Renewable Energy 163 (2021) 845e857

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Studies on application of vertical axis hydro turbine for sustainable

power generation in irrigation channels with different bed slopes
Shashikumar C M a, Ramesh Honnasiddaiah b, Vijaykumar Hindasageri a,
Vasudeva Madav a, *
a
Department of Mechanical Engineering, National Institute of Technology Karnataka, Surathkal, 575025, India
b
Department of Water Resources and Ocean Engineering, National Institute of Technology Karnataka, Surathkal, 575025, India

a r t i c l e i n f o a b s t r a c t

Article history: The present work is carried out to study the performance of a Savonius rotor for small-scale hydropower
Received 26 July 2019 generation. It has been observed that some of the irrigation channels available in the rural areas are
Received in revised form having enough bed slope to generate kinetic energy, which can be harnessed through a Savonius rotor.
29 August 2020
An in-house fabricated scale-down model of the Savonius rotor is tested at an inclination of the re-
Accepted 1 September 2020
Available online 4 September 2020
circulating indoor multipurpose tilting flume at 0
, 0.5
, 1.0
, 1.5
and 2.0
to determine performance
under controlled conditions. It is observed that at the tip speed ratio of 0.92 and channel inclination of
0.5
compared to 0
inclination, the coefficient of power and coefficient of torque improved to 40% and
Keywords:
Coefficient of power
10%, respectively. Furthermore, it is found that the torque and power developed by the turbine are
Coefficient of torque maximum at a bed slope of 2.0
owing to the maximum available energy.
Tip speed ratio © 2020 Elsevier Ltd. All rights reserved.
Hydropower generation
Bed slope inclination

1. Introduction
In the current scenario, most of the power generation is
accomplished using thermal power stations [1]. An increase in
greenhouse gas emissions from several thermal power stations
affects our environment and causes global warming. Moreover, the
electric power supply from the power stations to remote places
leads to enormous transmission and distribution losses [1,2]. Thus,
power generation using locally available renewable energy re-
sources can play a vital role in the remote areas. The extraction of
kinetic energy available in small irrigation channels for power
generation using the Savonius rotor can be a viable solution to meet
the green energy requirements in remote villages [2e6]. The
Savonius rotor works on drag difference between concave
(advancing blade), and convex blades (returning blade). Due to its
simple design, decent starting torque and ability to receive wind
from any direction, it has been used extensively for wind turbine
applications. Since its first use in 1920, tremendous efforts have
been put to enhance the performance. Various blade profiles have
* Corresponding author. Department of Mechanical Engineering, National Insti-
tute of Technology Karnataka, Surathkal, India.
E-mail address: vasu@nitk.edu.in (V. Madav).
been studied to improve efficiency. Fig. 1(a) and (b) shows sche-
matic of irrigation channel (top view) with Savonius hydrokinetic
turbine.
Roy and Saha [7] have reviewed experimental studies on the
Savonius turbine, which aim to improve performance parameters
such as CP, CT by varying different parameters like the shape of
deflector plate, guide vanes, and nozzle. Other parameters studied
are blade aspect ratio (AR), blade overlap ratio (OR), blade arc angle,
number of blades, flow Reynolds number, and blockage ratio. Wang
et al. [8] reviewed work on various performance augmentation
techniques like multi-staging, curtain design, windshields,
convergent nozzle, deflector plate, and guide-box tunnel.
The review articles by Akwa et al. [9] and Roy and Saha [10]
cover the theoretical, numerical, and experimental studies on
Savonius wind turbines. The coefficient of power of the Savonius
turbine varies from 0.12 to 0.3, where the maximum being 0.38 for
the rotor with curtain plates. The application of the Savonius rotor
extended to small hydropower applications is reported by Sarma
et al. [11]. Khan et al. [12] have tested single, double, and three
staged vertical axis Savonius turbines to extract hydropower en-
ergy. The CPmax reported is 0.038, 0.05, and 0.04 for single, double,
and three staged rotors. Study on the development of the innova-
tive design of rotor blades with varying aspect ratio (AR), overlap
ratio (OR), blade arc angle, and using endplates as well as deflector

https://doi.org/10.1016/j.renene.2020.09.015
0960-1481/© 2020 Elsevier Ltd. All rights reserved.
846 S. C M et al. / Renewable Energy 163 (2021) 845e857

Nomenclature g Acceleration due to gravity (m/s2)

Re Reynolds number (Dimensionless)
AR Area of the rotor (m2) Rshaft Radius of the shaft (m)
AW Wetted area of the channel (m2) RRope Radius of the rope (m)
BR Blockage ratio (Dimensionless) SSB Spring balance (g)
CT Coefficient of torque (Dimensionless) TSR (l) Tip speed ratio (Dimensionless)
CP Coefficient of power (Dimensionless) TRotor Rotor torque (N-m)
CTmax Maximum coefficient of torque (Dimensionless) TAvailable Available torque (N-m)
CPmax Maximum coefficient of power (Dimensionless) t Thickness of the turbine blade (mm)
Cd Coefficient of discharge (Dimensionless) Vw Velocity of the water (m/s)
DR Diameter of the rotor (m) WL weighing pan load (g)
Do Diameter of the endplate (m) WC Width of the channel (m)
Fr Froude number (Dimensionless) WR Width of the rotor (m)
HR Height of the rotor (m)
HW Depth of water in the channel (m) Greek symbols
HC Height of the channel (m) u Angular velocity (rad/s)
p Radius of the straight edge (m) mw Absolute viscosity of water (N-s/m2)
PRotor Power developed by the rotor (W) rw Density of water (kg/m3)
PAvailable Power available (W) j Blade arc angle (Degree)
q Radius of the arc (m) q Bed slope inclination of the channel (Degree)
QActual Actual discharge (m3/s)

