Experiments in Fluids

, Volume 50, Issue 6, pp 1671–1684

Lift enhancement and flow structure of airfoil with joint trailing-edge flap and Gurney flap

Research Article

DOI: 10.1007/s00348-010-1024-8

Cite this article as:
Lee, T. & Su, Y.Y. Exp Fluids (2011) 50: 1671. doi:10.1007/s00348-010-1024-8

Abstract

The impact of Gurney flaps (GF), of different heights and perforations, on the aerodynamic and wake characteristics of a NACA 0015 airfoil equipped with a trailing-edge flap (TEF) was investigated experimentally at Re = 2.54 × 105. The addition of the Gurney flap to the TEF produced a further increase in the downward turning of the mean flow (increased aft camber), leading to a significant increase in the lift, drag, and pitching moment compared to that produced by independently deployed TEF or GF. The maximum lift increased with flap height, with the maximum lift-enhancement effectiveness exhibited at the smallest flap height. The near wake behind the joint TEF and GF became wider and had a larger velocity deficit and fluctuations compared to independent GF and TEF deployment. The Gurney flap perforation had only a minor impact on the wake and aerodynamics characteristics compared to TEF with a solid GF. The rapid rise in lift generation of the joint TEF and GF application, compared to conventional TEF deployment, could provide an improved off-design high-lift device during landing and takeoff.

List of symbols

b

Wing span

c

Airfoil chord

Cd

Section drag coefficient

Cl

Section lift coefficient

Cl,max

Maximum lift coefficient

Cl

Lift-curve slope

Cm

Section pitching moment coefficient about ¼-chord

Cm,peak

Peak pitching moment coefficient

Cp

Surface pressure coefficient

d

Perforation hole diameter

h

Gurney flap height

Re

Reynolds number, = Uc

u

Mean streamwise velocity

u

Streamwise velocity fluctuation

U

Freestream velocity

x, y, z

Streamwise, transverse and spanwise direction

α

Angle of attack

αss

Static-stall angle

αzl

Zero-lift angle

δ

Trailing-edge flap deflection

σ

Flap porosity

ζ

Mean streamwise vorticity

ν

Kinematic viscosity

1 Introduction

The enhancement of airfoil lift and the maintenance of low drag and nose-down pitching moment have always posed a challenge to aerodynamicists and fluid dynamicists as well. Various passive and active control means have been attempted to maximize the airfoil lift generation with a minimum drag penalty. Among them, trailing-edge flow control schemes, via the use of, for example, Gurney flap and trailing-edge flap, have been studied extensively by researchers elsewhere to maintain the high lift required, especially, for approach and landing and takeoff. The Gurney flap, which was first introduced by Dan Gurney to increase the downforce on the racing car to improve tire adherence for lateral traction, required during high-speed turns, is considered to be promising for significantly increasing the lift coefficient with small changes in the drag and the stalling incidence (for example, Liebeck 1978; Neuhart and Pendergraft 1988; Storms and Jang 1994; Giguere et al. 1997; Myose et al. 1998; Jang et al. 1998; van Dam et al. 1999; Jeffrey et al. 2000; Stanewsky 2001; Zerihan and Zhang 2001; Li et al. 2003; Meyer et al. 2006; Troolin et al. 2006; Liu and Monteford 2007; Wang et al. 2008; Maughmer and Bramesfeld 2008). The Gurney flap is a simple device located at the trailing edge of the airfoil on the pressure side and perpendicular to the chord with its height remaining usually within 2–3% of the airfoil chord. It is now known that the flap leads to an increase in flow turning (aft camber) near the trailing edge, rendering a significant increase in the lift force. The downward turning of the mean flow also increases the airfoil circulation and shifts the location of rear stagnation point at a given angle of attack. The Gurney flap also decelerates the flow and thus increases the pressure on the pressure side of the airfoil. The pressure difference between upper and lower surfaces at the trailing edge also leads to an aft shift in loading and hence to an increase in effective chord and nose-down pitching moment. However, depending on the height of the flap, the sectional drag is increased, which can outweigh the increment in the sectional lift and result in a deteriorated lift-to-drag ratio in comparison with the unflapped airfoil.

