Accepted Manuscript

Effect of winglets induced tip vortex structure on the performance of subsonic wings

Gautham Narayan, Bibin John

PII: S1270-9638(16)30556-9
DOI: http://dx.doi.org/10.1016/j.ast.2016.08.031
Reference: AESCTE 3760

To appear in: Aerospace Science and Technology

Received date: 22 September 2015

Revised date: 26 July 2016
Accepted date: 28 August 2016

Please cite this article in press as: G. Narayan, B. John, Effect of winglets induced tip vortex structure on the performance of subsonic
wings, Aerosp. Sci. Technol. (2016), http://dx.doi.org/10.1016/j.ast.2016.08.031

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing
this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is
published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 Revised manuscript Submitted to
Aerospace Science and Technology

Effect of winglets induced tip vortex structure on

the performance of subsonic wings
Gautham Narayan∗ and Bibin John∗∗
∗Undergraduate student, ∗∗ Associate Professor
School of Mechanical Engineering
VIT University, Vellore, Tamil Nadu, India-632 014
Tel.: +91-9092512690
e-mail: bibin.john@vit.ac.in

Key words: Multi tip Winglet; Aspect ratio; Lift induced Drag; Tip vortices.

Abstract
This paper presents the comparative study of the effectiveness of three different winglet designs
in reducing lift induced drag by changing the number of vortices and vortex distribution at the
wingtip and correlating it to the aerodynamic characteristics of the baseline wing. The best three
winglet geometries are appended to wings of different aspect ratio to further study their effec-
tiveness in reducing drag and augmenting the lift coefficient. Computational simulations were
performed on Ansys Fluent V15 using the Reynolds Averaged Navier-Stokes equations coupled
with the K − ω SST turbulence model to study the three dimensional flow and vortex structure
about the half wing. The simulation shows that there is a strong relationship between the size
of the tip vortex and the aerodynamic parameters such as lift, drag and pitching moment of the
wing. The multi tipped wing is the most effective at dispersing the vortice energy and reducing
the induced drag. Optimisation of the number of tips was found to be crucial to increase the lift
coefficient while reducing the contributions of frictional and vortex drag. Although the lift was
found to be increasing with the number of tips, the rise in frictional drag due to wetted surface
area is a limiting factor towards aerodynamic efficiency. The results show that the multi-tip-4 is
most capable at increasing the lift coefficient, but is surpassed by multi-tip-3 in lift to drag ratio.
As aerodynamic efficiency is key for improving flight range and duration, it is concluded that
multi-tip-3 is the optimum winglet for the given design conditions. The baseline wings show
maximum performance at higher aspect ratios. Overall, winglets are found to be more effective
at lower aspect ratio and provide highest improvement in aerodynamic efficiency at a moderate
aspect ratio of 10.
 Gautham Narayan and Bibin John

1 Introduction

The lift generated by a wing can be explained using the Coanda effect [2], the jet of air that
flows around a circular or cylindrical profile follows the curved profile until subjected to an
adverse pressure gradient in which case flow would separate. The flow acceleration imparts
momentum to the flow and reduces the pressure on the top surface. This imparted momentum
has components in the longitudinal and normal directions classically known as lift and drag.

The three dimensional lift distributions across a wing can be explained using the Prandtl
lifting line theory [1], which says the lift generated by the wing can be computed by integrating
the circulation along the span of the wing. Due to the finiteness of the wing, the circulation
produces three dimensional effects at the wingtips known as tip vortices as in Fig. 1. These
vortices give rise to the vortex drag, lift induced drag. The lift induced drag is a by-product of
downwash which dissipates momentum to the air, resulting in a lifting force. Due to this three
dimensional effect there is deviation from Prandtl’s the optimum elliptical lift distribution. Thus
the wing region at the tip does not generate much lift, rather it leads to reduction in the total lift
produced and subsequent drop in the span efficiency factor.

One of the most efficient way to reduce wing tip vortices is by adding winglets capable of
preventing the flow on the upper surface of wing from leaking over. This would reduce the in-
tensity of wing tip vortices and the corresponding lift induced drag. However the aerodynamic
efficiency improvement with the integration of such wing tip devices mainly depends on the
configuration of such devices and overall size of the wing. The earliest research on the appli-
cability of winglets in reducing the lift induced drag can be seen in the works of Whitcomb
[5], who experimentally investigated the aerodynamic efficiency of a wing tip sail. Significant
improvement in L/D ratio with the prevention of tip cross flow has been noticed in his studies.
Computational studies of Heyson et al. [6] were focusing on the comparison of aerodynamic
advantages of large number of tip extensions and winglet shapes. From this study it has been
observed that winglets offer improved performance over tip extensions for the same bending
moment created due to the wing modification. In 1978 Montoya, Flechner and Jacobs [7] car-
ried out experimental investigations with almost similar objectives as that of Heyson et al. Their
wind tunnel measurements on four different configurations showed that relative aerodynamic
gain to structural weight penalty is more in case of winglets while comparing with the same
parameter of wing extension. In the year 1992, Gratzer[8] patented his novel design of Spiroid-
tipped wing. Gratzer claims that the winglet design put forth by him has superior performance
in reducing the lift induced drag. Moreover in this study it is also reported that spiroid-wing can
even eliminate the tip vortices completely. Studies of Kravchenco[9] revealed that the winglets
can enhance the aerodynamic benefits, but at the cost of additional structural problems. Through
the study author has demonstrated that the incorporated winglets enhances the bending stress at
the wing root. After Kravchenco’s work, many researchers [22, 11] dedicatedly carried out re-
search to reduce induced drag by means of wing tip devices. The evolution of high performance
computing capabilities has greatly accelerated the research work in this field. Aerodynamic
shape optimization for induced drag reduction can be seen in the works of Yamazaki et al. [12],
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

