Circulation Of Leading-edge Vortex Research Articles

Residual circulation budget analysis in a Wells turbine with leading-edge micro-cylinders

To further interpret the control mechanisms of leading-edge micro-cylinders for flow separation in a Wells turbine, the residual circulation budget was performed through an improved triple decomposition method that the computational cost was reduced by 38.2%.The intensity of the leading-edge vortex (LEV), free of shear effects, was tracked under typical conditions. Meanwhile, disturbance physics was dissected by splitting the contributions of shear, swirling, and straining motions to the flow from dominant modes, captured by the proper orthogonal decomposition method under typical working conditions. The influence of micro-cylinders on the spatiotemporal evolution of the LEV has been discussed in detail. The results indicate that the accumulative shaft power of the turbine is improved by 54.2% through the micro-cylinders operating by suppressing the swirling intensity near the leading edge and weakening the instability of the boundary layer. Moreover, low-frequency disturbances could be significantly weakened and tended to transfer to higher-order modes with lower energy. Under the stall condition, the shear layer flux and diffusive flux dominated the LEV circulation rate. In contrast to the referential model, as the cylinders blocked the LEV trajectory after falling, the annihilation slowed down with an intensified accumulative circulation of the LEV.

Flow development and leading edge vorticity in bristled insect wings

Small flying insects such as the tiny thrip Gynaikothrips ficorum have wings with bristles attached to a solid shaft instead of solid membranes. Air passing through the bristle fringe, however, makes bristled insect wings less effective for aerodynamic force production. In this study, we quantified the ability of bristled wings to generate a leading edge vortex (LEV) for lift support during wing flapping, scored its circulation during wing translation, and investigated its behaviour at the stroke reversals. The data were measured in robotic model wings flapping with a generic kinematic pattern at Reynolds number of ~ 3.4, while applying two-dimensional particle image velocimetry. We found that aerodynamic performance due to LEV circulation linearly decreases with increasing bristle spacing. The wings of Gynaikothrips ficorum might thus produce approximately 9% less aerodynamic force for flight than a solid membranous wing. At the stroke reversals, leading and trailing edge vortices dissipate quickly within no more than ~ 2% of the stroke cycle duration. This elevated dissipation makes vortex shedding obsolete during the reversals and allows a quick build-up of counter-vorticity when the wing reverses flapping direction. In sum, our findings highlight the flow conditions associated with bristled wing design in insects and are thus significant for assessing biological fitness and dispersal of insects flying in a viscosity-dominated fluid regime.

Insights into leading edge vortex formation and detachment on a pitching and plunging flat plate

The present study is a prelude to applying different flow control devices on pitching and plunging airfoils with the intention of controlling the growth of the leading edge vortex (LEV); hence, the lift under unsteady stall conditions. As a pre-requisite the parameters influencing the development of the LEV topology must be fully understood and this constitutes the main motivation of the present experimental investigation. The aims of this study are twofold. First, an approach is introduced to validate the comparability between flow fields and LEV characteristics of two different facilities using water and air as working media by making use of a common baseline case. The motivation behind this comparison is that with two facilities the overall parameter range can be significantly expanded. This comparison includes an overview of the respective parameter ranges, control of the airfoil kinematics and careful scrutiny of how post-processing procedures of velocity data from time-resolved particle image velocimetry (PIV) influence the integral properties and topological features used to characterise the LEV development. Second, and based on results coming from both facilities, the appearance of secondary structures and their effect on LEV detachment over an extended parameter range is studied. A Lagrangian flow field analysis based on finite-time Lyapunov Exponent (FTLE) ridges allows precise identification of secondary structures and reveals that their emergence is closely correlated to a vortex Reynolds number threshold computed from the LEV circulation. This threshold is used to model the temporal onset of secondary structures. Further analysis indicates that the emergence of secondary structures causes the LEV to stop accumulating circulation if the shear layer angle at the leading edge of the flat plate has ceased to increase. This information is of particular importance for advanced flow control applications, since efforts to strengthen and/or prolong LEV growth rely on precise knowledge about where and when to apply flow control measures.Graphical abstract

Experimental investigation on the leading-edge vortex formation and detachment mechanism of a pitching and plunging plate

Abstract

Lift enhancement strategy and mechanism for a plunging airfoil based on vortex control

