You are on page 1of 7

Gurney Flaps on Slender and Nonslender Delta Wings

D. I. Greenwell

City University, London, England EC1V 0HB, United Kingdom
DOI: 10.2514/1.46610
A parametric low-speed wind-tunnel study has been undertaken of the effects of Gurney flap height on the
aerodynamic characteristics of delta wings with sweep angles of 40, 60, and 70

. Flap effects on lift and drag are
consistent with previous two-dimensional data when wing geometry is accounted for by using the change in zero-lift
incidence rather than lift and using the relative flap area rather than flap height. Flap deployment primarily affects
the attached-flow potential lift on delta wings, with vortex lift being almost unchanged. Induced-drag factors are
considerably reduced as flap height is increased, but this is due to the use of flat-plate models with near-zero leading-
edge suction. For control applications, pitch/lift ratios are similar to those for delta wings with conventional trailing-
edge flaps, but the trim-drag penalties are much higher. An aft shift in loading associated with Gurney flap
deployment leads to large aft movements of the aerodynamic center, which significantly reduces control capability.
Nomenclature
a
0
= lift-curve slope at zero lift
AR = aspect ratio, b
2
=S
b = wing span
c = mean aerodynamic chord
c
r
= root chord
C
D
= drag coefficient, D=qS
C
Dmin
= minimum drag coefficient
C
DV
= vortex (induced) drag coefficient
C
D0
= drag coefficient at zero lift
C
L
= lift coefficient, L=qS
C
L
= lift-curve slope, dC
L
=d
C
M
= pitching-moment coefficient, M=qcS
C
M0
= pitching-moment coefficient at zero lift
C
N
= normal force coefficient, N=qS
C
NP
= potential flow component of normal force
C
NV
= vortex lift component of normal force
D = drag
h = projected flap height (measured from lower surface)
K
P
= potential lift factor, ÷ C
L
K
V
= vortex lift factor
k = induced-drag factor, (C
D
÷ C
Dmin
)AR=C
2
L
L = lift
q = dynamic pressure,
1
2
V
2
q = lift increase factor on flap height
S = wing area
S
f
= flap frontal area, hb
t = wing thickness
= angle of attack

0L
= zero-lift angle of attack

0P
= zero-lift angle of attack for potential lift component

0V
= zero-lift angle of attack for vortex lift component
= leading-edge sweep angle
I. Introduction
M
any recent unmanned combat air vehicle (UCAV) and micro
air vehicle (MAV) concepts are tailless configurations with
delta or diamond wing planforms of relatively low (40–50
·
) sweep
angle. In terms of flow characteristics, these planforms sit between
low-aspect-ratio slender delta wings with a leading-edge vortex
system over the wing and conventional medium-aspect-ratio swept
wings with fully attached flow. The resulting aerodynamic and flight
dynamic behavior is complex and still not fully understood,
particularly at low Reynolds number [1]. For these tailless config-
urations, pitch control using conventional trailing-edge flaps causes
difficulties due to the coupling of pitching moment with lift. Sellars
et al. [2] postulated that using tangential trailing-edge blowing on a
delta wing could generate pitching moments without any change in
lift, due to the interaction between the potential and vortex lift
components. This turned out not to be the case [3], but results
presented by Li et al. [4] indicated that a Gurney flap could have
favorable pitch control characteristics when applied to a nonslender
delta wing. The published data on Gurney flaps on delta wings is
of rather limited extent, and somewhat contradictory, so an
experimental study was undertaken to clarify the effects of sweep
angle on the changes in lift, drag, and pitching moment due to flap
deployment.
II. Gurney Flaps Applied to Delta Wings
AGurney flap is a small trailing-edge flap (typically less than 5%
chord), deflected at 60–90
·
to the chord line. The wake flow behind
the flap directly alters the trailing-edge Kutta condition [5], leading to
a large lift increment despite its small size (for example, a C
L
of
~0:8 for a 5%flap on a 2-Dairfoil [6]). There is anassociated parasite
drag penalty, of the order of the frontal area of the flap, but this is
dependent on the flap height relative to the local boundary-layer
depth and can be negligible for flaps of 2% height or smaller [7].
The vast majority of experimental and theoretical work on Gurney
flaps has been on 2-D airfoils, along with a small number of experi-
mental investigations of medium-aspect-ratio (5–6) rectangular
wings. Comparatively little work on delta wings with Gurney flaps
has been published, with the literature consisting of three relatively
comprehensive data sets presenting lift, drag, and pitching moment
[4,8–11], along with a handful of papers reporting more limited
results [12–14].
Traub and Galls [8] report on a very thorough study of a number of
leading-edge and trailing-edge flap configurations on a 70
·
delta
wing. A thin flat-plate wing model was used, with a thickness/chord
ratio t=c
r
of 0.38%and a blunt leading edge. Constant-height Gurney
flaps with height/chord ratios h=c
r
of 0.5 and 0.95% were tested at a
Reynolds number based on root chord of 1:12 × 10
6
. Traub and Galls
also present a detailed comparison with a range of previous studies on
70
·
delta wings, giving a high level of confidence in the experimental
arrangements and data-reduction procedures. The effects of the
0.95% flap on lift, drag, and pitching moment also correspond very
closely to the data presented later in this paper.
Presented as Paper 3468 at the 25th AIAA Applied Aerodynamics
Conference, San Francisco, CA, 5–8 June 2006; received 7 August 2009;
accepted for publication 29 September 2009. Copyright © 2009 by the
American Institute of Aeronautics and Astronautics, Inc. All rights reserved.
Copies of this paper may be made for personal or internal use, on condition
that the copier pay the $10.00 per-copy fee to the Copyright Clearance Center,
Inc., 222 Rosewood Drive, Danvers, MA01923; include the code 0021-8669/
10 and $10.00 in correspondence with the CCC.