It is noteworthy that the Gurney flap concept has also been extended to oscillating wings (Kentfield and Clavelle 1993; Gerontakos and Lee 2006a; Yee et al. 2007) and dynamically deployed microtabs (Tang and Dowell 2007; Chow and van Dam 2006). Chow and van Dam (2006) computationally investigated the flow around a NACA 0012 airfoil equipped with a dynamically deployed microtab device. Tang and Dowell (2007) studied the aerodynamic loadings on a NACA 0012 airfoil with an oscillating trailing-edge Gurney flap via dynamic pressure distributions on the airfoil surface at Re = 3.48 × 105. Their measurements showed that an increase in the oscillating frequency increased the aerodynamic loadings and that the oscillating small flap can be a useful tool for active aerodynamic flow control. On the other hand, Meyer et al. (2006) investigated the effects of Gurney flaps (of a height of 1% chord) with three-dimensional modifications by introducing irregularities into the Gurney surface consisting of segments, V-shapes cut-outs, etc. These modifications led to a reduced lift and drag but an improved lift-to-drag ratio compared to a solid Gurney flap. Limited data for Gurney flaps with two sizes of holes (i.e., 0.3 and 0.5% chord in diameter) were also reported. Upstream Gurney flaps were also explored by Meyer et al. It should also be noted that Traub et al. (2006) also performed a preliminary study that included a porous Gurney flap with a height of less than or equal to 1.5% of the airfoil chord and a porosity of 22% at Re = 5.7 × 105. Recently, the effects of large Gurney-type flaps (with a flap height up to 12% of the airfoil chord), of different perforations, on the aerodynamic loading coefficients of the airfoil were examined by Lee and Ko (2009). The large flaps acted as an asymmetric bluff body and produced a large separated vortex shedding wake behind the two-dimensional flapped airfoil. The addition of flap perforation significantly reduced the wake size and unsteadiness compared to the solid flap. The flap perforation also reduced the positive aft camber effects (introduced by an otherwise solid flap) and destructed the recirculation flow (forming on the upstream face of the flap), which resulted in a decreased upper- and lower-surface pressure differential compared to the solid flap. The reduction in the drag, however, outweighed the loss in the lift and rendered an improved lift-to-drag ratio, compared to the solid flap.

On the other hand, trailing-edge flaps have been employed extensively as a routine practice of controlling the lift, by temporarily altering airfoil camber, especially during takeoff and landing, without penalizing cruise performance (see, for example, Smith 1975; Gai and Palfrey 2003; Perry and Mueller 1985; Vipperman et al. 1998). The trailing-edge flap concept has also been used in an attempt to control the unsteady lift including flutter suppression and gust alleviation (Theodorsen 1935). The effect of conventional trailing-edge flap deflection is to shift the lift curve to the left, due to the fact that when the flap was deflected downward, the effective camber of the airfoil was increased. The lift-curve slope of the flapped airfoil, however, remained essentially the same as that for the baseline, or unflapped, airfoil. Meanwhile, the zero-lift angle of attack was simply shifted to a lower value. The flapped airfoil also produced a larger Cl,max which generally occurred at a smaller angle of attack than that for the unflapped airfoil. More recently, trailing-edge flaps had also been used as unsteady aerodynamic control devices for the control of the transient lift on maneuvering fighter aircraft and the large reduction in the negative pitch moment on airfoils undergoing dynamic-stall oscillations (see, for example, Gerontakos and Lee 2006b).

The objective of this experimental study was to investigate the aerodynamic characteristics of a NACA 0015 airfoil equipped with a trailing-edge flap in conjunction with Gurney flaps, of different heights and perforations. The lift and pitching-moment coefficients were obtained through the integration of surface pressures measured at a finite number of locations of the airfoil surface. Near-wake velocity and vorticity flowfield, measured with hot-wire anemometry and particle image velocimetry, were also obtained to supplement the surface pressure measurements. The hot-wire wake measurements were also used to determine the values of drag coefficient.

2 Experimental methods

2.1 Flow facility and test model

The experiment was conducted in a 0.9 m × 1.2 m × 2.7 m low-speed, suction-type wind tunnel, with a freestream turbulence intensity of 0.1% at U = 15.2 m/s, in the Aerodynamics Laboratory in the Department of Mechanical Engineering at McGill University. A rectangular wing of a NACA 0015 airfoil section, fabricated from solid aluminum, with a chord c = 25 cm and a span b = 38 cm, was used as the test model. The origin of the coordinate system was located at the leading edge of the airfoil. The wing was equipped with end disks, of a diameter of 50 cm, to mitigate the free end effects. The gap between the wing model and the end disk was kept at less than 1 mm to minimize the leakage flow through the gap. The two-dimensionality of the flow distribution over the wing was checked by traversing a 5-μm normal hot-wire probe located at 30%c downstream from the leading edge of the wing and y = 5 mm above the airfoil. The non-uniformity was found to be ±3% of the freestream velocity.