A new flow control strategy based on leading-edge vortex (LEV) manipulation is proposed to improve the aerodynamic performance of a plunging airfoil. It has been found that the low pressure region produced by the LEV contributes to the high lift during dynamic stall, while the growth of the secondary vortex would weaken the LEV and result in a decrease in lift. Accordingly, the vortex control hypothesis is that we change the evolution of the secondary vortex and LEV, thus achieving a higher lift coefficient with a longer duration. The suction actuator is placed at different positions on the upper surface of the airfoil to test the control hypothesis. When the suction actuator is near the leading edge, the LEV detaches from the shear layer earlier and it can only enhance the lift slightly while not delay stall time. When the suction actuator is near the middle region, it could inhibit the growth of the secondary vortex and, thus, reduce its strength greatly. Therefore, the LEV circulation could continue to increase. As a result, the suction control could increase the lift coefficient and also prolong the high-lift duration. When the suction actuator is near the trailing edge, an increase in lift could also be achieved by an increase in the negative pressure over the upper surface as well as the LEV circulation. Thus, we present and validate the lift enhancement strategy for an unsteady airfoil based on vortex control.

On the stabilization of leading-edge vortices with different reduced frequencies over finite-aspect-ratio pitch-up wings

An implicit large-eddy simulation method is used to numerically simulate the flowfield around a finite-span rectangular wing in pitch motion. The effect of the reduced frequency (k) on the stability of the leading-edge vortex (LEV) is studied. The stability of LEVs is considered to include adhesion and structural stability. The wing is pitched at a position of around 1/4c (chord length), and the flow has a Reynolds number of 9053. Reduced frequencies of 0.2, 0.4, 0.8, and 1.6 are studied. For k = 0.2, the calculated results are in good agreement with experimental measurements, which demonstrates the reliability of the calculation method. The results show that the structural stability of LEVs can be significantly enhanced by increasing k. Larger values of k reduce the spatial scale of the LEVs, as well as prevent them from growing too fast and bursting. Moreover, a larger reduced frequency also delays the decrease in the LEV circulation, allowing the LEV structure to maintain stability at a larger angle of attack (α). In addition, a larger value of k also helps to enhance the adhesion stability of the LEVs. Numerical simulation results show that smaller values of k encourage the formation of secondary vortices. These accelerate the backflow between the LEVs and the wing, thus promoting the upward movement of the LEVs and reducing their adhesion stability. However, analysis of the convection terms in the vortex dynamic equations indicates that lower k is beneficial for the vorticity in the LEV to be transported outwards, but very small k will cause the reverse transport of vorticity, which is harmful to the stability of the LEV.

Leading-edge vortex formation and transient lift generation on a revolving wing at low Reynolds number

Both the transient formation and the stable attachment of leading-edge vortex (LEV) contribute to the high lift generation of an insect wing when it revolves at high angles of attack. In this study, we examined the LEV formation and the transient lift generation on a revolving wing with Reynolds number at 1500 and aspect ratio at 3, using combined computational and experimental methods. The wing started from stationary with a constant angular acceleration within a prescribed chord length of travel at its radius of gyration (i.e. λa), ranging from 0.25 to 2, and then revolved at a constant velocity. Our results showed that the rate of LEV development saturated when λa<1, resulting in a minimum chord length of travel of 2.5 to arrive at a steady LEV. This minimum travel distance agreed well with the prediction based on the vortex formation criterion proposed in other vortex generators. On the other hand, the instantaneous lift generation was strongly dependent on the instantaneous wing velocity and acceleration and therefore exhibited no saturation as λa decreases, which is seen in the LEV formation. The lift generation was attributed to the evolution of tangential vorticity according to the vortex impulse theory. Finally, the instantaneous volume-averaged LEV circulation at different angles of attack scaled well with the instantaneous velocity normal to the leading edge.

Adjoint-Based Optimization for Thrust Performance of Three-Dimensional Pitching–Rolling Plate

An adjoint-based optimization is applied to study the thrust performance of a pitching–rolling ellipsoidal plate in a uniform stream at Reynolds number 100. To achieve the highest thrust, the optimal kinematics of pitching–rolling motion is sought in a control space including the pitching amplitude, the rolling amplitude, and the phase delay between the pitching and rolling motions. A continuous adjoint approach with boundary motion being handled by noncylindrical calculus is developed as a computationally efficient optimization algorithm to deal with large control space with a morphing domain. The comparison between the optimal motion and other reference motions shows a significant improvement of thrust from the increase of the rolling amplitude and an optimal phase delay of 122.6 deg between the pitching and rolling motions. The combination of these two factors impacts the overall thrust performance through their strong effects on the angle of attack, the leading-edge vortex circulation, and the pressure distribution on the plate. Further wake structure analysis suggests that the optimal motion improves its propulsive performance by generating a stronger leading-edge vortex and straightening the wake deflection.