Liersch et al. [13] and Takenaka et al.[14]. The multidisciplinary optimization study of Take-
naka et al. concentrated on designing an optimized winglet for a commercial jet aircraft. High
fidelity computational fluid dynamic solution method, finite element based weight calculation
and multi-objective genetic algorithms were integrated together to perform this optimization
study. As the outcome of this optimization study, Takenaka proposed a blended winglet design
with a leading edge aft of the reference wing leading edge. Span length and cant angles were
found as the dominant parameters that control the drag reduction. Later Bourdin et al. used
winglets with variable cant angles to study the capability of winglet in controlling the aircraft
[17]. This investigation showed that the pair of winglets with adaptive cant angle can provide a
coupled control mechanism. Ning and Kroo[19] compared tip extension, winglets and C-wings
and showed that Winglets wing offer significant drag reduction at higher lift coefficients than tip
extensions. Computationally, Toor et al. [15] investigated the effects of varying the winglet pa-
rameters on overall winglet performance. It was found that winglet effectiveness decreased with
increasing cant angle and remained fairly constant for all winglet root toe angles. For different
sweepback angles and taper ratios, winglet performance increased upto an optimum value, after
which it started to decrease. The taper ratio of the wing was found to have the highest variation
on lift to drag ratio, close to 7%, while Toe angle has the least variation upto 2% on lift to drag
ratio.
Efforts for implementing multi-winglets for induced drag reduction have been noticed in the
work of Smit et al. [16]. In this study the authors performed analysis of the flow over a rectan-
gular wing constructed using NACA 0012 aerofoil fitted with number of flat plates as winglets.
Out of different winglet configurations tested in this study a few were found to be useful in in-
creasing the L/D ratio. In one of the latest studies in this field Guerrero et al. [18] re-investigated
the spiroid winglet numerically. from this study it has been noticed that, integration of spiroid
winglet to a clean wing reduces lift induced drag by 75% at maximum lift generation. De-
layed wing stall and improved post-stall performance were also observed for wing with spiroid
winglet. The investigation of Joel et al. [20] focused on finding an optimized blended winglet
configuration for the KC-135R aircraft capable of offering an increase in endurance and po-
tential range of 5.62% and 3.55% respectively, and a raked wingtip with 20 degree leading
edge sweep that showed an increase of 8.32% and 4.69% respectively. Due to this the KC-135
aircraft observed a significant decrease in fuel burn by up to 8%. In a recently published ar-
ticle, Panagiotou et al. [21] presented a winglet optimization procedure for unmanned aerial
vehicles (UAVs). The flow around several winglet configurations has been investigated by solv-
ing Reynolds averaged Navier-Stokes equations along with Spalart Allmaras turbulence model.
They analysed aerodynamic parameters along with vorticity contours and tip vortex images to
determine the optimum winglet design which had a cant angle 50 degrees and corresponded to
a maximum L/D ratio of 22.85. They also concluded that the improvement in L/D is caused
mainly by the increase in lift force, than by the decrease in drag force and that the new winglet
had no effect on the stall angle. They could gain 10% enhancement in total flight time by incor-
porating the optimized winglet geometry.

Investigations focused on the gain of aerodynamic advantages through implementation of

multi tipped winglets are also available in the open literature. Most of such experimental and
 Gautham Narayan and Bibin John

Figure 1: Formation of tip vortices due to secondary flows.

numerical studies were devoted to the analysis of vortices behind multi tipped winglets and
discussion of the orientation of those vortices with respect to the incoming flow. One of the
earlier research on multi tipped winglet is the one reported by Roche and Palffy [22]. In this
study, with the aid of Prandtl-Munk linear model authors showed that a wing grid can modify
the planar vortex sheet in the streamwise direction by preventing roll up. Comparison of induced
drag reduction caused due to introduction of wing grid and split wing are also reported in this
paper. Later Catalano and Ceron [23] used a NACA 653018 airfoil to show the comparison
between various configurations of a three tipped multi winglet for drag reduction using hot
wire anemometry. Experimental studies of Luis et al. [24] were focusing on the design of an
adaptive wingtip that is capable of improving take-off parameter and stall speed by 20% so as
to reduce the runway requirement. Potential improvements with the use of multi-winglets were
presented by Cosin et al.[25]. Almost 7.3% gain in the maximum aerodynamic efficiency has
been observed with the integration of this kind of wingtip device. Also significant improvement
in stall characteristics and reduction in structural loading were noticed.
Recently Reddy et al. [26] performed a multi objective optimization to study the effectiveness
of parameterized multi tipped winglet geometries for different angles of each tip while taking
into account the entire realistic 3D winglets-wing-body-tail configuration. They have presented
optimised multi-winglet flow simulation for low (take-off) speeds over a range of angles of
attack. Results indicate a 12.8% increment in lift, 4.5% reduction in drag and 60% reduction in
pitching coefficient while using a Pareto front optimization technique. Work of Gavrilovic et al.
[27] is the latest published article in this field that deals with the design optimization of winglet
geometry. They have used Fluent to analyse around 100 winglet shapes to create a response
surface. Non-Dominated Sorting Genetic algorithm has been used to arrive at pareto optimal
solutions of lift and drag. Almost 15% improvement in lift-to-drag ratio and 7.1% enhancement
in range factor have been claimed for the optimized winglet design. It also demonstrates in an
unbiased manner that several concepts for winglets are inferior, despite the fact that they were
promising when first published.

Over the past five decades substantial amount of research has occurred in this field [4][6]
[7][9]. Existing literature in this field explains the performance characteristics of conventional
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

winglet geometries and their optimization towards lift and drag [5][20]. Many research groups
around the globe have proposed different winglet designs and each of them has claimed superior
performance of their designs [8][22]. Multi tipped winglets were also investigated by many
researchers [16][23][26]. However there have been no studies undertaken to understand the
consequences of having more than or less than three individual tips interacting with the flow. In
cases where a certain parameter such as maximum lift is of prime importance, more number of
wing tips have shown to be extremely beneficial in reducing the vortex strength as discussed in
this study.

