Materials Today: Proceedings xxx (xxxx) xxx

Contents lists available at ScienceDirect

Materials Today: Proceedings

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

Savonius wind turbines: A review of recent advances in design and

performance enhancements
Anupam Dewan a,⇑, Adesh Gautam a, Rahul Goyal b
a
Department of Applied Mechanics, Indian Institute of Technology Delhi, New Delhi 110016, India
b
Centre for Energy Studies, Indian Institute of Technology Delhi, New Delhi 110016, India

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

Article history: The energy crisis due to a rapid globalization and adverse effects of global warming has led to an
Available online xxxx increased need for the non-conventional sources of energy. In the recent times, significant research activ-
ities are carried out in the renewable energy field, mostly in solar and wind power generation. Wind tur-
Keywords: bines are categorised into vertical axis wind turbines (VAWT) and horizontal axis wind turbines (HAWT).
Savonius rotor Savonius rotor (S-type) is a type of VAWT and its function depends on the drag force. It has many advan-
Power coefficient tages of easy installation, simplicity in design, good self-starting capability, low speeds operation and
Tip speed ratio
independency of direction of wind. Although, the negative torque produced on the returning blade leads
Overlap ratio
Augmentation techniques
to its low efficiency. Researchers have conducted several investigations to improve the functioning of
Savonius rotor for satisfying large-scale energy demands. The parameters affecting the performance, such
as, overlap ratio, aspect ratio, tip speed ratio and blade shapes, are improved by undergoing various
designs. Several augmentation techniques, such as, end plates, shielding obstacle, guide box tunnel,
deflecting plate, curtain plate, quarter blades and nozzles, have been used to enhance the performance
of S-type VAWT. With the installation of these augmentation devices, the negative torque acting on
the blades was reduced and the starting ability of the rotor was improved. These devices in turn help
in increasing the power output as well as the torque coefficient of various rotor configurations. It has
been reported in the literature that the coefficient of power of the conventional Savonius rotor lies in
the span of 0.12 to 0.18. By using different methods for optimizing its design or installing the augmen-
tation devices, the coefficient of performance can be improved to a value of 0.52. In this article various
performance influencing parameters and the power augmentation techniques used for S-type VAWT have
been studied and reported studies on these techniques have been reviewed. The article concludes with
proposed directions for future research.
Ó 2021 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the 3rd International Con-
ference on Advances in Mechanical Engineering and Nanotechnology.

1. Introduction non-conventional sources of energy, such as, solar, hydroelectric,

biomass, tidal and wind energy. Significant research activities have
A rapid increase in population has led to a faster depletion of been carried out recently in the fields of solar and wind power, in
natural fuel resources, such as, coal, petroleum and natural gas. order to meet the need of an increasing energy and reduce pollut-
There has been an increased demand for energy due to globaliza- ing emissions. The technology of extracting energy from wind has
tion and a rapid expansion in a wide spectrum of industries. Solar been beneficial in satisfying industrial and domestic needs. Wind
energy has been an important source of renewable energy. The turbines are used to convert wind energy to electricity. They can
conventional energy sources for example coal and petroleum are be categorized into two main groups depending on orientation of
limited in availability and cause huge pollution and severe dam- the axis of rotation, that is, horizontal-axis wind turbines (HAWT)
ages to the environment. Hence there has been a rising need for and vertical-axis wind turbines (VAWT). HAWTs have higher effi-
ciency than vertical axis wind turbines and are mostly utilized
for the production of electricity. However, VAWTs have advan-
⇑ Corresponding author. tages, such as, low cut-in speed ratio, the vertical orientation of
E-mail address: adewan@am.iitd.ac.in (A. Dewan).

https://doi.org/10.1016/j.matpr.2021.05.205
2214-7853/Ó 2021 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the 3rd International Conference on Advances in Mechanical Engineering and Nanotechnology.

Please cite this article as: A. Dewan, A. Gautam and R. Goyal, Savonius wind turbines: A review of recent advances in design and performance enhance-
ments, Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2021.05.205
A. Dewan, A. Gautam and R. Goyal Materials Today: Proceedings xxx (xxxx) xxx