Pitching effect on transonic wing stall of a blended flying wing with low aspect ratio

Numerical simulation of the pitching effect on transonic wing stall of a blended flying wing with low aspect ratio was performed using improved delayed detached eddy simulation (IDDES). To capture the discontinuity caused by shock wave, a second-order upwind scheme with Roe’s flux-difference splitting is introduced into the inviscid flux. The artificial dissipation is also turned off in the region where the upwind scheme is applied. To reveal the pitching effect, the implicit approximate-factorization method with sub-iterations and second-order temporal accuracy is employed to avoid the time integration of the unsteady Navier–Stokes equations solved by finite volume method at Arbitrary Lagrange–Euler (ALE) form. The leading edge vortex (LEV) development and LEV circulation of pitch-up wings at a free-stream Mach number M = 0.9 and a Reynolds number Re = [Formula: see text] is studied. The Q-criterion is used to capture the LEV structure from shear layer. The result shows that a shock wave/vortex interaction is responsible for the vortex breakdown which eventually causes the wing stall. The balance of the vortex strength and axial flow, and the shock strength, is examined to provide an explanation of the sensitivity of the breakdown location. Pitching motion has great influence on shock wave and shock wave/vortex interactions, which can significantly affect the vortex breakdown behavior and wing stall onset of low aspect ratio blended flying wing.

Unsteady aerodynamics of a pitching-flapping-perturbed revolving wing at low Reynolds number

Due to adverse viscous effects, revolving wings suffer universally from low efficiency at low Reynolds number (Re). By reciprocating wing revolving motion, natural flyers flying at low Re successfully exploit unsteady effects to augment force production and efficiency. Here we investigate the aerodynamics of an alternative, i.e., a revolving wing with concomitant unsteady pitching and vertical flapping perturbations (a pitching-flapping-perturbed revolving wing). The current work builds upon a previous study on flapping-perturbed revolving wings (FP-RWs) and focuses on combined effects of pitching-flapping perturbation on force generation and vortex behaviors. The results show that, compared with a FR-RW, pitching motion further (1) reduces the external driving torque for rotating at 0° angle of attack (α0) and (2) enhances lift and leads to a self-rotating equilibrium at α0 = 20°. The power loading of a revolving wing at α0 = 20° can be improved using pitching-flapping perturbations with large pitching amplitude but small Strouhal number. Additionally, an advanced pitching improves the reduction of external driving torque, whereas a delayed pitching weakens both the lift enhancement and the reduction of external driving torque. Further analysis shows that pitching effects can be mainly decomposed into the Leading-Edge-Vortex (LEV)-mediated pressure component and geometric projection component, together they determine the force performance. LEV circulation is found to be determined by the instantaneous effective angle of attack but could be affected asymmetrically between upstroke and downstroke depending on the nominal angle of attack. Pitching-flapping perturbation thus can potentially inspire novel mechanisms to improve the aerodynamic performance of rotary wing micro air vehicles.

Ground effect on the aerodynamics of three-dimensional hovering wings

This paper reports the results of combined experimental and numerical studies on the ground effect on a pair of three-dimensional (3D) hovering wings. Parameters investigated include hovering kinematics, wing shapes, and Reynolds numbers (Re). The results are consistent with the observation by another study (Gao and Lu, 2008 Phys. Fluids, 20 087101) which shows that the cycle-averaged aerodynamic forces generated by two-dimensional (2D) wings in close proximity to the ground can be broadly categorized into three regimes with respect to the ground clearance; force enhancement, force reduction, and force recovery. However, the ground effect on a 3D wing is not as significant as that on a 2D flapping wing reported in (Lu et al 2014 Exp. Fluids, 55 1787); this could be attributed to a weaker wake capture effect on 3D wings. Also, unlike a 2D wing, the leading edge vortex (LEV) remains attached on a 3D wing regardless of ground clearance. For all the wing kinematics considered, the three above-mentioned regimes are closely correlated to a non-monotonic trend in the strength of downwash due to the restriction of root and tip vortex formation, and a positional shift of wake vortices. The root vortices in interaction with the ground induce an up-wash in-between the two wings, causing a strong ‘fountain effect’ (Maeda and Liu, 2013 J. Biomech. Sci. Eng., 8 344) that may increase the body lift of insects. The present study further shows that changes in wing planform have insignificant influence on the overall trend of ground effect except for a parallel shift in force magnitude, which is caused mainly by the difference in aspect ratio and leading edge pivot point. On the two Reynolds numbers investigated, the results for the low Re case of 100 do not deviate significantly from those of a higher Re = 5000 except for the difference in force magnitudes, since low Reynolds number generates lower downwash, weaker LEV, and lower rotational circulation. Additionally, lower Re leads to a weaker fountain effect.

