© 2018 IJSRSET | Volume 4 | Issue 1 | Print ISSN: 2395-1990 | Online ISSN : 2394-4099

Themed Section : Engineering and Technology

Review on Performance Evaluation of Helical Savonius Wind

Turbine
S. H. Chaklasiya, R. R. Saxena, Dr. P. V. Ramana
Mechanical Department, Sardar Vallabhbhai Patel Institute of Technology, Vasad, Gujarat, India

ABSTRACT

Large Scale Wind Turbines (LSWTs) have been extensively examined for decades but very few studies have
been conducted on the small-scale wind turbines (SSWTs) especially for the applications where artificial wind
speed is of order of few meters per second. Energy can be harvested with the help of helical savonius turbine
where air is exhausted to atmosphere at high velocity. Such wind turbine is more efficient and economical
compare to conventional which are based upon natural wind energy.
Keywords: Helical Savonius turbine, Vertical Axis Wind Turbine, Wind Energy

I. INTRODUCTION Selection of wind turbine is highly depending on

compatibility with available wind source. HAWT is
Carbon dioxide plays important role in greenhouse based upon lift force, design of this type of blade is
effect (GHGs) that is a global concern about reducing based upon air foil. HAWT is widely used on
carbon emissions. In this regard, to harvest energy commercial level as they found more economical
form exit duct which exhaust air at high velocity can solution compare to VAWT on commercial level.
be used to generate electricity with the help of While, VAWT is drag type wind turbine, such
generator. turbine is widely used where high velocity wind is
available onsite during whole year. VAWT is very
History of wind power starts from Persia (present-day efficient in turbulence wind as well, while HAWT
Iran) about 500–900 AD ago. Finnish engineer Sigurd needs yaw system to maintain its performance.
Johannes has invented Savonius turbine in 1922. Turbine is also classified based on power rating which
However, Europeans had carried out experiment with are as follows:
curved blades on vertical axis wind turbine(VAWT). 1. Non-Commercial
Two patent US1697574 and US1766765 have been 1.1. Micro :- This type of wind turbine can produce
registered earlier on 1925 and 1928 respectively. electricity from 4 W to 250 W.
1.2. Mini :- This type of wind turbine can produce
II. WIND TURBINE electricity from 250 W to 1400 W.
1.3. Household :- This type of wind turbine can
Wind Turbine is a device which converts kinetic produce electricity from 1400 W to 16000 W.
energy of wind into mechanical energy which 2. Commercial
eventually into electric energy. There are two type of 2.1. Small Commercial :- This type of wind turbine
wind turbine based upon their axis configuration. can produce electricity from 25 KW to 100 KW.
1. Horizontal Axis Wind Turbine (HAWT) 2.2. Medium Commercial :- This type of wind turbine
2. Vertical Axis Wind Turbine (VAWT) can produce electricity from 100 KW to 1000 KW.

IJSRSET1841305 | Received : 21 Feb 2018 | Accepted : 28 Feb 2018 | January-February-2018 [(4) 1 : 1464-1471] 1464
-
 2.3. Large Commercial :- This type of wind turbine
can produce electricity from 1000 KW to 3000 KW.

III. SAVONIUS TURBINE

The Savonius turbine is one of the simplest turbines.

It is a drag type turbine, consisting of two or three
scoops. Its looks like “S” shape when viewed from top
view. If it is swept over helix profile, it’s called helical Figure 1. Comparison of the average coefficients of
savonius turbine. power obtained at different TSRs [2].