shaft for easy placement of generator, ease in design and low cap- 1.3. Performance of Savonius turbine
ital investment. There is still need of technology development or
identification of some augmentation techniques to increase the The power coefficient C P and torque coefficient C T are the main
efficiency of the VAWTs. The present work provides a perspective parameters which define the performance of a Savonius turbine,
review of these augmentation techniques. both defined at a specific tip speed ratio (TSR) k. Tip speed ratio
is the fraction of speed of blade tip and the wind speed through
the blades and is given as
1.1. Types of vertical axis wind turbines
V Rotor xR
k¼ ¼ ð1Þ
V V
VAWTs are divided into two main categories, namely, Savonius
and Darrieus wind turbines. Savonius rotors can operate in regions The ratio of net power output to the power available at the rotor
of low wind speed and have a better starting ability than Darrieus is defined as the power coefficient which can be given as
rotors. Darrieus wind turbines are lift based rotors, whereas, the P turbine P turbine
drag force acting on the blade is the main reason for operation of CP ¼ ¼ ð2Þ
P av ailable 12 qAV 3
the Savonius rotor which produces low noise levels operating at
lower speeds compared to other wind turbines. where P turbine ¼ T  xs ¼ 2p60NT and P av ailable ¼ 12 qAV 3 .
The ratio of torque developed to the torque available because of
the wind is defined as the torque coefficient which can be given as
1.2. Working principle
T turbine T turbine F r
CT ¼ ¼ ¼ ð3Þ
Savonius rotors are composed of two half-cylinders fixed on the T av ailable 12 qAV 2 R 12 qAV 2 R
opposite sides of a vertical shaft [Fig. 1(a)]. The wind blowing into
the turbine comes in contact with two oppositely facing surfaces, C P and C T can be related using the following equations
one concave and the other convex [Fig. 1(b)]. P turbine T  xs T Rxs
The primary driving mechanism responsible for the rotation of Cp ¼ ¼ ¼  ¼ C T  TSR ð4Þ
P av ailable 1 qAV 3 1 qAV 2 R V
2 2
Savonius type rotors arises from the drag force difference between
the concave and convex sides of the rotor blades [Fig. 1(b)]. Thus A Savonius wind turbine has the limitations that it produces a
the main driving component of the S-type rotors is based on drag lower efficiency compared to other turbines. During the recent
force [1]. The working principle based on the drag force gives rise decades a significant research has taken place with attempts to
to limit that the revolving speed of rotor cannot exceed the speed incorporate suitable design modifications to increase the efficiency
of the wind. This is the primary reason that their tip speed ratio of the Savonius rotor. Various experimental and numerical studies
cannot gain a magnitude more than 1. were conducted to notice the improvement in the efficiency of the

Fig. 1. (a) A typical two-bladed Savonius rotor [1], (b) forces acting on two-bladed Savonius rotor [1] and (c) schematic of conventional Savonius rotor with geometry
specifications [2].