found that CP for two-staged conventional rotors is lesser than that

Fig. 1. (a) Schematic (top) view of irrigation channel and (b) Savonius rotor in Irri-
gation channel [3].
plates is reported by Kumar and Saini [2]. Nakajima et al. [13]
studied the effect of direction of rotation of a horizontal axis
Savonius water turbine on the power coefficient. It was observed
that the CP was always larger in counter-clockwise rotation (see
Fig. 1) as compared to the clockwise rotation until the clearance
ratio remains less than 0.73. In a further study by Nakajima et al.
[14], two double staged horizontal axis rotors (without and with
90
degree phase change) were studied to compare performance
with a single staged horizontal axis conventional rotor. The authors
for a single staged rotor. However, for two-staged rotors with phase
change, 10% higher CP is observed than the single staged rotor.
Yaakob et al. [15] numerically investigated conventional Savonius
water turbines and reported a maximum CP of 0.275 at TSR of 0.7,
which is in good agreement with the reported experimental results.
Golecha et al. [16] tested a modified Savonius rotor with a
Reynolds number of 1.32  105 using a single deflector plate and
found that CPmax is 0.21 at TSR of 0.82. Furthermore, in the study on
two deflector plates by Golecha et al. [17], for an optimum position,
CP increased to 0.35. Birjandi et al. [18] conducted an experimental
investigation of hydrokinetic turbines using the aerofoil sections of
NACA-0021 at different water levels. The results showed degrada-
tion of CP due to partial immersion for the lift-based vertical-axis
Darrieus turbines.
Patel et al. [6] studied the influence of overlap ratio (OR) and
aspect ratio (AR) on the performance of the conventional Savonius
water turbine. They recommended using endplates in the turbine
design, having an OR from 0.1 to 0.15 and AR higher than 1.8. Sahim
et al. [19] studied a combination of Darrieus and Savonius type of
water turbine. They observed that the Darrieus turbine gets easily
self-started in the presence of a Savonius rotor even at low speed. A
combined water turbine was studied with a deflector plate in a
further study by Sahim et al. [20]. It is reported that both CP and CT
could be improved by keeping the deflector plate in the rotor’s
upstream. Golecha et al. [21] studied the effect of the separation
gap ratio on the interaction between two rotors placed in an inline
fashion. It was concluded that placing the turbine near to the first
turbine increases the interaction and reduces the value of CPmax for
both the turbines compared to the arrangement of placing the
downstream turbine far away from the upstream turbine. Patel
et al. [22] further studied the hydrodynamic output of a vertical axis
Savonius turbine with semi-circular vanes theoretically and pro-
posed a concept to choose the best Savonius turbine system layout
configuration. Kumar and Saini [23] carried out numerical studies
on twisted blades having twist angles 0
, 12.5
, and 25
for a ve-
locity range of 0.5e2.0 m/s. The CPmax value of 0.39 was found with
a twist angle of 12.5
at a velocity of 2 m/s. Kumar and Saini [24]
also carried out numerical studies on twisted blades modified with
various arc angles keeping the same twist angle of 12.5
and
 S. C M et al. / Renewable Energy 163 (2021) 845e857
reported that CPmax ¼ 0.426 at arc angle 150
. Elbatran et al. [25]
numerically investigated the modified Savonius rotor with ducted
nozzle and reported the values of CPmax and CTmax 0.27 and 0.37,
respectively.
Talukdar et al. [26] have carried out experiments on two and
three-bladed semi-circular rotors for the rotor’s different immer-
sion levels. It is reported that a two-bladed semi-circular rotor has a
better performance compared to two blades of the elliptical rotor.
Furthermore, the rotor immersed 100% in water has better perfor-
mance. Mosbahi et al. [27] used the augmentation method to work
on the Savonius water turbine’s helical shape and found that
CPmax ¼ 0.14 at TSR of 0.7. Patel et al. [28] studied the effect of canal
geometric parameters on the efficiency of Savonius hydro turbines
and introduced the methodology for velocity correction to deter-
mine the actual turbine performance. Their experimental findings
show that if velocity correction is not considered, there is a wide
variation in the maximum power coefficient for all cases studied.
This literature review reveals that several experimental and
numerical studies have been performed by different authors, using
different types of turbines in a laboratory-scale open channel with
a horizontal bed (bed slope q ¼ 0). The only reported study by
Sivakumar et al. [29] on varying bed slopes was carried out by
numerical simulation. The authors predicted that the total power
generation increased at the highest slope due to the increase in
flow velocity. There are no experimental studies reported in the
literature on the performance of the Savonious rotor in an open
channel with varying slopes. Hence, in the present study, an
experimental investigation is conducted on a Savonius rotor to
study the influence of inclination of bed slope in a laboratory-scale
multipurpose tilting flume with the following objectives:
(a) To study the performance of scaled-down modified Savonius
rotor with various angles of inclination of the channel from
q ¼ 0
to 2
with an increment of 0.5
.
(b) To study the effect of the Froude number (Fr) on the per-
formance of the turbine concerning different bed slope
inclination (q) of the channel.
Fig. 2. (a) Schematic of the experimental setup and (b) pictorial view of experimental
setup
1. Spring balance, 2. Pulley, 3. Rope, 4. Weighing Pan, 5. Shaft, 6. Flange bearing, 7. End Plates, 8. Savonius rotor, 9. Supporting rods.
847
2. Experimental setup
Fig. 2 (a and b) shows the schematic and pictorial view of the
rotor test setup used in the present work. It consists of four
threaded studs, which are used as pillars to mount the test rig. A
mild steel plate is bolted to the threaded rods at its four ends.
Flange bearing (SKF make) is fitted to the upper side of the mild
steel plate with four fasteners. The scaled-down Savonius rotor
with end plates is fabricated using a galvanized iron sheet of 1 mm
thickness (t) fitted to flange bearing by 12 mm shaft. Rope dyna-
mometer equipped with weighing pan, spring balance (2.5 g ac-
curacy), and pulley are used to apply the load to the Savonius rotor
shaft. Nylon string of 1 mm diameter is wound around the rotor
shaft and connected to a weighing pan and spring balance using
three pulleys attached to the structure. Digital tachometer (Pyle
make) is used to measure the shaft RPM with the help of a reflector
attached to the shaft. Important geometric parameters of the
Savonius rotor are shape factor (p/q), arc angle (j), and aspect ratio
(AR) [16], which are fixed values of 0.2, 124
and 0.7 respectively.
Fig. 3 (a and b) describes the geometric parameters which are
considered for the present experimental investigation. Further-
more, parameters like straight edge, circular arc (p), arc angle (q),
the diameter of the rotor (DR), the height of the rotor (HR), and
endplate diameter (Do) of the rotor are also shown in Fig. 3 (a and
b).
2.2. Dimensions of scaled model and procedure
Golecha et al. [16] carried out the experimental investigations in
an open channel having a cross-sectional area of
730 mm  330 mm to study the influence of the deflector plate on
the performance of modified Savonius water turbine. In the present
study, experiments were performed in laboratory-scale water re-
circulating indoor flume with a flatbed having a cross-sectional
area of 215 mm  350 mm using a scale-down model of Savo-
nious water turbine used by Golecha et al. [16]. The cross-sectional
area of the experimental setup used by Golecha et al. [16] and the
848 S. C M et al. / Renewable Energy 163 (2021) 845e857

RPM and torque developed by the rotor. A similar procedure is

adopted for other cases with inclination angle q ¼ 0
e2
with an
increment of q ¼ 0.5
. The discharge is kept constant for both
horizontal and inclined cases. The velocity of water (VW) is calcu-
lated by finding the channel’s wetted area (AW) and discharge
through the rectangular notch. The measured values of torque and
power are used to calculate the performance parameters such as
coefficient of torque (CT) and coefficient of power (CP), respectively.
The details of the instruments used to experiment are listed in
Table 2.