The effect of following parameter among the main Whereas J.L. Menet et al. [4] has found in
parameters affect the performance of a savonius wind experimental analysis that rotor having overlap ratio
turbine. of r/6, where r is radius of the rotor, having zero space
A. Tip Speed Ratio (TSR) between two opposite buckets have maximum
B. No. of blade (n) coefficient of power is found 0.3 at TSR = 1, whereas
C. Aspect ratio (a) maximum torque coefficient is found 0.425 at TSR =
D. Helix angle (α) 0.4. M.A. Kamoji et al. [5] has experimentally found
E. Overlap ratio (e) that, The Tip Speed Ratio (TSR) for maximum power
F. No. of stage (N) coefficient is maximum for helical Savonius rotor
G. With or without endplate having 90º helix angle and aspect ratio= 0.88 and
H. Effect of Blade Curve Profile overlap ratio = 0, then that of the conventional
I. Effect of with and without shaft Savonius rotor.
J. Reynold’s No.
K. Turbulence model B. Effect of No. of Blade (n)
Zhenzhou Zhao et al. [6] found in result of numerical
A. Effect of Tip Speed Ratio (TSR) analysis that power coefficient is 0.165 and 0.12 for
Tip speed ratio (TSR) is defined as ratio of rotor tip two and three blade rotors respectively at TSR = 0.8,
speed (u) to the free wind speed (v). which is higher compare to conventional rotor in case
of two blade rotor.
Jae-Hoon Lee et al. [1] had concluded in numerical
analysis CT has highest value of 0.34 at a TSR of 0.45
for a twist angle of 45º, fluctuation of CT decreases
with increase in twist angle of 135º. In experimental
result it has been found that CT have negative value in
range of 60º -150º and 240º - 330º for a twist angle of
0º, 45º and 135º with a TSR of 0.88. Konrad Kacprzak
et al. [2] has carried out numerical investigation and Figure 2. Air flow distribution of two and three blade
found that the highest power coefficients are obtained [6]
at TSR ≈ 0.81 which reduces with further increment
in Tip Speed Ratio as shown in figure 2. In U.K. Saha et al. [7] has carried experimental analysis
computational model G. Ferrari et al. [3] concluded considering two and three bladed rotors in which
savonius rotor exhibits higher torque which values of they found that coefficient of power is 0.32 and 0.28
0.5 Nm at lower TSR of 0.6. respectively. From fig. 2 it can be visualized that air
International Journal of Scientific Research in Science, Engineering and Technology (ijsrset.com)
1465
 which strikes on a blade get reflected back to
upcoming blade which create negative torque thus
with increase in rotor diameter coefficient of power is
decreased. Frederikus Wenehenubun et al. [8] has
concluded that three blade savonius rotor produces
higher rotational speed compared to two and four
blades from their experimental study and simulation.
They had also found that four blade rotor performs
better at lower tip speed ratio while three blade rotor
are best performing for higher tip speed ratio.
Figure 3. Variation of coefficient of power at aspect
Mohammed Ali et al. [9] also conclude in
ratio of (0.88, 0.93 & 1.17) [5].
experimental analysis that the two blades Savonius
wind turbine is more efficient, it has higher power
M.A. Kamoji et al. [5] found in experiment that, effect
coefficient under the same test condition than that of
of aspect ratio on the coefficient of power for a helical
three blades Savonius wind turbine.
Savonius rotor with a 90º twist at a Reynolds number
of 120,000 shown in fig. 4. Performance of helical
C. Effect of Aspect Ratio (a)
Savonius rotor with an aspect ratio of 0.88 is
Aspect ratio is given by ratio of height to width of the
marginally higher (Cpmax = 0.165 at a TSR = 0.7)
turbine.
compared to the helical rotor with a rotor aspect ratio
of 0.93 (Cpmax = 0.16 at a TSR = 0.74). Coefficient of
Parag K. Talukdar et al. [10] had reported rotor
static torque varies from 0.27 to 0.0038 for a rotor
having lower aspect ratio is more preferable as it is
with an aspect ratio of 0.88 and from 0.17 to 0.04 for a
easy to provide structural stability. Rotor having
rotor with an aspect ratio of 1.17. Coefficient of static
higher aspect ratio creates more bending moment on
torque varies from 0.08 to 0.33 for a helical Savonius
shaft which require heavy shaft which eventually
rotor with an aspect ratio of 0.93.
increase inertia of the turbine which is unfavourable
to its performance. Zhenzhou Zhao et al. [6] shows
D. Effect of Helix Angle (α)
that performance of rotor (two blade, helical angle =
Jae-Hoon Lee et al. [1] has concluded in numerical
180º and overlap ratio = 0.3) is found optimum in CFD
work that; in conventional savonius turbine helix
analysis having aspect ratio of 6 whose performance is
angle is zero in that case rate of fluctuation of torque
Cpmax = 0.2 at TSR = 0.75. For lower aspect ratio
is considerably high. In order to overcome and to
coefficient of torque is considerably high at low RPM,
make turbine more stable helix angle is provided. For
while rotor having higher aspect ratio have higher
more than 90º twist angle, it was found that the
RPM but its have lesser coefficient of torque. Rotor
torque coefficients stabilized and remained constant.
having higher aspect ratio would cause eddy airflow
Helix angle also reduce variation in projection area
which leads to performance dropping.
during its entire revolution as shown in figure 5.