Therefore the objective of this paper is to present a fundamental study correlating the num-
ber vortices, vortice distribution and their vorticity with the standard aerodynamics coefficients
such as lift, drag, lift drag ratio and pitching moment. In order to quantitatively describe the
effect of multiple vortices generated from the multi tipped winglet as opposed to conventional
winglets, the study has been divided into three stages. In the first stage, the conventional winglet
geometries such as Blended and BMAX winglets are compared with the baseline wing to pro-
vide a basis to quantify the performance gains with such types of winglets. In the second stage,
the focus is shifted towards multi tipped winglets consisting of two to four tips to understand
the interaction between smaller vortices emanating from the individual tips of the winglet in
the presence of the major counter-rotating vortices produced by the wing on the lift and drag
forces. This provides us an overall understanding of the optimum multi tipped winglet that can
augment the lift while maintaining sufficiently low drag for practical applications. In the third
stage, the Baseline wing, Blended, BMAX and the optimum Multi tipped winglet are studied
for their effectiveness when attached to wings of different aspect ratio. The last stage is im-
portant to quantify the physical implications of adding winglets to long range aircrafts such as
sailplanes[11] with high aspect ratio and also for faster low aspect ratio wings used on other
commercial aircraft.

2 Model design and mesh generation

A planar three dimensional wing was modeled using NACA 2412 airfoil [3] with a 3 meter
span, 0.25m chord at the root and 0.2m chord at wingtip in order to achieve high aspect ratio.To
attain the lift requirement ( CL value of 0.2 at a cruise angle of attack of 20 [3]) the wing area has
been taken as 0.6 m2 . This wing section is expected to generate 120N of lift at a flow Reynolds
number of 7 × 105 . The model consists of only the wing section devoid of the fuselage and
tail section. It may be noted that almost all the works in this research field followed the same
strategy for geometry design. The baseline wing is then appended with Blended, BMAX and
Multi tipped winglets described in Fig.2. All the winglet heights have been maintained at 10%
of the semi wingspan for all cases as suggested by [4] for maximum efficiency. The Blended
winglet has been designed with a single vertical tip aligned with the direction the flow while
maintaining a cant angle of 75 degrees based on the whitcomb design [5]. The BMAX winglet
consists of two vertical tips each with an equal cant angle of 65 degrees. This is because the
bottom tip is expected to provide equivalent performance improvement as the upper tip of the
 Gautham Narayan and Bibin John

Figure 2: Different winglet designs 1. Blended (side view and front view) 2. BMAX (side view
and front view) 3. a) Multi-tip-2; b) Multi-tip-3; c) Multi-tip-4

winglet [5]. The multi tip winglets have been named as multi-tip-2, multi-tip-3 and multi-tip-4
to signify the number of the tips present. All three multi tip geometries have tips between a
lower cant angle of 10 degrees and upper angle of 60 degrees in order to maintain a constant
sweep area for quantitative comparison. All the winglets and their individual tips are modeled
using the NACA 2412 airfoil while maintaining a zero toe angle with respect to the oncoming
flow. This ensures that all the winglets have similar geometric parameters which help to quantify
their performances in the later stages of the study.

The mesh generation for the present computational domain considers tetrahedral elements
that are refined from an initially coarse mesh of 0.15 million elements to a final mesh of 2 mil-
lion elements. The surface mesh over the wing section is shown in Fig.3. The mesh refinement
has been carried out based on a grid independent study while iterating the solution until all the
forces are well captured and there is negligible improvement in convergence on further refin-
ing the mesh. The evidence for the grid independent nature of the solution can be seen from
Fig. 4. It has been observed that domain with elements of 1.6 million or above gives a change
in drag coefficient only in fourth decimal. So a domain with 1.6 million of elements has been
chosen as the converged grid for the present study. Same kind of mesh independent tests were
carried out for other wing geometries as well. Such grid independent results are presented for
the subsequent discussions.
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

Figure 3: Sample surface mesh over the wing section of a refined computational domain

Figure 4: Variation of drag coefficient with grid refinement

 Gautham Narayan and Bibin John

3 Solver details and solution methodology

Numerical simulation of the external flow aerodynamics has been carried out using commercial
CFD package Ansys Fluent 15. Solution of the present case is sought by solving steady
Reynolds averaged three dimensional Navier Stokes equations. Various RANS CFD models
including Spalart Allmaras, k − , k − ω SST were considered before choosing the k omega
shear stress model. This decision was made by considering the capability of the k-omega model
to capture the turbulence effects in greater clarity. This is due to the accurate prediction of the
higher order stress relaxation terms. The k − ω SST model developed by F.A Menter [28, 29]
uses the more computationally intensive k − ω model in the near field region adjacent to the
wall to accurately capture the near field viscous sub layer. However it uses the k −  model
which is relatively less resource intensive for far field applications thus facilitating better flow
resolution for the computational resources in hand. The mathematical formulation of the two
equation model considered for the present study is given as,

Turbulence Kinetic Energy

 
∂(ρk) ∂(ρuj k) ∂ ∂k
+ = P − β ∗ ρωk + (μ + σk μt ) (1)
∂t ∂xj ∂xj ∂xj

Specific Dissipation Rate

 
∂(ρω) ∂(ρuj ω) γ ∂ ∂ω 1 ∂k ∂ω
+ = P − βρω 2 k + (μ + σω μt ) + 2(1 − F1 )σω2 (2)
∂t ∂xj vt ∂xj ∂xj ω ∂xi ∂xi

The detailed definitions of terms contained in the above equations can be seen in the original
articles [28, 29].