2.2. Data reduction

The performance of Savonius water turbines expressed in terms

of the coefficient of power (CP) and coefficient of torque (CT) are
given by the following equations [16]:

TRotor TRotor
Coefficient of torque ðCT Þ ¼ ¼

Fig. 3. Geometric parameters of Savonius rotor [16] (a) schematic Savonius rotor and TAvailable
2rW AR VW
1 2 DR
(b) pictorial view of Savonius rotor. 2

TRotor
¼
 (1)
present study are different. Therefore, scaled-down dimensions are 1r H D2 V 2
4 W R R W
calculated and presented in Table 1.
Fig. 4 (a and b) show the schematic and pictorial view of
multipurpose tilting flume having a length, width, height of 4.75m, Where AR ¼ HRDR
0.215m, and 0.35m, respectively. The sidewalls of the tilting flume  
are made of acrylic material. The flume has an inlet and outlet with Rotor torque ðTRotor Þ ¼ ðWL  SSB Þg Rshaft þ RRope (2)
a collecting tank at its extreme ends. This tilting flume is having a
recirculation type of flow employing a two-monoblock pump
PRotor ðT  uÞ
(SHARP make) of 2.0 HP capacity with a maximum discharge ca- Coefficient of power ðCP Þ ¼ ¼
Rotor 
PAvailable 1r ðA ÞV 3
pacity of 0.01544 m3/s, fixed to the flume bottom structure, and two 2 W R W
independent control valves provided for each pump to regulate the
discharge. A pre-calibrated rectangular notch attached at the flume ðTRotor  uÞ
¼
 (3)
outlet having a coefficient of discharge (Cd) of 0.6. The lifting 1r ðH D ÞV 3
2 W R R W
mechanism is employed by two screw threads placed at the inlet
and outlet side of the tilting flume, giving a slope of ±2
with an
increment of 0.5
. The pictorial view of the side view, front view of uDR
the experimental setup in a multipurpose tilting flume with height Tip Speed Ratio ðTSRÞ l ¼ (4)
2VW
gauge shown in Fig. 4 (c, d, and e).
Fig. 5 (a) and (b) show the schematic of horizontal tilting flume
rW VW DR
inclined at an angle of (q) to the horizontal, where HC and HW are Reynolds number ðReÞ ¼ (5)
mW
the height of the channel and height of the water, respectively.

 pffiffiffiffiffiffi
2
2.1. Experimental procedure Actual discharge ðQ Actual Þ ¼ Cd WC 2g ðHW Þ3=2 (6)
3
The rotor test rig is placed inside the flume, and two monoblock
QActual QActual
pumps generate the flowrate of water in the flume. The water flow Velocity of water ðVw Þ ¼ ¼
rate is measured by a rectangular notch fixed at the outlet of the AW ðWC HW Þ

 pffiffiffiffiffiffi
flume. The kinetic energy of the flowing water initiates the rotation
Cd 23 WC 2g ðHW Þ3=2
of the rotor due to energy transfer. A Rope dynamometer is used for
¼ (7)
applying load on the rotor shaft using a weighing pan, spring bal- ðWC HW Þ
ance, and pulleys to determine the torque developed by the rotor.
The power generated is determined by measuring the rotor shaft’s Where AW ¼ WC HW

Table 1
The details of the experimental setup and dimensions of the Savonius rotor.

Sl. No Parameter Golecha et al. [16] Present Study

1. Diameter of the rotor (DR) mm 245 72

2. Endplate Diameter (Do) mm 269.5 79.2
3. Height of the actual turbine (HR) mm 170 51
4. Aspect ratio (AR) 0.7 0.7
5. Blade arc angle (j) Degree 124 124
6. Blade shape factor (p/q) 0.2 0.2
7. Width of the channel (WC) mm 730 215
 S. C M et al. / Renewable Energy 163 (2021) 845e857 849

Fig. 4. (a) Schematic and (b) pictorial view of multipurpose tilting flume with (c) rotor test rig (side view) and (d) rotor test rig (front view) (e) height gauge.
850 S. C M et al. / Renewable Energy 163 (2021) 845e857

Fig. 6. Validation of CP with respect to TSR for q ¼ 0
of inclination of the channel.

0sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1



DCT DTRotor 2 DTAvailable 2 A
¼@ þ (10)
CT TRotor TAvailable
Fig. 5. Schematic of the experimental setup in a horizontal (a) and inclined flume (b).
1. Savonious rotor, 2. Top plate, 3. Rotor shaft, 4. Rotor upper-end plate, 5. Rotor bottom 0sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1


endplate.
DCP DCT 2 Dl2 A
¼@ þ (11)
CP CT l

WR HR
Blockage ratio ðBRÞ ¼ (8)
WC HW
The blockage ratio (BR) is defined as the ratio of the rotor’s front 3. Results and discussion
area to the front wetted area of the channel and is equal to 24.3%.
From the study of Alexander and Holownia [30], it is found that The results are validated by comparing the same for the modi-
blockage correction is not required for the blockage ratio of less fied Savonius rotor of Golecha et al. [16]. Influence of bed slope
than 30%, and hence blockage correction is not included in the (q ¼ 0
e2
) on the performance of the Savonius rotor is studied
present work. experimentally and compared with the results of horizontal
(q ¼ 0
) and inclined conditions ranging from q ¼ 0.5
e2
with an
2.3. Experimental uncertainty increment of 0.5
. Figs. (6) to (11) shows the comparison of the
modified Savonius rotor [16] with a scaled-down Savonius rotor.
Uncertainties are calculated using Eqs. (9)e(11) and the method
suggested by Moffat [31]. The uncertainty values for TSR, CT, and CP 3.1. Validation of the experimental procedure
are found to be 2.02%, 6.04%, and 5.7%, respectively.
0sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 Figs. (6-7) show the comparison of variation of CP and CT con-



Dl @ Du2 DDR 2 DVW 2 A cerning TSR for scaled-down rotor (present study) with modified
¼ þ þ (9)
l u DR VW Savonius rotor [16] for q ¼ 0
bed slope. The present study’s
experimental results follow a similar trend with the experimental
results reported by Golecha et al. [16]. It is found that CPmax ¼ 0.1515

Table 2
The instruments used to carry out the experiments in multipurpose tilting flume.

Sl. No Instrument used Details

1. Multipurpose tilting flume WC ¼ 0.25m, length of the flume ¼ 4.75m

2. Starter Independent starter for each pump, 1 Phase
3. No. of control valves 2.0
4. Pump make SHARP
5. Pump capacity and number of pumps 2 HP and 2.0
6. Tachometer Specifications (a) RPM measurement,
PLYE Make-PLT26 (b) Professional Digital Non-contact laser tachometer
(c) Resolution: 2.5e1000 RPM
(d) Accuracy: ±0.05%
7. Spring balance To measure net load acting on the shaft
SALTER make Accuracy: 2.5 g accuracy
8. Height gauge To measure the depth of water (Hw)
9. Scaled-down modified Savonius turbine blade Diameter of the turbine (DR) ¼ 72 mm
Height of the turbine blade (HR) ¼ 51 mm
 S. C M et al. / Renewable Energy 163 (2021) 845e857 851

Fig. 7. Validation of CT with respect to TSR for q ¼ 0
of inclination of the channel. Fig. 10. Variation of CT with respect to TSR for q ¼ 0o-2o of inclination of channel.

Fig. 8. Comparison of CT with respect to TSR for q ¼ 0
and 0.5
of inclination of the
channel with the literature [16,26] under the fully submerged condition. Fig. 11. Variation of CP with respect to TSR for q ¼ 0o-2o of inclination of channel.