International Journal of Scientific Research in Science, Engineering and Technology (ijsrset.com)

1466
 The maximum coefficient of power and the
corresponding tip speed ratio with different overlap
ratios modelled in this study. It may be seen that the
tip speed ratio at which the maximum coefficient of
power is observed decreases with the increase in the
overlap ratio for helical Savonius rotors. Maximum
coefficient of power for a helical Savonius rotor is
around 0.18 for a 0.3 overlap ratio at a TSR of around
0.73. Joao Vicente Akwa et al. [11] has concluded
rotor having overlap ratio of 0.15 gives averaged
Figure 4. Variation of projection area at different power coefficient equals to 0.3161 for tip speed ratio
azimuths angles [1]. of 1.25 from CFD analysis.

A. Kamoji et al. [5] had concluded after performing

several experiments that torque coefficient during
complete revolution is positive, whereas in
conventional rotor at several rotor angles torque
coefficient is negative. In helical savonius rotor
having 90º helix angle coefficient of power increases
with increase in Reynold number. Figure 5. The effect of overlap on pressure around
rotor [6].
E. Effect of Overlap Ratio (e)
Overlap ratio can be defined as ratio of distance F. Effect of No. of stage (N)
between two blades at centre of the turbine to the N.H. Mahmoud et al. [13] has found in experimental
diameter of the blade. Zhenzhou Zhao et al. [6] has analysis that; the two stages rotor gives higher specific
found in numerical analysis that, overlap distance of power than single stage rotor as shown in Fig.7. The
savonius rotor also effects on the power coefficient specific power is defined as the power obtained from
pressure contour of savonius rotor having two blades unit projected area of the rotor. In order to verify this
with an overlap distance at rotating condition has result, the static torque affected on rotor blades for
been shown in Fig.6. It can be visualized that a both single and double stages rotors is measured at the
pressure is high in the upwind rotor because of same angle of rotation and at different wind speeds.
heading-on the wind or because of collecting the Double stages rotor has higher static torque and
wind. Whereas pressure is very low on the counter consequently higher static torque coefficient
blade, because of lacking of the air blowing. As the compared to the single stage rotor. Tsutumo Hayashi
difference of pressure between two blades is more, it et al. [14] had found in experimental analysis rotor
leads to higher push and power coefficient get lower. having three stage have almost same torque
Passage between two buckets is provided to allow the coefficient thorough out complete rotation of rotor at
wind to flow in counter blade so it will reduce TSR = 0.043 at a wind speed of 12 m/s. But with
pressure difference between two blade which increase of TSR of 0.259 again the variation in torque
eventually increase torque. If ratio is more it will not coefficient increase, with further increase of TSR
allow wind to thrust in a upwind blade and if ratio is variation in torque coefficients is again decreases.
too small it will not allow to flow the wind in counter
blade. So, it requires optimum value which leads to a
maximum power coefficient.
International Journal of Scientific Research in Science, Engineering and Technology (ijsrset.com)
1467
 Figure 8. Savonius Wind Turbine Modified by
Sukanta Roy [15]

