Shape Memory Alloy as Actuator to Deflect a Wing Flap

Abstract

This project is a feasibility study to control a wing flap using shape memory alloy (SMA). Two key issues were addressed: (1) how to use SMA to effectively actuate the wing flap and (2) how to improve the actuation frequency once the initial conceptual design proves to be feasible. The 1 task involves the conceptual design, and compatibility between SMA and the other structural components and the arrangements of the SMA wires. The 2 task involves optimization of the cooling rate of SMA through materials selection and testing or through a drastic design improvement. The project begins with rib design, fabrication and tests. A section of the wing, spanning 150mm, was built after the rib successfully demonstrated the required deflection. The prototype consists of 2 ribs with aluminum rods as spars, and instead of a conventional hinge, metallic shims were used to allow the deflections. Flexible skins were also explored and experimented. Glass fiber composite bonded to 0.5mm thick silicon rubber sheet and overlapping glass fiber composite were made and tested on the prototype. Calculation to handle the aerodynamic load was also performed. It is known that the cooling rate of SMA depends on several parameters including the materials characteristics, the geometric factor and the environmental condition. Two approaches were taken to tackle the problem of slow cooling rate of SMA in order to increase its actuation frequency, (1) selecting SMAs with low transformation hysteresis, and (2) drastic improvement in the conceptual design through the use of a novel mechanism that could engage the SMA wire for actuation and to disengage during cooling. In this way the actuation rate can be further improved. Finally the controllability of the prototype was also investigated, including the measurement and estimation of power consumption, and dependence of heating and cooling time on input variations. A microcontroller was employed to accomplish the task of heating/cooling the SMA in sequence, and this could be expanded to include the control of the engage/disengage mechanism.