Leading Edge Vortex Circulation Development on Finite Aspect Ratio Pitch-Up Wings

We study the leading edge vortex (LEV) development and LEV circulation of pitch-up wings at a low Reynolds number of 500. Wings of different aspect ratios are linearly pitched up from 0 to 90 deg at two reduced pitch rates of 0.1 and 0.2. The flowfield is described by solving the unsteady three-dimensional incompressible Navier–Stokes equations on composite overlapping grids. The -criterion is used to isolate the LEV structure from shear layer. The simulation shows that forces increase with the aspect ratio and the pitch rate. The stall angle increases with the pitch rate but decreases with the aspect ratio. also It is shown that the LEV circulation depends on the wing aspect ratio: increasing wing aspect ratio increases the rate of LEV circulation generation during the pitch-up motion. It also shows that the reduced pitch rate could delay the LEV circulation development. Furthermore, the study shows that the LEV circulation distribution better matches the spanwise force distribution for higher aspect ratio wings.

Aspect-ratio effects on rotating wings: circulation and forces

AbstractWe employ experiments to study aspect ratio ($\def\AR{A\mkern-8muR}\AR$) effects on the vortex structure, circulation and lift force for flat-plate wings rotating from rest at 45° angle of attack, which represents a simplified hovering-wing half-stroke. We use the time-varying, volumetric $\AR =2$ data of Carr et al. (Exp. Fluids, vol. 54, 2013, pp. 1–26), reconstructed from phase-locked, phase-averaged stereoscopic digital particle image velocimetry (S-DPIV), and an $\AR =4$ volumetric data set matching the span-based Reynolds number ($\mathit{Re}$) of $\AR =2$. For $\AR =1{-}4$ and $\mathit{Re}_{\mathit{span}}$ of $O$($10^{3}$–$10^{4}$), we directly measure the lift force. The total leading-edge-region circulation for $\AR =2$ and 4 compares best overall using a span-based normalization and for matching rotation angles. The total circulation increases across the span to the tip region, and is larger for $\AR =2$. After the startup, the total circulation for each $\AR$ has a similar slope and a slow growth. The first leading-edge vortex (LEV) and the tip vortex (TV) for $\AR =4$ move past the trailing edge, followed by substantial breakdown. For $\AR =2$ the outboard, aft-tilted LEV merges with the TV and resides over the tip, although breakdown also occurs. Where the LEV is ‘stable’ inboard, its circulation saturates for $\AR =2$ and the growth slows for $\AR =4$. Aft LEV tilting reduces the spanwise LEV circulation for each $\AR$. Both positive and negative axial flow are found in the first LEV for $\AR =2$ and 4, with the positive component being somewhat larger. This yields a generally positive (outboard) average vorticity flux. The average lift coefficient is essentially constant with $\AR$ from 1 to 4 during the slow growth phase, although the large-time behaviour shows a slight decrease in lift coefficient with increasing $\AR$. The S-DPIV data are used to obtain the lift impulse and the spanwise and streamwise components contributing to the lift coefficient. The spanwise contribution is similar for $\AR =2$ and 4, due to similar trailing-edge vortex interactions, LEV saturation behaviour and total circulation slopes. However, for $\AR =2$ the streamwise contribution is much larger, because of the stronger, coherent TV and aft-tilted LEV, which will create a relatively lower-pressure region over the tip.

Three-dimensional flow structures and unsteady forces on pitching and surging revolving flat plates