Sukanta Roy et al. [15] has experimentally found that,

Figure 6.Variation of static torque coefficient with
with the newly developed SSWT, a noticeable
wind speed for single and double stage rotors [13].
improvement in the maximum power coefficient is
observed over other tested models. Performance gains
G. Effect of with or without End-plate
of 3.3%, 6.9%, 19.2% and 34.8% are achieved over
N. H. Mahmoud et al. [13] has found in experimental
modified Bach, Benesh, semi-elliptical, and
analysis that, variation in mechanical power with
conventional SSWTs, respectively.
wind speed for rotors with and without end plates is
given in Fig. 8. Rotors with end plates give higher
They also found that, the CPmax values are
mechanical power than rotors without end plates.
experienced at an optimum range of TSR such as
This is because the existence of end plates increases
0.75–0.82, 0.72–0.80, 0.72–0.80, 0.70–0.75 and 0.66–
the amount of air which strikes the blades of Savonius
0.73 for the modified Bach, Benesh, semi-elliptical
rotor.
and conventional blades respectively.

Figure 7. Variation of mechanical power with wind

speed for rotor with and without end plates [13].

H. Effect of Blade Curve Profile

Figure 9. Different Type of Savonius Wind Turbine
There are five turbines are considered for
[15]
investigation as follows conventional semi-circular,
As shown in figure 1, Konrad Kacprzak et al. [2] has
semi-elliptic, benesh, modified bach type, newly
found that power coefficient the bach-type rotor is
developed.
superior to the other tested geometries and at the

International Journal of Scientific Research in Science, Engineering and Technology (ijsrset.com)

1468
 same time the elliptical savonius turbine exhibits
better power characteristics than the conventional
one.

Wenlong Tian et al. [16] had modified curve profile of

savonius blade, which exhibit higher average power
of 0.2580 (4.41% higher than conventional) at high
tip speed ratio of 0.7.

I. Effect of with and without shaft Figure 10. Effect of Reynolds No. on coefficient of
In CFD analysis Rajat Gupta et al. [17] studied two- Power with respect to Tip Speed Ratio [12]
bucket helical Savonius rotor with shaft at 45º twist
angle. From numerical analysis they concluded that K. Effect of Turbulence Model
the highest values of dynamic pressure and velocity Turbulence is considered as critical which should be
magnitude were obtained at the chord ends with 45º accounted precisely in the case of air. There are a
bucket twist and 90º rotor angle, which would ensure number of models available to take care of turbulence
improved performance of the rotor as a whole by factor in the simulation. Reynolds number and flow
increasing the aerodynamic torque production of the geometry are the prevailing parameters to select an
rotor. Bachu Deb et al. [18] has carried out numerical appropriate turbulence model.
analysis which is designed in gambit software having
rotor diameter of 17 cm, helical angle of 45º and M.H. Mohamed et al. [13] has carried flow simulation
diameter of blade is 17 cm. Blade is attached with using various governing equation, as shown in fig. 11
each other with the help of end plate. In CFD analysis it can be visualized that realizable k-ϵ model gives
it has been found that helical savonius rotor without precious result compare to SST, standard k-ϵ and RSM
shaft obtained maximum power coefficient is 0.462 model over wide range of speed ratio.
for 45º rotor angle at TSR of 1.636.
C Kang et al. [19] conclude that among the various
J. Effect of Reynolds No. turbulence models, the Spalart - Allmaras (SA) model
Reynolds no. can be defined as a dimensionless is a simple one-equation turbulence model, where the
number used in fluid mechanics to indicate whether near wall gradients of the transported variable are
fluid flow past a body or in a duct is steady or much smaller than the turbulent kinetic energy
turbulent. equation based (k - ϵ) models. This might make this
model less sensitive in the near walls treatment
In experimentally analysis it has been found that by A. around the Savonius rotor. However, a 3D CFD
Damak et al. [12] with increase of Reynold No. curve simulation using SA model was carried out to
profile of coefficient of power increase as shown in compare the performance of vertical-axis spiral rotor
figure. Kamoji [5] and Akwa et al. [11] had also with two end plates and one middle plate with a
reported the same trend of Reynold no. at different conventional Savonius rotor. Results of the simulation
tip speed. depicted that torque performance of the spiral rotor
was more favourable during its whole rotation cycle
[6].