Figs. (10) and (11) shows the variation of coefficient of torque

(CT) and coefficient of power (CP) with respect to tip speed ratio
(TSR) while varying bed slopes from q ¼ 0
e2
with an increment
of 0.5 for a scaled-down modified Savonius turbine blade used in
the present study. The scaled-down turbine model was tested in a
multipurpose tilting flume by maintaining a 70 mm depth of the
water with an inlet velocity (Vw) of 0.3088 m/s for q ¼ 0
bed slope
under complete immersion condition. The value of CPmax and CTmax
with respect to TSR were found to be 0.1515 and 0.2179 at a TSR of
0.7.
Talukdar et al. [26] obtained higher CPmax and CTmax values with
respect to TSR while conducting experiments using a Savonius rotor
at a fully submerged condition with an inlet velocity 0.8 m/s. The
high-pressure region on the concave side and low-pressure region
at the convex side of advancing blade profile aids in producing a
pressure drop that results in turbine rotation. This higher pressure
Fig. 9. Comparison of CP with respect to TSR for q ¼ 0
and 0.5
of inclination of at the concave side of the advancing blade contributes to a more
channel. significant drag coefficient, which in turn improves overall
efficiency.
and CTmax ¼ 0.2179 at a TSR of 0.7. Table 3 shows performance
parameters such as CPmax and CTmax of Golecha et al. [16] and the
Table 3
present study and the percentage increase of CPmax and CTmax. This
Important dimensions and values of CTmax and CPmax along with percentage error.
variation in the present study is mainly because of the change in the
turbine blade’s dimensions, top, and bottom endplates. In the Sl. Performance Golecha et al. Present Percentage
No parameters [16] Study increase
present study, galvanized iron with 1 mm thickness was used,
whereas in the study proposed by Golecha et al. [16], an acrylic 1. CPmax 0.14 0.1515 8.21%
2. CTmax 0.2 0.2179 6.7%
sheet with a 10 mm thickness is used.
852 S. C M et al. / Renewable Energy 163 (2021) 845e857

Table 4
Comparison of performance parameters of the turbine under various conditions.

Sl. No Researcher Type of turbine Bed slope (q) degree Immersion level (%) VW (m/s) CPmax CTmax TSR

1. Talukdar et al. [26] Two bladed 0
100 0.8 0.28 0.31 0.89
Savonius 80 0.10 0.17 0.61
60 0.07 0.10 0.57
2. Golecha et al. [16] Modified two-bladed 0
100 0.45 0.14 0.20 0.70
Savonius rotor


3. Present Study Modified two-bladed 0 100 0.3088 0.15 0.22 0.70
Savonius rotor (Scaled-down) 0.5
100 0.3266 0.21 0.23 0.92
1
73 0.4239 0.12 0.14 0.84
1.5
33 0.9827 0.03 0.08 0.33
2
31 1.3511 0.02 0.06 0.32

In the present study, the value of CPmax and CTmax with respect to
TSR were found to be higher than the experimental studies carried
out by Golecha et al. [16] using modified Savonious rotor at fully
submerged condition with an inlet velocity of 0.45 m/s. For the bed
slope (q) of 0.5
, the depth of water (Hw) was reduced to 67 mm
(4.3% reduction), and the turbine blade is in a fully submerged state.
The value of CTmax and CPmax are found to be 0.2296 and 0.2119 at a
TSR equals to 0.9227. This observation is due to the increase in inlet
velocity. As the inlet velocity of water increases, the performance of
the Savonious turbine increases. Similar results were reported by
Anuj Kumar and Saini [23] while conducting three dimensional
transient CFD simulations using twisted blades by varying the twist
angle with different water velocities when the turbine blade is fully
submerged condition. Due to the high-pressure zone at the concave
side of the advancing blade and low-pressure zone at the convex
side of the advancing blade profile aids in producing a pressure
drop, which results in turbine rotation. Furthermore, Birjandi et al.
[18] also reported that the maximum power coefficient might be
obtained when the turbine is under fully submerged condition
while conducting experiments on squirrel-cage hydrokinetic tur-
bine model with two blades. The increase in the coefficient of po-
wer is due to a reduction in the positive clearance coefficient and
water level above the turbine blade. The coefficient of power
developed by the turbine blade is increased under a fully sub-
merged condition.
For the other bed slopes, q ¼ 1
, 1.5
, and 2
, the turbine’s im-
mersion level decreases to 73%, 33%, and 31% of the original height,
respectively. As the depth of water (Hw) decreases with respect to
bed slope (q), the velocity of water (Vw) increases, which results in a
decrease of tip speed ratio (TSR). The values CTmax and CPmax
decrease due to an increase in power available (PAvailable) in the
flume with respect to bed slope (q ¼ 1
, 1.5
, and 2
), as shown in
Figs. 10 and 11. When the turbine is partially submerged, the
portion of the blade is exposed to air in the advancing side, leading
to a decrease of the coefficient of power and coefficient of torque
[18,26].
The performance of a turbine increases with respect to the TSR
when the turbine is in a fully submerged condition (q ¼ 0.5
), as
shown in Figs. 10 and 11 and a submerged cross-sectional area of
the turbine is 0.0037 m2 (4.3% reduction in immersion level) with
respect to bed slope q ¼ 0.5
. As the turbine is in the fully sub-
merged condition the entire turbine blade surface is in contact with
water and higher extraction of kinetic energy is possible
(CPmax ¼ 0.2119) when compared to the turbine blade which is in
partially submerged condition, and submerged cross-sectional area
of the turbine blade is 0.0037, 0.0016 and 0.0012 m2 (immersion
level of turbine decreased to 73%, 33% and 31% of the original height
movement causes the rotor to rotate; however, the surface area of
the turbine rotor on which the water impacts is reduced. When the
turbine blade is exposed to air, the turbine blade hydrodynamics
will be disturbed due to the density difference between air and
water (density of water 850 times more massive than air) [12]. The
air entrainment decreases the density of water, which is in contact
with the turbine blade surface and leads to a reduction in the drag
forces created in this area which, in turn, reduces the turbine per-
formance [18,32].
The Savonious rotor is a unique fluid mechanical device that
works on the drag effect mechanism rather than a lift mechanism.
The drag coefficient of the concave surface (advancing blade) is
larger than the convex surface (returning blade), forcing the rotor
to rotate. This reduction in the drag coefficient of the concave and
convex surface, forces the rotor to rotate at a reduced torque co-
efficient, which affects the turbine efficiency.
The present study reveals that the turbine rotor submergence is
found to be a critical parameter that significantly affects the turbine
efficiency. The turbine performance parameters obtained for bed
slope of q ¼ 0
and 0.5
, recommend that the turbine be submerged
to a depth that provides a subcritical Froude number (Fr). Due to the
limitation in the current experimental setup, it was not possible to
carry out the experiments under the fully submerged condition at
bed slopes q ¼ 1
, 1.5
and 2
(immersion level of turbine decreased
to 73%, 33% and 31% of the original height). Further studies need to
be carried out at different bed slopes when the turbine rotor surface
area is in complete contact with the water to obtain increased
turbine performance.
Table 4 show a comparison of the performance parameters of the
turbine under various bed slope conditions. Figs. 8 and 9 shows
comparison of CT and CP with respect to TSR for q ¼ 0
and 0.5
of
inclination of the channel with the literature [16,26] under the fully
submerged condition. Figs. 8 and 9 show change in coefficient of
torque and coefficient of power for the bed slope of q ¼ 0
and 0.5

under fully submerged condition of the turbine blade and it is
compared to the case when the immersion level of the turbine blade
is 100%. The variation of coefficient of torque (CT) and power coef-
ficient (CP) with respect to tip speed ratio (TSR) for the bed slope
ranging from q ¼ 0
e2
is presented in Figs. 10 and 11, respectively.
The values of maximum coefficient of power (CPmax) and
maximum coefficient of torque (CTmax) with tip speed ratio (TSR)
for different bed slope (q) is presented in Table 5.with the literature
[16,26] under fully submerged condition.
Table 5
Maximum values of CP and CT with respect to TSR for different bed slopes.
Sl. No q HW VW CPmax CTmax TSR

respectively) with respect to bed slope q ¼ 1
, 1.5
and 2
1. 0
70 0.3088 0.1515 0.2179 0.6955
(CPmax ¼ 0.1208, 0.0278 and 0.0205). Under the partially submerged 2. 0.5
67 0.3266 0.2119 0.2296 0.9227
3. 1
51 0.4239 0.1208 0.1446 0.8356
condition, the turbine rotor is suspended above the water surface,
4. 1.5
22 0.9827 0.0278 0.0842 0.3298
and only a lower portion of the blades are submerged. The water 5. 2
16 1.3511 0.0205 0.0635 0.3235
 S. C M et al. / Renewable Energy 163 (2021) 845e857 853

3.3. Limitation of the present study Table 6

Typical values of Froude number (Fr) for different bed slope.