2
A. Dewan, A. Gautam and R. Goyal Materials Today: Proceedings xxx (xxxx) xxx

rotor due to the changes in the geometries. The important param- that optimized number of blades were independent of the number
eters affecting the rotor performance, such as, aspect ratio (AR), of stages. However, a two-stage Savonius rotor with twisted blades
overlap ratio (OR), number of blades, staging and modification in produced a better power coefficient of 0.31 compared to the case of
blade shapes were optimized in different studies and these are semi-circular one. Ali [11] also studied the number of blades of
summarized in the following sections. Savonius rotor and concluded that under the same test conditions
a two-bladed Savonius rotor was more productive than the three
2. Parameters affecting the performance of Savonius type wind blades.
turbine
2.4. Blade shape
The crucial parameters such as aspect ratio, overlap ratio, num-
ber of blades (staging) and shape of the blades are modified to Alom et al. [12] used numerical and experimental techniques to
improve in the efficiency of the turbine in the various studies. investigate the influence of varying the blade profiles on the per-
formance coefficients of Savonius wind turbines. Fig. 2 shows dif-
2.1. Aspect ratio ferent types of blade profiles of semi-circular, elliptical, Benesh
and modified Bach.
The aspect ratio can be defined as the ratio of the height of the 2-D unsteady computations were conducted using Shear-Stress
rotor H to the diameter D as AR = H/D, which is shown in Fig. 1(c). It Transport (SST) turbulence model in the ANSYS Fluent. C p;max
strongly affects the aerodynamic performance of the Savonius obtained using numerical method were 0.272, 0.294, 0.304, 0.34
wind turbine. Mahmoud et al. [3] reviewed various configurations and using experimental method (wind tunnel test) were 0.158,
at different aspect ratios (a) as 0.5, 1.0, 2.0, 4.0 and 5.0. It was 0.159, 0.162, 0.19, for semi-circular, Benesh, modified Bach and
observed that on increasing the aspect ratio the power coefficient elliptical profiles, respectively. Lee et al. [13] examined the perfor-
also increased, keeping other parameters as constants. The obser- mance of helical-shaped Savonius wind turbines by varying helical
vations of Kamoji [4] depict that a rotor with an aspect ratio of angles of blades from 0° to 135°. The power coefficient C p attained
0.7 produced a maximum power coefficient as 0.21. A good perfor- its maximum value close to 0.13 at a blade twist angle of 45° [Fig. 3
mance [5] is recorded by keeping AR within the range 1.5 – 2.0; (a)]. There was a decrease in C p;max by 25.5% at twist angle of 90°
however, for most of the existing rotors the aspect ratio was kept and 135°. On varying the azimuthal twist angle, the torque coeffi-
at lower values (close to 1.0) due to the structural reasons. cient gained its peak of 0.34 at a blade twist angle of 45° and TSR of
0.4.
2.2. Overlap ratio (OR) and phase shift Laws et al. [14] investigated three designs (one standard and
two modified) to compare the power and torque generated with
The overlap ratio can be defined as the ratio of distance of over- respect to the conventional turbine. Models A and B with thicker
lap of the two blades (e) to the chord length of the blades (d) mid-sections [Fig. 3(b)] had slightly flattened shape with more
(OR = e/d), as indicated in Fig. 1(c). The amount of the wind passing sharpened blade edge than conventional ones. The overset mesh
through the blades gap increases with an increase in the overlap of framework was used and the governing flow equations which were
the blades. Hence this leads to wind flowing through the concave coupled with the rotational equations of the turbine were solved
side of the returning blade which produces a strong thrust. Tania numerically. OpenFOAM solver was used for the numerical compu-
et al. [6] studied the optimum overlap ratio of semi-cylindrical tations in which Reynolds-averaged Navier–Stokes (RANS) equa-
blades of Savonius wind turbine by considering four S-type rotors tions were coupled with the SST model. Model-A showed a
having overlap ratio equal to 0.15, 0.20, 0.25 and 0.30. The suitable higher Cp than other blade shapes owing to its reduced surface area
overlap ratio was found to be 0.15 for wind speeds smaller than at the blade tip and less pressure difference on both sides of the
4.0 m/s, whereas it was 0.30 for wind speed exceeding 4.0 m/s. blade. Model-A and Model-B predicted a relative improvement in
The variation in performance on account of overlap ratio was pri- Cp by 28.12% and 10.53%, respectively, compared to the conven-
marily due to the turbulent character of wind at high speeds. Alom tional blades.
et al. [7] studied numerically the optimum OR of the elliptical blade
rotor considering the sectional cut angle of 47.5° for the elliptical 3. Augmentation devices for Savonius wind turbine
profile. Various elliptical profiles with overlap ratios ranging from
0.0 to 0.30 were studied by performing 2-D unsteady simulations To divert the wind flow towards the advancing blade of the
around the rotor. The unsteady simulations showed a maximum rotor augmentation techniques are important. This acts like a
C p of 0.34 for OR = 0.15. Hassanzadeh et al. [8] studied the effect
of outward overlap of blades on the performance of two blade
Savonius rotor by varying OR within the range ± 0.3. The worst case
of overlap occurred for e/d = 0.3. The static torque of a two-stage
rotor was influenced by phase shift [9] and with an increase in the
phase shift angle the curve of static torque coefficient C ts became
smoother. This helped in eliminating the negative torque acting
on the blade and also enhanced the self-starting characteristics
of the Savonius wind turbine.

2.3. Number of blades

With the changing working conditions, the efficiency of a Savo-

nius rotor also gets affected by the number of rotor blades. Gener-
ally, two, three and four bladed rotors have been studied in the
literature. Saha et al. [10] conducted experiments by using single,
two and three stages of the Savonius turbine rotors and concluded Fig. 2. Types of several blade profiles [12].

3
A. Dewan, A. Gautam and R. Goyal Materials Today: Proceedings xxx (xxxx) xxx

Fig. 3. (a) Top and side views of wind rotor shapes with twist angle u = 0° and u = 45° [13] and (b) model A and model B [14].

shielding effect of the returning blade from the incoming wind.

Hence the negative torque gets reduced which leads to more
power of the Savonius rotor. Different flow augmentation tech-
niques like endplates, guide vanes, shielding obstacles, deflectors,
curtain plates and vent augmenters are discussed in the following
section.