Anuj Kumar et al. [21] has considered Realizable k-ε

turbulence model for CFD analysis as it demonstrates

International Journal of Scientific Research in Science, Engineering and Technology (ijsrset.com)

1469
 the excellent agreement with obtained experimental BE Launder et al. [20] says that, the standard k - ϵ
result. Realizable k-ε turbulence simulate better model is the basic k–ε turbulence model and more
which involves the rotating behaviour of blades or air suitable when flow is fully turbulent and has given
foil, a boundary layer, separated flows or flow better results than SA model for turbine analysis.
through a channel.
IV. SUMMARY OF LITERATURE REVIEW

From the literature, it can be concluded that helical

savonius turbine is more efficient than conventional
type wind turbine as the helical savonius turbine
generates more power and torque coefficient. It is also
observed form the literature that the turbine having
helix angle of 180º makes the two blades rotor
advantageous then the three-blade rotor. Savonius
turbine having aspect ratio of about 0.88 having end
plate with an overlap ratio in range of 0.15 to 0.30 are
produces more torque and power. Sukanta Roy et al.
Figure 11. Effect of Governing Equation on results of has developed a modified blade profile which
CFD Analysis [21] develops more drag force hence eventually improves
the performance by 34.8% compare to conventional
The results of their study are shown in Figure 11 savonius wind turbine. For the numerical simulations,
which demonstrates the excellent agreement obtained the researchers applied different models for the
between CFD and experiments results for the simulation to understand the physical behaviour and
realizable k ε turbulence model. Accordingly, the flow conditions and it can be concluded from the
realizable k-ε turbulence model has been considered literature that standard k - ϵ model gives the better
for the present study. It was suggested under earlier results than the SA model for turbine analysis. The
studies that Realizable k-ε turbulence model is better results obtained from the Standard k - ϵ model
to simulate the rotating behaviour of blades or air foil, validates completely with experimental results.
flow through the channel, a boundary layer or
separated flows. Accordingly, the Realizable k-ε V. CONCLUSION
model has been employed as turbulence model under
the present study to simulate the rotating behaviour Large Scale Wind Turbines (LSWTs) have been
of turbine blades. This model has two major extensively examined for decades but very few studies
differences from the Standard k-ε. It comprises a new have been conducted on the small-scale wind turbines
formulation for the turbulent viscosity and a new (SSWTs). The performance improvement of a helical
transport equation for the dissipation rate ε, derived Savonius rotor is studied by using three dimensions
from an exact equation for the transport of the mean- CFD model & experimentally in the papers.
square vorticity fluctuations. Further, this model does Researches are carried out to study the influence of
not rely on the assumed relationship between the blade number, overlap ratio, helical angle, no. of stage
Reynolds stress tensor and the strain rate tensor. In and aspect ratio on the performance of helical
the realizable k-ε model, the coefficient of the model Savonius rotors. Through the work in the papers, the
is expressed as a function of mean flow and coefficient of the power of Savonius rotor is
turbulence properties, rather than assumed to be dramatically improved from the conventional
constant as in the standard model. Savonius rotor.

International Journal of Scientific Research in Science, Engineering and Technology (ijsrset.com)