Reader in Flight Dynamics, School of Engineering and Mathematical
Sciences, Northampton Square; d.greenwell@city.ac.uk.
JOURNAL OF AIRCRAFT
Vol. 47, No. 2, March–April 2010
675
Buchholz and Tso [9] summarize a project study [10] of a range of
larger Gurney flaps on a 60
·
delta wing. Aflat-plate wing model was
used, with a thickness/chord of 3% and a 45
·
lower-surface beveled
leading edge. Constant-height flaps with height/chord ratios of 1, 2,
3, and 5% were tested at a Reynolds number based on root chord of
8:6 × 10
5
. However, when the results presentedin [9] were compared
with previous data for 60
·
delta wings (e.g., [15]), it became clear that
the basic wing lift-curve slope was far too high, by a factor of around
÷30%. An analysis of the entire data set [16] showed that dividing all
lift, drag, and pitching-moment measurements by 1.3 gave a more
reasonable agreement with published 60
·
wing data and with the
effects of flap height reported later in this paper. This suggests a
possible error in the dynamic pressure measurement; whatever the
error source, this data set is clearly unreliable.
Li et al. [4,11] present work on a cropped 40
·
delta wing in two
complementary papers. A flat-plate wing model was used, with a
thickness/chord of 1.6% and a 60
·
symmetric beveled leading-edge.
Constant-height flaps with height/chord ratios of 1, 2, 3, and 5%were
tested at a Reynolds number based on root chord of 2:5 × 10
5
. When
the results presented in [4,11] were compared with previous data for
40
·
delta wings (e.g., [1,17]), it was noted that the lift-curve slope
was too lowby a factor of ÷20%(allowing for taper ratio) and that the
aerodynamic center was misplaced. No obvious reason could be
found for these discrepancies, so again this data set must be con-
sidered unreliable.
Figure 1 illustrates the typical effects of Gurney flap deployment
on lift, drag, and pitching moment, using data from the sole
remaining reliable and comprehensive data set [8]. Lift at zero
incidence C
L0
and maximum lift C
Lmax
both show a significant
increase, along with a small increase in lift-curve slope (Fig. 1a).
Zero-lift drag C
D0
increases, but induced-drag factor k reduces
(Fig. 1b), leading to a reduction in maximumlift/drag ratio (L=D)
max
and an increase in the lift coefficient for minimumdrag (Fig. 1c). Flap
deployment gives a nose-down change in zero-lift pitching moment
C
m0
, combined with an aft shift in aerodynamic center (Fig. 1d) as
dC
m
=dC
L
becomes more negative. Pitch/lift ratio C
m0
=C
L0
is
negative and of a similar magnitude to conventional trailing-edge
controls.
Most explanations of the effects of Gurney flaps suggest that the
change in the trailing-edge Kutta condition due to flap deployment
leads to a change in effective camber [18,19]. In addition, the
pressure difference between upper and lower surfaces at the trailing
edge due to the separated wake of the flap [20] leads to an aft shift in
loading and hence to an increase in effective chord [21,22]. The 3-D
wing behavior shown in Fig. 1 suggests that both factors are
operating. The increase in lift, maximum lift, and nose-down
pitching moment are all consistent with an increase in effective
camber. However, the drag polar remains symmetric about zero lift,
presumably because the increase in drag is due directly to the
separated wake flow behind the flap, rather than to an incidence-
dependent change in profile drag. The increase in lift-curve slope and
aft shift in aerodynamic center are consistent with an increase in
effective chord, but the reduction in induced-drag factor is less
straightforward. Reference [8] suggests that since the lift-dependent
induced drag is equal to C
L
tan for a flat-plate delta wing with zero
leading-edge suction, an increase in lift at fixed incidence leads to a
reduction in induced-drag factor.
III. Experimental Arrangement
Three sharp-edged flat-plate delta-wing models with leading-edge
sweep angles of 40, 60, and 70
·
were tested in the Bristol
University 0:8 × 0:6 m Low-Turbulence Wind Tunnel [23]. This
tunnel is of conventional closed-circuit design, with a large con-
traction ratio and flow control screens to give turbulence levels well
below 0.1%. Model planform area S was held constant at 0:072 m
2
,
giving a model/tunnel area ratio of 0.15 and a root chord c
r
of 0.246,
0.353, and 0.436 m, respectively. Models were made of 3 mm
aluminumsheet, with leading edges beveled on the lower surface at a
20
·
included angle (Fig. 2) and with a blunt trailing edge. Flaps were
made of 1 mm aluminum sheet and attached to the lower surface of
the wing. Forces and moments were measured using an AeroTech 3-
component overhead balance, with the models mounted inverted on a
single strut plus a pitch pushrod (Fig. 3). A small underwing fairing
covered the strut and pushrod attachment points.
Gurney flap heights of 1, 3, and 6%of root chord were tested, with
flap height h measured from the wing lower surface. Tests with
a)
b) c)
d)
Fig. 1 Typical effects of Gurney flap on delta-wing aerodynamics (data
from [1]).
Fig. 2 Delta-wing model geometry.
676 GREENWELL
inverted (i.e., upper surface) flaps established that the underwing
fairing had no significant effect on trends in aerodynamic behavior
with flap height. Tests were carried out at a constant Reynolds
number based on root chord of 0:7 × 10
6
. Pitching moments were
nondimensionalized using the mean aerodynamic chord c (2c
r
=3)
and taken about a moment reference center at the quarter-chord point
(0:5c
r
). Solid and wake blockage were corrected for using the
method of [24], with downwash corrections from [25]. Balance
accuracy was estimated at better than 0.15%in lift, drag, and pitching
moment, with pitch angle set to better than 0.1
·
. Further details of the
experimental arrangement can be found in [26,27].
In order to verify the accuracy of the experimental procedure and
data reduction, the baseline delta-wing data was compared with a
range of published results for similar wings. Figure 4 compares the
basic 60
·
sweep wing results with data from Flygtekniska
Försöksanstalten (FFA) in Sweden [28]. The FFA wing had a very
similar geometry (a strut-mounted flat-plate model with an under-
body fairing) and was tested at the same Reynolds number of
0:71 × 10
6
. Figure 4 also shows normal force data from the seminal
study of delta-wing aerodynamics by Wentz and Kohlman [15] and
the theoretical normal force given by the leading-edge suction
analogy (LESA) [29]. Angle of attack is given relative to the zero-lift
incidence
0L
, to allow for the effect of different leading-edge
chamfer angles [30] and support interference. Normal force and
pitching moments compare very well, with identical lift-curve slopes
and aerodynamic centers at low incidence. A similarly good agree-
ment with published data was obtained for the clean 40 and 70
·
wings
and for the 70
·
wing with a 1% flap [16].
IV. Effect of Gurney Flap Height
A. Lift
Looking first at the effect of flap height on lift, Fig. 5 shows the
increment in lift coefficient at zero incidence C
L0
as a function of
relative height h=c
r
(following previous convention in using the root
chord as a reference length for 3-D wing data). The filled symbols
denote increments derived from the Bristol data of [26,27], and the
open diamond symbols denote a rectangular wing with an aspect
ratio of 6 [31]. The crosses show data for a range of 2-D airfoils with
Gurney flaps [6,7,32–34]. The same symbols are used in the
following figures.
On the basis of thin-airfoil theory, [19] proposes that the lift
increment due to a Gurney flap is
C
L0
=q(h=c)
n
(1)
where q is supposedly a function of Reynolds number (varying
between 2 and 4) and n =0:5. However, plotting the data of [19] and
Fig. 5 in log–log form indicates that the variation for both 2-D and
3-Dwings is closer to n =0:7 (with q ranging between 3 and 6). The
experimental data reviewed in [19] also shows no systematic
variation of q with Reynolds number. The difference between the
2-D and 3-D data in Fig. 5 can be attributed to two factors: 1) the
effect of aspect ratio on lift-curve slope and 2) the effect of taper ratio
on the relative flap height. The effect of variations in lift-curve slope
a
0
can be accounted for by replacing lift increment C
L0
with
change in zero-lift incidence
0L
~÷C
L0
=a
0
. The effect of taper
ratio can be accounted for by weighting the flap height by the local
chord. For the constant-flap-height data of Fig. 5, the most
straightforward weighting method is to use the relative flap area
S
f
=S =(hb)=S. Figure 6 demonstrates that applying both factors
collapses the scattered 2-D and 3-D lift increment data of Fig. 5
reasonably well onto a single trend:

0L
~0:9(S
f
=S)
0:7
(2)
For the specific case of delta wings, where the lift may be split
into potential (or attached) and leading-edge vortex (or separated)
flow contributions [29], the question arises as to the relative effect
of a Gurney flap on the two lift components. Off-surface flow
Fig. 3 Model mounted in the Bristol Low-Turbulence Tunnel.
Fig. 4 Comparison of basic 60