3.1. Endplates

With the use of endplates, a significant increment in the coeffi-

cient of power of a Savonius rotor may be achieved. Endplates pre-
vent air to escape from the tips of the concave side of advancing
and returning blades to the atmosphere. This leads to high-
pressure variation between the two sides of the blades which leads
to increment in the extraction of energy from the wind. Alom et al.
Fig. 4. (a) Circular end plate with size ratio 1.1 [16] and (b) guide blades
[2] conducted experiments using end plates in an elliptical bladed arrangement [18].
Savonius rotor. On varying the tip speed ratios, the rotor with end
plates produced a C p;max of 0.232 at a TSR of 0.8. Hence on using end
plates without shaft C p;max increased by 26.31% compared to the Using blades with a twist angle of 180°, C p;max increased by 36%
case with the shaft. The use of end plates in improved geometry compared to the case with no end plates. The Savonius rotor with
of elliptical profile rotor was studied by Sanusi et al. [15]. The pro- circular end plates leads to maximum value of coefficients of
posed rotor design was a combination of semi-circular profile power and torque because of increment in the aerodynamic power
blade with an elliptical profile on the concave side. The combined with the end plate area.
blade increased C p;max to 0.27 for a TSR of 0.8. The use of elliptical
profile on the concave side moved the energy catching point away 3.2. Guide vanes
from the centre of rotation, which increased the net positive torque
acting on the rotor. In a study by Saad et al. [16] the effect of vary- Guide vanes reduce air flow to the returning blade by directing
ing the end plates size ratio was observed for a twisted Savonius the incoming wind towards the advancing blade. This results in a
rotor. A 3-D, incompressible and unsteady RANS model was decrease of the negative torque acting on the Savonius rotor. The
numerically simulated using SST turbulence model on varying guide blade mechanism produces a funneling effect, accelerating
the twist angle, overlap ratio and the end plates size ratio of the the wind in the upstream of the rotor. Kalluvila and Sreejith [18]
Savonius rotor. The rotor design with a twist angle of 45°, zero examined the impact of guide blade arrangement on changed rotor
overlap of blades, and end plates size ratio of Do/D = 1.1 (where geometry [Fig. 4(b)]. The rotor blade profile was an S-shaped curve
Do and D are the diameters of the lower and upper circular end followed by a linear section. The alternate guide blades were
plates, respectively) [Fig. 4(a)] produced the highest net output inclined at 6.7° and 20° to the upstream direction, respectively.
power among all the tested geometries. The flow leaving the Using the numerical computations, the maximum power coeffi-
advancing blade hit the concave side of the returning blade with cient of 0.31 was achieved at a TSR of 0.8. However, on performing
a high momentum, resulting in an increase of net positive torque experiments in open wind tunnel produced a C p;max of 0.28 at a TSR
and power output. The static torque coefficient of the suggested of 0.85. Manganhar et al. [19] optimized the guiding rotor house
design attained positive values throughout the rotation of the (RH) design comprising of four equidistant vertical walls tilted at
rotor, predicting better self-starting ability compared to other rotor 45⁰ clockwise with each reference axis. The venturi-effect pro-
designs studied in the literature. The shapes and sizes of the end duced due to the rotor house accelerated the wind velocity to
plates also has significant impact on the aerodynamic performance nearly 1.5 times towards the rotor zone, increasing the power coef-
as observed in the case of the helical Savonius wind rotor [17]. ficient from 0.125 to 0.218.
4
A. Dewan, A. Gautam and R. Goyal Materials Today: Proceedings xxx (xxxx) xxx