The wing was also equipped with a simple hinged full-span 25%c trailing-edge flap (referred to as TEF), which can be deflected downward with a maximum deflection amplitude δ of 18°. The gap between the top and bottom surfaces was sealed. The schematics of the airfoil model and the trailing-edge flap are shown in Fig. 1a. In addition, Gurney flaps (referred to as GF), of different heights (h = 0.7, 1.5, 2.1, 3, 4.5 and 6%c) and porosities (σ = 0, 23 and 40%c based on open to closed area of the flap surface), were also equipped to the wing trailing edge. The height was measured from the lower wing surface. The Gurney flaps were attached to the trailing edge by using double-sided Mylar adhesive film. For h ≥ 3%c, the perforated Gurney-type flaps were made from thin perforated sheets of aluminum, of a thickness of 1 mm, with staggered circular holes of d = 1.6 and 3.2 mm (models A-PMA-062 and A-PMA-125 distributed by Small Parts Inc.; see Fig. 1b). The perforated flaps were simply taped to simulate the solid Gurney flaps. For h < 3%c, the perforation was hand-drilled. The chord Reynolds number was fixed at 2.45 × 105. No wind tunnel wall corrections were made in the present paper. Wind tunnel wall corrections can be found in Pope and Rae (1984) and Ewald (1998).
Fig. 1

Schematics of a airfoil model and PIV field of view and b perforated GF. TEF trailing-edge flap, GF Gurney flap

The surface pressure distribution on the airfoil was obtained via 48 0.35-mm-diameter pressure orifices, covering up to x/c = 96.3%, distributed over the upper and lower surfaces of the airfoil model. The orifices were staggered 1.5 mm apart in the spanwise direction to avoid the wake effect from an upstream orifice on orifices further downstream. The surface pressures were scanned electronically through a 48-port scanivalve system, via 22-cm long and 0.75-mm i.d. plastic tubings that separated the surface orifice and the pressure transducer. The pressure signals were integrated numerically to compute the aerodynamic load and pitching moments. The dynamic range of the pressure transducer was on the order of 10 kHz. The transducer signals were low-pass filtered (250 Hz) and amplified with a multi-channel AA Lab model G3006 pressure measurement system. The effects of the plastic tubing on the unsteady pressure signals were examined by comparing the transducer output level and the phase with a controlled acoustic sound source. The effect of the length of the plastic tubing was a simple time constant delay on all pressure signals. Details of this method can be found in the work of Rennie and Jumper (1996) and Lee and Basu (1998). An uncertainty analysis gave a typical total uncertainty ±0.013 in Cp. The wake velocity profiles, at selected α, are also obtained by using a miniature hot-wire probe operated by an AA Lab Model AN-1000 constant-temperature anemometer. The probe was located at x/c = 0.6 downstream of the airfoil trailing edge. The probe was traversed along the y-axis, with an increment of 3.2 mm or ∆y = 1.25%c, via a computer-controlled traversing mechanism. The hot-wire signals were sampled at 2 kHz for 10 s at each y-location. The near-wake velocity and vorticity flowfields were also obtained by using particle image velocimetry. The Cd values were determined via the wake measurements (see, for example, Althaus and Wortmann (1981) and Maughmer and Bramesfeld (2008)).

2.2 Particle image velocimetry (PIV)

The PIV experiment was performed in a 20 cm × 20 cm × 85 cm wind tunnel at U = 9.5 m/s or Re = 6.1 × 104. A black anodized NACA 0015 wing model, with c = 10 cm and b = 19 cm, mounted horizontally in the center of the test section was used as the test model. The wing pitch axis was located at ¼-c location. The wing model was also equipped with a 25%c trailing-edge flap and Gurney flaps, of different h and σ. The flowfield was illuminated by two 1.7-mm-thick laser light sheets, generated by a dual head Continuum Nd:YAG laser (Model SureLite II), pulsed at 10 Hz, and separated by a time delay of 49.4 μs. The particle images were digitally acquired using a TSI PowerView 4MP Plus CCD camera (Model 630059), with a resolution of 2,048 × 2,048 pixels, via a 64-bit frame grabber installed in a Dell Precision 690 workstation. Timing and control of the PIV system, which included the lasers, CCD camera, frame grabber, and synchronizer, was accomplished by a programmable LaserPulse synchronizer (Model 610035) with a time resolution of 1 ns. Seeding of the flow with particles of propylene glycol (Fisher P355-1) was accomplished by using a TSI 6-jet atomizer (Model 9306) in conjunction with a custom-built diffuser. Also, in order to reduce the amount of reflection (by 70–80%) from the wing surface in PIV images, a fluorescent paint, prepared by mixing 3 g of Rhodamine 6G with 100 ml of ethanol and 500 ml of water-soluble transparent acrylic paint (as recommended by Troolin 2009), was used on the wing surface. A 532 ± 2 nm narrow band-pass interference filter (Edmund model NT 43-174) was also installed on the CCD camera to block the higher wavelength. The PIV images were analyzed by using Insight 3G (version 8.0.4.0) software package developed by TSI Inc. The size of the interrogation window was set to either 24 × 24 or 32 × 32 (representing a physical size of roughly 1.5 and 2 mm, or 2% and 2.6%c, respectively), with a 50% overlapping of the interrogation windows. The velocity was found to be accurate to within 1% of the free stream velocity. Vorticity values were obtained by centrally differentiating the velocity field, which were estimated to be within 5% accuracy (Moffat (1985)). Details of the PIV setup are given in Gerontakos and Lee (2008).