wing with theory and results from
Torlund [28] and Wentz and Kohlman [15]. Fig. 5 Effect of flap height on lift increment at zero incidence.
GREENWELL 677
measurements and visualizations reported in [8] indicate that a 1%
Gurney flap on a 70
·
delta wing has a negligible effect on the strength
of the primary leading-edge vortex and on the variation of vortex
breakdown position with angle of attack, but that the height of the
vortex core above the wing surface was reduced slightly. The
reduction in vortex height at a given angle of attack would be
expected to increase the vortex lift component. An indication of the
relative magnitudes of the potential and vortex lift components can
be obtained by fitting the LESA lift model [29]
C
N
=C
NP
÷C
NV
=K
P
sin() cos() ÷K
V
sin
2
() (3a)
to the experimental normal force data. Following the lead given by
Fig. 6, good agreement was obtained by modeling the effects of a
Gurney flap as shifts
0P
and
0V
in the effective incidences of the
potential and attached-flow components
C
N
=K
P
sin( ÷
0P
) cos( ÷
0P
)
÷K
V
sin( ÷
0V
)[ sin( ÷
0V
)[ (3b)
as demonstrated in Fig. 7. This shows the variation in normal force
for a 70
·
delta wing with 1 and 3% Gurney flaps, compared with
Eq. (3b). The corresponding lift factors K
P
and K
V
and zero-lift
offsets
0P
and
0V
, as determined from a nonlinear least-squares fit
to the data, are shown in Table 1. Although the results of fitting a
nonlinear equation to experimental data should be treated with
caution, it is clear that the Gurneyflap has a much greater effect on the
potential lift component than on the vortex lift, with the change in
zero-lift incidence
0P
of the order of six times bigger than the change
in the vortex zero-lift incidence
0V
. At a given incidence, the change
in lift due to flap deployment is largely due to the change in potential
lift, with a small contribution from the vortex lift. This is consistent
with the results reported in [8]. At lowangles of attack, the reduction
in local potential lift-curve slope
dC
NP
d
=K
p
cos(2)
as the potential lift curve moves to the left is counterbalanced by the
increase in the vortex lift-curve slope
dC
NP
d
=K
v
sin(2)
giving a small increase in local lift-curve slope at low incidences.
B. Drag
Figure 8 shows profile drag penalties for 2-D airfoils and 3-D
wings as a function of flap height. Plotting the change in minimum
drag C
Dmin
against relative flap area S
f
=S collapses 2-D and 3-D
data onto a single linear trend. As one might expect for a bluff-body
wake flow, the increase in parasite drag area D=
1
2
V
2
for both 2-D
and 3-D wings is approximately equal to the frontal area of the flap
hb, giving
D
min
1
2
V
2
=C
Dmin
S ~hb ÷C
Dmin
~
hb
S
=
S
f
S
(4)
Figure 9 shows that Gurney flap deployment also has a significant
impact on induced drag for flat-plate delta wings, due to the loss of
leading-edge suction for a sharp-edged wing. Applying the analysis
of [8], the lift-dependent drag component of the normal force
becomes
C
Di
=C
N
sin() =C
L
tan() (5)
so that the induced-drag factor k is
k =AR
C
Di
C
2
L
=AR
tan()
C
L
(6)
Since the effect of a Gurney flap on a delta wing is to increase lift C
L
at a fixed incidence , Eq. (6) shows that the corresponding induced-
drag factor must reduce as flap height increases. For the LESAmodel
of Eq. (3), the induced-drag factor is a nonlinear function of
incidence, with a limiting (maximum) value at zero incidence of
k
÷0
=
AR
K
P
(7)
Equation (7) gives induced-drag factors rather higher than those
shown in Fig. 9, due to the effects of nonzero incidence and partial
recovery of leading-edge suction, but it does indicate the general
effects of wing planform. As sweep angle is reduced, the induced
drag increases, because the increased aspect ratio outweighs the
effect of increased lift-curve slope.
Fig. 6 Effect of flap area on zero-lift incidence.
Fig. 7 Effect of Gurney flap height on potential and vortex lift
components for a 70

delta wing.
Table 1 Modified LESA fit to Fig. 7
h=c
r
K
P
K
V

0P

0V
0 1.75 3.14 ÷0:5
·
÷0:5
·
1% 1.75 3.14 ÷4:5
·
÷1:0
·
3% 1.75 3.14 ÷9:5
·
÷2:0
·
678 GREENWELL
The corresponding effects of flap on lift/drag ratio (e.g., Fig. 1c)
follow directly from Fig. 8 and 9, since

