Renewable and Sustainable Energy Reviews 42 (2015) 801–822

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

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

Review of aerodynamic developments on small horizontal axis wind

turbine blade
N. Karthikeyan a, K. Kalidasa Murugavel a,n, S. Arun Kumar a, S. Rajakumar b
a
Centre for Energy Studies, National Engineering College, K.R. Nagar, Tamil Nadu, India
b
Department of Mechanical Engineering, Regional Centre, Anna University, Tirunelveli, Tamil Nadu, India

art ic l e i nf o a b s t r a c t

Article history: Wind energy is innately renewable, abundant in the earth and can possibly reduce the dependency on
Received 1 February 2013 fossil fuels. Wind is an incarnation of sun and is always nourished by the latter. Approximately 10 million
Received in revised form MW of energy can be continuously generated from the wind sources. In contrast to the large horizontal
20 August 2014
axis wind turbines (HAWT), which are established in the area with optimum wind conditions, small
Accepted 21 October 2014
wind turbines are being installed to produce power irrespective of favourable wind conditions.
Available online 13 November 2014
Parameters associated with blade geometry optimization are important, because once optimized,
Keywords: shorter rotor blades could produce power comparable to larger and less optimized blades. A detailed
Small wind turbine review of various blade profiles and aerofoil geometry optimization processes to achieve high power
Low Reynolds number aerofoil
coefficient in small wind turbines that falls below Reynolds number 500,000 have been presented in
BEM
this paper.
& 2014 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 801
2. Improvement of aerodynamic performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 803
2.1. Design of low Reynolds number airfoil: roughness effect or leading edge flap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 806
2.2. Airfoil-trailing edge flap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 809
2.3. Design of low Reynolds number airfoils: flat back. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 810
2.4. Airfoil-camber line modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 810
3. Effect of pitch angle on starting performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 812
4. Effect of chord and twist distribution on performance of SWT rotor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815
5. Non-twisted-double pitches rotor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 819
6. Noise reduction in small wind turbine rotor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 819
7. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 820
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 822

1. Introduction major problems in the world that has created the awareness
among the people to seek renewable energy [1]. The differential
Energy is the most considerable constituent of the socio- heating of the earth's surface produces wind by the sun. A rough
economic development and economic growth. Rapid rise in the statistical estimation has been given that with the available wind
level of greenhouse gases in the atmosphere, gradual increase in energy as much as 10 million MW of power could be generated. It
the cost of the fossil fuels and shortage of basins are the present is clean, eco-friendly and prime national security at a time when
the decreasing global reserves of fossil fuels is an eminent danger
in the sustainability of global economy. Large scale wind turbines
n are normally located at high potential wind resource areas, which
Corresponding author. Tel.: þ 91 9677928940; fax: þ 91 4632232749.
E-mail addresses: kali_vel@rediffmail.com, are scarce in number. For areas of low wind potential, low cost,
kkmmech@nec.edu.in (K. Kalidasa Murugavel). simple, portable, low noise and maintenance free structured small

http://dx.doi.org/10.1016/j.rser.2014.10.086
1364-0321/& 2014 Elsevier Ltd. All rights reserved.
802 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822

Nomenclature θP pitch angle (rad)

c airfoil chord or blade chord (m)
ϕ inflow angle (rad) r radial distance (m)
CL lift coefficient (dimensionless) F tip loss factor (dimensionless)
CD drag coefficient (dimensionless) Vo free stream wind velocity at hub (m/s)
α angle of attack (rad) D overall rotor diameter (m)
λ tip speed ratio (dimensionless) d local diameter (m)
λr local speed ration at any station r (dimensionless) αt angle of attack at tip of the blade (rad)
a axial flow induction factor (dimensionless) α0 angle of attack at zero lift (rad)
a0 tangential flow induction factor (dimensionless) k acceleration factor
UC starting cut-in wind speed (m/s) LP sound pressure level (dBA)
TR resistive torque (N m) ω rotor speed
NB number of blades σ blade solidity ratio (dimensionless)
ρ air density (kg/m3) M distance from the turbine to measurement point.
R rotor radius (m) ηAero aerodynamic efficiency
Icp chord pitch integral θt blade pitch angle at tip (rad)

wind turbines are of crucial influence in the rural and urban areas
wind power extraction [2]. The approximate power coefficient of
small scale and large scale wind turbines are 0.25 and 0.45,
respectively. However, the earlier scientific community conducted
most of the investigations with a focus more on structural analysis
than on aerodynamic optimization but slowly the scientists are
moving towards the aerodynamic analysis of wind turbines [3].
In the mid of 1980s, National Renewable Energy Laboratory
(NREL) started to develop several families of aerofoils especially
for wind turbine blades, and in the later years such development is
continued by Delft University. The aerofoils having good perfor-
mance characteristics in the fluid flow with the Reynolds number
of range of 5  105 are the best for small scale wind turbines and
their aerodynamic characteristics are listed in literatures [4–6].
Small scale wind turbines are classified based on their operat-
ing parameters and applications [7]. Improvement on aerody-
namic performance and optimization of the turbine blades are
widely based on Blade Element Momentum (BEM) theory, Genetic
Algorithms and xfoil. The implementation of the above mentioned
techniques on the blade design has produced dramatic develop-
ments in the small wind turbine sector. This paper reviews the
recent developments in the aerodynamic profiles, starting char-
acteristics, aerodynamic noise reduction and efficiency improve-
ments of the small horizontal axis wind turbine blade.

Table 2
Performance parameters for the airfoils considered for free and fixed transition at
Re¼ 300, 000. [5].

Airfoil Free transition Fixed transition Percentage

difference

ðC l =C d Þmax C L C Lmax ðC l =C d Þmax C l C lmax ðC l =C d Þmax C lmax

A18 79.6 0.80 1.23 41.2 1.03 1.22 48.2 0.7

BW-3 69.6 1.05 1.44 39.6 0.89 1.24 43.1 1.9
Clark-Y 77.2 0.85 1.35 39.1 0.83 1.13 49.4 16.5
E387 81.7 0.93 1.29 – – – – –
Go471a 82.3 1.08 1.40 – – – – –
NACA2414 66.6 0.90 1.23 – – – – –
RG15 69.0 0.66 1.14 – – – – –
S822 69.4 0.88 1.22 32.9 0.68 1.18 52.6 3.8
S823 62.7 1.05 1.18 30.2 0.78 1.14 51.8 3.0
S6062 73.1 0.65 1.11 – – – – –
S7012 72.1 0.71 1.14 40.4 0.94 1.15 44.0 -0.7
SD6060 73.5 0.72 1.11 – – – – –
SD7032 83.4 1.00 1.39 – – – – –
SD7037 76.3 0.84 1.28 44.1 0.99 1.32 42.2 -3.1
SD7062 77.5 1.23 1.66 45.1 0.96 1.23 41.8 25.8
Fig. 1. Velocity diagram at radial station [11].

Table 1
Operating parameters of small wind turbines [7].

