A prosthetic swing-phase control mechanism simulates the action of leg musculature, aiding gait function by controlling the duration of swing, extent of heel-rise, and by allowing the shank to smoothly decelerate into full knee extension without excessive impact. Non-fluid-based mechanisms have the potential to provide a durable and affordable solution as required in many parts of the world, but the design variables that lead to improved performance of non-fluid-based swing-phase control technologies are not well established. Seven transfemoral amputees were fitted with a prosthetic knee joint and different non-fluid-based swing-phase setups were systematically assessed. Clinical testing included walk tests utilizing a potentiometer (to measure knee flexion angles) and accelerometer (to measure terminal impact decelerations) mounted on the prosthetic limb. As hypothesized, the friction and spring systems improved gait function. This includes an increased walking speed that closely matched high-end hydraulic prosthetic knee joints, decreased and more normal maximum prosthetic knee flexion, decreased flexion duration, and lower terminal impact. Further improvements were obtained using a dual spring system, two springs in series, over the more conventional single spring system. Non-fluid-based swing-phase control mechanisms are simple and significantly improve the performance of prostheses. Their application is ideal where size, weight and cost may be constrained.