L
D

max
·
1

C
D0
k
_ ; C
Lmd
·
 
C
D0
k
r
(8)
The lift coefficient C
L;md
for maximum L=D ratio increases rapidly
with flap deployment, as zero-lift drag C
D0
increases and induced-
drag factor k reduces. The effect on (L=D)
max
depends on whether
C
D0
increases more rapidly than k reduces, which in turn depends on
the clean-wing profile drag. For the data presented here, (L=D)
max
is
always reduced by flap deployment.
It should be noted that for practical wings with rounded leading
edges, partial recovery of leading-edge suction leads to markedly
lower levels of induced drag [35], and so reductions in k due to flap
deployment will be much smaller. For example, the rectangular wing
with a NACA23012 airfoil section reported in [31] shows very little
variation in induced-drag factor with flap deployment. Gurney flaps
on a UCAV-type planform with an airfoil cross section are therefore
likely to have a much more adverse impact on L=D than that
suggested by tests on flat-plate wings [8,10,16].
C. Pitching Moment
The ratio of pitching-moment change to lift change (pitch/lift
ratio) is a critical parameter for a control effector on a tailless aircraft.
In order to illustrate the effect of wing and flap geometry on this
parameter, Fig. 10 plots zero-lift pitching-moment change C
m0
against change in lift at zero incidence C
L0
for a range of wings
with Gurney flaps (symbols as Fig. 5). Figure 11 shows the corre-
sponding trim-drag penalty C
Dmin
as a function of lift increment.
Two-dimensional thin-airfoil theory gives a limiting maximum
value for pitch/lift ratio C
M0
=C
L0
of ÷0:25. For 2-Dairfoils with
Gurney flaps [6,7,32–34], Fig. 10 indicates that the pitch/lift ratio
tends to be slightly less than this theoretical value. For 3-D wings
with Gurney flaps, the ratio is greater than 2-D theory predicts and
increases with reducing aspect ratio up to about ÷0:35 for a 70
·
delta.
This is the opposite of the trend predicted by [36] for wings with fully
attached flow.
For comparison purposes, Fig. 10 also shows pitch/lift ratios for
two delta wings with trailing-edge flaps: a 41
·
delta with a split flap
[17] and a 63
·
delta with a trailing-edge flap [37]. These two wings
are not directly comparable with the data reported here, since they
had airfoil cross sections and were tested at much higher Reynolds
numbers; however, published data for simple delta wings with
Fig. 8 Effect of flap height on increment in minimum drag.
Fig. 9 Effect of flap height on induced-drag factor.
Fig. 10 Pitch/lift ratio for conventional controls and Gurney flaps.
Fig. 11 Trim drag for conventional controls and Gurney flaps.
GREENWELL 679
trailing-edge flaps is remarkably scarce. Bearing in mind the different
test conditions, these two data sets indicate that Gurney flaps on delta
wings have pitch control characteristics similar to conventional
trailing-edge flaps. However, the associated trim drag is on the order
of four to five times higher (Fig. 11).
The most significant way in which Gurney flaps differ from
conventional trailing-edge controls is in their effect on aerodynamic
center, as shown in Fig. 12. For both 2-Dand 3-Dwings, Gurney flap
deployment gives a significant aft shift of the aerodynamic center
relative to the clean wing. The magnitude of the shift is roughly
proportional to the lift increment. Figure 12 indicates that
conventional trailing-edge controls on delta wings can also cause an
aft shift in aerodynamic center, but the magnitude is much smaller,
and the shift changes sign at higher control deflections. The cause of
the aerodynamic-center movement with Gurney flap deployment
appears to be an aft shift in chordwise loading distribution, due to the
finite pressure difference between the upper and lower surfaces
induced by the flap wake at the trailing edge [20].
V. Conclusions
A parametric wind-tunnel study has demonstrated that the effects
of Gurney flap deployment on lift, drag, and pitching moment are
similar for delta wings with sweep angles of 40, 60, and 70
·
,
spanning vortex flowtypes fromnonslender to slender. The effects of
flap height on lift and profile drag are also consistent with the
extensive literature on 2-DGurney flaps, when appropriately param-
eterized. The influence of wing aspect ratio can be accounted for by
using the shift in zero-lift incidence
0L
rather than the increase in
lift C
L
, while the effect of wing taper can be accounted for by using
the relative flap area S
f
=S rather than the height/chord ratio h=c
r
.
Fitting experimental data to the LESAmodel indicates that on a delta
wing the Gurney flap acts primarily to change the zero-lift incidence
of the attached-flow potential lift component, with a much smaller
effect on the separated-flow vortex lift component. The Gurney flap
generates a bluff-body wake, giving an increase in profile drag C
D0
that is approximately equal to the relative frontal area S
f
=S of the
flap.
For the simple flat-plate delta-wing models tested here, Gurney
flaps give very large improvements in induced drag, which tends to
counteract the increase in profile drag due to flap frontal area and
hence gives a relatively small reduction in L=D. This is due to the
nearly 100% loss of leading-edge suction on a flat-plate wing with a
sharp leading edge, which results in unrealistically high induced-
drag factors for the clean wings. The induced-drag reduction with
flap deployment for a flat-plate wing is essentially a geometry effect,
as normal force is increased at a fixed incidence. For practical UCAV
and MAV wings with rounded leading edges, the clean-wing
induced-drag factors (and hence the relative improvements due to
Gurney flap deployment) will be much smaller, and so adverse
effects on L=D are likely to be much larger.
Pitch/lift ratios for delta wings with Gurney flaps are similar to
those for delta wings with trailing-edge control surfaces and are
significantly more negative than for conventional wings with
attached flow. Gurney flaps would appear to have a similar pitch
control capability to conventional flaps, but this is offset by a much
higher trim-drag penalty and by an associated aft (stabilizing) shift of
the aerodynamic center.
The usual effective camber explanation of the effects of Gurney
flaps on 2-D airfoils has some shortcomings when applied to delta
wings. The observed increases in lift, maximum lift, and nose-down
pitching moment are all consistent with a change in effective camber.
However, the drag polars remain symmetric about zero lift, and the
vortex lift component is relatively unaffected. The large aft shift in
aerodynamic center is due to the local separated-floweffects near the
trailing edge, where the pressure difference between upper and lower
surfaces at the trailing edge due to the flap wake leads to an aft shift in
loading and hence to an increase in effective chord.
Data comparisons have highlighted a number of issues with the
limited literature on delta wings with Gurney flaps. In general, many
papers do not clearly specify the moment reference center, the
moment reference length, or the point from which flap height is
measured, which makes comparisons and assessments of trends very
difficult. More specifically, two out of three of the most compre-
hensive data sets available have lift, drag, and pitching-moment data
that are clearly in error, but because no comparison was made with
other (readily available) experimental data sets for delta wings, or
with simple theoretical predictions, these errors were not recognized
before publication.
Acknowledgments
The author wishes to thank Patrick Wichert, Chris Dance, Louise
McFarlane, and David Delamore-Sutcliffe for their assistance with
the experimental work and data analysis.
References
[1] Gursul, I., Gordnier, R., and Visbal, M., “Unsteady Aerodynamics of
Non-Slender Delta Wings,” Progress in Aerospace Sciences, Vol. 41,
No. 7, 2005, pp. 515–557.
doi:10.1016/j.paerosci.2005.09.002
[2] Sellars, N. D., Wood, N. J., and Kennaugh, A., “Delta Wing Circulation
Control Using the Coanda Effect,” AIAAPaper 2002-3269, June 2002.
[3] Harley, C. D., Wilde, P. I. A., and Crowther, W. J., “Application of
Circulation Control Manoeuvre Effectors for Three Axis Control of a
Tailless Flight Vehicle,” AIAA Paper 2009-146, Jan. 2009.
[4] Li, Y. C., Wang, J. J., Tan, G. K., and Zhang, P. F., “Effects of Gurney
Flaps on the Lift Enhancement of a Cropped Non-Slender Delta Wing,”
Experiments in Fluids, Vol. 32, No. 1, 2002, pp. 99–105.
doi:10.1007/s003480200010
[5] Liebeck, R. H., “Design of Subsonic Airfoils for High Lift,” Journal of
Aircraft, Vol. 15, No. 9, 1978, pp. 547–561.
doi:10.2514/3.58406
[6] Gai, S. L., and Palfrey, R., “Influence of Trailing-Edge FlowControl on
Airfoil Performance,” Journal of Aircraft, Vol. 40, No. 2, March–
April 2003, pp. 332–337.
doi:10.2514/2.3097
[7] Giguere, P., Dumas, G., and Lemay, J., “Gurney Flap Scaling for
Optimum Lift-to-Drag Ratio,” AIAA Journal, Vol. 35, No. 12,
Dec. 1997, pp. 1888–1890.
doi:10.2514/2.49
[8] Traub, L. W., and Galls, S. F., “Effects of Leading and Trailing Edge
Gurney Flaps on a Delta Wing,” Journal of Aircraft, Vol. 36, No. 4,
July–Aug. 1999, pp. 651–658.
doi:10.2514/2.2507
[9] Buchholz, M. D., and Tso, J., “Lift Augmentation on Delta Wing with
Leading-Edge Fences and Gurney Flap,” Journal of Aircraft, Vol. 37,
No. 6, Nov.–Dec. 2000, pp. 1050–1057.
doi:10.2514/2.2710
[10] Buchholz, M. D., “Lift Augmentation on Delta Wing via Leading-Edge
stabilizing
Fig. 12 Aerodynamic-center shift due to Gurney flap deployment.
680 GREENWELL
Fences and the Gurney Flap,” M.Sc. Thesis, California Polytechnic
State University, Dec. 1992.
[11] Li, Y., and Wang, J., “Experimental Studies on the Drag Reduction and
Lift Enhancement of a Delta Wing,” Journal of Aircraft, Vol. 40, No. 2,
March–April 2003, pp. 277–281.
doi:10.2514/2.3120
[12] Langan, K. J., and Samuels, J. J., “Experimental Investigations of
Maneuver Performance Enhancements on an Advanced Fighter/Attack
Aircraft,” AIAA Paper 1995-0442, Jan. 1995.
[13] Zhan, J., and Wang, J., “Experimental Study on Gurney Flap and Apex
Flap on a Delta Wing,” Journal of Aircraft, Vol. 41, No. 6, Nov.–
Dec. 2004, pp. 1379–1383.
doi:10.2514/1.4044
[14] Greenblatt, D., Kastantin, Y., Nayeri, C. N., and Paschereit, C. O.,
“Delta Wing Flow Control Using Dielectric Barrier Discharge
Actuators,” AIAA Paper 2007-4277, June 2007.
[15] Wentz, W. H., and Kohlman, D. L., “Vortex Breakdown on Slender
Sharp-Edged Wings,” Journal of Aircraft, Vol. 8, No. 3, March 1971,
pp. 156–161.
doi:10.2514/3.44247
[16] Greenwell, D. I., Dance, C., McFarlane, L., Wichert, P., and Delamore-
Sutcliffe, D., “Gurney Flaps on Slender and Non-Slender Delta Wings,”
AIAA Paper 2006-3568, June 2006.
[17] Demele, F. A., “Effects of a Leading-Edge Slat and a Trailing-Edge
Split Flap on the Aerodynamic Characteristics of a Wing –Fuselage
Combination Having a Nearly Triangular Wing of Aspect Ratio 2.9 at
Mach Numbers from 0.60 to 0.92,” NACA RM A57H19, Jan. 1954.
[18] Wang, J. J., Li, Y. C., and Choi, K.-S., “Gurney Flap—Lift Enhance-
ment, Mechanisms and Applications,” Progress in Aerospace Sciences,
Vol. 44, No. 1, 2008, pp. 22–47.
doi:10.1016/j.paerosci.2007.10.001
[19] Liu, T., and Montefort, J., “Thin-Airfoil Theoretical Interpretation for
Gurney Flap Lift Enhancement,” Journal of Aircraft, Vol. 44, No. 2,
March–April 2007, pp. 667–671.
doi:10.2514/1.27680
[20] Jeffreys, D., Zhang, X., and Hurst, D. W., “Aerodynamics of Gurney
Flaps on a Single-Element High-Lift Wing,” Journal of Aircraft,
Vol. 37, No. 2, March–April 2000, pp. 295–301.
doi:10.2514/2.2593
[21] Traub, L. W., Miller, A. C., and Rediniotis, O., “Preliminary Parametric
Study of Gurney-Flap Dependencies,” Journal of Aircraft, Vol. 43,
No. 4, July–Aug. 2006, pp. 1242–1244.
doi:10.2514/1.13852
[22] Thompson, B. E., and Lotz, R. D., “Divergent-Trailing-Edge Airfoil
Flow,” Journal of Aircraft, Vol. 33, No. 5, Sept.–Oct. 1996, pp. 950–
955.
doi:10.2514/3.47040
[23] Barrett, R. V., “Design and Performance of a New Low Turbulence
Wind Tunnel at Bristol University,” The Aeronautical Journal, Vol. 88,
No. 873, March 1984, pp. 86–90.
[24] Shindo, S., “Simplified Tunnel Correction Method,” Journal of
Aircraft, Vol. 32, No. 1, Jan. 1995, pp. 210–213.
doi:10.2514/3.46705
[25] Pope, A., Rae, W. H., and Barlow, J. B., Low-Speed Wind Tunnel
Testing, Wiley, New York, 1999.
[26] Wichert, P., “Experimental Investigation of Gurney Flaps on Low
Aspect Ratio Delta Wings,” M.Sc. Thesis, Univ. of Stuttgart, Stuttgart,
Germany, and Univ. of Bristol, Bristol, England, U.K., July 2004.
[27] Dance, C., and McFarlane, L., “Gurney Flaps on Delta Wings,”
Department of Aerospace Engineering, Univ. of Bristol, Project
Rept. 1123, Bristol, England, U.K., May 2005.
[28] Torlund, P.-A., “Wind Tunnel Force Measurements and Visualisation
on a 60
·
Delta Wing in Oscillation, Stepwise Motion and Gusts,”
Manoeuvring Aerodynamics, CP-497, AGARD, Paper 10, Neuilly-sur-
Seine, France, May 1991.
[29] Polhamus, E. C., “A Concept of the Vortex Lift of Sharp-Edge Delta
Wings Based on a Leading-Edge-Suction Analogy,” NASA TN-D-
3767, Dec. 1966.
[30] Ericsson, L. E., “Explanation for Huge Differences Between Measure-
ments of Vortex Breakdown on 65
·
Delta-Wing Configurations,” AIAA
Paper 1999-1410, Aug. 1999.
[31] Cavanaugh, M. A., Robertson, P., and Mason, W. H., “Wind Tunnel Test
of Gurney Flaps and T-Strips on an NACA 23012 Wing,” AIAA
Paper 2007-4175, June 2007.
[32] Li. Y. C., Wang, J. J., and Zhang, P. F., “Effects of Gurney Flaps on a
NACA0012 Airfoil,” Flow, Turbulence and Combustion, Vol. 68,
No. 1, 2002, pp. 27–39.
doi:10.1023/A:1015679408150
[33] Myose, R., Papadakis, M., and Heron, I., “Gurney Flap Experiments on
Airfoils, Wings and Reflection Plane Models,” Journal of Aircraft,
Vol. 35, No. 2, March–April 1998, pp. 206–211.
doi:10.2514/2.2309
[34] Guzel, G., Sankar, L. N., and Rhee, M., “Computational Investigation
of the Effects of Gurney Flap on the Aerodynamic Performance of VR-
12 Airfoil,” AIAA Paper 2005-4960, June 2005.
[35] Carlson, H. W., and Mack, R. J., “Studies of Leading-Edge Thrust
Phenomena,” Journal of Aircraft, Vol. 17, No. 12, Dec. 1980, pp. 890–
897.
doi:10.2514/3.57981
[36] “Aerofoil and Wing Pitching Moment Coefficient at Zero Angle of
Attack Due to Deployment of Trailing-Edge Plain Flaps at Low
Speeds,” ESDU International, Data Item 98017, Nov. 2003.
[37] Stephenson, J. D., and Amuedo, A. R., “Tests of a Triangular Wing of
Aspect Ratio 2 in the Ames 12-Foot Pressure Wind Tunnel 2—The
Effectiveness and Hinge Moments of a Constant-Chord Plain Flap,”
NACA RM A8E03, Sept. 1948.
GREENWELL 681