3 Results and discussion

To facilitate the investigation into the impact of joint trailing-edge flap (TEF) and Gurney flap (GF) on the airfoil aerodynamic characteristics, the variation of aerodynamic load coefficient with independent GF and TEF deployment was examined first and serves as a comparison.

3.1 Effect of Gurney flap

Figure 2 illustrates the influence of solid Gurney flap (with h = 0.7–6%c) on the aerodynamic characteristics of the NACA 0015 airfoil. Also included in this figure are the baseline, or unflapped, airfoil data. Figure 2a shows that the Gurney flap always enhanced lift coefficient Cl, including its maximum value Cl,max, and generated a leftward shift of the lift curve, compared to the baseline airfoil, as a result of the flap-induced positive camber effects and the shifting of the location of rear stagnation point. The observed increase in lift can also be illustrated from the increase in both the upper surface suction (due to the downward turning of the streamwise flow) and the lower surface pressure (as a result of the flow deceleration in the upstream face of the flap), compared to the baseline airfoil (see, for example, at α = 12°; Fig. 3a). The static stalling mechanism of the baseline airfoil (characterized by the upstream progression of the turbulent trailing-edge flow separation and the bursting of the laminar separation bubble at α = αss = 16.5°; see Fig. 3b) was, however, not affected by the GF. The presence of a laminar separation bubble (for 10° < α < 16.5°) can be clearly seen from the Cp distribution along the airfoil upper surface. For α > αss, the Cp exhibited a flat distribution. The larger the flap height, the larger the pressure difference between the upper and lower surfaces of the airfoil, especially in the trailing-edge region, which can also be interpreted as equivalent to lengthening the airfoil and increasing flow turning (aft camber) near the trailing edge. An aft movement of the center of pressure behind the quarter-chord pitching moment reference, caused by the shifting of the load from the forward part of the airfoil to the rear part, resulted in an increase in the nose-down pitching moment (Fig. 2b). The larger the flap height, the larger the extent of this “shifting” and the higher the nose-down Cm and its peak negative value −Cm,peak. The variation of Cl,max, zero-lift angle αzl, and −Cm,peak with flap height is summarized in Table 1. The Cp distribution further suggests that at low-to-moderate angles of attack the Gurney flap provided protection against a trailing-edge separation by reducing the pressure recovery demand (an effect comparable to the increase in dumping velocity caused by a conventional slotted trailing-edge flap), which explains the longer linear portion in Cl-α curve, as shown in Fig. 2a, compared to the baseline airfoil. Similar to that of trailing-edge flap flow control, the lift-curve slope Clα remained largely affected. At high α, the flap, however, promoted a localized suction pressure peak in the leading-edge region, pushing the boundary layer closer to separation and promoting the static stall. The stall angle αss decreased further as a larger Gurney flap was deployed. The Cp data also indicate that at the same lift condition the aerodynamic loading was carried to the trailing edge such that the upper and lower surface pressure distributions were effectively farther apart for the flapped airfoils. The downward turning of the mean flow behind the flapped airfoil can also be demonstrated from the deflected wake centerline, measured with a hot-wire probe located at x/c = 0.6 for α = 2°, 6°, 9°, and 12°, presented in Fig. 4a, which further suggests that the Gurney flap works by increasing the effective camber of the airfoil. It should be noted that, especially for non-symmetric airfoils, the hysteresis effects, which can cause a non-negligible difference in the maximum Cl and Cd values, also need to be examined.
Fig. 2

Impact of Gurney flap height and perforation on airfoil aerodynamic load coefficient

Fig. 3

Representative surface pressure coefficient distributions. a At α = 10° for δ = 18°, h = 3%c, and σ = 23%, and b presence of laminar separation and stalling mechanism on baseline airfoil