3.1 Boundary conditions

Boundaries of computational domain are extended to six times the model dimensions. The ex-
ternal wing surface is treated as stationary wall with no slip condition. This gives rise to the
log-law of the wall and produces a logarithmic velocity distribution at the viscous sub layer.
Viscous forces originate from the shear stresses in the inflation layer adjacent to the wing sur-
face. The boundaries of the enclosure are treated as walls with zero shear. This further reduces
the computational error due to lack of resources for solving an infinite air domain. The normal
boundary of the flow domain in front of the leading edge of the aircraft wing is taken as velocity
inlet and is assigned with an inlet velocity of 40 m/s, which is standard for an aircraft of similar
wing size[21]. A turbulence intensity of 5% was chosen at STP to include low altitude flight
regimes, which could make the wing more sensitive to stall . At the exit of the flow domain
external air pressure is set to zero gauge pressure.
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

α = 40 α = 80 α = 120 α = 160

Base line design

Blended wing

BMAX
Figure 5: Vorticity contours obtained for Baseline, Blended and BMAX wing designs at differ-
ent angles of attack

4 Results and Discussions

4.1 Comparison between Baseline, Blended and BMAX winglets

Simulations are carried out to study the aerodynamic characteristics of Blended and BMAX
winglets and to compare their performances with baseline wing section. Simulations are re-
peated for different angles of attack uptill 160 before which the wing would stall, to understand
the changes in the tip vortex distribution and its consequence on the aerodynamic parameters.
The vorticity contours obtained over wing surface with and without wing tip devices are de-
picted in Fig. 5. Contours obtained from the numerical investigations on a baseline wing of
aspect ratio 14 indicate that the tip region contains areas of high vorticity that rapidly increases
in size and strength with angle of attack. The addition of a Blended winglet substantially re-
duces the strength of the vortice at the wing tips, while the BMAX winglet introduces two sets
of vortices that interact with the tip flow to show significantly improved lift and drag coefficients
at angle of attack 4 degrees and beyond.

Lift Coefficient: The lift coefficient increases in a linear trend with angle of attack up to 13
degrees as shown in Fig. 6, after which the wing stalls. The cambered nature of the 2412 airfoil
ensures a positive coefficient even at zero degrees angle of attack. The blended winglet has
provided a slightly higher stall angle due to its ability to delay flow separation emanating from
 Gautham Narayan and Bibin John

Figure 6: Variation of Lift coefficient with angle of attack for different winglet shapes

the tip. This is followed by the BMAX winglet which has the highest slope and stall angle.
However, accurate prediction of post stall performance requires transient simulation, which is
computationally intensive and also beyond the scope of this paper. It can be seen from Table.1
that the BMAX winglet offers an 10% increase in lift coefficient as compared to the 3% by the
blended winglet.

Drag coefficient: The influence of wing tip devices with varying vortice distributions can be
quantitatively observed from the coefficient of drag plotted against angle of attack in Fig. 7. The
coefficient of drag increases until the stall angle of attack after which there is a steep increase.
There exists a cross over point for drag curves of BMAX and Blended winglets between 2 and 4
degrees where onwards the induced drag significantly increases as compared to the skin friction.
The Blended winglet is able to maintain a lower CD as compared to baseline at angles greater
than 2 degrees due to the reduction in the tip vorticity. The BMAX winglet has a slightly higher
CD at lower angles due to higher wetted area, but has the least drag at higher angles of attack.
This can be attributed to the fact that there are two separate induced vortices that are of much
lower intensity in case of BMAX winglet. This shows that there is a correlation between the
intensity of the vortice and the vortex drag. The induced vortice by the winglet acts an energy
dissipation device that is able to produce a smaller vortices emanating from the wing tip and
preventing the formation of the main vortex, thereby reducing the vortex drag and the total drag
coefficient.
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

Figure 7: Variation of drag coefficient with angle of attack for different winglet shapes

Although individual variations of lift and drag coefficients for different winglet designs are
analysed, it is the ratio of lift and drag coefficient that signifies aerodynamic efficiency in the
wing design. As the L/D ratio improves the climb speed, take-off and landing performance along
with range. Therefore discussion of CL /CD variation is vital for the completeness of present
research. Hence to show the variation of lift with drag, drag polars of all the wing designs are
compared in Fig. 8. The lift is seen to be increasing with increase in drag till stall limit. Beyond
the stall limit the drag continues to increase, whereas lift reduces. Figure 8 clearly shows that
the drag corresponding to maximum lift is minimum in case of BMAX winglet design, while
the same is maximum for base line design. In addition to that, maximum lift coefficient is
significantly lower for base line design as compared to wings designed with different wing tip
devices. So this observation points towards the fact that the maximum lift requirement of a
UAV can be achieved at a much lower angle of attack upon designing the wing with proper
wing tip device. The lift to drag ratios plotted against angle of attack clearly speaks about the
aerodynamic advantages that the winglet designs bring in, especially in terms of flight range
and efficiency.