. by a factor of around ‡30%.11] were compared with previous data for 40 delta wings (e. Tests with III.g. with a thickness/chord of 1. and 5% were tested at a Reynolds number based on root chord of 8:6  105 . 2 Delta-wing model geometry.6% and a 60 symmetric beveled leading-edge. Flaps were made of 1 mm aluminum sheet and attached to the lower surface of Fig. combined with an aft shift in aerodynamic center (Fig. and 70 were tested in the Bristol University 0:8  0:6 m Low-Turbulence Wind Tunnel [23]. the pressure difference between upper and lower surfaces at the trailing edge due to the separated wake of the flap [20] leads to an aft shift in loading and hence to an increase in effective chord [21. Pitch/lift ratio Cm0 =CL0 is negative and of a similar magnitude to conventional trailing-edge controls.246. 3. and pitching moment. and nose-down pitching moment are all consistent with an increase in effective camber. In addition. Most explanations of the effects of Gurney flaps suggest that the change in the trailing-edge Kutta condition due to flap deployment leads to a change in effective camber [18. Gurney flap heights of 1. However. with leading edges beveled on the lower surface at a 20 included angle (Fig. [4.17]). so again this data set must be considered unreliable. This tunnel is of conventional closed-circuit design. 2. it became clear that the basic wing lift-curve slope was far too high. an increase in lift at fixed incidence leads to a reduction in induced-drag factor.11] present work on a cropped 40 delta wing in two complementary papers. [15]). [1. with a thickness/chord of 3% and a 45 lower-surface beveled leading edge. and 5% were tested at a Reynolds number based on root chord of 2:5  105 . 1b). using data from the sole remaining reliable and comprehensive data set [8].. drag. Flap deployment gives a nose-down change in zero-lift pitching moment Cm0 . The increase in lift-curve slope and aft shift in aerodynamic center are consistent with an increase in effective chord. rather than to an incidencedependent change in profile drag. 1a). leading to a reduction in maximum lift/drag ratio …L=D†max and an increase in the lift coefficient for minimum drag (Fig. and pitching-moment measurements by 1. However. No obvious reason could be found for these discrepancies.. This suggests a possible error in the dynamic pressure measurement. 1c). 2. When the results presented in [4. Constant-height flaps with height/chord ratios of 1. Lift at zero incidence CL0 and maximum lift CL max both show a significant increase.1%. Forces and moments were measured using an AeroTech 3component overhead balance. The 3-D wing behavior shown in Fig. A flat-plate wing model was used. and 0.353. 3. An analysis of the entire data set [16] showed that dividing all lift. drag. the drag polar remains symmetric about zero lift. Models were made of 3 mm aluminum sheet. presumably because the increase in drag is due directly to the separated wake flow behind the flap. giving a model/tunnel area ratio of 0. A flat-plate wing model was used. 1 suggests that both factors are operating. and 6% of root chord were tested. Figure 1 illustrates the typical effects of Gurney flap deployment on lift. Zero-lift drag CD0 increases. respectively. along with a small increase in lift-curve slope (Fig.19]. 0. with the models mounted inverted on a single strut plus a pitch pushrod (Fig. but the reduction in induced-drag factor is less straightforward. Li et al. whatever the error source. Constant-height flaps with height/chord ratios of 1. The increase in lift. Model planform area S was held constant at 0:072 m2 .676 GREENWELL Buchholz and Tso [9] summarize a project study [10] of a range of larger Gurney flaps on a 60 delta wing. 60. but induced-drag factor k reduces (Fig. Reference [8] suggests that since the lift-dependent induced drag is equal to CL tan for a flat-plate delta wing with zero leading-edge suction. Experimental Arrangement Three sharp-edged flat-plate delta-wing models with leading-edge sweep angles  of 40.3 gave a more reasonable agreement with published 60 wing data and with the effects of flap height reported later in this paper. with flap height h measured from the wing lower surface. 1d) as dCm =dCL becomes more negative. 3). maximum lift.22]. with a large contraction ratio and flow control screens to give turbulence levels well below 0. 2) and with a blunt trailing edge. a) b) c) d) Fig. the wing. it was noted that the lift-curve slope was too low by a factor of 20% (allowing for taper ratio) and that the aerodynamic center was misplaced. 1 Typical effects of Gurney flap on delta-wing aerodynamics (data from [1]).436 m.g. when the results presented in [9] were compared with previous data for 60 delta wings (e. A small underwing fairing covered the strut and pushrod attachment points.15 and a root chord cr of 0. this data set is clearly unreliable. 3.