In the present study, experiments are performed in a laboratory- Sl. Bed Depth of water Vw (m/ Froude Maximum Maximum
scale water re-circulating multipurpose tilting flume with a flatbed. No slope HW (mm) s) number (Fr) TDeveloped PDeveloped
The available flume was designed to vary from 0
to maximum up q
to 2
with a maximum discharge. Due to this limitation, the labo- 1. 0
70 0.3088 0.372 0.0063 0.0082
ratory experiments were carried out with a maximum bed slope up 2. 0.5
67 0.3226 0.397 0.0069 0.0616
3. 1
51 0.42389 0.599 0.0119 0.0169
to 2
with the available maximum discharge QActual 0.01544 m3/s. In
4. 1.5
22 0.98265 2.11 0.0245 0.0185
order to maintain the same depth of water (Hw) in all the bed slopes 5. 2
16 1.35115 3.41 0.0378 0.0292
(q), the supply of water from the pump should be high, i.e., more
discharge (QActual) is required from the pump. The variation of
water velocity and discharge with respect to bed slope for constant
velocity of water (VW) for the bed slope ranging from q ¼ 0
e0.5

depth of water (Hw) is presented in Fig. 12.
shown in Figs. 10 and 11. The type of flow is subcritical flow (Fr < 1).
Cleynen et al. [36] reported that a similar improvement in tip speed
3.4. Effect of Froude number (Fr) ratio with the increase in Froude number had been observed for
free stream water wheels. Similarly, for the other bed slopes, the tip
Froude number is a ratio of the inertia force to the gravitational speed ratio (TSR) of the turbine blade decreases due to an increase
force of the fluid [33e37]. The value of Froude number (Fr) depends in water velocity, and the value of CP and CT decreases. For Froude
on the depth (HW) and velocity (Vw) of water in the channel, and it number >1, the type of flow is Super critical. There is a loss of power
can be written as exhibited by the turbine blade due to the high flow velocity of
water, and a similar observation was reported by Quaranta [37].
Inertia force Vw As per the torque and power values compared in Table 6, it is
Fr ¼ ¼ pffiffiffiffiffiffiffiffiffiffi
Gravitationalforce gHw evident that bed slope of q ¼ 2
generates maximum value when
compared to other bed slope inclinations (q ¼ 0
, 0.5
, 1
, and 1.5
).
Where. Hence, the Froude number (Fr) reaches the supercritical flow con-
Fr ¼ 1-Flow is critical (Inertia force ¼ gravitational force). dition, indicating that inertia force is dominant, and the maximum
Fr < 1-Flow is subcritical (Inertia force < gravitational force, i.e., energy is available at the bed slope of q ¼ 2
.
gravitational forces dominate over inertia force). Fig. 13 shows the pictorial view of turbulence, splashes and
Fr > 1-Flow is supercritical (Inertia force > gravitational force,
i.e., inertia forces dominateover gravitational force).
Different types of flows occur in the irrigation channel, such as
critical flow (Fr ¼ 1), subcritical flow (Fr < 1), and supercritical flow
(Fr > 1) which depends mainly on the velocity of water (Vw) and
bed slope (q). As the bed slope increases, the velocity of the water
(Vw) in the channel increases, the depth of water (Hw) decreases,
and the Froude number (Fr) value increases. As the Froude number
increases the magnitude of force due to gravity reduces and inertial
force increases simultaneously. On the other hand, increase of bed
slope leads to Froude number (Fr) increase thereby load-carrying
capacity of the turbine blade increases. Whereas, increase in ve-
locity of water (VW) with respect to bed slope (q), rises the torque
and power values developed by the turbine blade to maximum for
q ¼ 2
and minimum for q ¼ 0
. The values of maximum torque and
power developed by the turbine with respect to Froude number (Fr)
for different values of velocity of the water (Vw), depth of water
(Hw), and bed slopes (q) are illustrated in Table 6.
As the Froude number (Fr) increases from 0.372 to 0.397, the tip
speed ratio (TSR) of the rotor blade increases due to an increase in

Fig. 13. Pictorial view of turbulence, separation of water and splashes developed
Fig. 12. Variation of the velocity of water and discharge with respect to bed slope (q) around the turbine blade for the supercritical flow (Fr > 1) for a bedslope of q ¼ 1.5
(a)
for a constant depth of water (Hw). and 2
(b).
854 S. C M et al. / Renewable Energy 163 (2021) 845e857

Fig. 16. Variation of maximum power available and developed by the turbine with
Fig. 14. Variation of Froude number (Fr) with respect to the velocity of water (Vw) and respect to Froude number (Fr) for different bed slope (q ¼ 0o-2o) inclinations of the
depth of water (Hw) for different bed slope (q ¼ 0o-2o) inclinations of the channel. channel.

separation of water which is developed around the turbine blade

for the higher Froude number (Fr > 1) at super critical flows for bed
slopes of q ¼ 1.5
and 2
. It is observed that with the increase of the
Froude number (Fr) from subcritical to critical and then to super-
critical, CT and CP of the turbine blade decreases. This may be owing
to the fact that as the Froude number (Fr) increases, Reynolds
number (Re) is also increases during interaction of turbine blade
with water.Therefore, more turbulence is created around the tur-
bine blade as shown in Fig. 13 (a and b). This in turn increases the
flow departure possibility across the turbine rotor neighborhood,
and results in lower CT and CP [38,39]. The turbine blade experi-
ences more turbulence which results in local velocity spikes or
drops at higher Froude number (Fr > 1) and separation of water
starts around the leading edge of the turbine blade as shown in
Fig. 13 (a and b) [32,39]. This causes the substantial splashing which Fig. 17. Variation of the velocity of water (VW) with respect to bed slope (q ¼ 0
e2
).
is developed around the turbine blade (Fig. 13 (a and b)) at super-
critical flow (Fr > 1) for a bed slope of q ¼ 1.5
and 2
and increases of water (Hw) and power (P) available in the flume with respect to
drag torque on the rotor shaft, which lowers the CT and CP turbine bed slope (q), were plotted using second order polynomial fit as
[32]. shown in Figs. 17e19.
The variation of Froude number (Fr) with respect to the depth of The relationship between the velocity of water (Vw), depth of
water (Hw) for different bed slopes (q ¼ 0o-2o) and velocity of water water (Hw) and power available (PAvailable) with respect to bed slope
(Vw) in the flume is shown in Fig. 14. The variation of torque (q) is expressed by following Eqs. 12e14
available (TAvailable), maximum torque developed, power available
(PAvailable), and maximum power developed with respect to Froude Vw ¼ 0:332ðqÞ2  0:1163ðqÞ þ 0:2966 (12)
number (Fr) for different bed slopes (q ¼ 0
e2
) of the flume is
presented in Figs. 15 and 16. Hw ¼ 0:00543ðqÞ2  0:01974ðqÞ þ 0:07309 (13)
When the bed slope increases (q ¼ 0
e2
), the depth of water
(Hw) in the flume reduces, and the velocity of water (Vw) along with
PAvailable ¼ 0:5347ðqÞ2  0:37634ðqÞ þ 0:06618 (14)
the available power (kinetic energy) increases as well. In order to
develop the relationship between the velocity of water (Vw), depth

