In contrast to larger flight-capable insects such as hawk moths and fruit flies, miniature flying insects such as thrips show the obligatory use of wing-wing interaction via "clap and fling" during the end of upstroke and start of downstroke. Although fling can augment lift generated during flapping flight at chord-based Reynolds number (Re) of 10 or lower, large drag forces are necessary to clap and fling the wings. In this context, bristles observed in the wings of most tiny insects have been shown to lower drag force generated in clap and fling. However, the fluid dynamic mechanism underlying drag reduction by bristled wings and the impact of bristles on lift generated via clap and fling remain unclear. We used a dynamically scaled robotic model to examine the forces and flow structures generated during clap and fling of: three bristled wing pairs with varying inter-bristle spacing, and a geometrically equivalent solid wing pair. In contrast to the solid wing pair, reverse flow through the gaps between the bristles was observed throughout clap and fling, resulting in: (a) drag reduction; and (b) weaker and diffuse leading edge vortices that lowered lift. Shear layers were formed around the bristles when interacting bristled wing pairs underwent clap and fling motion. These shear layers lowered leakiness of flow through the bristles and minimized loss of lift in bristled wings. Compared to the solid wing, peak drag coefficients were reduced by 50-90% in bristled wings. In contrast, peak lift coefficients of bristled wings were only reduced by 35-60% from those of the solid wing. Our results suggest that the bristled wings can provide unique aerodynamic benefits via increasing lift to drag ratio during clap and fling for Re between 5 and 15.

The smallest flying insects with body lengths under 2 mm show a marked preference for wings consisting of a thin membrane with long bristles, and the use of clap and fling kinematics to augment lift at Reynolds numbers (Re) of approximately 10. Bristled wings have been shown to reduce drag forces in clap and fling, but the aerodynamic roles of several bristled wing geometric variables remain unclear. This study examines the effects of varying the ratio of membrane area (A M ) to total wing area (A T ) on aerodynamic forces and flow structures generated during clap and fling at Re on the order of 10. We also examine the aerodynamic consequences of scaling bristled wings to Re = 120, relevant to flight of fruit flies. We analyzed published forewing images of 25 species of thrips (Thysanoptera) and found that A M /A T ranged from 14% to 27%, as compared to 11% to 88% previously reported for smaller-sized fairyflies (Hymenoptera). These data were used to develop physical bristled wing models with A M /A T ranging from 15% to 100%, which were tested in a dynamically scaled robotic clap and fling model. At all Re, bristled wings produced slightly lower lift coefficients (C L ) when compared to solid wings, but provided significant drag reduction. At Re = 10, largest values of peak lift over peak drag ratios were generated by wing models with A M /A T similar to thrips forewings (15% to 30%). Circulation of the leading edge vortex and trailing edge vortex decreased with decreasing A M /A T during clap and fling at Re = 10. Decreased chordwise circulation near the wing tip, vortex shedding, and interaction between flow structures from clap with those from fling resulted in lowering C L generated via clap and fling at Re = 120 as compared to Re = 10. Clap and fling becomes less beneficial at Re = 120, regardless of the drag reduction provided by bristled wings.

In contrast to larger species, little is known about the flight of the smallest flying insects, such as thrips and fairyflies. These tiny animals range from 300 to 1000 microns in length and fly at Reynolds numbers ranging from about 4 to 60. Previous work with numerical and physical models have shown that the aerodynamics of these diminutive insects is significantly different from that of larger animals, but most of these studies have relied on two-dimensional approximations. There can, however, be significant differences between two-and three-dimensional flows, as has been found for larger insects. To better understand the flight of the smallest insects, we have performed a systematic study of the forces and flow structures around a three-dimensional revolving elliptical wing. We used both a dynamically scaled physical model and a three-dimensional computational model at Reynolds numbers ranging from 1 to 130 and angles of attacks ranging from 0 • to 90 •. The results of the physical and computational models were in good agreement and showed that dimensionless drag, aerodynamic efficiency, and spanwise flow all decrease with decreasing Reynolds number. In addition, both the leading and trailing edge vortices remain attached to the wing over the scales relevant to the smallest flying insects. Overall, these observations suggest that there are drastic differences in the aerodynamics of flight at the scale of the smallest flying animals.

Tiny flying insects of body lengths under 2 mm use the “clap-and-fling” mechanism with bristled wings for lift augmentation and drag reduction at a chord-based Reynolds number (Re) on O(10). We examine the wing–wing interaction of bristled wings in fling at Re = 10 as a function of initial inter-wing spacing (δ) and degree of overlap between rotation and linear translation. A dynamically scaled robotic platform was used to drive physical models of bristled wing pairs with the following kinematics (all angles relative to vertical): (1) rotation about the trailing edge to angle θr, (2) linear translation at a fixed angle (θt), and (3) combined rotation and linear translation. The results show that (1) the cycle-averaged drag coefficient decreased with increasing θr and θt and (2) decreasing δ increased the lift coefficient owing to increased asymmetry in the circulation of leading and trailing edge vortices. A new dimensionless index, reverse flow capacity (RFC), was used to quantify the maximum possible ability of a bristled wing to leak the fluid through the bristles. The drag coefficients were larger for smaller δ and θr despite larger RFC, likely due to the blockage of inter-bristle flow by shear layers around the bristles. Smaller δ during early rotation resulted in the formation of strong positive pressure distribution between the wings, resulting in an increased drag force. The positive pressure region weakened with increasing θr, which in turn reduced the drag force. Tiny insects have been previously reported to use large rotational angles in fling, and our findings suggest that a plausible reason is to reduce drag forces.