Category W (kW) R (m) Maximum rotor speed (rpm) Typical uses Generator type (s)

Micro 1 1.5 700 Electric fences, yachts Permanent magnet (PM)

Mid-range 5 2.5 400 Remote houses PM or induction
Mini 20 þ 5 200 Mini grids, remote communities PM or induction
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822
2. Improvement of aerodynamic performance
The efficiency of rotor basically depends on the aerofoil, which
is used to reduce the pressure on the upper surface to increase the
lift which generates sufficient torque. The power coefficient and
the torque generated can be optimized by maximizing the lift
coefficient (CL) and the lift to drag (L/D) ratio for aerofoil [8,9].
High L/D ratio contributes to high values of torque and which is
desirable for small sized rotors and significant in gaining good
response at low wind speed to generate maximum power [5,8,
and10].
ηAero ¼ 11þððCCDD=C
=C L Þcot ϕ
L Þ tan ϕ
ð1  aÞ
tan ϕ ¼
λr ð1 þ a' Þ
where ϕ is the inflow angle and λr ¼ ωr=V o is local speed ratio at
any station represented in velocity diagram shown in Fig. 1 [11].
From the above Eq. (1), it can be seen that the aerodynamic
efficiency is substantially influenced by lift to drag ratio and
flow angle.
The small wind turbines are classified into three subsets; based on
the power output and radius of rotor are such as micro (o1 kW and
1.5 m radius), mid-range (o5 kW and 2.5 m radius) and mini
(20 kWþ and 5 m radius). Starting is more difficult in microturbines
Fig. 2. Lift curves for the airfoils at Re¼ 300,000 (free and fixed transition) [5].
Fig. 3. Drag polar for the airfoils (free transition) [5].
803
when compared with mid-range and mini turbines because of the
rotor radius (R) which is less than 1.5 m; starting solely depends on
the aerodynamic torque produced by the wind and the resistive
torque is directly proportional to R3 [7]. The rotor commences to
rotate when the aerodynamic torque exceeds the combined resistive
torque (TR) offered by drive train of the turbine system and generator.
The minimum wind speed that is required for the rotor rotation is
known as cut-in speed and it can be calculated by [12]
" #1=2
2T R
UC ¼ ð3Þ
N B ρR3 I cp
ð1Þ where
Z 1
I cp ¼ cr sin ð2θp Þð1  σ cos 2 θp Þdr ð4Þ
ð2Þ rs
where NB is the number of blades, ρ is the density of air (kg/m3), TR is
the resistive torque (N m), R is the blade radius, m and Icp is the chord
pitch integral.
The estimation of the cut-in wind speed of the small wind
turbine solely depends on the blade geometry like plan form and
pitch distribution (Icp). It does not depend much on R and N in
Eq. (3).
To avoid the starting problem, small wind turbines rarely have
variable pitch and often fixed to expose high flow angle. As the
blade begins to rotate, α can be 901 or greater since that is in the
804 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822

range where the lift and drag tends to become independent of the blade span.
detailed geometry such as thickness and camber. In other words,
aerofoils behave like a flat plate at high angle of attack [12]. 4π r F sin ϕ
C ðr Þ ¼       ð6Þ
N B C L f½ λr þ tan ϕ = 1  λr tan ϕ   C D =C L g
C L ¼ 2 sin α cos α and C D ¼ 2 sin 2 α ð5Þ
\where NB is the number of blades, CD/CL is the minimum drag-to-lift
Parameters associated with blade geometry optimization are ratio of the airfoil section, F is the tip loss factor, λr ¼ λ. r is the local
important, because once optimized, shorter rotor blades could speed ratio at radial distance r along the blade, λ ¼ ωR/Vo is the
produce similar power compared to one that is not optimized and design tip-speed ratio, ω is the rotational speed of the blade, Vo is the
larger in diameter. Increasing the power extraction from the low free stream wind velocity at hub height and R is the rotor radius.
wind areas, aerodynamic optimization of the rotor is very essen- The twist distribution based on the twist of the zero lift line
tial, which is to be coupled with optimization of chord and twist [13] is given by
distribution, number of blades, selection of airfoil geometry, and  
the tip speed ratio (TSR) [10]. θp ¼ ð Rαt =r  αt Þ  kð1  r=RÞ ð7Þ
Flapwise (chord) distribution and edgewise (blade thickness)
distribution and twist distribution based on blade element where θp is the pitch angle, R is the radius of the blade, r is the
momentum theory, BEM. Eq. (6) gives the chord distribution along radial location, αt is the angle of attack at the tip of the blade and k

Fig. 4. Drag polar for the airfoil considered (fixed transition) [5].

Fig. 5. (a) Leading-edge tape configurations and (b) contours of aerofoils tested [6].
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822 805

is the acceleration factor (a constant) such that k 4 0.

 
αt ¼ ϕ  θ t þ α0 ð8Þ

where, αt is the angle of attack at tip, α0 is the angle of attack at

zero lift, ϕ is the inflow angle, and θt is the pitch angle at the tip
(θt E 01).
Noise emission prediction and control is very important in
small wind turbines that are located close to dwelling areas. The
main sources of wind turbine noise may be divided into two:
mechanical noise which includes the noise from the fans, gen-
erator, gear box, etc. and the aerodynamic noise, which is
originated from the interaction between the rotor and the wind.
Therefore the main engineering effort is being made on the subject
of aerodynamic noise. The sound power produced by a turbine is
Fig. 7. Percentage loss in power coefficient from the addition of leading edge tape
approximately proportional to the output power of the turbine, layers [6].
with the constant of proportionality being 10–7 [14]. Small wind
turbine noise depends on the tip speed ratio (blade velocity), blade
diameter, changes in the wind orientation [15] and sources of
noise due to rotation of blades are caused by blunt trailing edge
and blade tip shape.

Lp ¼ 50 log 10 ω þ60 log 10 D ð9Þ

Sound pressure level Lp values correlated with other factors like Fig. 8. Boundary layer trip geometry used to simulate the effects of leading edge
propagation and sound reflection by the earth's surface to get debris and erosion (dimensions are in inches) [17].

Fig. 6. Drag polar showing the effect of varying the tape edge location on the suction surface of various airfoils (one layer of tape and Re¼ 300,000): (a) BW-3; (b) FX 63-137;
(c) S822; and (d) SG6051 [6].
806 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822

Eq. (9) [16].
Lp ¼ 50 log 10 ω þ60 log 10 D 27 20 log 10 M
where M is the distance from the turbine to measurement point.
2.1. Design of low Reynolds number airfoil: roughness effect or
leading edge flap
negatively affect the lift characteristics and thus aerodynamic
performance with simulated leading-edge roughness is likely to
ð10Þ
be useful in the airfoil selecting process as shown in Fig. 2. Among
the airfoils tested with the simulated leading-edge roughness,
except BW-3 and Go417a, which have a narrow drag bucket, the
remaining airfoils that were tested offer a wide selection as shown

The airfoil selection process for small horizontal axis wind

turbines with extensive database of 15 low Reynolds number
airfoils was generated and coefficient of lift and drag were
analysed by wind tunnel experiments in Tables 1 and 2 [5]. Low
Reynolds number airfoils generally loses its performance due to
laminar separation bubbles, which is encountered by adding small
trips or tabulators on the upper surface leading edge and it
reduces the length of laminar separation bubble, and quickly
produces the turbulent boundary layer.
The addition of trips causes an overall increase in drag except
for the airfoils that have suffered from severe laminar separation
effects under clean conditions. Also drag polar shows that for fixed
transition the performance is essentially Reynolds number inde-
pendent as the drag polar, forms almost a single curve. Roughness
affects a significant increase in the overall drag and can also Fig. 11. Lift and drag coefficient for various angle of attack [19].