3.3. Shielding obstacles plates. It was noticed that the Savonius rotor with curved shaped
deflectors [Fig. 5(b)] showed the best performance. The third
To reduce the drag force acting on the returning blades obsta- design showed a C p;max of 0.52 for an operating range of k  2.2.
cles are generally placed to shield it from the wind. Hence the neg- The rear deflector [25] used in a Savonius turbine was able to accli-
ative torque is decreased and this results in improving the net matize itself with respect to the changing direction of the wind.
power output of the Savonius rotor. The optimum angle of the The power coefficient was least affected at low length deflector
obstacle placed in the upstream was studied by Putri et al. [20]. kept farther from the centre of rotation. Hence, the optimal posi-
The aerodynamic capability of the turbine was observed by chang- tion of the rear deflector was found to be at S1 =D ¼ 0:82, S1 being
ing the orientations of the upstream obstacle at speeds of wind the distance of the deflector from the centre of rotation of the rotor.
0.48 m/s and 7.45 m/s corresponding to Reynolds numbers of Nimvari et al. [26] proposed a porous deflector to overcome the
30000 and 90000, respectively. The torque acting on the turbine problem of complex wake zone creation behind a conventional
was firmly affected by the obstacle opening angle and the corre- solid deflector and enhance the functioning of the Savonius rotor.
sponding maximum value was observed at h = 40° independent A computational investigation was conducted by utilizing the
of Re. The power generated attained its peak value at a slightly unsteady Reynolds-averaged Navier-Stokes equations using SST
greater angle h = 45°, concluding that the power output and model. It was observed that the porous deflector with deflector
dynamic torque acting on the shaft may attain peak values at dif- porosity = 0.9, deflector angle = 90°, porous deflector height ratio
ferent points. The optimum position of the shielding obstacle can (h/D) = 0.257 and a distance ratio (L/D) = 0.99 produced the maxi-
also be achieved by applying suitable optimization techniques like mum value of the average C P of 0.274. The presence of the porous
Evolutionary Algorithms [21]. At the obstacle position of deflector enhanced the performance since it resulted in a more reg-
b = 100.83° for a two blade Savonius rotor, the C p;max obtained ular flow of wind due to elimination of wake zone formed behind a
was 0.2503 corresponding to a tip speed ratio of 0.8 (a relative solid deflector.
increment of 27.3% in the output power coefficient). This optimum
position of the obstacle plate was used to study the blade shape for
3.5. Curtain plates
increasing the power generation of the Savonius rotor [22]. The
mathematical optimization (Evolutionary Algorithms) was
The aerodynamic functioning of a Savonius wind rotor is
deployed to arrive at the blade shape which was optimal. This opti-
improved by using cutain plates. Tartuferi et al. [27] examined
mization technique considered the output power coefficient as the
the application of a new curtain system in Savonius rotor [Fig. 6
target function by defining three intermediate points of the blade
(a)], with the ability of self-orienting itself with the direction of
using six degrees of freedom (Xp1, Yp1, Xp2, Yp2, Xp3, Yp3). Based
the wind flow. It comprised different aerodynamic appendages
on the study, the optimized blade shape with an obstacle resulted
coupled together in a manner such that the wind flow did not pro-
in an overall increment in the output power coefficient by almost
duce any resultant aerodynamic torque on system. Apart from
38.9% relative to a standard Savonius turbine at k = 0.7.
maximizing the extraction of energy from the incoming wind,
the curtain system also resulted in decrement of the drag acting
on the returning blade. This led to a decrease in the magnitude
3.4. Deflecting plates
of the negative torque acting on the returning blade of the rotor.
The maximum power coefficient obtained both experimentally
The deflector plate prevents the flow of wind to the returning
and numerically was 0.301 and 0.274 for k of 0.751 and 0.718,
blade and hence the wind resistance gets reduced. This helps in
respectively. Hence by using the new curtain system the increase
enhancing the performance of a Savonius rotor. Roy et al. [23] used
in net positive torque enhanced the power output by approxi-
an oriented jet deflector [Fig. 5(a)] to carry out experiments in a
mately 20% than that of the case without aerodynamic appendages.
wind tunnel at a low speed of wind that is 6.2 m/s. Using an ori-
ented jet, the maximum Cp of 0.41 was observed at a = 40° and
b = 10° and k = 0.87. For the case of a conventional Savonius wind 3.6. Vent augmenters
turbine without implementing any deflector, the C p;max was 0.23 at
TSR = 0.71. El-Askary et al. [24] studied new rotor designs in which Alom and Saha [28] examined the function enhancement of a
a wind jet was generated on the concave side of the advancing Savonius rotor by using vent augmenters. Three different designs
blade only. The simulations were performed by using SST turbu- were investigated by varying the position of vents on the blade
lence model. Three suggested designs with an overlap ratio of profiles, i.e., design I with vents on blade centre line, design II with
0.15, were prepared by varying the positions of three deflecting vents at 30° above and 30° below the blade centre line [Fig. 6(b)]

Fig. 5. (a) Oriented jet deflector [23] and (b) geometry of the deflector plates design with smooth turning [24].

5
A. Dewan, A. Gautam and R. Goyal Materials Today: Proceedings xxx (xxxx) xxx

performing experiments in wind tunnel and applying blockage cor-

rection, the C p;max for the corresponding designs were obtained as
0.146, 0.134 and 0.112, respectively for AR = 0.7 and TSR = 0.49.
Using 2-D unsteady simulations C p;max values obtained were
0.341, 0.33 and 0.27, respectively, at a TSR = 0.8. From 3-D
unsteady simulation at speed of wind 6.2 m/s, the vented elliptical
rotor give rise to the highest C P;max of 0.132 for a TSR of 0.8.

3.7. Miniature blades and quarter blades

Miniature blades and quarter blades when placed concentric

within the rotor blades [Fig. 7(a)] are useful in increasing the effi-
ciency of SSWT. An increase in surface area of a Savonius rotor
leads to better utilization of kinetic energy of the incoming wind.
Sharma and Sharma [30] investigated the effect of adding concen-
Fig. 6. (a) New conveyor deflector curtain system [27] and (b) vents in semicircular
blades [28]. tric multiple miniature blades [Fig. 7(a)] inside the rotor blades. 3-
D transient CFD computations were conducted using Ansys CFX 13.
An increase of around 11.34% was observed in the maximum COP
for a velocity of 9.21 m/s. Similarly multiple quarter blades were
added concentric to a conventional Savonius rotor to improve its
performance [Fig. 7(b)][31].On using multiple quarter blades, there
was an increment varying from 8.89% to 13.69% in the coefficient
of performance compared to the conventional semi-circular config-
uration. The unutilized portion of wind kinetic energy could be
captured efficiently by using an additional set of quarter blades.
Being simple and small in size, miniature and quarter blades do
not increase the weight and complexity of a Savonius rotor com-
pared to other augmentation techniques.