Miniature insects must overcome significant viscous resistance in order to fly. They typically possess wings with long bristles on the fringes and use clap-and-fling mechanism to augment lift. These unique solutions to the extreme conditions of flight at tiny sizes (< 2 mm body length) suggest that natural selection has optimized wing design for better aerodynamic performance. However, species vary in wingspan, number of bristles (n), and bristle gap (G) to diameter (D) ratio (G/D). How this variation relates to body length (BL) and its effects on aerodynamics remain unknown. We measured forewing images of 38 species of thrips and 21 species of fairyflies. Our phylogenetic comparative analyses showed that n and wingspan scaled positively and similarly with body length across both groups, whereas G/D decreased with BL, with a sharper decline in thrips. We next measured aerodynamic forces and visualized flow on physical models of bristled wings performing clap-and-fling kinematics at chord-based Reynolds number of 10 using a dynamically scaled robotic platform. We examined the effects of dimensional (G, D, wingspan) and non-dimensional (n, G/D) geometric variables on dimensionless lift and drag. We found that: (a) increasing G reduced drag more than decreasing D; (b) changing n had minimal impact on lift generation; and (c) varying G/D minimally affected aerodynamic forces. These aerodynamic results suggest little pressure to functionally optimize n and G/D. Combined with the scaling relationships between wing variables and BL, much wing variation in tiny flying insects might be best explained by underlying shared growth factors.

No abstract

The smallest flying insects such as thrips (body length < 2 mm) are challenged with needing to move in air at chord-based Reynolds number (Rec) on the order of 10. Pronounced viscous dissipation at such low Rec requires considerable energetic expenditure for tiny insects to stay aloft. Free-flying thrips flap their densely bristled wings at large stroke amplitudes, bringing both wings in close proximity of each other at the end of upstroke ('clap') and moving their wings apart at the start of downstroke ('fling'). From high-speed videos of free-flying thrips, we observed that their forewings remain clapped for approximately 10% of the wingbeat cycle before start of fling. We sought to examine if there are aerodynamic advantages associated with pausing wing motion after clap and before fling at Rec=10. A dynamically scaled robotic clap-and-fling platform was used to measure lift and drag forces generated by physical models of non-bristled (solid) and bristled wing pairs for pause times ranging between 0% to 41% of the cycle. In both solid and bristled wings, varying pause time showed no effect on average force coefficients generated within each half-stroke. This was supported by nearly identical time-variation of circulation of the leading and trailing edge vortices for different pause times. At smaller pause times, bristled wings showed larger reduction of cycle-averaged drag coefficient as compared to that of solid wings. For a given wing design (solid or bristled), the ratio of cycle-averaged lift coefficient to cycle-averaged drag coefficient was unchanged across different pause times. We observed 13.5% drop in cycle-averaged power coefficient and 3% drop in cycle-averaged lift coefficient when moving from 0% pause to 9% pause duration. Our results suggest that pausing at the end of clap can be beneficial for reducing the power required to fling, with a small reduction in lift.

The smallest flying insects such as thrips (body length < 2 mm) are challenged with needing to move in air at chord-based Reynolds number (Re c ) on the order of 10. Pronounced viscous dissipation at such low Re c requires considerable energetic expenditure for tiny insects to stay aloft. Free-flying thrips flap their densely bristled wings at large stroke amplitudes, bringing both wings in close proximity of each other at the end of upstroke ('clap') and moving their wings apart at the start of downstroke ('fling'). From high-speed videos of free-flying thrips, we observed that their forewings remain clapped for approximately 10% of the wingbeat cycle before start of fling. We sought to examine if there are aerodynamic advantages associated with pausing wing motion after clap and before fling at Re c =10. A dynamically scaled robotic clap-and-fling platform was used to measure lift and drag forces generated by physical models of non-bristled (solid) and bristled wing pairs for pause times ranging between 0% to 41% of the cycle. In both solid and bristled wings, varying pause time showed no effect on average force coefficients generated within each half-stroke. This was supported by nearly identical time-variation of circulation of the leading and trailing edge vortices for different pause times. At smaller pause times, bristled wings showed larger reduction of cycle-averaged drag coefficient as compared to that of solid wings. For a given wing design (solid or bristled), the ratio of cycle-averaged lift coefficient to cycle-averaged drag coefficient was unchanged across different pause times. We observed 13.5% drop in cycle-averaged power coefficient and 3% drop in cycle-averaged lift coefficient when moving from 0% pause to 9% pause duration. Our results suggest that pausing at the end of clap can be beneficial for reducing the power required to fling, with a small reduction in lift.