1470
 VI. REFERENCES Savonius wind rotor using computational fluid
dynamics. Renewable energy. 2012 Feb
[1]. Lee JH, Lee YT, Lim HC. Effect of helical angle 1;38(1):141-9.
on the performance of Savonius wind turbine. [12]. Damak A, Driss Z, Abid MS. Experimental
power. 2014 Aug 24;5:1. investigation of helical Savonius rotor with a
[2]. Kacprzak K, Liskiewicz G, Sobczak K. twist of 180. Renewable Energy. 2013 Apr
Numerical investigation of conventional and 1;52:136-42.
modified Savonius wind turbines. Renewable [13]. Mahmoud NH, El-Haroun AA, Wahba E, Nasef
energy. 2013 Dec 1;60:578-85. MH. An experimental study on improvement of
[3]. Ferrari G, Federici D, Schito P, Inzoli F, Mereu Savonius rotor performance. Alexandria
R. CFD study of Savonius wind turbine: 3D Engineering Journal. 2012 Mar 1;51(1):19-25.
model validation and parametric analysis. [14]. Hayashi T, Li Y, Hara Y. Wind tunnel tests on a
Renewable Energy. 2017 May 1;105:722-34. different phase three-stage Savonius rotor.
[4]. Menet JL. Aerodynamic behaviour of a new JSME International Journal Series B Fluids and
type of slow-running VAWT. InWind Energy Thermal Engineering. 2005;48(1):9-16.
2007 (pp. 235-240). Springer, Berlin, [15]. Roy S, Saha UK. Wind tunnel experiments of a
Heidelberg. newly developed two-bladed Savonius-style
[5]. Kamoji MA, Kedare SB, Prabhu SV. wind turbine. Applied Energy. 2015 Jan
Performance tests on helical Savonius rotors. 1;137:117-25.
Renewable Energy. 2009 Mar 1;34(3):521-9. [16]. Tian W, Mao Z, Zhang B, Li Y. Shape
[6]. Zhao Z, Zheng Y, Xu X, Liu W, Hu G. Research optimization of a Savonius wind rotor with
on the improvement of the performance of different convex and concave sides. Renewable
Savonius rotor based on numerical study. Energy. 2018 Mar 1;117:287-99.
InSustainable Power Generation and Supply, [17]. Gupta R, Deb B, Misra RD. Performance
2009. SUPERGEN'09. International Conference analysis of a helical Savonius rotor with shaft at
on 2009 Apr 6 (pp. 1-6). IEEE. 45 twist angle using CFD. Mechanical
[7]. Saha UK, Thotla S, Maity D. Optimum design Engineering Research. 2013 Mar 1;3(1):118.
configuration of Savonius rotor through wind [18]. Deb B, Gupta R, Misra RD. Performance
tunnel experiments. Journal of Wind analysis of a Helical Savonius rotor without
Engineering and Industrial Aerodynamics. 2008 shaft at 45° twist angle using CFD. Journal of
Aug 1;96(8-9):1359-75. Urban and Environmental Engineering. 2013
[8]. Wenehenubun F, Saputra A, Sutanto H. An Jan 1;7(1):126-33.
experimental study on the performance of [19]. Kang C, Zhang F, Mao X. Comparison study of a
Savonius wind turbines related with the vertical-axis spiral rotor and a conventional
number of blades. Energy procedia. 2015 Apr savonius rotor. InPower and Energy
1;68:297-304. Engineering Conference (APPEEC), 2010 Asia-
[9]. Ali MH. Experimental comparison study for Pacific 2010 Mar 28 (pp. 1-4). IEEE.
Savonius wind turbine of two & three blades at [20]. Launder BE, Spalding DB. The numerical
low wind speed. International Journal of computation of turbulent flows. InNumerical
Modern Engineering Research (IJMER). 2013 Prediction of Flow, Heat Transfer, Turbulence
Sep;3(5):2978-86. and Combustion 1983 (pp. 96-116).
[10]. Talukdar PK, Sardar A, Kulkarni V, Saha UK. [21]. Kumar A, Saini RP. Performance analysis of a
Parametric analysis of model Savonius single stage modified Savonius hydrokinetic
hydrokinetic turbines through experimental turbine having twisted blades. Renewable
and computational investigations. Energy Energy. 2017 Dec 1;113:461-78.
Conversion and Management. 2018 Feb
15;158:36-49.
[11]. Akwa JV, da Silva Junior GA, Petry AP.
Discussion on the verification of the overlap
ratio influence on performance coefficients of a

International Journal of Scientific Research in Science, Engineering and Technology (ijsrset.com)

1471