Fig. 9. Drag curve for E387 clean and F trip cases [17].

Fig. 10. (a) Low Reynolds number and (b) high Reynolds number [18].
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822 807

in Figs. 3 and 4 for constant-speed HAWTs. The A18, BW-3, S7012,
S822, S823 and SD7037 airfoils are essentially CLmax insensitive to
roughness and therefore author suggested these airfoils for stall
regulated HAWTs.
The wind tunnel study was carried out for lift and drag measure-
ments and aerodynamic performance of leading-edge tape, which is
used on small wind turbine blade leading edge to protect from blade
erosion over the Reynolds number range from 150,000 to 500,000 for
airfoils BW-3, FX 63-137, S822, SG6042 and SG6051 in Fig. 5(a) and (b)
[6]. The magnitudes of the aerodynamic effects are caused by the tape
and it is aerofoil dependent. For BW-3 aerofoil is insensitive to the
tape, natural transition occurs in the leading edge, and the tape had
negligible effect on the drag polar in Fig. 6(a). The S822 and SG6051
airfoils benefit from extending the tape beyond 5% chord, while there
is a trade-off between drag in the middle and upper end of the lift
range in the case of the FX 63-137 airfoil in Fig. 6(b) and (d). Laminar
separation effects are particularly strong for the S822 airfoil owing to
its high thickness, which caused reductions in maximum lift-to-drag
ratio from 5% to 18% in Fig. 6(c). Even though the drag of the SG6042
airfoil was highly sensitive to the tape, it had lower drag with tape
rather than other airfoils tested owing partly to its lower thickness.
The drag penalties are noted particularly while increasing tape
thickness and Reynolds number as shown in Fig. 7.
For variable speed HAWTs operating at constant tip-speed
ratio, constant power coefficient, and the different tape configura-
tions tested on the SG6042 airfoil reduced the power coefficient by
approximately 2% or less. The losses in power coefficient, though
small can be expected to increase for wind turbines operating at
Reynolds numbers above 500,000. As a practical consequence,
special attention should be given to the application of the tape for
wind turbines with ratings of the order of 5 kW, particularly if two
layers of tape are used as shown in Fig. 7.
From the wind tunnel study, authors suggested minimizing the
performance losses due to leading edge tape by the chord wise
extension of the tape on pressure side and increasing the size of
the length at least by 5% chord and 15–30%chord is aerodynami-
cally better. Multiple tape layers are most preferable and systema-
tic investigation is needed.
A detailed wind tunnel tests of lift, drag and moment measure-
ments on the six airfoils are E387, FX 63-137, S822, S834, SD2030,

Fig. 12. Variation of lift and drag coefficient vs. angle of attack at Re¼148,000 [20].

Fig. 13. Comparison of lift and drag characteristics for various angle of attack [20].
808 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822

and SH3055 were conducted [17], to be use in small wind turbines
at Reynolds numbers of 100,000, 200,000, 350,000 and
500,000,150,000 for both clean and rough conditions. The airfoils
were tested with three dimensional zigzag boundary-layer trips of
length 0.013 in. and height 0.11% of chord. The upper and lower
surfaces are fixed near the leading edge (Fig. 8). The main effect of
the trip is to promote transition shortly downstream of the trip.
In the lowest Reynolds number, as the angle of attack increased
the drag curve increased from lower range and adverse pressure
gradient of the upper surface becomes stronger and as a result of
bubble drag grows until the drag is a maximum that form a “drag
bucket”. As the Reynolds number increases the length of the
bubble decreases which lowers the drag. If a bubble is present in
a clean case where Re ¼100,000 the trip has the effect of short-
ening the laminar separation bubble. When there is no bubble,
where Re4 100,000, transition is forced to occur sooner than it
otherwise would. The trade-off between having lower bubble drag
and higher turbulent skin friction drag leads to lower net drag,
while for all other points the added turbulent skin friction drag
outweighs the reduction in the bubble drag, resulting in higher
drag with the trip as shown in Fig. 9. This trade-off, leading to
higher drag, is particularly true for the higher Reynolds number
cases. For FX 63-137 airfoil maximum lift performance suffers
from the addition of simulated roughness. Another difference is
that while the E387 had a highly unsteady stall leading to a sharp
break, the FX 63-137 exhibited a soft stall with little unsteadiness.
For S822 and S834 airfoils, unsteadiness at stall prevented taking
high angle of attack data for Re¼ 500,000. The degree of the
unsteadiness during the tests was observed to be somewhat less
than that for the E387. Of the six airfoils tested, this airfoil
displayed the most unsteadiness install, and this limited the angle
of attack range for Re¼500,000 case.
The new vortex generator that consist of an airfoil section to
reduce the drag of the conventional vortex generator is especially

Fig. 14. (a) Lift and drag characteristics and (b) pressure distribution of AF300 airfoil at Re¼75,000 and angle of attack¼ 101 [20].

Fig. 15. (a) Convergence and (b) pressure coefficient for the T.Urban 10/193 blade section at the design condition [10].
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822 809

suitable for small wind turbine airfoils and was tested on wind
tunnel experiments to investigate the effects of Reynolds number,
height and interval of vortex generator [18]. The flat plate of the
vortex generator enhances lift and the drag also increases simul-
taneously especially at low Reynolds number of Re¼100,000–
400,000. As a result, it was concluded that the performance drop
was due to the drag of the vortex generators caused by its
modification without losing the primary function of the vortex
generator. The modified vortex generators considerably improved
the lift-to-drag ratio especially in the range of the low Reynolds
number. Therefore, it was verified that the modified vortex
generator is suitable for small wind turbines shown in Fig. 10.
attack from 01 to 401 and the heights of the flaps used were 1.0%,
1.5%, 2.0% and 2.5% of the airfoil chord [19]. The effects of different
deflection angles (01, 451, 901, 1351) of the trailing edge flaps are
compared. All tested types of flaps, attached to the airfoil on the
lower surface near the trailing edge, significantly increased the lift
coefficient, with little or no change in drag coefficients. The best
increment in performance is by the addition of gurney flaps, for
which the trailing edge angle is 901 to the airfoil chord and 2%C

2.2. Airfoil-trailing edge flap

The feasibility study of power and efficiency augmentation of

the horizontal axis wind turbine is done by adding small flaps to
its trailing edge of blades which consists of NACA632-215 airfoil
tested at Reynolds number range of 2.4  105, varying the angle of Fig. 17. Comparison of thin airfoil and NACA4418 [21].