To find the optimum design that gives maximum lift to drag ratio, CL /CD vs angle of attack is
presented in Fig. 9. The CL /CD ratio initially increases in a linear fashion with increase in angle
of attack and attains its maximum value at about 3 degrees. All three wing designs deliver their
best performance at roughly this angle of attack. However their relative magnitude of CL /CD
differs. At 3 degrees angle of attack the CL /CD ratio obtained with BMAX design is almost
15% higher than that obtained for base line design.
 Gautham Narayan and Bibin John

Figure 8: CL vs CD comparison with angle of attack for different winglet shapes

Figure 9: CL by CD vs angle of attack for different winglet shapes

 Effect of winglets induced tip vortex structure on the performance of subsonic wings

Table 1: Comparison of aerodynamic parameters obtained with different wing designs at 40

angle of attack
CL %increase CLmax % increase CD %decrease CL /CD %increase in
in CL in CLmax in CD CL /CD
Baseline 0.47 N/A 1.35 N/A 0.0220 N/A 21.36 N/A
Blended 0.48 3.13 1.40 3.86 0.0217 1.36 22.12 3.54
BMAX 0.52 10.64 1.45 7.57 0.0212 3.64 24.53 14.81

5 Comparison between Multi Tipped Winglets

Studies presented in the previous section about Blended and BMAX winglets have shown a
considerable improvement in various performance parameters through a range of angles of at-
tack. It is possible to correlate the parameters with the geometry of the winglet as well as the
number of tips that is contributing to the vortex drag reduction. While we can add more num-
ber of tips for drag alleviation and lift augmentation, the increased wetted area may increase
the drag coefficient which in turn reduces the aerodynamic efficiency. Thus it is important to
find an optimum number of tips that can produce smaller vortices while keeping the frictional
drag within acceptable limits. With this objective, three different multi tipped winglet geome-
tries were further investigated to understand their relative performance in terms of aerodynamic
coefficients. Results of these studies are summarized in the subsequent sub-sections.

5.1 Vortices

The multi tipped winglet geometries have been simulated using a wing of aspect ratio 14 to
compare their performance with baseline and BMAX winglets. The vorticity contours in Fig.10
indicate that the each of the tips produce their individual vortices which are much lower in
vorticity magnitude as compared to the Baseline wing. Multi-Tip-4 shows the highest reduction
of vorticity while Multi-Tip-2 the least, at high angles of attack.

5.2 Lift Coefficient

There is significant improvement in lift coefficient at all angles of attack with multi tipped
winglets. This is because of the ability to alter the lift distribution of the wing by influencing the
tip vortices. Although the winglet tips are aligned in the direction of the flow, the lift induced
counter rotating vortices interact with individual tips of the winglet to produce smaller vortices
that ultimately help in reducing the overall vorticity near the wing tips. Also, this limits the cross
flow occurring due to the leakage of air from the region of high pressure beneath the wing to
the upper surface where low pressure exists. By increasing the number of tips it is observed in
Fig.11(a) that this interaction is stronger and more efficient at improving the lift. The multi-tip-4
shows an 19% improvement in CL from the baseline wing as compared to the 17% by multi-tip-
3 and 8% by multi-tip-2. This clearly shows that having a higher number of tip vortices directly
improves the lifting performance.
 Gautham Narayan and Bibin John

α = 40 α = 80 α = 120 α = 160

Base line design

Multi-Tip-2

Multi-Tip-3

Multi-Tip-4
Figure 10: Vorticity contours obtained for different multi tipped winglet designs at different
angles of attack
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

5.3 Drag coefficient

Comparison of drag coefficient values obtained with different multi tipped winglets are de-
picted in Fig. 11(b). The Multi tipped winglet is able to reduce the total drag coefficient due
to a reduction in the vortex intensity arising from the individual vortices. But a higher number
of tips in the winglet lead to a significant increase in the frictional drag thus making it coun-
terproductive. The multi-tip-2 winglet has a CD of 0.215 as compared to the BMAX winglet
of 0.212. Even though both the winglets have two tips, disparity in the orientation and shape
of the winglet tips incurred variation in aerodynamic parameters. BMAX winglet has two tips
vertically opposite to each other as compared to the 50 degree sweep of the multi-tip-2 winglet.
The multi-tip-3 winglet has shown a significant reduction in drag over the BMAX winglet, this
due to the three vortices producing a vortex drag alleviating effect as compared to the two vor-
tices in BMAX. The multi-tip-4 has a higher drag coefficient than that of multi-tip-3 due to the
significant increase in wetted area. This has led to the conclusion that the multi-tip-3 has the
best performance and most suitable for drag reduction.

5.4 CL /CD

Overall, the multi tipped winglets have performed better than the BMAX and Blended winglets,
except the multi-tip-2 which has shown reduced performance in Fig.11(c). The lift coefficient
has proportionately increased with the number of tip vortices. However the drag coefficient
has reduced when having two and three vortices due lesser vortex drag, but has increased with
multi-tip-4 due to high wetted surface area. With reference to Table.2, multi-tip-4 shows an 20%
increase in CL /CD from baseline as compared to 22% by multi-tip-3 and 11% by multi-tip-2
at angle of attack of 40 . Thus the CL /CD ratio which is an accepted estimate of aerodynamic
efficiency shows that the multi-tip-3 is the optimum winglet.

Table 2: Comparison of aerodynamic parameters obtained with different multi tipped winglet
designs at 40 angle of attack
CL %increase CLmax % increase CD %decrease CL /CD %increase in
in CL in CLmax in CD CL /CD
Baseline 0.47 N/A 1.35 N/A 0.022 N/A 21.36 N/A
Multi-Tip-2 0.51 8.51 1.43 5.79 0.0215 2.27 23.72 11.03
Multi-Tip-3 0.55 17.02 1.47 9.35 0.0210 4.55 26.19 22.59
Multi-Tip-4 0.56 19.15 1.52 12.76 0.0218 0.91 25.55 20.24

5.5 Coefficient of pitching moment

The coefficient of pitching moment has important influence on the longitudinal stability of the
aircraft. The winglets have shown a reduction in the negative pitching moment, thus implying
that there is a sense of increased stability and reduced upward pitching tendency. Figure 12
shows that the pitching moment is roughly reduced by 18% by the BMAX winglet as com-
pared to 7% by the Blended winglet. Pitching moment coefficient is observed to be the least for
 Gautham Narayan and Bibin John

(a) (b)

(c) (d)
Figure 11: Comparison of aerodynamic parameters: (a) Lift coefficient, (b) Drag Coefficient,
(c) Drag polar and (d) Lift to drag ratio

multi-tip-4 with an improvement of 28% from baseline as compared to 21% for multi-tip-3 and
13% for multi-tip-2. The moment coefficient shows direct correlation to ability of the winglet
to reduce the vorticity at the tips. A larger number of wing tips increase the interaction of the
winglet with the oncoming flow and thus each tip contributes towards a positive moment coef-
ficient. Hence, having a larger number of wingtips helps to reduce the overall negative moment
coefficient.