Fig. 15. Variation of torque available and developed with respect to Froude number
(Fr) for different bed slope (q ¼ 0o-2o) inclinations of the channel. Fig. 18. Variation of the depth of water (HW) with respect to bed slope (q ¼ 0
e2
).
 S. C M et al. / Renewable Energy 163 (2021) 845e857 855

4. Conclusions

In the present investigation, a Savonius rotor of a geometrically

scaled-down model is designed, and the experimental result is
compared with the previous investigation [16]. Based on the pre-
sent study, the following conclusions are made.

 For a bed slope of q ¼ 0.5
it is observed that there is a increase of

CP and CT values at a TSR of 0.92.Whereas for q ¼ 1
, 1.5
and 2
,
CP and CT values decreased due to partial submergence of the
rotor.
 Due to the decrease in water depth in the flume, the velocity of
water increases, and TSR reduces, i.e., the water velocity is
higher than the rotational speed of the turbine blade. The value
of CP and CT decreases with respect to bed slope (q ¼ 1
, 1.5
and
Fig. 19. Variation of power available (Kinetic energy available) with respect to bed 2
)
slope (q ¼ 0
e2
).  The velocity of water (Vw) in the flume increases due to change
in bed slope angles (q), the kinetic energy available in the
channel increases, and inertia flow is dominating the gravity
flow as the Froude number (Fr) reaches supercritical flow.
 Since PAvailable in the channel is more than the power developed
Table 7 by the rotor, so the CP and CT values decreased for bed slope
Average discrepancy between experimental and curve fit values. angles q ¼ 1
, 1.5
and 2
due to decrease in TSR. The Froude
Parameter/Sl. No 1 2 3 4 5 6
number (Fr) reaches supercritical value for at bed slope angle
q ¼ 1.5
and 2
.
Velocity of water (VW) vs Bed slope (q)
 As the bed slope angle increases, torque and power developed
Bed slope (q) 0
0.5
1
1.5
2

Experimental parameter VW 0.3088 0.3266 0.4239 0.9827 1.3511 by the turbine blade increases. At bed slope angle q ¼ 2
the
Curve fit parameter VW 0.29658 0.32161 0.51197 0.8692 1.39286 torque and power developed by the turbine is maximum due to
Average discrepancy % 8.18 increase in Froude number (Fr > 1).
Depth of water (HW) vs Bed slope (q)
Bed slope (q) 0
0.5
1
1.5
2

Experimental parameter HW 70 67 51 22 16
Curve fit parameter HW 73.09 61.84 47.88 31.2 11.89 Author contributions
Average discrepancy % 17.14
Power available (PAvailable) vs Bed slope (q)
Bed slope (q) 0
0.5
1
1.5
2

Shashikumar Channarayapatna Manjunatha.
Experimental parameter PAvailable 0.05408 0.06164 0.13984 0.75149 1.42064 Conceived and designed the analysis.
Curve fit parameter PAvailable 0.06618 0.0107 0.22099 0.69461 1.43536 Designed experimental setup and performed the various ex-
Average discrepancy % 34.33 periments with the support from all other authors.
Collected the data.
Acquired and processed the experimental data.
Table 8
Performed the analysis.
Typical values of Vw, Re, and TSR at the various bed slopes (q) of the channel. Worked out almost all the technical details and performed nu-
merical calculations.
Sl. No Bed slope inclination Vw in m/s Reynolds number (TSR) at

 Wrote the paper.
of the channel (q) r V D no load
Re ¼ W W R
 Drafted the manuscript and designed the figures with the input
mW uDR
2VW from all authors.
1. 0
0.3088 2.95  104 1.3177 Ramesh Honnasiddaiah.
2. 0.5
0.3266 2.99  104 1.4367 Other contribution.
3. 1
0.4239 3.65  104 1.1556 Aided in interpreting the results and worked on the manuscript.
4. 1.5
0.9827 9.13  104 0.5215 Vijaykumar Hindasageri.
5. 2
1.3511 1.25  105 0.4853
Other contribution.
Contributed to the design and implementation of research, To
the analysis of research and writing of manuscript.
The average discrepancy values between the results obtained by Vasudeva Madav.
using equations (12)e(14) (2nd order polynomial fit) and the Other contribution.
experimental values are listed in Table 7 below. Helped to supervise the project and contributed to preparation
Table 8 shows the values of the velocity of water in the channel of the manuscript.
(Vw), Reynolds number (Re), and tip speed ratio (No load condition)
for different bed slope inclinations (q ¼ 0o-2o) of the channels with Declaration of competing interest
an increment of 0.5
. It is observed that the increase in bed slope (q)
inclination increases the velocity of the water (Vw), however, it The authors declare that they have no known competing
decreases the TSR. financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
856 S. C M et al. / Renewable Energy 163 (2021) 845e857

Appendix

Datasheet for the experiments carried out at different bed

slopes (q ¼ 0
e0.5
) and values of various parameters (TDeveloped,
TAvailable, PDeveloped, PAvailable, CT, CP, and TSR) is presented in
Tables (A 1 to A-5).

Table A-1
Experimental values for the bed slope q ¼ 0
.

Rshaft ¼ 6 mm, Rrope ¼ 1 mm, Dr ¼ 72 mm, HR ¼ 51 mm and VW ¼ 0.3088 m/s

Sl. No TDeveloped TAvailable CT N PDeveloped PAvailable TSR CP

1. 0.000000 0.0063 0.0000 108 0.0000 0.0541 1.3177 0.0000

2. 0.000343 0.0063 0.0545 96 0.0034 0.0541 1.1713 0.0638
3. 0.000687 0.0063 0.1089 84 0.0060 0.0541 1.0249 0.1116
4. 0.001030 0.0063 0.1634 74 0.0080 0.0541 0.9029 0.1475
5. 0.001373 0.0063 0.2179 57 0.0082 0.0541 0.6955 0.1515

Table A-2
Experimental values for the bed slope q ¼ 0.5
.