Fig. 16. Lift coefficient as function of the incident angle (a), drag polar (b), transition in the suction side (c) and pressure side (d) as function of incidence angle [10].
810 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822
Fig. 18. Relationship between (a) lift–drag coefficients and (b) lift–drag ratios with angle of attack [21].
Fig. 19. Comparison of lift, drag coefficients and lift–drag ratio of seagull airfoil and NACA 4412 airfoil with angle of attack [22].
height of GF. The stall angle of the airfoil with the Gurney Flap is 21
earlier than that of the airfoil without flap as shown in Fig. 11.
2.3. Design of low Reynolds number airfoils: flat back
The aerodynamic based design of low Reynolds number airfoil
in the range of 38,000–205,000 for the small wind turbine
application was conducted [20].
The AF300 airfoil was optimized from existing low Re airfoils,
initially taking S1210 and S1223 as base airfoil, and modifying by
percentage increase in the trailing edge thickness, S1210 (3%) variant is
selected based on its aerodynamic performance lift coefficient values
close to 2 (1.91 and 1.97) at Re¼55,000 and 148,000 through xfoil
code can be seen in Fig. 12. Compared with other low Reynolds
number airfoils suited for small wind turbine rotor blades at
Re¼55,000, 100,000 and 148,000, the performance of the AF300
airfoil showed good progress attaining the highest combinations of
optimum CL and L/D ratios as shown in Fig. 13. Extensive wind tunnel
tests results of lift distribution showed that at Re¼ 38,000, CL values
ranges from 0.41 to 1.05 for α 0–181 as in Fig. 14. At Re¼ 75,000,
128,000 and 205,000, CLmax ¼1.72, 1.81 and 1.86 respectively for the
angle of attack of 141. The flat back trailing edge of the AF300 airfoil
delays flow separation and increases CL and improves the aerodynamic
properties and adds strength to the airfoil structure and would require
lighter and less expensive materials for the blades, decreasing the
inertia and improving start-up and allowing the rotors to operate at
lower cut-in wind speeds. A high stall angle of 141 means that the
rotor could operate at a wider range of angle of attack without being in
danger of stalling and the inner rotor sections near the root region that
experience these low range of Re would be greatly benefitted from the
increase in the performance.
2.4. Airfoil-camber line modification
Pressure load inverse method was successfully applied in the
design of new wind turbine airfoil that achieves high performance
in urban environment by increasing the maximum lift with
constant pressure load along the chord of blade at Reynolds
number ranging from 3  105 to 1  106 and design inlet angle
β ¼41. In this method, mean camber line only required 8 iterations
to attain the prescribed convergence, which has been named T.
Urban 10/193. The pressure load distribution in the designed blade
section, increases gradually up to 20% of the axial chord from
leading edge, and from 20% to 80% of axial chord is almost
constant and reduces smoothly towards the trailing edge as shown
in Fig. 15. It can be seen that there is a closer relationship between
the transition points and the evolution of lift and drag. In the
suction side of the airfoil transition occurs near the trailing edge at
minimum angle of attack α. As α increases, transition moves
towards the trailing edge at three Re ranges until the beginning of
stall as shown in Fig. 16. T.urban 10/193 blade section, as an
isolated airfoil, confirmed maximum lift and moderate drag with
soft-stall characteristics, due to a reduction of adverse pressure
gradient on the upper side [10].
The three bladed 3 kW wind turbine with thin airfoil and tip
speed ratio of 3 was developed [21]. From Fig. 17, thin airfoil
section was developed from the NACA4418 airfoil, upper surface is
close to base airfoil, although their wing profiles are entirely
different from one another. Thin airfoil has 11% camber (3 times
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822 811

bigger) at 0.35 chords and 8.5% thickness (less than 12 ) at 0.10 airfoil of the seagull is suitable for special work circumstances of
chords compared with NACA4418, which has 4% camber at 0.40 the small wind turbine that can be used for the design of the
chord and 18% thickness at 0.30 chord. blades, especially in the tip. Further research can be taken out
The lift coefficient of thin airfoil CL is higher than that of the about the family of seagull airfoil. The results may provide
NACA 4418 airfoil and the maximum CL value of 2.04 exists near technical reference for the wind turbine designing.
angle of attack α ¼121. On the other hand, the difference between
the drag coefficients CD in both airfoil cases is very small from
attack angle α ¼0  101; otherwise, the CD of the thin airfoil is
slightly larger than that of the NACA airfoil. The maximum L/D
ratio of thin airfoil obtained at angle of attack α ¼ 51 and CLmax is
1.63 whereas the L/D ratio of NACA4418 at α ¼71 with CLmax is only
of 1.19. Moreover, for a negative attack angle α, the lift–drag ratio
of the thin airfoil is virtually zero, whereas values are negative for
the NACA airfoil. The new thin airfoil showed good performance
over the base airfoil as shown in Fig. 18.
The aerodynamic performance of seagull airfoil was studied
using the results of foreign bionic experimental airfoil and the data
were analysed by CFD to make a suitable airfoil for small wind
turbine blade in the range of Re¼100,000–200,000 [22]. The
results show that the performance of seagull airfoil is superior to
NACA 4412 airfoil and flow separation is harder to identify than in
NACA 4412 as shown in Fig. 19. Through transforming the shape of
seagull airfoil, it can be found that the original parameters are
relatively better as shown in Fig. 20. The analysis shows that the Fig. 21. The CL of the tested airfoils against α for Re¼ 38,000 and 205,000 [23].