6 Effects of Aspect ratio on aerodynamic performance

Having placed the aerodynamic efficiency as the most important parameter for this study,
the next stage of this investigation considers the optimized multi-tip-3, Blended and BMAX
winglets towards the study of variation of aerodynamic parameters when appended to wings of
different aspect ratios. Three different aspect ratios are considered for this study, viz. 6, 10 and
14. Here aspect ratio 6 can be considered as a representative of low aspect ratio wings, whereas
14 belongs to the class of high aspect ratio wings [3]. Wing of aspect ratio 10 can be considered
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

Figure 12: Comparison of pitching moment coefficient with angle of attack

as a medium aspect ratio wing. Hence computational simulations are carried out to reveal the
winglet offered improvements in aerodynamic parameters of afore mentioned wing sections.

6.1 Effect of aspect ratio on lift coefficient

Comparison of lift coefficient distributions obtained with different wing geometries and aspect
ratios are depicted in Fig. 13. It is evident from Fig. 13(a) that a baseline wing of higher aspect
ratio has better lifting characteristics. There is a proportional rise in the lift curve slope with
aspect ratio, while the corresponding stall angle drops. The maximum lift coefficient is seen to
be reducing with decrease in aspect ratio. The wing is expected to stall when the effective angle
of attack is more than the local stall angle of the airfoil. The effective angle of attack is the
difference between the geometric angle of attack and the lift induced angle of attack caused by
the downwash. In the case of high aspect ratio wings the downwash tends to gradually reduce
from the wingtip to the root, thereby the lift induced angle of attack is lower and the effective
angle of attack is higher. Due to this the lift coefficient rises steeply with angle of attack and
stalls quickly. In low aspect ratio wings, the change in downwash is sharper from the wing
tip to the root, thus the effective angle of attack is relatively lower due to which the wing lift
coefficient tends to have a more gradual increase and a delayed stall.
 Gautham Narayan and Bibin John

(a) (b)

(c) (d)
Figure 13: Comparison of lift coefficient variation with aspect ratio for (a) Baseline, (b)
Blended, (c) BMAX and (d) Multi-Tip-3
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

By adding a winglet at the wing tip, we are able to reduce the variation of CL and CLmax with
angle of attack. It can be seen from Fig.13(a) that the difference between the slopes of the CL
curves is maximum for the baseline wing and reduces in Blended, BMAX and Multi-tip-3 in
the corresponding order. Also stall angle in wings of aspect ratio 14 remain the highest for all
the winglet designs. The percentage improvement in CL at angle of attack of 4 degrees can be
seen from Table.3.

Table 3: Comparison of CL values obtained with different aspect ratios at 40 angle of attack
Baseline Blended BMAX BMAX
AR CL %increase in CL CL %increase in CL CL %increase in CL CL %increase in CL
14 0.47 N/A 0.48 3.13 0.52 10.64 0.55 17.02
10 0.39 N/A 0.42 7.69 0.46 17.95 0.50 28.21
6 0.35 N/A 0.38 8.57 0.42 20.01 0.45 28.57

6.2 Effect of aspect ratio on Drag Coefficient

Figure.14(a) shows that the drag coefficient for a baseline wing of AR-14 is lower than that of
AR-10 and AR-6 up till the stall angle of attack. After stall the CD of AR-14 case increases
steeply and there exists a cross over point where the CD of the baseline wing with AR-14
becomes more that of the lower aspect ratio wings. The wings of lower aspect ratio tend to
stall later and have a much higher drag coefficient during the pre-stall operation. This can be
attributed to the effect of induced drag arising from the downwash of the wing. Thus when the
aspect ratio is lower, the downwash tends to be stronger at the tips leading to a higher induced
drag coefficient. This can be observed for angles of attack beyond 2 degrees after which the
induced drag becomes more significant than the frictional drag.

It can be observed from the results that the addition of a winglet effectively reduces the lift
induced drag arising from the downwash for wings of lower aspect ratio. Fig.14 shows that for
each winglet design, the wing with the lower aspect ratio tends to have a higher CD until stall
angle after which it rises steeply. It can be observed from Table.4 that the Multi-tip-3 winglet
has an overall lower drag coefficient as compared to that of Blended and BMAX winglets in
all the cases. Also the Multi-tip-3 winglet is the most effective at improving the CD for lower
aspect ratio wings.

Table 4: Comparison of CD values obtained with different aspect ratios at 40 angle of attack
Baseline Blended BMAX BMAX
AR CD %decrease in CD CD %decrease in CD CD %decrease in CD CD %decrease in CD
14 0.0220 N/A 0.0217 1.36 0.0212 3.64 0.0210 4.55
10 0.029 N/A 0.027 6.90 0.026 10.34 0.025 13.79
6 0.034 N/A 0.033 2.94 0.032 5.88 0.030 11.76
 Gautham Narayan and Bibin John

(a) (b)

(c) (d)
Figure 14: Comparison of drag coefficient variation with aspect ratio for (a) Baseline, (b)
Blended, (c) BMAX and (d) Multi-Tip-3
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

(a) (b)

(c) (d)

Figure 15: Comparison of CL /CD variation with aspect ratio for (a) Baseline, (b) Blended, (c)
BMAX and (d) Multi-Tip-3

6.3 CL /CD

Figure.15, shows that the on overall CL /CD of the wing is lesser for wings of lower aspect
ratio. The maximum CL /CD also tends to be lower and tends to occur over a wider range of
angles with low aspect ratio wings. The steeper lift and drag curve’s in wings of higher aspect
ratio also contribute towards making the wing very sensitive to angle of attack for aerodynamic
efficiency. In Fig. 15 it can be observed that in all geometries the peak aerodynamic efficiency
is attained between 3 to 5 degrees after which the CL /CD ratio quickly reduces for all other
angles of attack.