Rshaft¼ 6 mm, Rrope ¼ 1 mm, Dr ¼ 72 mm, HR ¼ 51 mm and VW ¼ 0.3226 m/s,

Sl. No TDeveloped TAvailable CT N PDeveloped PAvailable TSR CP

1. 0.000000 0.0069 0.0000 123 0.0000 0.0616 1.4367 0.0000

2. 0.000343 0.0069 0.0499 118 0.0042 0.0616 1.3783 0.0688
3. 0.000687 0.0069 0.0998 109 0.0078 0.0616 1.2731 0.1271
4. 0.001030 0.0069 0.1497 99 0.0107 0.0616 1.1563 0.1732
5. 0.001373 0.0069 0.1997 90 0.0129 0.0616 1.0512 0.2099
6. 0.001579 0.0069 0.2296 79 0.0131 0.0616 0.9227 0.2119

Table A-3
Experimental values for the bed slope q ¼ 1
.

Rshaft ¼ 6 mm, Rrope ¼ 1 mm, Dr ¼ 72 mm, HR ¼ 51 mm and VW ¼ 0.4329 m/s

Sl. No TDeveloped TAvailable CT N PDeveloped PAvailable TSR CP

1. 0.000000 0.0119 0.0000 130 0.0000 0.1398 1.1556 0.0000

2. 0.000343 0.0119 0.0289 124 0.0045 0.1398 1.1022 0.0319
3. 0.000687 0.0119 0.0578 117 0.0084 0.1398 1.0400 0.0601
4. 0.001030 0.0119 0.0867 110 0.0119 0.1398 0.9778 0.0848
5. 0.001373 0.0119 0.1156 104 0.0149 0.1398 0.9245 0.1069
6. 0.001717 0.0119 0.1446 94 0.0169 0.1398 0.8356 0.1208

Table A-4
Experimental values for the bed slope q ¼ 1.5
(only submerged area of the turbine is taken for calculating CT and CP).

Rshaft ¼ 6 mm, Rrope ¼ 1 mm, Dr ¼ 72 mm, HR ¼ 22 mm and VW ¼ 0.9827 m/s

Sl. No TDeveloped TAvailable CT N PDeveloped PAvailable TSR CP

1. 0.000000 0.0245 0.0000 136 0.0000 0.7515 0.5215 0.0000

2. 0.000343 0.0245 0.0140 130 0.0047 0.7515 0.4985 0.0070
3. 0.000687 0.0245 0.0281 126 0.0091 0.7515 0.4832 0.0136
4. 0.001030 0.0245 0.0421 113 0.0122 0.7515 0.4333 0.0182
5. 0.001373 0.0245 0.0561 104 0.0149 0.7515 0.3988 0.0224
6. 0.001717 0.0245 0.0701 96 0.0172 0.7515 0.3681 0.0258
7. 0.002060 0.0245 0.0842 86 0.0185 0.7515 0.3298 0.0278

Table A-5
Experimental values for the bed slope q ¼ 2
(only submerged area of the turbine is taken for calculating CT and CP).