Fig. 20. Comparison chart of aerodynamic characteristics of seagull and the transformation airfoils at Re¼ 1  105 [22].
812 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822
Wata et al. [23] designed and developed the new airfoil from
set of low Reynolds number airfoils based on their aerodynamic
performance, which are combined with each other to create a new
one, which is tested at Re¼ 38,000, 28,000 and 205,000, α ¼0–221
and validated by xfoil and wind tunnel tests. WhenSG6043 airfoil
was combined with GOE15, E422, Eppler 560 and S1223, the
performance of SG6043_Eppler 422 was better than other combi-
nations as shown in Fig. 21.
The numerical results agreed well with experimental results at
higher Reynolds numbers, giving CL values close to 1.7. Experi-
mentally, the pressure distributions showed a stalling angle of 151.
The maximum L/D ratio was obtained at α ¼ 81, which can be
considered as the design angle of attack as shown in Fig. 22. It
showed that the airfoil achieved very good aerodynamic charac-
teristics at different Reynolds numbers and can be used as an
efficient airfoil in small HAWTs.
The numerical analysis of aerodynamic performance of
NACA0018 wind turbine airfoil was conducted at Re ¼5  105, with
α varying from  81 to 131 by fluent software [24]. Aerodynamic
performance like lift and drag of the NACA0018 airfoil varies with
different angles of attack ranging from  81 to 131. Four Lift and
Drag curves obtained by different turbulence models, which
coincide with experimental curve but five equations Reynolds
stress showed better results, drag differed due to leading edge
roughness as shown in Fig. 23. Pressure differences between upper
and lower surfaces of the airfoil increases with the AOA increasing.
At front edge of airfoil, pressure difference is higher, compared to
rear edge, where the lift is produced. At α ¼ 0, pressure of upper
and lower surfaces is equal. When the angle of attack was large,
the flow separated in the front edge of airfoil, then combined in a
certain distance of the rear airfoil as shown in Fig. 24. The
Fig. 22. Lift coefficient (CL) vs. angle of attack for various Re values [23].
Fig. 23. Lift and drag coefficient with various angle of attack [24].
calculation results provided a reference for the research and
development of wind turbine airfoils.
3. Effect of pitch angle on starting performance
The starting characteristics was analysed in the 5 kW wind
turbine for two different pitch angles. At a large pitch angle, the
blade takes 10 s to reach their maximum speed of 50 rpm at wind
speed of 8 m/s and at a lower pitch the blade takes 50 s to start at
same wind speed for the operational rotor speed of about
300 rpm. A significant trade-off between starting and operating
performance has been identified [25].
The starting performance improvement of 1 m and 600 W
microturbine rotor was designed by scaling down of 2.5 m and
5 kW small wind turbine. The new blades consist of larger chord
(40% more) compared to 5 kW turbine, with pitch angle of 51 and
to operate as three bladed rather than pairs. These blades, in
combination with a new generator, produce an 80% improvement
in power extraction over the original which is not optimized and
cut-in wind speed also reduced to 2.5 m/s [26].
The starting performance of a small horizontal-axis wind
turbine by varying blade pitch angles between 01 and 351 in 51
increments was investigated [9]. At 01, the angle required for
maximum power, the turbines with 5 m diameter blades produce
5 kW power at a wind speed of 10 m/s. At this pitch, initiation is
characterised by a long “idling period” in which the blade's
angular velocity increased only slowly because of the very high
angles of attack. Increasing the pitch angle of the blades resulted
in excessive rapid starting, decrease in the maximum angular
acceleration, and a better agreement between calculation and
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822 813

Fig. 24. Distribution of pressure coefficient for different angle of attacks [24]. (a)  81 down, (b)  41 down, (c) 01, (d) 61 up, (e) 81 up, and (f) 131 up.
814 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822

Fig. 24. (continued)

measurement. The pitch angle that gives the best starting beha- The starting cut-in wind speed of two different small wind
viour was 201, suggesting that any pitch adjustment mechanism turbines (5 kW [25] and 600 W [26]) were estimated based on
should operate only over that range. Chord-pitch integral by field test [12]. Comparison of field test
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822
measurement with predicted cut-in wind speed in Eqs. (3) and (4)
shows that predicted results exceed over the actual cut-in wind
speed. The high angle of attack occurs at starting, the lift and drag
is predicted by using Eq. (5). The insensitivity of pitch angle and
short of information is the causes for overestimation. The chord–
pitch integrand (Icp) values are higher in root and lower in tip
region, hence it is clear that, high solidity near the hub contributes
to generate starting torque and power producing torque comes
from near the blade tip in Fig. 25.
The starting performance of a three-bladed, 600 W, 2 m dia-
meter horizontal axis wind turbine was investigated and validated
through the measured field test and compared with calculations
employing a quasi-steady blade element analysis [27]. During the
idling period the Re is very low, α is higher than 401 and the hub
region generates more torque than the blade tip. The blades
started rotating at a wind speed of 4.6 m/s on average but they
varied between 2.5 and 7 m/s and generally coincided with
increasing wind speed in as shown in Fig. 26. Besides, most
starting torque is generated near the hub, and most power
extracting torque comes from the tip region [26,30], so it should
be possible to optimize starting performance while maintaining
good power performance as shown in Fig. 27.
Fig. 25. The chords pitch integrand of two small wind turbine blades [12].
Fig. 26. Predictions of (a) Reynolds number and (b) angle of attack variation for low wind speed start [27].
815
Performance improvement of small wind turbine at low wind
speed, pitch of the blades varied over a range of 151, 181 and 201 to
study the performance and the start-up wind speed. It was found
that the turbine performed best at 181 pitch angle [28]. The chord
distribution of the blades resulted in the outer portion to have
higher solidity compared to the baseline blades to have a fast start
up and low cut-in wind speed. The cut-in wind speed based on
10 s averaged data at the optimum pitch angle was 3.24 m/s
whereas the instantaneous cut-in wind speed was 2.34 m/s,
compared to the cut-in wind speed of 3.58 m/s measured for the
rotor with the baseline blades as shown in Fig. 28.
4. Effect of chord and twist distribution on performance of
SWT rotor
The design and fabrication of small wind turbine blade with
mixed airfoils (NACA63-621 and FX66-S-196) applied on inboard
and outboard region based on their aerodynamic characteristics
were done [29]. Rotors chord (thickness) distribution and twist
distribution were calculated by Eqs. (6) and (7) on the blade design
and the two extreme thickness of the blade were 210 mm at the
root, and 25 mm at the tip where thickness distribution was linear
between these two values. Also, from Eq. (7) the pitch angle at the
tip was between 01 and 21 and root was 221 in Fig. 29.
In pitch regulated wind turbines, the performance of this
designed blade can only produce power up to rated power. Beyond
this range the performance of the blade is controlled by control
Fig. 27. Prediction of torque generated by inner and outer blade elements for low
wind speed start [27].
816 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822

system which literally dumps all excess power.The rotor blade can
produce a wide range of power outputs within the same wind
speed class, which is largely due to the production history. After
cut-in, the rotor will produce more power following an increase in
the wind speed. This will always produce less power than
expected for the given wind speed class. However, after a while
in operation in high winds, the rotor will produce more power
than expected for that wind speed class shown in Fig. 30. This
explains the large scatter of data from cut-in to rated values. The
maximum power coefficient of the rotor blade Cpmax was 0.412 and
was considered extraordinary, being the first prototype designed
and manufactured in Jordan.
For small wind turbine systems with multi-purposes such as
quiet running, safety, and simple installation, the rotor blade
consisting of NACA2404 airfoil with fan wing configuration,
500 mm rotor diameter and 181 pitch angle of rotor was developed
[2]. The evaluations of the energy output, turbine speed, power
coefficient, and torque of turbine were also taken along for a wide
range of free stream velocity. It was confirmed that mF500 shows a
good efficiency in the wind range of 8–U–12 m/s. The net effi-
ciency and power coefficient were 0.25 and 0.36 in average,
respectively. Therefore, the integrated system with μF500 should
be available to use for wind power resource as it shows a good
performance with the low tip speed ratio. The maximum power
coefficient was about 0.40 when the tip speed ratio was 2.7 as
shown in Fig. 31. Therefore it is advantageous to introduce this
system into the life space or near urban space. Compared with the
other commercial turbines, the performance excelled at a low
wind speed.
The rotational rates and power coefficients of small wind
turbine blade fabricated with NACA airfoils such as NACA 0012,
NACA 4412, NACA 4415, and NACA 23012 profiles, containing
310 mm diameter and made from “Balsa Wood” was developed
[30]. Blades on the model rotors have some properties such as
blade numbers, twisting angle, and blade angle to hub. Varying
these properties may produce 180 different combinations for each
rotor. Rotation rates for each rotor form were determined based on
wind speed. This investigation indicated a strong correlation
between rotor rotation rate and blade angle, between power
coefficient and blade angle, and between power coefficient and
rotor blade number. An increase in wind speed rate resulted in a
higher correlation between rotor rotation rate and wind profiles,

Fig. 28. (a) Average power output of the turbine at various wind speeds and pitch angles. (b) Average power coefficient of the turbine as a function of wind velocity at
various pitch settings [28].