From Table.5, it can be seen that the addition of the winglet makes a more significant contri-
bution towards improving the aerodynamic efficiency of the wing at lower aspect ratio. Due to
the nature of the wing to have a stronger downwash near the wing tips, the winglet acts as an
imaginary span extension device that is capable of increasing the effective aspect ratio of the
 Gautham Narayan and Bibin John

wing. Also, The multi-tip-3 winglet is the most capable at augmenting the lift at lower aspect
ratios followed by the BMAX and Blended winglets. This can be explained by the fact that
the multi-tip-3 shows a steeper lift curve as explained in the previous section coupled with a
higher reduction in drag from vorticity reduction. Interestingly the comparison of CL /CD ra-
tios obtained with different aspect ratios at 4 degrees of angle of attack presented in Table. 5
evidence that the percentage improvement in lift to drag ratio is maximum in case of moderate
aspect ratio of 10. The trend in winglet effectiveness improvement with reduction in aspect ratio
ceases at a particular aspect ratio value, after which the winglet effectiveness drops. This leads
to an important conclusion that there exist an optimum aspect ratio at which the integrated effect
of improved lift and reduced induced drag along with enhanced parasite drag offers maximum
aerodynamic efficiency in terms of lift to drag ratio. Below this aspect ratio the higher para-
sitic drag and lower lift coefficient diminish the effectiveness of all winglets, which is counter
intuitive with respect to the trend with CL /CD .

Table 5: Comparison of CL /CD values obtained with different aspect ratios at 40 angle of attack
Baseline Blended BMAX BMAX
AR CL /CD %increase in CL /CD %increase in CL /CD %increase in CL /CD %increase in
CL /CD CL /CD CL /CD CL /CD
14 21.36 N/A 22.12 3.54 24.53 14.81 26.19 22.59
10 13.45 N/A 15.56 15.67 17.69 31.56 20.03 48.72
6 10.29 N/A 11.52 11.86 13.14 27.50 15.01 45.71

7 Conclusions

Computational investigation of three dimensional, subsonic flow over a half wing is presented
in this work. Detailed comparative study is performed to converge to an effective winglet shape
that can give maximum aerodynamic performance. First stage of the present study focused on
the comparison of aerodynamic parameters of the Blended and BMAX winglets against the
baseline design. The BMAX winglet outperforms in aerodynamic efficiency by 14% as com-
pared to 3.5% by the Blended winglet. Most shrunken tip vortex core is noticed in case of
the BMAX winglet, which shows better performance coefficients as compared to the Blended
winglet. The relative advantage of BMAX winglet in terms of improvement in CL /CD ratio
can be attributed to the fact that the BMAX winglet creates two distinct tip vortex cores as
opposed to one in case of Blended winglet. It is observed that the reduction of wing vortic-
ity by introducing smaller vortices into the stream is an effective method to augment the lift
while reducing drag. This observation provoked the thought of implementation of multi-tipped
winglets to improve aerodynamic performance of wing of aspect ratio 14. Hence the second
stage of the present study devoted to the investigation on multi-tipped winglets. Winglets hav-
ing tips of 2, 3 and 4 were considered for this study. The geometry of multi-tip-2 differs from
the BMAX winglet in terms of their sweep angles and tip chord length. Investigations on multi
tipped winglets proved that the lift increases with number of tips whereas the lift induced drag
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

decreases with increase in number of tips. Moreover, the obvious observation of increase in
parasite-drag with increase in number of tips has also been made though this study. Although
increase in number of tips reduces the lift induced drag, the increase in parasite drag resulted
from the added wetted area is found to be increasing the overall drag coefficient. At higher
angles of attack the drag coefficients of Multi-tip-4 winglet is significantly greater than the
multi-tip-3 winglet. The comparison of lift to drag ratios obtained with three different multi
tipped winglets proves that muti-tip-3 is the most aerodynamically efficient among the winglets
considered in this study. The percentage improvement in CL /CD compared at a realistic angle
of attack 4 degrees indicated 22.5% improvement in case of Multi-tip-3, while 11% and 20%
were noticed with multi-tip-2 and muti-tip-4 respectively. Hence the multi-tip-3 winglet with
three vortices proves to be the most optimum multi tipped winglet due to a better lift slope than
the BMAX winglet while having the least total drag coefficient of the three multi tips.

The third stage of the discussion focused on the analysis of effectiveness of winglets on wings
with different aspect ratios. Three different wing aspect ratios viz. 14, 10 and 6 were considered
in this phase of study. This study showed that the winglets provide improvement in aerody-
namic efficiency at all aspect ratios. However the percentage of improvement in aerodynamic
efficiency measured in terms of lift to drag ratio initially increases with decrease in aspect ratio
and later below a critical aspect ratio it starts reducing. Hence this observation makes a key
finding of this present study that there exist an optimum aspect ratio at which winglets offer
maximum effectiveness for a given flight condition.

References

[1] J.D. Anderson, Fundamentals of Aerodynamics, McGraw-Hill, New York, 2011.

[2] Auerbach, D., Why aircraft fly, European Journal of Physics, Vol. 21,pp. 289-296. 2000

[3] Abbott, Ira H., Albert E. Von Doenhoff, and Louis Stivers Jr. Theory of Wing Sections,
Including a Summary of Airfoil Data. (1945).