1 . Fig. Balance accuracy was estimated at better than 0. Normal force and pitching moments compare very well.32–34].27]. Solid and wake blockage were corrected for using the method of [24]. Lift Fig. the question arises as to the relative effect of a Gurney flap on the two lift components.7. In order to verify the accuracy of the experimental procedure and data reduction. with pitch angle set to better than 0. Figure 4 compares the basic 60 sweep wing results with data from Flygtekniska Försöksanstalten (FFA) in Sweden [28]. The difference between the 2-D and 3-D data in Fig. where the lift may be split into potential (or attached) and leading-edge vortex (or separated) flow contributions [29]. 5. The crosses show data for a range of 2-D airfoils with Gurney flaps [6. 5 in log–log form indicates that the variation for both 2-D and 3-D wings is closer to n ˆ 0:7 (with q ranging between 3 and 6). Off-surface flow Fig. The effect of variations in lift-curve slope a0 can be accounted for by replacing lift increment CL0 with change in zero-lift incidence  0L  CL0 =a0 . inverted (i. with downwash corrections from [25]. Figure 6 demonstrates that applying both factors collapses the scattered 2-D and 3-D lift increment data of Fig. Figure 4 also shows normal force data from the seminal study of delta-wing aerodynamics by Wentz and Kohlman [15] and the theoretical normal force given by the leading-edge suction analogy (LESA) [29]. drag. Tests were carried out at a constant Reynolds number based on root chord of 0:7  106 . and the open diamond symbols denote a rectangular wing with an aspect ratio of 6 [31]. IV. A similarly good agreement with published data was obtained for the clean 40 and 70 wings and for the 70 wing with a 1% flap [16]. 5 reasonably well onto a single trend:  0L  0:9…Sf =S†0:7 (2) For the specific case of delta wings. with identical lift-curve slopes and aerodynamic centers at low incidence. The experimental data reviewed in [19] also shows no systematic variation of q with Reynolds number. upper surface) flaps established that the underwing fairing had no significant effect on trends in aerodynamic behavior with flap height. The same symbols are used in the following figures. 5 can be attributed to two factors: 1) the effect of aspect ratio on lift-curve slope and 2) the effect of taper ratio on the relative flap height. [19] proposes that the lift increment due to a Gurney flap is CL0 ˆ q…h=c†n (1) where q is supposedly a function of Reynolds number (varying between 2 and 4) and n ˆ 0:5..15% in lift. The FFA wing had a very similar geometry (a strut-mounted flat-plate model with an underbody fairing) and was tested at the same Reynolds number of 0:71  106 . 5 shows the increment in lift coefficient at zero incidence CL0 as a function of relative height h=cr (following previous convention in using the root chord as a reference length for 3-D wing data).27]. Fig. Pitching moments were nondimensionalized using the mean aerodynamic chord c (2cr =3) and taken about a moment reference center at the quarter-chord point (0:5cr ).e. the baseline delta-wing data was compared with a range of published results for similar wings. 3 Model mounted in the Bristol Low-Turbulence Tunnel. and pitching moment. Further details of the experimental arrangement can be found in [26. 5 Effect of flap height on lift increment at zero incidence. 4 Comparison of basic 60 wing with theory and results from Torlund [28] and Wentz and Kohlman [15]. the most straightforward weighting method is to use the relative flap area Sf =S ˆ …hb†=S. The effect of taper ratio can be accounted for by weighting the flap height by the local chord. Effect of Gurney Flap Height A. to allow for the effect of different leading-edge Looking first at the effect of flap height on lift. plotting the data of [19] and Fig. However. For the constant-flap-height data of Fig. The filled symbols denote increments derived from the Bristol data of [26. Angle of attack is given relative to the zero-lift incidence 0L . . On the basis of thin-airfoil theory.GREENWELL 677 chamfer angles [30] and support interference.

An indication of the relative magnitudes of the potential and vortex lift components can be obtained by fitting the LESA lift model [29] CN ˆ CNP ‡ CNV ˆ KP sin… † cos… † ‡ KV sin2 … † (3a) Fig. the induced drag increases. giving Dmin hb Sf ˆ 1 2 ˆ CD min S  hb ! CD min  V S S 2 (4) as demonstrated in Fig. with a small contribution from the vortex lift.75 Modified LESA fit to Fig. Following the lead given by Fig. the lift-dependent drag component of the normal force becomes CDi ˆ CN sin… † ˆ CL tan… † so that the induced-drag factor k is k ˆ AR CDi tan… † ˆ AR 2 CL CL (6) (5) Since the effect of a Gurney flap on a delta wing is to increase lift CL at a fixed incidence . 7. it is clear that the Gurney flap has a much greater effect on the potential lift component than on the vortex lift. the increase in parasite drag area D= 1 V 2 for both 2-D 2 and 3-D wings is approximately equal to the frontal area of the flap hb. measurements and visualizations reported in [8] indicate that a 1% Gurney flap on a 70 delta wing has a negligible effect on the strength of the primary leading-edge vortex and on the variation of vortex breakdown position with angle of attack. For the LESA model of Eq. good agreement was obtained by modeling the effects of a Gurney flap as shifts 0P and 0V in the effective incidences of the potential and attached-flow components CN ˆ KP sin… ‡ KV sin… 0P † cos… 0V †j sin… 0P † 0V †j (3b) Figure 8 shows profile drag penalties for 2-D airfoils and 3-D wings as a function of flap height. (3b). dCNP ˆ Kv sin…2 † d giving a small increase in local lift-curve slope at low incidences. Although the results of fitting a nonlinear equation to experimental data should be treated with caution. At a given incidence. Eq.75 1. As sweep angle is reduced. due to the effects of nonzero incidence and partial recovery of leading-edge suction. The corresponding lift factors KP and KV and zero-lift offsets 0P and 0V . This is consistent with the results reported in [8]. the induced-drag factor is a nonlinear function of incidence. are shown in Table 1. as determined from a nonlinear least-squares fit to the data. 7 Effect of Gurney flap height on potential and vortex lift components for a 70 delta wing.14 0P 0:5 4:5 9:5 0V 0:5 1:0 2:0 Figure 9 shows that Gurney flap deployment also has a significant impact on induced drag for flat-plate delta wings. but that the height of the vortex core above the wing surface was reduced slightly. B. Applying the analysis of [8]. with the change in zero-lift incidence 0P of the order of six times bigger than the change in the vortex zero-lift incidence 0V . 6.678 GREENWELL Fig. because the increased aspect ratio outweighs the effect of increased lift-curve slope. (6) shows that the corresponding induceddrag factor must reduce as flap height increases. . Plotting the change in minimum drag CD min against relative flap area Sf =S collapses 2-D and 3-D data onto a single linear trend. compared with Eq. As one might expect for a bluff-body wake flow. The reduction in vortex height at a given angle of attack would be expected to increase the vortex lift component. (3). At low angles of attack.14 3. with a limiting (maximum) value at zero incidence of k !0 ˆ AR KP (7) Equation (7) gives induced-drag factors rather higher than those shown in Fig. Drag to the experimental normal force data. the reduction in local potential lift-curve slope dCNP ˆ Kp cos…2 † d as the potential lift curve moves to the left is counterbalanced by the increase in the vortex lift-curve slope Table 1 h=cr 0 1% 3% KP 1. the change in lift due to flap deployment is largely due to the change in potential lift. 9.14 3.75 1. but it does indicate the general effects of wing planform. due to the loss of leading-edge suction for a sharp-edged wing. This shows the variation in normal force for a 70 delta wing with 1 and 3% Gurney flaps. 6 Effect of flap area on zero-lift incidence. 7 KV 3.