Fig. 29. (a) Blade thickness distribution and (b) graph of the zero lift line twist [29].
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822 817

between rotor rotation rates and wind speed, between power

coefficient and blade profiles, and between power coefficient and
blade twisting. Rotor models manufactured by using NACA 4412
profiles with 0 grade twisting angle, 5 grade blade angle, double
blades had the highest rotation rate, compared with those man-
ufactured by using NACA 4415 profiles with 0 grade twisting angle,
18 grade blade angle, 4 blades which had the highest power
coefficient. Rotor models gave a rotation rate up to 3077 rpm, with
a power coefficient rate up to 0.425.
Jialin et al. [31] investigated to improve the low-speed wind
turbine blade performance by using NREL's low speed airfoils S822
and S823, which are used on tip and root of the rotors respectively.
Radial variation of the chord length in optimized blade is
smaller than the original blade in blade root area. Twist angle of
the optimized blade radically declines like a curve whereas the
initial one contains inflexion line. These changes are reflected
in the modelling on the root area and the optimized blade
consists of softer exterior curve. The root area of optimized blade
changes from flat surface to circular surface, which improves the

Fig. 30. (a) Power–wind velocity curve [29]. (b) CP–λ for WEC [29]. Fig. 33. Comparison of performance of the optimized wind rotor with the original [31].

Fig. 31. (a) Variation of power coefficient with tip speed ratio. (b) Net efficiency characteristics of microwind turbine with wind velocity [2].

Fig. 32. Comparison of optimized chord distribution and twist of the blade with the original [31].
818 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822

aerodynamic performance of the blade. Twist angle of optimized

blade is vaguely outsized than initial blade at middle and tip as
shown in Fig. 32. The genetic algorithms, used to maximize the
chord length and installation angle of the blades are based on the
BEM theory.
The optimized rotor's power coefficient is improved from the
original rotor in each case of tip speed ratio. Remarkable changes
occur above the tip speed ratio of 6, which means that when the
wind speed is less than 5 m/s at 50 rpm, the aerodynamic
performance obviously improved by optimal blade as shown in
Fig. 33. The power coefficient of the optimized wind rotor blade is
0.43 higher than original blade.
Gómez et al. [32] designed a 3.5 m two bladed rotors which
consist of NACA5317 airfoil to maximize the power coefficient
when it is operating at variable speed based on BEM method. The
geometric parameters obtained from design are in a good range
for manufacturing. The pitch angle has a net variation of 13.31, and

Fig. 34. Geometry parameters of the blades [32].

Fig. 35. Variation of power coefficients in terms of tip speed ratio [32]. Fig. 37. Distribution of chord length and twist angle [21].

Fig. 36. (a) Flap wise and edgewise taper distributions along the rotor radius. (b) Modified blade twist distribution compared with twist of zero life line equation with k¼ 0.5 [28].
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822 819

it is almost linear in Fig. 34. The rotor performance is in the 65% refers to that outer portion of blade which furnishes more
acceptable range according to existing wind turbines, the max- power.
imum power coefficient being 0.5. Therefore it is used to reduce The new Phase VI non-twisted blade works only 50% near to
the complication of SHAWT rotor fabrication shown in Fig. 35. optimum (CL/CD)max, although remaining half is away from opti-
A 2-bladed rotor was designed to work with the Air-X wind mum as shown in Fig. 38. It shows modified WT is generally not as
turbine to operate in low wind speed ranges of 3–6 m/s which good as the original WT. The new wind turbine blade is subdivided
consists of AF300 flat back airfoil, specially designed to avoid into two parts, each non-twisted, with a variable chord along the
laminar separation bubbles to achieve high lift at low Reynolds radius, a particular pitch angle, and a winglet to connect the two
numbers and provide structural stability to the blades, thus parts. The optimum choice of pitch angle provides the best blade
improve the overall performance of the turbine [28]. The taper aerodynamic condition and maximizes the power coefficient. The
and twist were incorporated to the low Re AF300 airfoil section new blade reveals some energy loss due to the tip vortices of each
based on BEM theory Eqs. (5) and (6), to increase the aerodynamic blade part (which can be minimized by winglets), yet proves that
performance at low wind speeds in Fig. 36. The 2-bladed rotor it is possible to create a wind turbine with high power coefficients
achieved CP values of 0.l, 0.217 and 0.255 at the wind speeds of 4, as shown in Fig. 39. It offers optimum aerodynamic flow, and
5 and 6 m/s respectively whereas the baseline 3-bladed rotor presents a power output gain in all the wind-speed ranges (from
achieved 0.052, 0.112 and 0.15 at these wind speeds. Peak power cut-in to cut-out velocity).
coefficient attained by the 2-bladed rotor design at 6 m/s wind
speed was 0.29 as shown in Fig. 27.
In [21], for easy blade pitch angle control authors have tailored
the chord length distribution by decreasing its size and weight
6. Noise reduction in small wind turbine rotor
while attempting to maintain the high blade power coefficient.
The blade shapes have been modified based on BEM theory that is
Noise measurement investigation was carried out on small
blade chord length of the thin airfoil becomes shorter than that of
wind turbine for the capacity of 5 kW with 2.5 m long blade at the
optimum design. The chord length of the tailored thin airfoil blade
wind speed range and angular velocity respectively 4–10 m/s and
is 230 mm from the blade tip to r ¼1.6 m, 190 mm at cut off
400 rpm. The investigation demonstrated that turbine produces
r ¼0.4 m, and then linearly increases up to 230 mm at r ¼1.6 m.
half the sound power of human voice at 5 m/s and noise output is
The thickness has been modified from 8.5% of chord at the blade
apparently close to that predicted by simple data correlations
tip to 18.5% of chord at the cut off in order to keep the stiffness of
obtained from larger turbines. These data correlations in Eqs. (8)
the blade. The optimum twist angle of blade was from 341 to 71 at
and (9) were used to identify the optimum from the proposed sites
blade tip while the relative twist angle was 271 as shown in Fig. 37.
for installation of small wind turbines [16].
In the case of average wind speeds of 10 m/s and a maximum of
19 m/s, the automatically controlled wind turbine ran safely in
rough wind conditions and showed an average generator output of
1105 W and a power coefficient 0.14.