[4] Jason E. Hicken and David W. Zingg, Induced-Drag Minimization of Nonplanar Geome-
tries Based on the Euler Equations, AIAA Journal Vol. 48, No. 11, November 2010

[5] R.T. Whitcomb, A design approach and selected wind-tunnel results at high subsonic
speeds for wing-tip mounted winglets, NASA TN D-8260, NASA Lang-ley Research Cen-
ter, 1976.

[6] H.H. Heyson, G.D. Riebe, and C. L. Fulton, Theoretical parametric study of the rela-
tive advantages of winglets and wing-tip extensions. Vol. 1020. National Aeronautics and
Space Administration, Scientific and Technical Information Office, 1977.

[7] L.C. Montoya, S.G. Fiechner and P.F. Jacobs, Effect of an alternate winglet on the pressure
and spanwise load distributions of a first-generation jet transport wing. NASA Technical
Memorandum, 78786, 78786, 1978.
 Gautham Narayan and Bibin John

[8] L.B. Gratzer, Spiroid-tipped wing, U.S. Patent 5,102,068, 7 April 1992.

[9] S. Kravchenco, The application of the wing tip lifting surfaces for practical aerodynamic.
ICAS- 96-4.6.4, Sorrento Italy, pp. 1338-1349, 1996.

[10] L. U. Roche, and S. Palffy, WING-GRID, a Novel Device for Reduction of Induced Drag
on Wings, Proceedings of ICAS 96, Sorrento, September, 1996.

[11] M. Maughmer, T. Swan and S. Willits, Design and testing of a winglet airfoil for low-speed
aircraft. Journal of Aircraft, Vol 39, No 4, pp. 654-661, 2002.

[12] Yamazaki, W. Matsushima, K. and Nakahashi, K. Aerodynamic Design Optimization Us-

ing the Drag-Decomposition Method, AIAA Journal, Vol. 46, No. 5, pp. 1096-1106, 2008.

[13] Liersch, C. M., Streit, T., and Visser, K. D., Numerical Implications of Spanwise Camber
on Minimum Induced Drag Configurations, 47th AIAA Aerospace Science Meeting and
Exhibit, AIAA Paper 2009-898, Jan. 2009.

[14] K. Takenaka, K. Hatanaka, W. Yamazaki and K. Nakahashi, Multidisciplinary design ex-

ploration for a winglet, Journal of Aircraft, 45(5), 1601-1611, 2008.

[15] Z. Toor and J Masud: Part I: Uncertainty Analysis of Various Design Parameters on
Winglet Performance, In AIAA SciTech, 54th AIAA Aerospace Sciences Meeting, AIAA
Paper- 2016-0556, 2016

[16] M. J. Smith, N. Komerath, R. Ames, O. Wong, and J. Pearson, Performance Analysis of a

Wing with Multiple Winglets, AIAA Paper- 2001-2407, 2001.

[17] P. Bourdin, A. Gatto, and M. I. Friswell, The application of variable cant angle winglets
for morphing aircraft control. In 24th Applied Aerodynamics Conference, pp: 5-8, 2006.

[18] J.E Guerrero, D. Maestro, and A. Bottaro, Biomimetic spiroid winglets for lift and drag
control. Comptes Rendus Mecanique, 340(1), 67-80, 2012.

[19] S. A. Ning and I. Kroo, Tip extensions, winglets, and c-wings: conceptual design and
optimization. AIAA Paper, 7052, 2008.

[20] Joel F. Halpert, Daniel H. Prescott, Thomas R. Yechout and Michael Arndt: Aerodynamic
Optimization and Evaluation of KC-135R Winglets, Raked Wingtips, and a Wingspan
Extension, In 48th AIAA Aerospace Sciences Meeting, AIAA Paper- 2010-57, 2010

[21] P. Panagiotou, P. Kaparos and K. Yakinthos, Winglet design and optimization for a MALE
UAV using CFD, Aerospace Science and Technology 39 pp: 190-205, 2014.

[22] La Roche, Ulrich, and H. Lucas La Roche," Induced Drag Reduction Using Multiple
Winglets, Looking Beyond the Prandtl-Munk Linear Model",AIAA Paper No. AIAA-
2004-2120, 2004.
 Effect of winglets induced tip vortex structure on the performance of subsonic wings

[23] F. M. Catalano, and H. D. Ceron-Munoz, Experimental analysis of aerodynamics char-

acteristics of adaptive multi-winglets, In 43rd AIAA Aerospace Sciences Meeting and
Exhibit, pp. 2005-1231. 2005.

[24] F. Luis, A. Gomes, and A. Suleman. "Design and Analysis of an Adaptive Wingtip." 52nd
AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Confer-
ence, 2011.

[25] Cosin, R., F. M. Catalano, L. G. N. Correa, and R. M. U. Entz. "Aerodynamic analysis of

multi-winglets for low speed aircraft." In Proceedings of 27th International Congress of
the Aeronautical Sciences (ICAS 2010). 2010.

[26] S. R. Reddy, G. S. Dulikravich, A. Abdoli, and H. Sobieczky, Multi-Winglets: Multi-

Objective Optimization of Aerodynamic Shapes, In 53rd AIAA Aerospace Sciences Meet-
ing, p. 1489. 2015.

[27] Gavrilovic, N.N., Rasuo, B.P., Dulikravich, G.S. and Parezanovic, V.B., 2015. Commercial
aircraft performance improvement using winglets. FME Transactions, 43(1), pp.1-8, 2014.

[28] F. R. Menter, "Zonal Two Equation k-ω Turbulence Models for Aerodynamic Flows",
AIAA Paper 93-2906, 1993.

[29] F. R. Menter, Two-Equation Eddy-Viscosity Turbulence Models for Engineering Applica-

tions, AIAA Journal, vol. 32, no 8. pp. 1598-1605, 1994.