Table 1

Test configurations and aerodynamic characteristics

Case

h

 σ

δ

Cl,max

αzl

Cm,peak

Case

h

σ

δ

Cl,max

αzl

Cm,peak

BA

1.13

0

−0.061

TEF + GF

0.7

0

18

1.68

−9.6

−0.181

GF

0.7

0

0

1.23

−1

−0.096

TEF + GF

1.5

0

18

1.75

−10.6

−0.220

GF

1.5

0

0

1.37

−1.7

−0.127

TEF + GF

1.5

23

18

1.70

−10.1

−0.195

GF

1.5

23

0

1.32

−1.4

−0.117

TEF + GF

2.1

0

18

1.79

−11.6

−0.231

GF

2.1

0

0

1.47

−3

−0.150

TEF + GF

2.1

23

18

1.73

−11.1

−0.222

GF

2.1

23

0

1.38

−2.7

−0.141

TEF + GF

3

0

18

1.85

−12.8

−0.242

GF

3

0

0

1.56

−3.2

−0.165

TEF + GF

3

23

18

1.76

−12.2

−0.232

GF

3

23

0

1.46

−2.4

−0.153

TEF + GF

3

40

18

1.70

−11.7

−0.226

GF

3

40

0

1.35

−1.7

−0.140

TEF + GF

4.5

0

18

1.95

−13.6

−0.256

GF

4.5

0

0

1.72

−4.7

−0.203

TEF + GF

4.5

23

18

1.83

−13.0

−0.247

GF

4.5

23

0

1.56

−3.8

−0.165

TEF + GF

4.5

40

18

1.77

−12.6

−0.241

GF

4.5

40

0

1.46

−3.3

−0.150

TEF + GF

6

0

18

2.05

−15

−0.271

GF

6

0

0

1.97

−6.8

−0.231

TEF + GF

6

23

18

1.92

−13.8

−0.261

GF

6

23

0

1.82

−5

−0.193

TEF + GF

0.7

0

9

1.58

−7.9

−0.174

TEF

0

0

6

1.27

−3

−0.088

TEF + GF

1.5

0

9

1.56

−8.3

−0.188

TEF

0

0

9

1.39

−4

−0.120

TEF + GF

1.5

23

9

1.57

−8.0

−0.180

TEF

0

0

12

1.43

−5.5

−0.127

TEF + GF

2.1

0

9

1.66

−9.7

−0.219

TEF

0

0

18

1.56

−7.5

−0.150

TEF + GF

2.1

23

9

1.64

−9

−0.183

       

TEF + GF

3

0

9

1.80

−9.7

−0.232

       

TEF + GF

3

23

9

1.67

−9.0

−0.205

       

TEF + GF

4.5

0

9

1.92

−11.1

−0.240

       

TEF + GF

4.5

23

9

1.78

−10.4

−0.222

       

TEF + GF

6

0

9

2.04

−12.4

−0.252

       

TEF + GF

6

23

9

1.88

−11.3

−0.231

BA denotes baseline airfoil; h in %c; σ in %; δ in °; αzl in °

Fig. 4

Mean and fluctuating wake velocity profiles at x/c = 0.6 for selected α with δ = 18°, h = 6%c, and σ = 23%

Figure 2c shows that the increase in Cl of the flapped airfoil was also at a cost of increasing drag. The larger the flap height, the higher the drag. Similar to the increase in Cl and the peak negative Cm with flap height, the drag coefficient Cd was also found to increase non-linearly with the flap height. The most effective enhancement in Cl exhibited at smaller flap heights (see Fig. 5a, b). Figure 5a clearly shows that as expected, the value of Cl,max increased, in a linear manner, with increasing Gurney flap height, compared to the baseline airfoil. The most effective lift enhancement was, however, noticed by plotting the Cl,max, normalized by h/c, against the non-dimensional flap height h/c, as shown in Fig. 5b. It is evident that the effectiveness of the lift enhancement decreased non-linearly with increasing flap height, with the most effective lift enhancement occurred at the smallest flap height. Note that the smallest flap height was also accompanied by a smallest increase in Cd and Cm, compared to the larger flap heights tested in the present study. The Gurney flap also caused a larger wake (Fig. 4a), of increased unsteadiness or urms intensity (Fig. 4b), at the selected angles of attack (α = 2°, 6°, 9° and 12°), compared to the baseline airfoil, which also indicates a higher mean drag. The rapid increase in Cl, however, outweighed the rise in Cd and led to an improved lift-to-drag ratio.
Fig. 5

Variation of Cl,max with flap height for independent GF and joint TEF (deflected at δ = 18°) and GF deployment. ∆Cl,max = Cl,max of joint TEF and GF − Cl,max of TEF