Fig. 7. (a) Multiple miniature blades [30] and (b) multiple quarter blades [31]. 3.8. Recommendations on augmentation devices

Different augmentation devices discussed in the previous sec-

and design III with vents at 30° below and 30° above the blade cen- tion are end plates, guide vanes, shielding obstacle, deflector
tre line. 2-D unsteady simulation was performed around the Savo- plates, curtain plates, vent augmenters, miniature blades and mul-
nius rotor by using the SST model in ANSYS Fluent. Design-II tiple quarter blades. The best technique is chosen to be the one that
resulted in a maximum Cp of 0.292 compared to 0.27 for the case results in the maximum coefficient of power. The values of the per-
of the conventional semi-circular bladed profile without vent. formance coefficient for various augmentations techniques consid-
The total pressure and velocity at the returning blade was found ered in the present study are presented in Table 1. Hence
to be smaller compared to those for other cases and the maximum considering the maximum power output, the deflecting plates with
velocity occurred at the advancing blade in design II. In another curved passage with a C P;max of 0.52 is best suited among all the
case, Alom and Saha [29] studied the function enhancement of rotor designs. However, we should also take into consideration
elliptical blades using venting slots. 2-D and 3-D unsteady simula- the dynamic and static torque coefficients. A high positive value
tions using SST k-x turbulence model were performed around the of static torque coefficient leads to a good starting ability of a Savo-
vented elliptical profile Savonius rotor. Three different designs of nius rotor. Based on these considerations a rotor with twisted
vented elliptical profile with cut angle h = 47.5°, non-vented ellip- blades gives good results. Hence a Savonius rotor with twisted
tical and semi-circular profiles were considered for simulations. On blades can be further modified using augmentation techniques,

Table 1
Performance coefficients of various augmentation techniques reported in the literature.

Author Year Flow deflector MaximumC P

Numerical Experimental
Alom et al. [2] 2019 End plates in elliptical – 0.232
Sanusi et al. [15] 2016 End plates in combined semi-circular& elliptical – 0.27
Saad et al. [16] 2020 End plates in twisted blades 0.231 –
Kalluvila and Sreejith [18] 2018 Guide vanes 0.31 0.28
Manganhar et al. [19] 2019 Guiding rotor house – 0.218
Mohamed et al. [21] 2010 Obstacle shielding 0.2503 –
Mohamed et al. [22] 2011 Obstacle shielding 0.298 –
Roy et al. [23] 2014 Flat plate deflectors – 0.41
El-Askary et al. [24] 2015 Curved shaped deflectors 0.52 –
Tartuferi et al. [27] 2015 Curtain plates 0.274 0.301
Alom and Saha [28] 2016 Vent augmenters in conventional blade 0.292 –
Alom and Saha [29] 2018 Vent augmenters in elliptical blades 0.132 0.146
Sharma and Sharma [30] 2017 Miniature blades 0.226 –
Sharma and Sharma [31] 2016 Multiple quarter blades 0.2266 –

