Experimental characterization of prosthetic mechanisms with one-degree of freedom

This paper deals with experimental tests performed on three mechanisms with one degree of freedom employed as powered articulations of prostheses for upper limb amputees. The purpose of the experimental characterization was to evaluate the mechanism performance and quality by means of proper parameters in order to objectively assess the effects of possible modifications of the mechanism design. To this aim two parameters were chosen: the Global Efficiency ηg, intended as the ratio of the total output energy (Lu) to the total input energy (Le) supplied during a test trial, and the Mechanical Efficiency ηm, that is the ratio of the output energy Lu to the mechanical energy Lm provided by the motor that drives the mechanism. By defining the Electrical Efficiency ηel, as the ratio between the mechanical energy Lm provided by the motor and the input electrical energy Le it holds: ηg = Lu/Le = (Lu/Lm) (Lm/Le) = ηel ηm. During the test trials, the mechanisms were driven to execute specific operative load cycles, corresponding to a trajectory of the followers from an initial position to a final different one, with initial and final velocity equal to zero, and with a mono-directional flow of the energy, i.e. the mechanism actuator works exclusively as a motor to equilibrate the external load which always has a resistant role. The procedure to experimentally evaluate ηg and ηm of mechanisms working in such operative conditions is shown and discussed. In particular, it is based on the theoretical calculation of Lu (that is straightforward if the loads acting on the mechanism, its kinematic scheme and the trajectory of the operative load cycles are known), on the experimental measure of Le, and finally on the implementation of a proper model of the motor Electrical Efficiency (that depends on the motor characteristics, to be known, and its working conditions to be experimentally detected). The test bench used to apply the method is only required to acquire the supply voltage and current of the motor, and the mechanism follower velocity. Therefore, the bench is simple and rather cheap and the procedure to integrate the experimental data and theoretical calculations is straightforward. The paper also outlines in some detail the application of the method to the three prosthetic devices mentioned above, i.e. two mechanisms for a shoulder articulation (which is at a prototype stage) and an elbow joint (which is a more mature product, but still in evolution). Different test sessions were performed on each mechanism, varying the follower mean velocity and the maximum value of the applied load (that is not constant during a given trajectory) with the purpose of investigating the effects of increasing work rate and speed of movement on the mechanism efficiency. The corresponding results are reported and discussed.