Figure 2a also shows that the addition of flap perforation (with σ = 23 and 40%) reduced Cl, including its maximum value, compared to the solid flap at the same height and angle of attack, as a result of the jet flows created by the perforated holes in the flap which, in turn, lessened the favorable flap-induced camber effect (Lee and Ko 2009). The flap with smaller porosity created weaker jet flows, compared to those of larger perforation, and thus caused a smaller loss in lift. The flap perforation-induced lift loss can also be demonstrated from the Cp distribution and wake profiles presented in Figs. 3a and 4. The flap perforation lessened the degree of downward turning of the mean flow, leading to a reduced positive camber effect (compared to that associated with a solid flap) and thus a reduced suction pressure on the airfoil upper surface (Fig. 3a). Meanwhile, the recirculation flow region, or vortex, formed in the upstream face of an otherwise solid flap, was also disrupted and led to a reduced surface pressure on the lower surface of the airfoil, especially in the trailing-edge region. The lift-curve slope and the stall angle were only slightly affected by the addition of flap perforation, compared to the solid flap. The flap perforation caused a leftward shift of the lift curve but with a lesser extent compared to that of the solid flap. More importantly, the flap perforation also led to a reduced Cd (Fig. 2c) and a recovery in the nose-down pitching moment coefficient and its peak negative value (Fig. 2b), compared to the solid flap. The reduction in drag of the airfoil with flap perforation can also be directly reflected from the reduced wake deficit and fluctuation displayed in Fig. 4a, b. The larger the flap porosity, the smaller the wake velocity deficit and fluctuation. Typical aerodynamic characteristics of the airfoil with perforated Gurney flap are also summarized in Table 1.

3.2 Effect of joint trailing-edge flap and Gurney flap

Before the discussion of the impact of the attachment of GF to the trailing-edge flap (TEF), the influence of TEF (with δ = 6°–18°) on the aerodynamic characteristics of the NACA 0015 airfoil was also re-examined. Figure 6a shows that the simple hinged trailing-edge flap created more lift simply by mechanically increasing the effective camber of the airfoil and that the effect of TEF deflection was to shift the lift curve to the left and αzl to a more negative value with increasing flap deflection. The lift-curve slope of the flapped airfoil remained essentially unchanged compared to the baseline airfoil. Similar to the Gurney flap, the TEF-induced aft camber also produced a large increase in nose-down pitching moment compared to the baseline airfoil (see Fig. 6c). However, in contrast to the Gurney flaps (with h < 3%c), the TEF-generated drag penalty (Fig. 6b) outweighed the lift enhancement and rendered a deteriorated lift-to-drag ratio. Furthermore, the magnitude of Cl,max, and −Cm,peak of the TEF airfoil, with δ ≤ 18°, was comparable to that produced by GF with h < 4.5%c (see Table 1), which implies that the pronounced lift enhancement achieved by TEF deployment can also be obtained via simple GF application. Note that in addition to the Gurney flap-induced further increase in Cl, the most significant way in which Gurney flaps differ from conventional trailing-edge flap control is in their effects on the center of pressure. For h > 3%c, the Gurney flap deployment gave a more significant aft shift of the aerodynamic center or a larger increase in the nose-down pitching moment, as a result of a further aft shift in chordwise loading distribution (due to the Gurney flap-induced increased pressure difference between upper and lower surfaces; see Fig. 3a), compared to the TEF airfoil. The Cp data also indicate that, in contrast to the observed increase in the suction pressure in the leading-edge region of the GF airfoil, the suction peak of the TEF airfoil was also decreased over a much wider extent in the leading-edge region with increasing flap deflection. The wake velocity profiles also indicate downward turning of the flow behind the trailing-edge flap, regardless of α (see Fig. 4), which also suggests that TEF works by increasing the effective camber of the airfoil. The trailing-edge flap, however, caused a relatively smaller wake compared to that of large Gurney flaps (e.g., with h > 3%c; Fig. 4a). The size of wake flow, including its unsteadiness, or urms intensity (Fig. 4b), also increased with increasing trailing-edge flap deflection.
Fig. 6