6
A. Dewan, A. Gautam and R. Goyal Materials Today: Proceedings xxx (xxxx) xxx

such as, guide vanes, deflectors, vent augmenters, etc. to improve References
its power output.
[1] S. Roy, U.K. Saha, Review of experimental investigations into the design,
performance and optimization of the Savonius rotor, J. Power and Energy 227
4. Conclusions
(4) (2013) 528–542.
[2] N. Alom, N. Kumar, U. Saha K., Analyzing the effect of shaft and end-plates of a
The Savonius rotor is a drag based vertical axis wind turbine newly developed elliptical-bladed Savonius rotor from wind tunnel tests, in:
having simple design, easy establishment, better self-starting abil- Proc. of 38th ASME Int Conf. on Ocean, Offshore and Arctic Engineering, ASME,
2019, pp. 1–8.
ity at speeds which are low compared to other wind turbines. [3] N.H. Mahmoud, A.A. El-Haroun, E. Wahba, M.H. Nasef, An experimental study
However, the negative static torque is produced on the returning on improvement of Savonius rotor performance, Alex. Eng. J. 51 (1) (2012) 19–
blade and it results in its lower efficiency compared to other verti- 25.
[4] M.A. Kamoji, S.B. Kedare, S.V. Prabhu, Experimental investigations on single
cal axis wind turbines. Over the last few decades, extensive stage modified Savonius rotor, Appl. Energy 86 (7-8) (2009) 1064–1073.
research activities have been reported to optimize its geometric [5] R. Pudur and S. Gao, Performance analysis of Savonius rotor on different aspect
configuration and augmentation techniques were utilized to have ratio for hydropower generation, Proc. First Conf. on Power, Dielectric and
Energy Management at NERIST2015.
an increment in power output and efficiency. The objective of the [6] R. Tania, R.L. Florin, I.D. Adriana, M. Roxana, A. Ancuta, Experimental
present paper wasto inspect the effect of various affecting variables investigation on the influence of overlap ratio on Savonius turbines
as well as number of augmentation techniques utilized in a Savo- performance, Int. J. Renewable Energy Res. 8 (2018) 1791–1799.
[7] A. Nur, U. Saha K., Arriving at the optimum overlap ratio for an elliptical-
nius rotor. The recent advances in the augmentation devices to
bladed Savonius rotor, in: Proc. of ASME Turbomachinery Technical Conf. and
improve the functioning of the conventional Savonius rotor are Exposition GT2017, ASME, 2017, pp. 1–10.
also discussed in the present study. The power augmentation [8] R. Hassanzadeh, M. Mohammadnejad, Effects of inward and outward overlap
ratios on the two-blade Savonius type of vertical axis wind turbine
devices help in overcoming the problem of negative torque gener-
performance, Int. J. Green Energy 16 (15) (2019) 1485–1496.
ation by controlling the wind direction which prevents air from [9] C. Jian and J. Kumbernuss, Influence of phase-shift and overlap ratio on
hitting the convex side of the returning blade. Based on the results Savonius Wind Turbine’s Performance, J. of Solar Energy Engineering 134
discussed in the present study, the power coefficient of a Savonius (2012) 1-9.
[10] U.K. Saha, S. Thotla, D. Maity, Optimum design configuration of Savonius rotor
rotor using deflecting plates with curved passage can obtain a through wind tunnel experiments, J. Wind Eng. Indust. Aerodyn. 96 (8-9)
rather high value of 0.52. However, the use of this power augmen- (2008) 1359–1375.
tation device makes the design more complex and increases the [11] M.H. Ali, Experimental comparison study for savonius wind turbine of two &
three blades at low wind speed, Int. J. of Modern Eng. Res. (IJMER) 3 (2013)
dependency on wind flow direction. In recent times, the blade 2978–2986.
shape has been an important aspect of research on a Savonius [12] N. Alom, U.K. Saha, Influence of blade profiles on Savonius rotor performance:
rotor. The parameters, such as, aspect ratio, overlap ratio, twisting Numerical simulation and experimental validation, Energy Convers. Manage.
186 (2019) 267–277.
angle of blades, should be investigated to further enhance the per- [13] J.-H. Lee, Y.-T. Lee, H.-C. Lim, Effect of twist angle on the performance of
formance. The multi-staging with other VAWTs, such as, lift-based Savonius wind turbine, Renewable Energy 89 (2016) 231–244.
Darrius wind turbine has also been an important topic of research. [14] P. Laws, J.S. Saini, A. Kumar, S. Mitra, Improvement in Savonius wind turbine
efficiency by modification of blade designs -anumerical study, J. Energy
The augmentation techniques should be capable of increasing the
Resour. Technol. 142 (2020) 1–12.
power output as well as the self-starting characteristics of a Savo- [15] A. Sanusi, S. Soeparman, S. Wahyudi, L. Yuliati, Experimental study of
nius rotor. There is also a need to have a smooth value of the coef- combined blade Savonius wind turbine, Int. J. Renewable Energy Res. 6
(2016) 614–619.
ficient of torque during the rotation to reduce the mechanical
[16] A.S. Saad, I.I. El-Sharkawy, S. Ookawara, M. Ahmed, Performance enhancement
vibrations. Further, the use of augmentation device should not of twisted-bladed Savonius vertical axis wind turbines, Energy Convers.
increase the weight or design complexity of a Savonius wind tur- Manage. 209 (2020) 112673, https://doi.org/10.1016/j.
bine. From the present study we also conclude that most numerical enconman.2020.112673.
[17] K.S. Jeon, J.I. Jeong, J.-K. Pan, K.-W. Ryu, Effects of end plates with various
simulations have been investigated by utilizing RANS equations shapes and sizes on helical Savonius wind turbines, Renewable Energy 79
using the SST model and recently results obtained were compared (2015) 167–176.
utilizing the k-e turbulence model. The simulation environment [18] J.B.S. Kalluvila, B. Sreejith, Numerical and experimental study on a modified
Savonius rotor with guide blades’, Int. of Green Energy 15 (2018) 744–757.
was mostly performed in ANSYS Fluent. However, different [19] A.L. Manganhar, A.H. Rajpar, M.R. Luhur, S.R. Samo, M. Manganhar,
advanced models, such as, Reynolds-Stress Transport Model [32] Performance analysis of a savonius vertical axis wind turbine integrated
or advanced modelling techniques, such as, Partially-Averaged with wind accelerating and guiding rotor house, Renewable Energy 136 (2019)
512–520.
Navier-Stokes (PANS) [33,34] and large eddy simulation (LES) [20] N.P. Putri, T. Yuwono, J. Rustam, P. Purwanto, G. Bangga, Experimental studies
[35] and advanced RANS turbulence modelling [36] approaches, on the effect of obstacle upstream of a Savonius wind turbine, SN Appl. Sci. 1
can also be used. Further, numerical simulations can also be con- (2019) 1–7.
[21] M.H. Mohamed, G. Janiga, E. Pap, D. Thévenin, Optimization of Savonius
ducted using FVM based open source software OpenFOAM [14],
turbines using an obstacle shielding the returning blade, Renewable Energy 35
[33], [34] and [35]. Hence this review has been able to present (2010) 2618–2626.
the recent advances in the augmentation techniques and formed [22] M.H. Mohamed, G. Janiga, E. Pap, D. Thévenin, Optimal blade shape of a
modified Savonius turbine using an obstacle shielding the returning blade,
an effective stage for the research in future in enhancing perfor-
Energy Convers. Manage. 52 (2011) 236–242.
mance of a Savonius rotor. [23] S. Roy, P. Mukherjee, U.K. Saha, Aerodynamic performance evaluation of a
novel Savonius-style wind turbine under an oriented jet, in: Proc. of ASME
CRediT authorship contribution statement 2014 Gas Turbine India Conf., 2014, pp. 1–7.
[24] W.A. El-askary, M.H. Nasef, A.A. Abdel-hamid, H.E. Gad, Harvesting wind
energy for improving performance of Savonius rotor, J. Wind Eng. Indust.
Anupam Dewan: Conceptualization, Methodology, Investiga- Aerodyn. 139 (2015) 8–15.
tion, Writing - Reviewing and Editing, Supervision. Adesh Gautam: [25] F. Guo, B. Song, Z. Mao, W. Tian, Experimental and numerical validation of the
influence on Savonius turbine caused by rear deflector, Energy 196 (2020)
Methodology, Investigation, Writing - Reviewing and Editing.
117–132.
Rahul Goyal: Writing - Reviewing and Editing. [26] M.E. Nimvari, H. Fatahian, E. Fatahian, Performance improvement of a
Savonius vertical axis wind turbine using a porous deflector, Energy
Convers. Manage. 220 (2020) 113062.
Declaration of Competing Interest [27] M. Tartuferi, V.D. Alessandro, S. Montelpare, R. Ricci, Enhancement of Savonius
wind rotor aerodynamic performance : a computational study of new blade
The authors declare that they have no known competing finan- shapes and curtain systems, Energy 79 (2015) 371–384.
[28] N. Alom, U.K. Saha, Numerical optimization of semicircular-bladed Savonius
cial interests or personal relationships that could have appeared rotor using vent augmenters, Proc. Asian Congress Gas Turbines (2016) 1–9.
to influence the work reported in this paper.
7
A. Dewan, A. Gautam and R. Goyal Materials Today: Proceedings xxx (xxxx) xxx

