Large-displacement Morphing Wing "Droop Nose" for high-lift: Lessons learned in design, manufacture and testing

This presentation covers the design evolution, manufacture and testing of a 1.1 m span full-scale 3D segment of a morphing wing leading edge droop nose as part of the German project "SFB880". The droop-nose potentially enables very high-lift coefficients (in the order of 6.5) when combined with an internally-blown Coanda trailing edge flap. This droop nose undergoes a large and smooth change in curvature (90° change in the camber-line) from the clean cruise configuration to the droop low-speed configuration and this low-speed shape has been aerodynamically optimized to delay stall and reduce power requirements of the compressor system for the internal flap-blowing system. The structural design consists of a flexible hybrid skin composed of HexPly913 glass fiber reinforced plastic (GFRP) and the elastomer ethylene propylene diene monomer (EPDM) and integral stringers, two sets of kinematic driving and supporting mechanisms, and independent harmonic drive actuators. The skin, in more detail, features a spatially variable-thickness central GFRP laminate with outer layers of EPDM with embedded stiffeners running spanwise to resist spar-bending loads. Failure criteria for such GFRP-EPDM hybrid materials were established to determine maximum allowable curvature changes for both static and dynamic (i.e. fatigue cycles) cases. The design was synthesized through optimization routines, starting with determining the mechanism joint positions to meet target output trajectories with minimal error, followed by determining the best skin thickness distribution to meet the outer 3D-surface displacement and curvature targets as best as possible. The optimization routines featured global and local search methods by means of genetic algorithms and the Simplex method respectively. The results of the design-through-optimization procedure showed realizable designs with the outer surface in an acceptable error range. The manufacturing of the droop-nose was performed through mostly conventional composite production methods. A negative mold was wire-EDM machined with the 3D outer contour shape and additional metallic tooling was manufactured to support the integral stringers. The EPDM layers with embedded GFRP bundles were premanufactured and laid onto the main mold in sequence with the main thickness-variable GFRP laminate. The kinematic mechanism members were laser-cut from a 2D aluminium sheet of 5 mm and low-friction bushings inserted into the joint locations. An auxiliary spar was machined from aluminium and the kinematic assemblies with rotational actuators first mounted to this spar, followed by the composite skin. The demonstrator was instrumented with 32 strain gauges and external wire displacement sensors (as a safety measure to prevent asynchronous rotation of the independent actuators) and was also externally measured with digital image correlation techniques. The strain results show maximum strains (~0.7%) within the allowable dynamic limits (~1.3%) and qualitative shape results show generally good agreement although one undesirable "bump" exists in an aerodynamically unfavorable position. It is assumed that this is caused by higher-than-expected flexibility in the aluminium kinematics and further DIC experiments are underway to measure the output trajectories with and without the skin to determine the skin stiffness effects on the kinematics. The kinematics will potentially be remanufactured from steel to increase kinematic stiffness. Calculated torque values show values in the order of 330 Nm (skin only, no aerodynamic load) and work is also underway to measure torque values. Experiments with other samples show that integrating carbon fiber bundles in the EPDM layers could be used for electrically resistive thermal de-icing, integrating metallic lightning strike protection meshes and polyethylene erosion protection layers is feasible, and that waviness and roughness are aerodynamically acceptable. Work is also underway to test the droop nose in a spanwise-bending large test rig, and the experimental results with combined droop and spanwise bending loads are expected to be disseminated