Determining the Viability of the Use of Magneto-Rheological Fluid in a Low Cost Stroke Rehabilitation Device

Abstract

There are over 15 million people affected by strokes worldwide with a third left disabled. It is estimated that only 5 to 20 % regain upper limb functionality. However, research has shown that repetitive movement on the affected limb improves motor relearning. With the number of people affected by strokes rising each year the demand has begun straining hospital resources, therefore there is a need for some therapy to be moved away from clinical settings and into a person’s home. Robot assisted therapy is a growing field aiming to meet this demand. However currently there are no low cost devices able to actively exercise and strengthen a person’s hand during the acute (early) stage of stroke rehabilitation. This study is a part of a larger project involving the development of a low cost, assistive stroke rehabilitation device requiring a controllable damper. The aim of the study is to determine whether the use of magnetorheological fluid in a controllable damper is viable for use in the planned rehabilitation device. A rotary damper configuration was chosen as it can be made compact and avoid fluid leakage. To be deemed suitable for the application, the viscous torque of the damper needed to be controllable with varying input current. The required damping torques produced must be repeatable and needs to be generated below 34 C, the specified maximum operating temperature of the system. The performance of three vane designs for the rotary damper were investigated. These three designs were layered discs, a paddle and a helix. A test rig using a pulley configuration was designed and constructed to quantify the performance of the vane designs. The test rig recorded the opposing force and temperature measurements for each damper design. The measurements of interest were the off-state (no input current) torque, the achievable torque range, and also the consistency of the measurements. Experiments were conducted with the damper containing air to determine the pre-existing friction between the vane and housing, and water and motor oil were used as the damper fluid to investigate the performance of the designs with known fluid viscosities. Lastly experiments containing magneto-rheological fluid were conducted to determine the controllability and consistency of the viscous torque of each design. The paddle design was selected based on its range and consistency of produced torque, simplicity of the design and expected economical manufacture. With an input current of 0 to 2 A the damper produced a viscous torque range of 0.0036 Nm to 0.044 Nm, which was the equivalent opposing force of approximately 7.3 N. During testing of the various damper designs, a few imperfections were found. A modified version of the chosen damper was constructed to determine whether those features were manufacturing artifacts. It was found that the force measurements became smoother and previous periodic oscillations in the measurements were eliminated. The viscous torque of the paddle design was found to be controllable within the given operational conditions and therefore the use of magnetorheological fluid is a viable solution for use in a low cost stroke rehabilitation device.