The effect on …L=D†max depends on whether CD0 increases more rapidly than k reduces. 8 and 9. For 3-D wings with Gurney flaps. It should be noted that for practical wings with rounded leading edges. since   L 1 / p . These two wings are not directly comparable with the data reported here. as zero-lift drag CD0 increases and induceddrag factor k reduces. which in turn depends on the clean-wing profile drag. Gurney flaps on a UCAV-type planform with an airfoil cross section are therefore likely to have a much more adverse impact on L=D than that suggested by tests on flat-plate wings [8. The corresponding effects of flap on lift/drag ratio (e. Fig. however. 1c) follow directly from Fig. the ratio is greater than 2-D theory predicts and increases with reducing aspect ratio up to about 0:35 for a 70 delta. This is the opposite of the trend predicted by [36] for wings with fully attached flow. Fig. 8 Effect of flap height on increment in minimum drag. For example. and so reductions in k due to flap deployment will be much smaller. D max CD0 k CLmd  r CD0 / k (8) The lift coefficient CL. .. 5). Fig. …L=D†max is always reduced by flap deployment. Fig. published data for simple delta wings with Fig. Fig. 10 indicates that the pitch/lift ratio tends to be slightly less than this theoretical value. Two-dimensional thin-airfoil theory gives a limiting maximum value for pitch/lift ratio CM0 =CL0 of 0:25. since they had airfoil cross sections and were tested at much higher Reynolds numbers.g.32–34].GREENWELL 679 Pitching Moment C. the rectangular wing with a NACA23012 airfoil section reported in [31] shows very little variation in induced-drag factor with flap deployment. 10 plots zero-lift pitching-moment change Cm0 against change in lift at zero incidence CL0 for a range of wings with Gurney flaps (symbols as Fig. Fig. For 2-D airfoils with Gurney flaps [6.md for maximum L=D ratio increases rapidly with flap deployment. In order to illustrate the effect of wing and flap geometry on this parameter. The ratio of pitching-moment change to lift change (pitch/lift ratio) is a critical parameter for a control effector on a tailless aircraft. For comparison purposes. 10 also shows pitch/lift ratios for two delta wings with trailing-edge flaps: a 41 delta with a split flap [17] and a 63 delta with a trailing-edge flap [37]. For the data presented here.7. partial recovery of leading-edge suction leads to markedly lower levels of induced drag [35]. 9 Effect of flap height on induced-drag factor.16]. Fig. 11 Trim drag for conventional controls and Gurney flaps.10. Figure 11 shows the corresponding trim-drag penalty CD min as a function of lift increment. 10 Pitch/lift ratio for conventional controls and Gurney flaps.

“Application of Circulation Control Manoeuvre Effectors for Three Axis Control of a Tailless Flight Vehicle. 2. J. June 2002. 15. two out of three of the most comprehensive data sets available have lift.. N.” Journal of Aircraft. Gordnier. Tan. Wood.” Journal of Aircraft. maximum lift. F. S.002 [2] Sellars. 11). Dumas.2514/2. The cause of the aerodynamic-center movement with Gurney flap deployment appears to be an aft shift in chordwise loading distribution.49 [8] Traub. J. The usual effective camber explanation of the effects of Gurney flaps on 2-D airfoils has some shortcomings when applied to delta wings. I.–Dec. No. C.680 GREENWELL stabilizing Fig. which tends to counteract the increase in profile drag due to flap frontal area and hence gives a relatively small reduction in L=D. 651–658. R. P. spanning vortex flow types from nonslender to slender.” Journal of Aircraft.. “Design of Subsonic Airfoils for High Lift. No. and Crowther. “Influence of Trailing-Edge Flow Control on Airfoil Performance.” AIAA Journal. Bearing in mind the different test conditions. P. and pitching moment are similar for delta wings with sweep angles of 40. and pitching-moment data that are clearly in error. 40. the drag polars remain symmetric about zero lift. [3] Harley.. giving an increase in profile drag CD0 that is approximately equal to the relative frontal area Sf =S of the flap. The most significant way in which Gurney flaps differ from conventional trailing-edge controls is in their effect on aerodynamic center.. D. The observed increases in lift. However. “Effects of Leading and Trailing Edge Gurney Flaps on a Delta Wing. and Lemay. 99–105. “Lift Augmentation on Delta Wing via Leading-Edge V. Data comparisons have highlighted a number of issues with the limited literature on delta wings with Gurney flaps.2514/2. However. “Gurney Flap Scaling for Optimum Lift-to-Drag Ratio. W. Figure 12 indicates that conventional trailing-edge controls on delta wings can also cause an aft shift in aerodynamic center. trailing-edge flaps is remarkably scarce. 4. 36.” AIAA Paper 2002-3269. and David Delamore-Sutcliffe for their assistance with the experimental work and data analysis. and MAV wings with rounded leading edges. Vol. The Gurney flap generates a bluff-body wake. drag.1007/s003480200010 [5] Liebeck. while the effect of wing taper can be accounted for by using the relative flap area Sf =S rather than the height/chord ratio h=cr . 1.. J. References [1] Gursul. 32.. which makes comparisons and assessments of trends very difficult. No. doi:10.2514/2. No. and Zhang.. No. these errors were not recognized before publication. Vol. For the simple flat-plate delta-wing models tested here. Dec. as normal force is increased at a fixed incidence. and Visbal. 12 Aerodynamic-center shift due to Gurney flap deployment.” Progress in Aerospace Sciences. Gurney flaps would appear to have a similar pitch control capability to conventional flaps. Fitting experimental data to the LESA model indicates that on a delta wing the Gurney flap acts primarily to change the zero-lift incidence of the attached-flow potential lift component.. In general. 547–561.1016/j. 41.. The large aft shift in aerodynamic center is due to the local separated-flow effects near the trailing edge. drag. 37. “Effects of Gurney Flaps on the Lift Enhancement of a Cropped Non-Slender Delta Wing. The magnitude of the shift is roughly proportional to the lift increment.. H. and Kennaugh.58406 [6] Gai. “Delta Wing Circulation Control Using the Coanda Effect... No. and Galls. July–Aug. pp.09. 1997. doi:10. J. R. with a much smaller effect on the separated-flow vortex lift component.. doi:10. More specifically. doi:10. but because no comparison was made with other (readily available) experimental data sets for delta wings. G. and the shift changes sign at higher control deflections. many papers do not clearly specify the moment reference center. I. M. 12.3097 [7] Giguere.2005. which results in unrealistically high induceddrag factors for the clean wings. S. doi:10. when appropriately parameterized.” Experiments in Fluids.” AIAA Paper 2009-146. F... 515–557. [4] Li. Nov. N. doi:10. but the magnitude is much smaller. 2009. Vol. D. but this is offset by a much higher trim-drag penalty and by an associated aft (stabilizing) shift of the aerodynamic center. 1999. and Tso. and Palfrey.2710 [10] Buchholz. “Lift Augmentation on Delta Wing with Leading-Edge Fences and Gurney Flap. Vol. L. pp. 2005. G. J.2514/2. L. A. 6. due to the finite pressure difference between the upper and lower surfaces induced by the flap wake at the trailing edge [20]. and 70 . Gurney flaps give very large improvements in induced drag. The induced-drag reduction with flap deployment for a flat-plate wing is essentially a geometry effect. where the pressure difference between upper and lower surfaces at the trailing edge due to the flap wake leads to an aft shift in loading and hence to an increase in effective chord. pp. W. No. D. as shown in Fig.. 7. P. M. or the point from which flap height is measured.. pp. Wilde. Gurney flap deployment gives a significant aft shift of the aerodynamic center relative to the clean wing. “Unsteady Aerodynamics of Non-Slender Delta Wings. 9.2514/3. pp.. This is due to the nearly 100% loss of leading-edge suction on a flat-plate wing with a sharp leading edge. Y. D. Louise McFarlane. or with simple theoretical predictions. 60. For both 2-D and 3-D wings. Conclusions A parametric wind-tunnel study has demonstrated that the effects of Gurney flap deployment on lift. For practical UCAV . Vol.. the moment reference length. 2002. the clean-wing induced-drag factors (and hence the relative improvements due to Gurney flap deployment) will be much smaller. pp.. pp. Wang.” Journal of Aircraft. Acknowledgments The author wishes to thank Patrick Wichert.paerosci. doi:10. Jan. 332–337. March– April 2003. The effects of flap height on lift and profile drag are also consistent with the extensive literature on 2-D Gurney flaps. Chris Dance. the associated trim drag is on the order of four to five times higher (Fig. M. 12.. J. Pitch/lift ratios for delta wings with Gurney flaps are similar to those for delta wings with trailing-edge control surfaces and are significantly more negative than for conventional wings with attached flow. The influence of wing aspect ratio can be accounted for by using the shift in zero-lift incidence  0L rather than the increase in lift CL . Vol... C. 1888–1890. and the vortex lift component is relatively unaffected.2507 [9] Buchholz. 1978.. R. 2000. 35. A. Vol. and nose-down pitching moment are all consistent with a change in effective camber. and so adverse effects on L=D are likely to be much larger. K. 1050–1057. these two data sets indicate that Gurney flaps on delta wings have pitch control characteristics similar to conventional trailing-edge flaps.

