Aeroelasticity Testing for HALE and Long-Endurance UAS

A wing built for maximum lift-to-drag ratio on a long-endurance aircraft is, almost by definition, a flexible wing: high aspect ratio and minimum structural weight are exactly what cruise efficiency demands, and both work directly against stiffness. On this class of aircraft, the wing's shape in flight is not the shape on the drawing — it bends and twists continuously, and that deformation feeds back into the aerodynamics in ways a rigid-wing model cannot represent.

Aeroelasticity — the coupling between aerodynamic loads, structural flexibility and flight dynamics — stops being a secondary check on this class of aircraft and becomes a primary design driver.

1. What actually happened to Helios

The clearest, most documented case is NASA's own. On 26 June 2003, the Helios Prototype's long-endurance configuration (HP03) encountered atmospheric turbulence about 30 minutes into flight. The wing's dihedral — its tip deflection, a direct measure of how much the flexible wing had bent upward — grew to an unusually persistent 30 to 40 feet, well beyond the roughly 11 to 17 feet typical in normal flight. NASA's investigation found that beyond that dihedral, the aircraft's pitch oscillation became dynamically unstable: airspeed excursions roughly doubled with every cycle of the oscillation. In the final nose-down dive, the aircraft exceeded its maximum design airspeed by more than a factor of two. The resulting dynamic pressure caused progressive failure of the wing leading-edge secondary structure, the solar array and skin tore away, and the aircraft lost the ability to generate lift and fell into the Pacific Ocean. From the first attempt to diagnose the problem to the aircraft breaking up in the air: 91 seconds .

2. Why bending fed the instability instead of correcting it

NASA's post-mishap analysis identified the mechanism, and it is a genuinely aeroelastic one, not a simple structural failure:

Spanwise lift redistribution. As the wing bent upward, lift shifted from the centre of the aircraft toward the outboard panels — and that redistribution was found to be highly sensitive to gusts as small as half a foot per second, far smaller than the turbulence actually encountered.
A self-sustaining loop, not a self-correcting one. The flight control system responded to increasing dihedral by deflecting the elevators trailing-edge-up; because the outer elevators were fixed, this response removed more lift from the centre than from the tips — which pushed the wing further into the high-dihedral shape rather than back toward its nominal one.
Increasing static stability fed phugoid instability. As dihedral grew, the aircraft's centre of aerodynamic force moved further above its centre of gravity, increasing longitudinal static stability — a property that is normally reassuring, but on this vehicle it was found to increase the instability of the long-period pitch oscillation.
Pitching inertia grew roughly fivefold over the dihedral range seen in the mishap, fundamentally changing the dynamics of the flight mode as the wing bent.

None of these are properties of a rigid airframe. They exist only because the wing's shape and the aircraft's aerodynamics were coupled — which is exactly what aeroelasticity means.

3. What NASA's own investigation says should have caught this

The mishap report's root causes are directly relevant to how this class of aircraft should be tested and designed: a lack of adequate analysis methods led to an inaccurate risk assessment of the effects of a configuration change (adding concentrated masses — a fuel cell pod and hydrogen tanks — to a wing originally designed as a distributed "spanloader"), which significantly reduced the aircraft's structural and aeroelastic margins without that reduction being properly predicted beforehand.

The investigation's own technical recommendations include developing ground-test procedures and techniques appropriate to this class of vehicle to validate new analysis methods and predictions — a direct call for physical testing, not simulation alone, wherever a design departs meaningfully from prior flight-proven configurations.

4. What an aeroelastic test campaign for a flexible wing actually needs

A stiffness-scaled model, not just a geometry-scaled one. For aeroelastic behaviour to scale correctly, the model's structural stiffness distribution has to be matched to the full-scale aircraft, not just its outer shape — a geometrically accurate but over-stiff model will simply hide the coupling the test exists to find.
Gust response testing, not just steady-state polars. The Helios instability was triggered by gusts on the order of half a foot per second — small enough that steady, non-gusting wind tunnel data alone would never have revealed the sensitivity.
Synchronised structural and aerodynamic instrumentation. Wing deflection, twist and dynamic loads need to be measured alongside forces and moments, in the same run, so the coupling between them is captured rather than inferred separately.
Testing across the mass and load-distribution configurations the aircraft will actually fly , not just the baseline case — concentrated versus distributed payload mass changes the aeroelastic margin, as Helios demonstrated directly.
Real-time monitoring of wing shape during test , mirroring NASA's own recommendation to measure dihedral in real time with a visual display — the instability that destroyed Helios was invisible to the flight crew until it was already unrecoverable.

5. Five mistakes we see

Treating stiffness as a structural sign-off, separate from the aerodynamic test programme. On a flexible-wing aircraft, structural and aerodynamic testing are the same problem, not two separate ones to check off in sequence.
Testing only steady, gust-free conditions. A configuration that looks entirely stable in smooth flow can be sensitive to gusts far smaller than intuition suggests.
Assuming a configuration change is "minor" because it doesn't change the outer mould line. Helios's fatal change was adding mass, not changing shape — and it still cut deeply into the aircraft's margins.
Validating a flight control law against a rigid-body model. A control system that responds correctly to a rigid airframe can actively worsen an aeroelastic instability, exactly as the elevator response did on HP03-2.
Skipping real-time shape monitoring because "the model doesn't move that much." The margin between normal and unstable dihedral on Helios was measured in tens of feet on a 247-foot wing — a deviation easy to underestimate until it's instrumented.

6. What this means for your programme

Tell us your wing's structural configuration and any planned mass changes , and we'll scope a stiffness-scaled aeroelastic test campaign — including gust response — built to catch the coupling a purely aerodynamic or purely structural test would miss. Related reading: our hub article on

long-range fixed-wing UAS testing

, and

HALE high-altitude aerodynamics

for the same aircraft.

Sources and notes

Full account of the Helios Prototype HP03-2 mishap (26 June 2003): dihedral growth to 30–40 ft versus a normal 11–17 ft, pitch oscillation with airspeed excursions doubling per cycle, exceedance of design airspeed by more than a factor of two, progressive structural failure, 91 seconds from first diagnosis to breakup, spanwise lift redistribution sensitivity to gusts as small as 0.5 ft/sec, the elevator/dihedral feedback loop, the static-stability-versus-phugoid-instability mechanism, the roughly fivefold growth in pitching inertia, root causes, and the recommendation for vehicle-class-appropriate ground-test procedures: Noll, T.E., Ishmael, S.D., Henwood, B., Perez-Davis, M.E., Tiffany, G.C., Madura, J., Gaier, M., Brown, J.M., Wierzbanowski, T., "Technical Findings, Lessons Learned, and Recommendations Resulting from the Helios Prototype Vehicle Mishap," NASA/Langley, Dryden, Glenn, Ames and Kennedy Research Centers with AeroVironment Inc. (NASA Technical Reports Server, document 20070022260), citing the full investigation in Noll et al., "Investigation of the Helios Prototype Aircraft Mishap," Volume I – Mishap Report, January 2004.
Aircraft background (247-foot wingspan, aspect ratio 30.9:1, flying-wing configuration): consistent with our published article on HALE high-altitude aerodynamics, which covers the same aircraft's aerodynamic design in detail.
Test-methodology recommendations (stiffness-scaled models, gust response testing, synchronised structural/aerodynamic instrumentation, real-time shape monitoring) are TunnelTech engineering practice, directly responsive to the ground-test recommendations in the cited NASA investigation.

Aeroelasticity and Gust Response for Long-Endurance, High-Aspect-Ratio Wings