Rshaft ¼ 6 mm, Rrope ¼ 1 mm, Dr ¼ 72 mm, HR ¼ 16 mm and VW ¼ 1.35115 m/s

Sl. No TDeveloped TAvailable CT N PDeveloped PAvailable TSR CP

1. 0.000000 0.0378 0.0000 174 0.0000 1.4206 0.4853 0.0000

2. 0.000343 0.0378 0.0091 168 0.0060 1.4206 0.4685 0.0043
3. 0.000687 0.0378 0.0181 159 0.0114 1.4206 0.4434 0.0080
4. 0.001030 0.0378 0.0272 151 0.0163 1.4206 0.4211 0.0115
5. 0.001373 0.0378 0.0363 144 0.0207 1.4206 0.4016 0.0146
6. 0.001717 0.0378 0.0454 137 0.0246 1.4206 0.3821 0.0173
7. 0.002060 0.0378 0.0544 125 0.0270 1.4206 0.3486 0.0190
8. 0.002403 0.0378 0.0635 116 0.0292 1.4206 0.3235 0.0205
 S. C M et al. / Renewable Energy 163 (2021) 845e857
References
[1] M.B. Salleh, N.M. Kamaruddin, Z. Mohamed-kassim, Savonius hydrokinetic
turbines for a sustainable river-based energy extraction : a review of the
technology and potential applications in Malaysia, Sustain. Energy Technol.
Assessments. 36 (2019), 100554, https://doi.org/10.1016/j.seta.2019.100554.
[2] A. Kumar, R.P. Saini, Performance parameters of Savonius type hydrokinetic
turbine - a Review, Renew. Sustain. Energy Rev. 64 (2016) 289e310, https://
doi.org/10.1016/j.rser.2016.06.005.
[3] B. Gunawan, Mar (2017), https://doi.org/10.1044/leader.ppl.22032017.18,
2017.
[4] C. Hill, J. Kozarek, F. Sotiropoulos, M. Guala, Hydrodynamics and sediment
transport in a meandering channel with a model axial-flow hydrokinetic
turbine, Water Resour. Res. 52 (2016) 860e879, https://doi.org/10.1002/
2015WR017949.
[5] I. Loots, M. Van Dijk, B. Barta, S.J. Van Vuuren, J.N. Bhagwan, A review of low
head hydropower technologies and applications in a South African context,
Renew. Sustain. Energy Rev. 50 (2015) 1254e1268, https://doi.org/10.1016/
j.rser.2015.05.064.
[6] T.I. Eldho, S.V.P. Vimal Patel, Ganapathi Bhat, Influence of overlap ratio and
aspect ratio on the performance of Savonius hydrokinetic turbine, Int. J. En-
ergy Res. 31 (2016) 135e147, https://doi.org/10.1002/er.
[7] S. Roy, U.K. Saha, Review of experimental investigations into the design,
performance and optimization of the Savonius rotor, Proc. Inst. Mech. Eng.
Part A J. Power Energy. 227 (2013) 528e542, https://doi.org/10.1177/
0957650913480992.
[8] K.H. Wong, W.T. Chong, N.L. Sukiman, S.C. Poh, Y.C. Shiah, C.T. Wang, Per-
formance enhancements on vertical axis wind turbines using flow augmen-
tation systems: a review, Renew. Sustain. Energy Rev. 73 (2017) 904e921,
https://doi.org/10.1016/j.rser.2017.01.160.
[9] J.V. Akwa, H.A. Vielmo, A.P. Petry, A review on the performance of Savonius
wind turbines, Renew. Sustain. Energy Rev. 16 (2012) 3054e3064, https://
doi.org/10.1016/j.rser.2012.02.056.
[10] S. Roy, U.K. Saha, Review on the numerical investigations into the design and
development of Savonius wind rotors, Renew. Sustain. Energy Rev. 24 (2013)
73e83, https://doi.org/10.1016/j.rser.2013.03.060.
[11] N.K. Sarma, A. Biswas, R.D. Misra, Experimental and computational evaluation
of Savonius hydrokinetic turbine for low velocity condition with comparison
to Savonius wind turbine at the same input power, Energy Convers. Manag. 83
(2014) 88e98, https://doi.org/10.1016/j.enconman.2014.03.070.
[12] M.N.I. Khan, M. Tariq Iqbal, M. Hinchey, V. Masek, Performance of savonius
rotor as a water current turbine, J. Ocean Technol. 4 (2009) 71e83.
[13] S.I.A.T.I. Miyoshi Nakajima, Performance of savonius rotor for environmentally
friendly hydraulic turbine, J. Fluid Sci. Technol. 3 (2008) 420e429, https://
doi.org/10.1299/jfst.3.420.
[14] S.I.A.T.I. Miyoshi Nakajima, Performance of double-step savonius rotor for
environmentally friendly hydraulic turbine, J. Fluid Sci. Technol. 3 (2008)
410e419, https://doi.org/10.1299/jfst.3.420.
[15] O. Yaakob, Y. Ahmed, M. Ismail, Validation study for savonius vertical axis
marine current turbine using CFD simulation, 6th Asia-Pacific Work, Mar.
Hydrodyn. (2012), 0e5, http://www.academia.edu/download/29677406/
Validation_Study_for_Savonius_Vertical_Axis_Marine_Current_Turbine_
Using_CFD_Simulation.pdf.
[16] K. Golecha, T.I. Eldho, S.V. Prabhu, Influence of the deflector plate on the
performance of modified Savonius water turbine, Appl. Energy 88 (2011)
3207e3217, https://doi.org/10.1016/j.apenergy.2011.03.025.
[17] G. Kailash, T.I. Eldho, S.V. Prabhu, Performance study of modified savonius
water turbine with two deflector plates, Int. J. Rotating Mach. 2012 (2012)
1e12, https://doi.org/10.1155/2012/679247.
[18] A.H. Birjandi, E.L. Bibeau, V. Chatoorgoon, A. Kumar, Power measurement of
hydrokinetic turbines with free-surface and blockage effect, Ocean. Eng. 69
(2013) 9e17, https://doi.org/10.1016/j.oceaneng.2013.05.023.
[19] K. Sahim, D. Santoso, A. Radentan, Performance of combined water turbine
with semielliptic section of the savonius rotor, Int. J. Rotating Mach. 2013
(2013), https://doi.org/10.1155/2013/985943.
857
[20] K. Sahim, K. Ihtisan, D. Santoso, R. Sipahutar, Experimental study of darrieus-
savonius water turbine with deflector: effect of deflector on the performance,
Int. J. Rotating Mach. 2014 (2014), https://doi.org/10.1155/2014/203108.
[21] K. Golecha, T.I. Eldho, S.V. Prabhu, Study on the interaction between two
hydrokinetic Savonius turbines, Int. J. Rotating Mach. 2012 (2012), https://
doi.org/10.1155/2012/581658.
[22] S.V.P. Vimal Patel, T.I. Eldho, Theoretical study on the prediction of the hy-
drodynamic performance of a Savonius turbine based on stagnation pressure
and impulse momentum principle, Energy Convers. Manag. 168 (2018)
545e563, https://doi.org/10.1016/j.enconman.2018.04.065.
[23] A. Kumar, R.P. Saini, Performance analysis of a Savonius hydrokinetic turbine
having twisted blades, Renew. Energy 113 (2017) 502e522, https://doi.org/
10.1016/j.renene.2017.06.020.
[24] A. Kumar, R.P. Saini, Performance analysis of a single stage modified Savonius
hydrokinetic turbine having twisted blades, Renew. Energy 113 (2017)
461e478, https://doi.org/10.1016/j.renene.2017.06.020.
[25] A.H. Elbatran, Y.M. Ahmed, A.S. Shehata, Performance study of ducted nozzle
Savonius water turbine, comparison with conventional Savonius turbine,
Energy 134 (2017) 566e584, https://doi.org/10.1016/j.energy.2017.06.041.
[26] P.K. Talukdar, A. Sardar, V. Kulkarni, U.K. Saha, Parametric analysis of model
Savonius hydrokinetic turbines through experimental and computational in-
vestigations, Energy Convers. Manag. 158 (2018) 36e49, https://doi.org/
10.1016/j.enconman.2017.12.011.
[27] M. Mosbahi, A. Ayadi, Y. Chouaibi, Z. Driss, T. Tucciarelli, Performance study of
a Helical Savonius hydrokinetic turbine with a new deflector system design,
Energy Convers. Manag. 194 (2019) 55e74, https://doi.org/10.1016/
j.enconman.2019.04.080.
[28] S.V.P. Vimal Patel, T.I. Eldho, Velocity and performance correction method-
ology for hydrokinetic turbines experimented with different geometry of the
channel, Renew. Energy 131 (2019) 1300e1317, https://doi.org/10.1016/
j.renene.2018.08.027.
[29] A. Sivakumar, R. Anandh, A. Anantharaman, M. Suresh, Multi stage power
generation in an open canal system by accelerated flow, J. Informatics Math.
Sci. 10 (2018) 313e320, https://doi.org/10.26713/jims.v10i1-2.1058.
[30] A.J. Alexander, B.P. Holownia, Wind tunnel tests on a savonius rotor, J. Ind.
Aerod. 3 (1978) 343e351, https://doi.org/10.1016/0167-6105(78)90037-5.
[31] R.J. Moffat, Describing the uncertainties in experimental results, Exp. Therm.
Fluid Sci. 1 (1988) 3e17, https://doi.org/10.1016/0894-1777(88)90043-X.
[32] S. Runge, T. Stoesser, E. Morris, M. White, Technology readiness of a vertical-
Axis hydro-kinetic turbine, J. Power Energy Eng. 6 (2018) 63e85, https://
doi.org/10.4236/jpee.2018.68004.
[33] V. Patel, T.I. Eldho, S.V. Prabhu, Performance enhancement of a Darrieus hy-
drokinetic turbine with the blocking of a specific flow region for optimum use
of hydropower, Renew. Energy 135 (2019) 1144e1156, https://doi.org/
10.1016/j.renene.2018.12.074.
[34] V. Patel, T.I. Eldho, S.V. Prabhu, Experimental investigations on Darrieus
straight blade turbine for tidal current application and parametric optimiza-
tion for hydro farm arrangement, Int. J. Mar. Energy. 17 (2017) 110e135,
https://doi.org/10.1016/j.ijome.2017.01.007.
[35] W.M.J. Batten, G.U. Müller, Potential for using the floating body structure to
increase the efficiency of a free stream energy converter, in: 34th IAHR Congr.
2011 - Balanc. Uncertain. Water a Chang. World, Inc. 33rd Hydrol. Water
Resour. Symp. 10th Conf. Hydraul. Water Eng., 2011, pp. 2364e2371.
[36] O. Cleynen, E. Kerikous, S. Hoerner, D. Thevenin, Influence of flotation bodies
on the power characteristics of a free-stream water wheel, WIT Trans. Eng.
Sci. 115 (2017) 55e60, https://doi.org/10.2495/MPF170071.
[37] E. Quaranta, Stream water wheels as renewable energy supply in flowing
water: theoretical considerations, performance assessment and design rec-
ommendations, Energy Sustain. Dev. 45 (2018) 96e109, https://doi.org/
10.1016/j.esd.2018.05.002.
[38] S.M. Hashemi, M. Moghimi, S. Derakhshan, Experimental and numerical study
of a flapping-blade vertical-axis hydrokinetic turbine under free surface
deformation and blockage effects, Int. J. Environ. Sci. Technol. (2020), https://
doi.org/10.1007/s13762-020-02642-y.
[39] G. Saini, R.P. Saini, A computational investigation to analyze the effects of
different rotor parameters on hybrid hydrokinetic turbine performance,
Ocean. Eng. 199 (2020), 107019, https://doi.org/10.1016/
j.oceaneng.2020.107019.