[11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] Fences and the Gurney Flap.2514/3. and Rediniotis.2514/3. and Mason. March–April 2007. No. Zhan.. W. L. J.” Progress in Aerospace Sciences. Wentz. J. R.. H. of Bristol. W.” Manoeuvring Aerodynamics.. England.. “Gurney Flaps on Delta Wings. Vol.2309 Guzel. 1948.. 1. and Heron. R. Y. Wichert.. P. 2002. Thesis.2514/2. Jan..1023/A:1015679408150 Myose. H..10.. Vol.Sc. Sept.” The Aeronautical Journal. “Experimental Studies on the Drag Reduction and Lift Enhancement of a Delta Wing. L. 1123.K. 890– 897. pp. July 2004. No. No. M. 1. “Tests of a Triangular Wing of Aspect Ratio 2 in the Ames 12-Foot Pressure Wind Tunnel 2—The Effectiveness and Hinge Moments of a Constant-Chord Plain Flap.2593 Traub. England. 88. Jan. Vol.2514/2. Wang. “Thin-Airfoil Theoretical Interpretation for Gurney Flap Lift Enhancement. Project Rept. and Kohlman.” Journal of Aircraft...” Journal of Aircraft. Stephenson.. Sept. “Simplified Tunnel Correction Method. I.. pp.. C.1016/j. Li. “Gurney Flap Experiments on Airfoils. Dance. Vol. 1996. doi:10. 40. L. 8. Rae.... March 1984.” AIAA Paper 1995-0442. C. Vol. D. and McFarlane. 1999. and Paschereit. doi:10. “Divergent-Trailing-Edge Airfoil Flow. Torlund. Low-Speed Wind Tunnel Testing. pp. and Samuels. No. No. No. M. pp. “A Concept of the Vortex Lift of Sharp-Edge Delta Wings Based on a Leading-Edge-Suction Analogy. D. 27–39.. 2. Mechanisms and Applications.” Flow. May 2005. Aug. 1995. C.2514/3. pp.” AIAA Paper 2006-3568. pp. Vol.” M. and Univ.-A.9 at Mach Numbers from 0. J. A.3120 Langan.. Stuttgart. doi:10. “Gurney Flaps on Slender and Non-Slender Delta Wings. J. 1980.57981 “Aerofoil and Wing Pitching Moment Coefficient at Zero Angle of Attack Due to Deployment of Trailing-Edge Plain Flaps at Low Speeds.–Oct. Demele. No. 4. 17. March–April 2003. Turbulence and Combustion. 6. P. and Hurst. “Experimental Investigation of Gurney Flaps on Low Aspect Ratio Delta Wings.2514/1.” NACA RM A8E03.. pp. 277–281.. S.” Department of Aerospace Engineering.. March–April 1998. F. Miller. A. U. J. No. W. O. and Barlow. Stepwise Motion and Gusts.. Neuilly-surSeine.. 1. and Wang. 68. W. 37.. C. Dance. A. Wiley. L. J. 1995.. 2. Shindo. F. “Effects of a Leading-Edge Slat and a Trailing-Edge Split Flap on the Aerodynamic Characteristics of a Wing –Fuselage Combination Having a Nearly Triangular Wing of Aspect Ratio 2. pp. 873.” Journal of Aircraft. “Effects of Gurney Flaps on a NACA0012 Airfoil. March–April 2000. K. H. of Bristol. “Preliminary Parametric Study of Gurney-Flap Dependencies.2514/3. E. D. R. “Wind Tunnel Force Measurements and Visualisation on a 60 Delta Wing in Oscillation... 156–161. No. W. pp. 86–90.” AIAA Paper 2005-4960. “Gurney Flap—Lift Enhancement. Vol. Sankar.. 1379–1383. doi:10.60 to 0. Wang.2514/2.. Robertson.. “Studies of Leading-Edge Thrust Phenomena.47040 Barrett. pp..” NACA RM A57H19.” M. J. T. “Aerodynamics of Gurney Flaps on a Single-Element High-Lift Wing.. “Vortex Breakdown on Slender Sharp-Edged Wings. Dec. 2. 2.” AIAA Paper 1999-1410. June 2007.. June 2007. Vol. L. 44. K.2514/1. 1999. E.” Journal of Aircraft.13852 Thompson. AGARD.GREENWELL 681 Wind Tunnel at Bristol University. H. P. and Zhang. doi:10. Nov.paerosci. May 1991.” Journal of Aircraft.” AIAA Paper 2007-4175. Univ.. and DelamoreSutcliffe. Y. and Lotz.. and Choi. and Montefort. Dec. M. and Rhee. B. Vol. 12.” Journal of Aircraft. D. “Explanation for Huge Differences Between Measurements of Vortex Breakdown on 65 Delta-Wing Configurations. D. 22–47. July–Aug. McFarlane. C. “Design and Performance of a New Low Turbulence [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] . Carlson.. New York. No. J. Univ. U. 950– 955. doi:10. and Mack. No. G. 2008. D.-S...4044 Greenblatt. pp.27680 Jeffreys. Bristol. doi:10. E. X. doi:10. of Stuttgart...” ESDU International. California Polytechnic State University. doi:10. Y. R.” Journal of Aircraft. 32. C. I.” NASA TN-D3767.. D. Wings and Reflection Plane Models.. 1966. No. 43. Vol. J. J.K. N. V. Vol. N. CP-497. Nayeri.” AIAA Paper 2007-4277.92..2514/1. doi:10. O. Nov. B. Zhang. Germany. “Wind Tunnel Test of Gurney Flaps and T-Strips on an NACA 23012 Wing. Polhamus.. Cavanaugh.. and Amuedo. Dec. March 1971.001 Liu. Papadakis.. J.. D. Jan. 2003..44247 Greenwell. L. pp. A. C..2007. doi:10. 44. Li.. and Wang. Bristol. 33. pp. P.” Journal of Aircraft. Vol.” Journal of Aircraft. J. 1242–1244.. P. 206–211.. 5. R. June 2005. Thesis. 1954. “Experimental Investigations of Maneuver Performance Enhancements on an Advanced Fighter/Attack Aircraft. 2004. Y..Sc.. “Delta Wing Flow Control Using Dielectric Barrier Discharge Actuators. 35. 2006. Data Item 98017. Kastantin. 667–671. W. 210–213. 3. C. J. “Experimental Study on Gurney Flap and Apex Flap on a Delta Wing. Ericsson. 1992.46705 Pope.. Wichert.. Li.. A. doi:10. 41.. France. J.. No. June 2006. 295–301. “Computational Investigation of the Effects of Gurney Flap on the Aerodynamic Performance of VR12 Airfoil. Vol. Paper 10.” Journal of Aircraft.– Dec.