5. Non-twisted-double pitches rotor

A new cost-efficient wind turbine blade with high power

coefficient was designed by Lanzafame and Messina [33] from
Phase VI rotor with S809 aerodynamic airfoil configuration of two
blades and rotor diameter of 10.6 m (hub radius equal to 1.258 m)
and pitch angle is 31. They optimized the rotor with absence of
twist carrying a variable chord along the blade.
The new and simple approach which avoids the twist along the
radius of the blades, while design of non-twisted blade, the
aerodynamic performance losses occurred. In the original blade
65% of blade performance works near to the (CL/CD)max values, and
the remaining 35% was less far. Moreover 35% of the power loss Fig. 39. Performance comparison between original, non-twisted, and the newly
occurs near the hub, which causes some power loss and inversely proposed WTB [33].

Fig. 38. S809 airfoil polar curve: Phase VI and Phase VI—not twisted WTB and double pitch operational range at V0 ¼ 8 m/s [33].
820 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822

Fig. 40. (a) Optimal chord and (b) twist distribution from altering the design CLmax and αmax in the BEM optimum rotor [34].

Table 3
CP and SPL values for the rotors in Fig. 24 [34].

Retip CP SPL (dBA)

BEM CL ¼ 1.20, α ¼41 E 280,000 0.500 53.39

BEM CL ¼ 0.86, α ¼11 E 390,000 0.495 51.39
DE an ¼0.25 E 400,000 0.497 51.44

Numerical optimization method called differential evolution of

noise reduction with retention of power performance for small
wind turbine was achieved with the help of empirical noise
prediction model, coupled with a blade element momentum
method [34]. Optimization of chord and twist of blade was
implemented for 5 kW small wind turbine blade at tip speed ratio
of 5.5 as shown in Fig. 40. The 2 dBA drop in the total sound
pressure level was to be expected for a 1% drop in CP. The noise
reduction was achieved as a result of a decrease in angle of attack
and increase in Reynolds number. As extension of this optimiza-
tion, both starting time and noise levels were simultaneously Fig. 41. Sensitivity to tip shape and trailing edge bluntness h [34].
reduced in SPL and TS of 4% (2 dBA) and 6% or reduction in SPL and
TS of 1% (0.7 dBA) and 16% respectively for a maximum drop in CP
of 1% as shown in Table 3. Important noise source of small wind are the most critical locations particularly for aerodynamic and
turbines is turbulent boundary layer trailing edge noise, which aero-acoustic concerns, respectively. The difference between the
coincides with large turbines that depends on trailing edge baseline and optimized versions of relatively thinner airfoils are
thickness and tip shapes. Aerodynamic noise of small wind turbine not seen significantly because the geometrical changes made on
also depends on the tip shape and trailing edge thickness. Sound those thin profiles are not allowed to be substantial. However,
pressure level is significant with trailing edge thickness with an even for the thin profiles, the improvement in noise and perfor-
increase from 1 mm to 3 mm whereas noise increased by 2 dB and mance characteristics is clearly shown. It can be seen in Fig. 42
negligible with tip shape as shown in Fig. 41. Tip shape can be with the geometrical optimization applied, the lift to drag ratios
allowed to be designed with maximum power extraction or are increased whereas the noise emission levels are decreased up
manufacturer's point of view and trailing edge should be in to 5 dB. Each of the optimized airfoils has been shown to have
reasonable thickness. better aerodynamic performance and lower noise emission than
Six airfoils (FX 63-137, S822, S834, SD2030, SG6043 and their baseline profiles shown Fig. 43.
SH3055) which are widely deployed on small scale wind turbines
are augmented with a view to reduce noise emission and increase
the aerodynamics performance criteria by adjusting the shape of 7. Conclusion
the airfoil and the numerical computations are performed for a
typical 3 bladed, 10 kW wind turbine which has a radius of 7 m The performance improvement of wind turbine was achieved
[35]. In order to achieve that goal, the flow analysis tool XFOIL and by implementation of various concepts by the researchers. The
airfoil self-noise analysis tool NAF Noise have been used. While new approach on airfoil geometry and wind turbine blade was
changing the geometry of the airfoils, especially the pressure side performed and the small modifications on small wind turbine
and the trailing edge characteristics have been modified since they blades which are working at less than 500,000 Reynolds number is
 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822 821

Fig. 42. (a) CL/CD ratio vs. angle of attack at Re¼ 300,000 and (b) SPL spectra [35].

Fig. 43. Six small wind turbine airfoil baselines and modified for noise emission characteristics [35].

Table 4
Improvement of aerodynamic performance on small wind turbine blade.

Element Parameters Result and remarks

Airfoils Trailing edge  Thinner with curved surface [35]

▪ Lift and drag ratio increases
▪ Noise emission reduces up to 5 dB.

 Flat back [20]

▪ structural strength rises
▪ reducing material cost
▪ lowering cut-in wind speeds
▪ wider range of angle of attack.
▪ CLmax ¼1.72, 1.81and 1.86 correspondingly for the AOA 141 at Re¼75,000, 128,000, and 205,000

Camber line  Pressure load inverse method [10].

▪ high maximum lift and a moderate drag

 On thickness for viscous flow

▪ reduces the drag without changing the lift

 Thin airfoil with shorter chord length [21]

▪ Increasing the performance
▪ Decreasing size and weight
▪ High blade power coefficient of 0.14
▪ Easy pitch angle control.

 Seagull airfoil is suitable for especially in blade tip [22].

822 N. Karthikeyan et al. / Renewable and Sustainable Energy Reviews 42 (2015) 801–822

Table 4 (continued )

Element Parameters Result and remarks

Blade Chord, twist distribution and  Best starting behaviour at 201, pitch adjustment mechanism operates above this range [9]
pitch angle  While designing
▪ Starting performance depends on hub region.
▪ Maximizing the power depends tip region [26,27].

 Best performance of AF300 with 2 bladed at 181, Cpmax of 0.29 at 6 m/s over 3-bladed rotor [28].
 NACA 63-621(root) & F66-S-196 (tip) airfoils with 5 m long blades gives best performance at 161 twist angle rotor power
output of 19.03 kW at 10 m/s and Cpmax ¼ 0.412 [29]
 Two bladed with NACA4412 achieve highest rotation rate of 3077 rpm.
 Four bladed with NACA4415 profile 181 of pitch achieves power coefficient up to 0.425 [30]
 Circular root area with radial change of twist in rotor achieves Cpmax of 0.430 [31]
 NACA 5317 with 3.5 m 2 bladed rotor Linear twist of 13.31 achieves maximum power coefficient of 0.5 [32]

Fan wing shape  NACA2404 airfoil achieves net efficiency and power coefficient is 0.25 and 0.36 in average Cpmax ¼0.40 at λ ¼ 2.7
 wide wind range of 8–12 m/s [2]