Tomographic particle image velocimetry was used to explore the evolution of three-dimensional flow structures of revolving low-aspect-ratio flat plates in combination with force measurements at a Reynolds number of 10,000. Two motion kinematics are compared that result in the same terminal condition (revolution with constant angular velocity and $$45^{\circ }$$ angle of attack) but differ in the motion during the buildup phase: pitching while revolving at a constant angular velocity; or surging with a constant acceleration at a fixed angle of attack. Comparison of force histories shows that the pitching wing generates considerably higher forces during the buildup phase which is also predicted by a quasi-steady model quite accurately. The difference in the buildup phases affects the force histories until six chords of travel after the end of buildup phase. In both cases, a vortex system that is comprised of a leading-edge vortex (LEV), a tip vortex and a trailing edge vortex is formed during the initial period of the motion. The LEV lifts off, forms an arch-shaped structure and bursts into substructures, which occur at slightly different phases of the motions, such that the revolving–surging wing flow evolution precedes that of the revolving–pitching wing. The delay is shown to be in accordance with the behavior of the spanwise flow which is affected by the interaction between the tip vortex and revolving dynamics. Further analysis shows that the enhanced force generation of the revolving–pitching wing during the pitch-up phase originates from: (1) increased magnitude and growth rate of the LEV circulation; (2) relatively favorable position and trajectory of the LEV and the starting vortex; and (3) generation of bound circulation during the pitching motion, whereas that of the revolving–surging wing is negligible in the acceleration phase.

Parameter dependence of vortex interactions on a two-dimensional plunging plate

The structure and dynamics of the flow field created by a plunging flat-plate airfoil are investigated at a chord Reynolds number of 10,000 while varying plunge amplitude and Strouhal number. Digital particle image velocimetry measurements are used to characterize the shedding patterns and the interactions between the leading- and trailing-edge vortex structures (LEV and TEV), resulting in the development of a wake classification system based on the nature and timing of interactions between the leading- and trailing-edge vortices. The streamwise advancement of the LEV during a plunge cycle and its resulting interaction with the TEV is primarily dependent on reduced frequency; however, for Strouhal numbers above approximately 0.4, significant changes are observed in the formation of vortices shed from the leading and trailing edges, as well as the circulation of the leading-edge vortex. The functional form of the relationship between leading-edge vortex circulation and Strouhal number suggests that the Strouhal number dependence is more specifically a manifestation of the effective angle of attack. Comparison with low-Reynolds-number studies of plunging airfoil aerodynamics reveals a high degree of consistency and suggests applicability of the classification system beyond the range examined in the present work.

Multiple leading edge vortices of unexpected strength in freely flying hawkmoth

The Leading Edge Vortex (LEV) is a universal mechanism enhancing lift in flying organisms. LEVs, generally illustrated as a single vortex attached to the wing throughout the downstroke, have not been studied quantitatively in freely flying insects. Previous findings are either qualitative or from flappers and tethered insects. We measure the flow above the wing of freely flying hawkmoths and find multiple simultaneous LEVs of varying strength and structure along the wingspan. At the inner wing there is a single, attached LEV, while at mid wing there are multiple LEVs, and towards the wingtip flow separates. At mid wing the LEV circulation is ~40% higher than in the wake, implying that the circulation unrelated to the LEV may reduce lift. The strong and complex LEV suggests relatively high flight power in hawmoths. The variable LEV structure may result in variable force production, influencing flight control in the animals.

Unsteady force generation and vortex dynamics of pitching and plunging aerofoils

AbstractExperimental studies of the flow topology, leading-edge vortex dynamics and unsteady force produced by pitching and plunging flat-plate aerofoils in forward flight at Reynolds numbers in the range 5000–20 000 are described. We consider the effects of varying frequency and plunge amplitude for the same effective angle-of-attack time history. The effective angle-of-attack history is a sinusoidal oscillation in the range$\ensuremath{-} 6$to$2{2}^{\ensuremath{\circ} } $with mean of${8}^{\ensuremath{\circ} } $and amplitude of$1{4}^{\ensuremath{\circ} } $. The reduced frequency is varied in the range 0.314–1.0 and the Strouhal number range is 0.10–0.48. Results show that for constant effective angle of attack, the flow evolution is independent of Strouhal number, and as the reduced frequency is increased the leading-edge vortex (LEV) separates later in phase during the downstroke. The LEV trajectory, circulation and area are reported. It is shown that the effective angle of attack and reduced frequency determine the flow evolution, and the Strouhal number is the main parameter determining the aerodynamic force acting on the aerofoil. At low Strouhal numbers, the lift coefficient is proportional to the effective angle of attack, indicating the validity of the quasi-steady approximation. Large values of force coefficients (${\ensuremath{\sim} }6$) are measured at high Strouhal number. The measurement results are compared with linear potential flow theory and found to be in reasonable agreement. During the downstroke, when the LEV is present, better agreement is found when the wake effect is ignored for both the lift and drag coefficients.