[29] N. Alom, U.K. Saha, Performance evaluation of vent-augmented elliptical- two different turbulence models, Int. J. Heat Mass Transfer 49 (21-22) (2006)
bladed savonius rotors by numerical simulation and wind tunnel experiments, 3914–3928.
Energy 152 (2018) 277–290. [33] P. Ranjan, A. Dewan, Partially Averaged Navier Stokes simulation of turbulent
[30] S. Sharma, R.K. Sharma, CFD investigation to quantify the effect of layered heat transfer from a square cylinder, Int. J. Heat Mass Transfer 89 (2015) 251–
multiple miniature blades on the performance of Savonius rotor, Energy 266.
Convers. Manage. 144 (2017) 275–285. [34] P. Ranjan, A. Dewan, A PANS study of fluid flow and heat transfer from a square
[31] S. Sharma, R.K. Sharma, Performance improvement of Savonius rotor using cylinder approaching a plane wall, Int. J. Thermal Sci. 120 (2017) 321–336.
multiple quarter blades – A CFD investigation, Energy Convers. Manage. 127 [35] R. Kumar, A. Dewan, A study of LES-SGS closure models applied to a square
(2016) 43–54. buoyant cavity, Int. J. Heat Mass Transfer 98 (2016) (2016) 164–175.
[32] M. Pathak, A. Dewan, A.K. Dass, Computational prediction of a slightly heated [36] M. Pathak, A. Dewan, A.K. Dass, Effect of streamline curvature on flow field of a
turbulent rectangular jet discharged into a narrow channel crossflow using turbulent plane jetin cross flow, Mechanics Res. Com. 34 (3) (2007) 241–248.