed that the separated flow is forced to
the blade tip.
6. THE EFFECT OF THE SLOT’S WIDTH
Fig 11. The effect of the slot locations on the fan efficiency and the
In the following section, the effect of the slot width on
fan static pressure rise at flow coefficient 0.0321 the static pressure rise and fan efficienc was tested using
four slot widths of 1.5 mm, 2.5mm, 3.5 mm, and 5 mm
Figure 11. demonstrates the effect of changing the slot while keeping the best slot location S3 at (Ds/D of 0.65) as
location along the blade span on the efficiency and the shown in fig. 14. Numerical simulation methods and same
static pressure difference of a backward centrifugal fan at boundary conditions were applied to compare the unsteady
the BEP (best efficiency point) at flow coefficient 0.0321. performance, which includes the static pressure difference,
The slot is shown to reduce the pressure rise and fan efficiency, and internal flow field distribution.
efficiency while it is known to reduce fan noise and
pulsation outflows [6]. The figure shows that the location
TABLE.3 Slot’s specifications.

Configurations S0 S1 S2 S3 S4 S5 S6
Ds/DImpeller - 0.5 0.6 0.65 0.7 0.8 0.9
Ds - 361 433.2 469.8 505.4 577.6 649.8
Width - 2.5 2.5 2.5 2.5 2.5 2.5

Fig 12. Streamlines comparisons colored by velocity magnitudes at flow coefficient 0.0321 between Datum Blades (S0) and slotted blades at six
different locations

169
Engineering Research Journal (ERJ) S. A. Beskales et al Vol.52, No1 January 2023, pp:164-171

Fig 13. A close view of velocity streamlines comparisons in the blade passages at flow coefficient 0.0321 between the datum blade (S0) and slotted
blades at six different locations

Fig 14. Fan blades with slot configuration S3 and different slot widths

Figure 15 represents the effect of slot width on static

pressure and efficiency at the BEP (flow coefficient of
0.0321). It is clear from the figure, that the slot
configuration W2.5 corresponding to the location Ds/D of
0.65 and 2.5 mm width found to be the optimum
configuration compared with the other ones (W1.5,3.5,5)
and lower than the base model W0 (without slot) by 1%
in efficiency. Additionally, as the slot width increase
above than 2.5 mm, the fan efficiency and static pressure
decreased. slot configuration W5 reduced the efficiency
by 3% compared with the model W0.
Figure 16 illustrates the flow behavior near the
suction side at different slot widths. The magnitude of the
velocity of the jet of fluid coming out of the slot
Fig 15. The effect of the slot widths on the fan efficiency and the decreased as the slot width increased from 1.5 to 5 mm,
fan static pressure rise at BEP (flow coefficient 0.0321) at location resulting in a reduction in fluid momentum at the suction
Ds/DImpeller of 0.65 side. At Configuration W5, the boundary layer thickness
increased, the reversed flow appeared and increased the