Stroke plane angle controls leading edge vortex in a bat-inspired flapper

The present interest in micro air vehicles has given the research on bat flight a new impulse. With the use of high speed cameras and improved PIV techniques, the kinematics and aerodynamics of bats have been studied in great detail. A robotic flapper makes it possible to do measurements by systematically changing only one parameter at a time and investigate the parameter space outside the natural flight envelope of bats without risking animal safety. For this study, a robotic flapper (RoBat), inspired by Leptonycteris yerbabuenae was developed and tested over the speed range 1–7 m/s, with variable maximum angles of attacks ( AoA max = 55 ° and 15°, respectively) and constant AoA max = 55 ° . These measurements show the presence of a leading edge vortex (LEV) for low speeds and a fully attached flow for high speeds at low AoA max , which is in line with natural bat flight. A LEV occurs for AoA max = 55 ° throughout the complete flight speed range, and throughout which the LEV circulation coefficient remains rather constant. This implies that bats and micro air vehicles could use LEVs for high load maneuvers also at relatively high flight speeds. However, at high flight speeds the LEV bursts, which causes increased drag, most likely due to a decrease in Strouhal number.

The aerodynamic effects of wing–wing interaction in flapping insect wings

We employed a dynamically scaled mechanical model of the small fruit fly Drosophila melanogaster (Reynolds number 100-200) to investigate force enhancement due to contralateral wing interactions during stroke reversal (the ;clap-and-fling'). The results suggest that lift enhancement during clap-and-fling requires an angular separation between the two wings of no more than 10-12 degrees . Within the limitations of the robotic apparatus, the clap-and-fling augmented total lift production by up to 17%, but depended strongly on stroke kinematics. The time course of the interaction between the wings was quite complex. For example, wing interaction attenuated total force during the initial part of the wing clap, but slightly enhanced force at the end of the clap phase. We measured two temporally transient peaks of both lift and drag enhancement during the fling phase: a prominent peak during the initial phase of the fling motion, which accounts for most of the benefit in lift production, and a smaller peak of force enhancement at the end fling when the wings started to move apart. A detailed digital particle image velocimetry (DPIV) analysis during clap-and-fling showed that the most obvious effect of the bilateral ;image' wing on flow occurs during the early phase of the fling, due to a strong fluid influx between the wings as they separate. The DPIV analysis revealed, moreover, that circulation induced by a leading edge vortex (LEV) during the early fling phase was smaller than predicted by inviscid two-dimensional analytical models, whereas circulation of LEV nearly matched the predictions of Weis-Fogh's inviscid model at late fling phase. In addition, the presence of the image wing presumably causes subtle modifications in both the wake capture and viscous forces. Collectively, these effects explain some of the changes in total force and lift production during the fling. Quite surprisingly, the effect of clap-and-fling is not restricted to the dorsal part of the stroke cycle but extends to the beginning of upstroke, suggesting that the presence of the image wing distorts the gross wake structure throughout the stroke cycle.

The aerodynamics ofManduca sexta: digital particle image velocimetry analysis of the leading-edge vortex

Here we present the first digital particle image velocimetry (DPIV) analysis of the flow field around the wings of an insect (the tobacco hawkmoth Manduca sexta, tethered to a 6-component force-moment balance in a wind tunnel). A leading-edge vortex (LEV) is present above the wings towards the end of the downstroke, as the net upward force peaks. Our DPIV analyses and smoke visualisations match the results of previous flow visualisation experiments at midwing, and we extend the experiments to provide the first analysis of the flow field above the thorax. Detailed DPIV measurements show that towards the end of the downstroke, the LEV structure is consistent with that recently reported in free-flying butterflies and dragonflies: the LEV is continuous across the thorax and runs along each wing to the wingtip, where it inflects to form the wingtip trailing vortices. The LEV core is 2-3 mm in diameter (approximately 10% of local wing chord) both at the midwing position and over the centreline at 1.2 m s(-1) and at 3.5 m s(-1) flight speeds. At 1.2 m s(-1) the measured LEV circulation is 0.012+/-0.001 m(2) s(-1) (mean +/-S.D.) at the centreline and 0.011+/-0.001 m(2) s(-1) halfway along the wing. At 3.5 m s(-1) LEV circulation is 0.011+/-0.001 m(2) s(-1) at the centreline and 0.020+/-0.004 m(2) s(-1) at midwing. The DPIV measurements suggest that if there is any spanwise flow in the LEV towards the end of the downstroke its velocity is less than 1 m s(-1). Estimates of force production show that the LEV contributes significantly to supporting body weight during bouts of flight at both speeds (more than 10% of body weight at 1.2 m s(-1) and 35-65% of body weight at 3.5 m s(-1)).