Airfoil load coefficient with trailing-edge flap deployment

Figures 7, 8, 9, 10, and 11 and Table 1 summarize the impact of joint trailing-edge flap and Gurney flap deployment on the airfoil aerodynamic and wake characteristics. To better exemplify the additional changes caused by the Gurney flap, the deflection of the trailing-edge flap was fixed at δ = 18° first. The discussion is then followed by the effects of flap perforation and TEF deflection. Figure 7a–c reveal that, for a TEF deflected at δ = 18°, the addition of solid Gurney flaps produced a significant additional increase in the lift, pitching moment, and drag coefficient compared to that produced by TEF deflection alone. Similar to the baseline-airfoil Gurney flap concept, as discussed in Fig. 2a–c, the Cl,max, of the joint TEF and GF application, was also found to increase with increasing Gurney flap height (see also Fig. 5a). An additional 10, 12, 15, 19, 25, and 31.5% (or 49, 55, 58.5, 64, 72.5, and 81.5%) increment in Cl,max for h = 0.7, 1.5, 2.1, 3, 4.,5 and 6%c of the joint TEF and GF application, compared to TEF deployment (or the baseline airfoil), was observed. The most effective Cl,max enhancement was, however, observed at the smallest Gurney flap height tested (see Fig. 5b). Figure 5b also reveals that, similar to that of independent GF, this effectiveness decreased non-linearly with h/c of the airfoil with both TEF and GF. Also shown in Fig. 5c is the variation of the increment in the maximum lift coefficient ∆Cl,max (=Cl,max of joint TEF and GF − Cl,max of TEF) of the joint TEF and GF deployment, relative to that produced by the TEF alone, with the flap height. The ∆Cl,max increased linearly with flap height. The larger the flap height, the larger the ∆Cl,max. The most effective increase in ∆Cl,max, however, exhibited at the smallest flap height. The value of Cl,max/(h/c) decreased rapidly and non-linear with increasing flap height (denoted by solid square in Fig. 5c). The additional Gurney flap-induced Cl increment can also be reflected from the Cp distribution along both the upper and lower surfaces of the airfoil (Fig. 3). As can be seen in Fig. 3, the flow was further accelerated and decelerated on the upper and lower surfaces, respectively, due to presence of the Gurney flap, in comparison with that associated with the trailing-edge flap. The larger the Gurney flap height, the higher the Cl, |−Cm,peak| and Cd, as shown in Fig. 7b, c, of the joint TEF and GF deployment. The additional drag penalty (caused by the presence of the Gurney flap with h < 3%c) was, however, relieved by the rapid rise in lift and gave rise to a virtually unaffected lift-to-drag ratio, compared to that of the TEF. The substantial increment in Cl of the joint TEF and GF deployment could therefore serve as a potential off-design high-lift device during landing and takeoff. Similar change in the aerodynamic performance with the flap height was also observed at δ = 9° (see Table 1). The results discussed in Fig. 7 can also be better understood via the PIV flowfield measurements. For clarity, only the typical PIV results for α = 5° with δ = 16° and h = 6%c are presented in Figs. 8, 9, 10, 11, and 12.
Fig. 7

Impact of GF, of different heights, on Cl, Cm, and Cd values with TEF deflected at δ = 18°

Fig. 8

Ensemble-averaged iso-u/U contours for α = 5°. a TEF + solid GF with δ = 16° and h = 6%c, b TEF with δ = 16°, c baseline airfoil, d solid GF with h = 6%c, and e TEF + perforated GF with δ = 16°, h = 6%c and σ = 40%

Fig. 9

Ensemble-averaged iso-ζc/U contours for α = 5°. See Fig. 8 for figure captions

Fig. 10

Normalized mean a wake velocity and b vorticity profiles as a function of x/c for α = 5°, δ = 16°, h = 6%c, and σ = 0 and 40%. black line baseline airfoil, green line TEF, dark blue line TEF + solid GF, red line TEF + perforated GF, light blue line GF

Fig. 11

Ensemble-averaged iso-v/U contours for α = 5°. See Fig. 8 for figure captions

Fig. 12

Normalized instantaneous iso-contours of streamwise (ae) velocity and (fj) vorticity. See Fig. 8 for figure captions