170
 Engineering Research Journal (ERJ) S. A. Beskales et al Vol.52, No1 January 2023, pp:164-171

loss at the blade tip. Configuration w1.5 also increased REFERENCES

the flow separation at the blade tip compared with the [1] Schlichting, “Boundary Layer Theory,” Seventh Ed. McCraw-
base model. On the other hand, models W2.5 and W3.5 Hill, 1968.
streamlined the fluid over the blade surface and reduced [2] T. L. Le, T. T. Nghia, H. D. Thong, and M. H. K. Son,
“Numerical Study of Aerodynamic Performance and Flow
the loss at the impeller outlet. Characteristics of a Centrifugal Blower,” Int. J. Intell. Unmanned
Syst., 2022, doi: 10.1108/IJIUS-07-2021-0076/FULL/PDF.
[3] Y. Li, D. H. Rowinski, K. Bansal, and K. R. Reddy, “CFD
Modeling and Performance Evaluation of a Centrifugal Fan Using
a Cut-Cell Method With Automatic Mesh Generation and
Adaptive Mesh Refinement,” in 24th International Compressor
Engineering Conference at Purdue, Jan. 2018. [Online].
Available: https://docs.lib.purdue.edu/icec/2622
[4] H. Ding, T. Chang, and F. Lin, “The influence of the blade outlet
angle on the flow field and pressure pulsation in a centrifugal
fan,” Processes, vol. 8, no. 11, pp. 1–14, 2020, doi:
10.3390/pr8111422.
[5] M. S. Kassim, F. A. Saleh, and M. A. Kadhum, “Experimental
and Numerical Investigation of Blades Slots on Rotating Stall
Phenomenon in Centrifugal Blower,” Univers. J. Eng. Sci., vol. 3,
no. 2, pp. 24–37, May 2015, doi: 10.13189/UJES.2015.030203.
[6] A. Pieruszka, T. Siwek, W. Kalawa, Ł. Lis, S. Stefański, and K.
Sztekler, “The Effect of Blade Slots on Flow Behaviour in a Fan
Impeller,” EPJ Web Conf., vol. 02067, pp. 1–4, 2019, doi:
Fig 16. A close view of velocity streamlines comparisons in the 10.1051/epjconf/201921302067.
blade passages at BEP between the confirmation (S3) and slotted blades [7] H. G. Zhang, F. Y. Dong, X. D. Zhang, and W. L. Chu,
with Different widths at the same location “Numerical Investigation of Effect of Blade Surface Slot on the
Performance and Flow Field of a Low Speed Centrifugal Fan,”
7. Conclusions Proc. ASME Turbo Expo, vol. 2E-2020, Jan. 2021, doi:
10.1115/GT2020-14472.
The present study numerically investigated the effect
of the location and width of a single cut slot in the [8] N. Madhwesh and K. V. Karanth, “Effects of Innovative Suction
impeller blades on flow behavior inside the blade Slots on the Performance of a Radial Bladed Impeller of a
Centrifugal Blower - A Numerical Transient Analysis,”
passages and the fan performance. The results are espublisher, vol. 17, pp. 91–100, Dec. 2021, doi:
concluded as the following: 10.30919/ES8D587.
 slot cut near the mid-span of the blade at [9] R. Rong, K. Cui, Z. Li, and Z. Wu, “Numerical Study of
(configurations S2, S3 and S4), the jet of fluid Centrifugal Fan with Slots in Blade Surface,” Procedia Eng., vol.
126, pp. 588–591, Jan. 2015, doi:
efflux from the pressure side to the suction side 10.1016/J.PROENG.2015.11.309.
through the slot increased the fluid velocity and [10] Y. Xie, J. Chen, H. Qu, G. Xie, D. Zhang, and M. Moshfeghi,
forced the recirculated flow toward the blade tip “Numerical and experimental investigation on the flow separation
and let the boundary layer build up over the control of S809 airfoil with slot,” Math. Probl. Eng., vol. 2013,
blade suction side. S3 is considered as the best doi: 10.1155/2013/301748.
location among all the tested locations. [11] Q. Ke and L. Tang, “Performance Optimization of Slotted Blades
for Low-Specific Speed Centrifugal Pumps,” Adv. Civ. Eng., vol.
conversely, slots cut near the blade leading edge 2023, doi: 10.1155/2023/9612947.
(configuration S1) and near the trailing edge [12] H. K. Versteeg and W. Malalasekera, An Introdaction to
(configuration S5 and S6) boosted the boundary Computational Fluid Dynamics: The Finite Volume Method, vol.
layer separation over the blade and resulting an Second Edi. 1996.
early separation compared with the base model [13] F. R. Menter, “Zonal Two Equation Κ- ω Turbulence Models for
S0 (without slots). Aerodynamic Flows,” AIAA 23rd Fluid Dyn. Plasmadynamics,
Lasers Conf., 1993, doi: 10.2514/6.1993-2906.
 Increased the slot width up to 5 mm reduced the [14] S. Schönwald, F. Kameier, and M. Böhle, “Influence of the
fluid momentum at the slot outlet and the Casing Width and the Impeller Position on Centrifugal Fan
efficiency dropped by 2% compared with Performance - A CFD-Based Study on Cause and Effect,” Proc.
configuration S3. ASME Turbo Expo, vol. 1A, 2014, doi: 10.1115/GT2014-25319.
 The optimum slot configuration for the used [15] M. Sosnowski, J. Krzywanski, and R. Gnatowska, “Polyhedral
Meshing as an Innovative Approach to Computational Domain
impeller is S3 with 2.5 mm width and at location Discretization of a Cyclone in a Fluidized Bed CLC Unit,”
at Ds equal to 0.65 of the impeller diameter and Energy and Fuels, vol. 14, p. 1027, Jan. 2017, doi:
at 37 % of the span from the blade leading edge. 10.1051/e3sconf/20171401027.

 Slot cut in the impeller blades from the hub to

the shroud at any location decreased the
efficiency and pressure rise coefficient across
compared with the base model S0 while sloted
blades are known to reduce noise and pulsating
outlet flow [6].

171