Double pitch  Non-twisted blade of two parts connected by winglets with optimized pitch angle blades performance is almost equal to
twisted blade [33]

summarized in Table 4. It is inferred that modifications of airfoil
trailing edge, thickness, and camber line have a significant effect
on the performance of noise, starting characteristics and lifts and
drag ratios of airfoil. The blade parameters such as pitch angle,
chord and twist distribution have also been discussed.
References
[1] Cooper P, Kosasih PB, Ledo L. Roof mounting site analysis for micro-wind
turbines. Renew Energy 2010;36:1379–91.
[2] Hirahara H, Hossain MZ, Nonomura Y. Testing basic performance of a very
small wind turbine designed for multi-purposes. Renew Energy
2005;30:1279–97.
[3] Fasel HF, Gross A. Numerical investigation of different wind turbine airfoils.
Tucson, Orlando, FL: The University of Arizona; 2011.
[4] Tangler W, Van Rooij R. Summary of the Delft University wind turbine
dedicated airfoils. J Sol Energy Eng—Trans ASME 2003;125(4):488–96.
[5] Giguere P, Selig MS. Low Reynolds number airfoils for small horizontal axis
wind turbines. Wind Eng 1997;21:367–80.
[6] Giguere P, Selig MS. Aerodynamic effects of leading-edge tape on aerofoils at
low Reynolds numbers. Wind Energy 1999;2:125–36.
[7] Clausen PD, Wood DH. Research and development issues for small wind
turbines. Renew Energy 1999;16:922–7.
[8] Lissaman PBS. Low-Reynolds-number airfoils. Annu Rev Fluid Mech
1983;15:223–39.
[9] Mayer C, Bechly ME, Hampsey M, Wood DH. The starting behaviour of a small
horizontal-axis wind turbine. Renew Energy 2001;22:411–7.
[10] Henriques JCC, Silva M, Estanqueiro AI, Gato LMC. Design of a new urban wind
turbine airfoil using a pressure-load inverse method. Renew Energy
2009;34:2728–34.
[11] Lanzafame R, Messina M. Fluid dynamics wind turbine design: critical analysis
optimization and application of BEM theory. Renew Energy 2007;32:
2291–305.
[12] Wood DH. A blade element estimation of the cut-in wind speed of a small
turbine. Wind Eng 2001;25(4):249–55.
[13] Habali SM, Saleh IA. Local design, testing and manufacturing of small mixed
airfoil wind turbine blades of glass fiber reinforced plastics part I: design of
the blade and root. Energy Convers Manag 2000;41:249–80.
[14] Lawson MV. Applications of aero-acoustic analysis to wind turbine noise. In:
Clayton BR (Ed.), Proceedings of the 14th BWEA conference on wind energy
conversion. London: M.E.P.; 1992. p. 91–9.
[15] Wagner S, Baresis R, Gidati G. Wind turbine noise. Germany: Springer Verlag;
1996.
[16] Wood DH. Noise measurement and prediction for small wind turbines. In:
Proceedings of solar '97.paper 153. Australian and New Zealand Solar Energy
Society; 1997.
[17] Selig MS, McGranahan BD. Wind tunnel aerodynamic tests of six airfoils for
use on small wind turbines. J Sol Energy Eng-Trans ASME 2004;126
(4):986–1001.
[18] Kogaki Tetsuya, Matsumiya Hikaru, Kieda Kaori, Yoshimizu Naofumi, Yama-
moto Yuusuke. Performance improvement of airfoil for wind turbine by Vortex
Generator. Wind Energy 2002;28(1):73–6.
[19] Bao Nengsheng, Ma Haoming, Zhiquan Ye. Experimental study of wind turbine
blade power augmentation using airfoil flaps, including the gurney flap. Wind
Eng 2000;24(1):25–34.
[20] Singh RK, Ahmed MR, Zullah MA, Lee YH. Design and testing of a low Reynolds
number airfoil for small horizontal axis wind turbines. Renew Energy
2012;42:66–76.
[21] Kazumasa Ameku, Nagai Baku M, Roy JitendroNath. Design of a 3 kW wind
turbine generator with thin airfoil blades. Exp Therm Fluid Sci 2008;32:
1723–30.
[22] Limin, Qiao, Wei Sidong, Gu, R.u.i., Quan, Xiaolin, Yang, Yingjun. The investi-
gation of the airfoil for the small wind turbine based on the seagull airfoil.
Power and Energy Engineering Conference (APPEEC), Asia-Pacific 2011.
[23] Wata J, Faizal M, Talu B, Vanawalu L, Sotia P, Ahmed MR. Studies on a low
Reynolds number airfoil for small wind turbine applications. Sci China Technol
Sci 2011;54(7):1684–8.
[24] Ji Yaoa, Weibin Yuanb, Wanga Jianliang, Jianbin Xiec, Zhoub Haipeng, Peng
Mingjun, et al. Numerical simulation of aerodynamic performance for two
dimensional wind turbine airfoils. Procedia Eng 2012;31(1):80–6.
[25] Bechly M.E., Clasen P.D., Ebert P.R., Pemberton A., Wood D.H. Field testing of a
prototype 5 kW wind turbine. In: Anderson MB (Ed.), Proceedings of the 18th
BWEA conference on wind energy conversion. London: M.E.P.; 1996. p. 103–
10.
[26] Wood DH, Robotham T. Design and testing of high performance blades for a
600 W horizontal axis wind turbines. In: Proceedings of the first Australian
wind energy conference, Newcastle, June 28–30 1999. p. 91–6.
[27] Wright AK, Wood DH. The starting and low wind speed behaviour of a small
horizontal axis wind turbine. Wind Eng Ind Aerodyn 2004;92:1265–79.
[28] Singh RK, Ahmed MR. Blade design and performance testing of a small wind
turbine rotor for low wind speed applications. Renew Energy 2013;50:812–9.
[29] Habali SM, Saleh IA. Design and testing of small mixed airfoil wind turbine
blades. Renew Energy 1995;6(2):161–9.
[30] Vardar Ali, Ilknur Alibas. Research on wind turbine rotor models using NACA
profiles. Renew Energy 2008;33:1721–32.
[31] Zhang Jialin, Zhou Zhenggui, Lei Yansheng. Design and research of high-
performance low-speed wind turbine blades. In: Proceedings of the World
Non-Grid-Connected Wind Power and Energy Conference (WNWEC) 2009.
[32] Carlos Gómez G, Bulmaro Macías G, Honorato Azucena C, Amparo Palomino
M. Design of a 3.5 m s rotor two bladed horizontal axis wind turbine. In:
Proceedings of the 20th International Conference on Electronics, Communica-
tions and Computer (CONIELECOMP); 2010.
[33] Lanzafame R, Messina M. Design and performance of a double-pitch wind
turbine with non-twisted blades. Renew Energy 2009;34:1413–20.
[34] Clifton-Smith MJ. Aerodynamic noise reduction for small wind turbine rotors.
Wind Eng 2010;34(4):403–20.
[35] Göçmen Tuhfe, Barıs Özerdem. Airfoil optimization for noise emission
problem and aerodynamic performance criterion on small scale wind turbines.
Energy 2012;46:62–71.