Figure 8a illustrates the normalized iso-contour of the mean streamwise velocity u/U in the vicinity of the trailing-edge flap equipped with a GF of a height of h = 6%c. The free stream flow direction is from left to right. The mean flowfield information was obtained by ensemble averaging 480 PIV images. It can be seen that there was a further increase in the extent of downward turning of the near wake (i.e., indicative of a further increment in aft camber), which is also responsible for the additional increase in the suction pressure on the upper side of the trailing-edge flap (as discussed earlier in Fig. 3), and thus the lift, compared to sole TEF application (Fig. 8b). The near-wake patterns behind the baseline airfoil and the Gurney flap-equipped airfoil are also presented in Fig. 8c, d for comparison. The iso-u/U contours also indicate that the width of the near wake, behind the joint TEF and GF (with h = 6%c), were considerably widened and the maximum velocity deficit was increased, compared to that associated with individual TEF or GF deployment, which suggests an increase in the mean drag (as also demonstrated in Fig. 6c). The existence of a small region of recirculation flow on the upstream face of the Gurney flap can also be seen in Fig. 8a, d. The normalized iso-contours of the ensemble-averaged streamwise vorticity ζc/U, corresponding to Fig. 8a–d, are also displayed in Fig. 9a–d. The negative and positive vortices are denoted by clockwise-rotating (CW) and counterclockwise (CCW) rotation, respectively. The high concentration of positive vorticity contained in the shear layer separated from the tip of the Gurney flap can be clearly seen in Fig. 9a, d. For a more quantitative comparison, the downstream development of the wake velocity and vorticity profiles with x/c is also reproduced in Fig. 10. The reversed flow region existed immediately downstream of the Gurney flap, and TEF is evident at x/c = 1.012, 1.0326, and 1.0483 in Fig. 10a. Note also the large discrepancy in u/U exhibited immediately behind the GF, TEF, and joint GF and TEF deployment. Further downstream, this discrepancy between the cases of TEF and GF vanished. The discrepancy, however, persisted for the joint TEF and GF case. The normalized iso-contours of the transverse velocity v/U are also presented in Fig. 11. It is of importance to note that the best strength of PIV would be getting the unsteady instantaneous flow fields rather than average flow fields and that the ensemble-averaged values tend to miss out the details. An excellent study of the instantaneous flow behavior behind a Gurney flap is given by Troolin et al. (2006) by using a time-resolved PIV system, incorporating a Nd:LIF laser pulsed at 1 kHz. The ensemble-averaged flow fields, obtained with a double-exposure Nd:YAG PIV system, reported in the present study should be of importance for CFD validations. Nevertheless, typical normalized instantaneous iso-contours of streamwise velocity and vorticity are included in Fig. 12 for comparison. The wake unsteadiness and the vortices shed downstream of the airfoil can be more readily seen. Aerodynamic characteristics of the airfoil with TEF (deflected at δ = 9°) and GF are also summarized in Table 1.

Finally, the impact of the Gurney flap perforation on the wake flow structure behind the joint TEF and GF deployment was also examined and is illustrated in Figs. 8e, 9e, 10, and 11e. The results show that the addition of the Gurney flap perforation produced a “jet flow” immediate downstream of the flap (for x/c < 1.08), which caused a more rapid mixing of the near wake and, subsequently, a reduced wake deficit and, subsequently, a smaller drag generation, compared to TEF with a solid Gurney flap (Figs. 8a, 9a). The strength of the jet flow increased with flap perforation. However, in contrast to the larger degree of flap perforation-induced reduction in the wake width and the mean drag reported by Lee and Ko (2009), compared to that of a solid Gurney flap, a smaller change in the wake width and fluctuations (see Fig. 4) and drag was noticed in the joint TEF and perforated Gurney flap application. In other words, the flap-perforation-generated “jet flow” had a diminishing influence on the large wake existing behind the trailing-edge flap. That is, for the joint TEF and GF application, the favorable flap-perforation-generated jet flow was limited to the lower part of the large separated wake behind the deflected trailing-deg flap. Note that GF works best if the airfoil wake is kept small. The small discrepancy in the wake velocity profiles and aerodynamic performance between the joint TEF and GF deployment (with and without flap perforation) are self-explained in Figs. 4 and 10, as well as for a Gurney flap with a smaller height (h = 3%c and σ = 0, 23, and 40%; see Fig. 13a–c).
Fig. 13

Impact of independent and joint TEF and GF deployment on airfoil aerodynamic characteristics for δ = 18°, h = 3%, and σ = 0, 23, and 40%

4 Conclusions

The aerodynamic and wake characteristics of a NACA 0015 airfoil equipped with a full-span 25%c plain trailing-edge flap, incorporating Gurney flaps of different flap heights and perforations, were investigated at Re = 2.54 × 105. The Gurney flap-equipped trailing-edge flap produced a further increase in the downward turning of the mean flow (increased aft camber), leading to a significant increase in the lift, drag, and pitching moment coefficient compared to that produced by independently deployed TEF or GF. The Cl,max increased with flap height, with the maximum effectiveness (in terms of Cl,max/(h/c) vs. h/c) exhibited at the smallest flap height. There was an additional positive shift of the lift curve with its slope remaining virtually unaffected. The rapid rise in lift and drag generation and the basically unchanged lift-to-drag performance (at least for h < 3%c) of the joint TEF and GF application, compared to conventional trailing-edge flap deployment, could provide an improved off-design high-lift device during landing and takeoff. In addition, the near wake behind the joint TEF and GF became wider and had a larger velocity deficit and fluctuation, compared to those of independent GF and TEF deployment, suggesting an increased mean drag. The addition of Gurney flap perforation, however, had only a minor impact on the wake and aerodynamics characteristics of the TEF airfoil.

Acknowledgments

This work was supported by the Natural Science and Engineering Research Council (NSERC) of Canada. L.S. Ko is thanked for his help with the PIV experiment.

Copyright information

© Springer-Verlag 2010

Authors and Affiliations

  1. 1.Department of Mechanical EngineeringMcGill UniversityMontrealCanada

Personalised recommendations