On the Design of Cylindrical Magnetorheological Clutches

Magnetorheological fluids (MRFs) exhibit variable mechanical properties in response to magnetic stimuli. Thanks to their rapid and reversible viscosity changes, MRFs can be utilized in a variety of applications including torque transmission devices such as clutches. In this work, the geometrical design of cylindrical MR clutches is investigated with the aim of optimizing the torque transmission capability. Effects of design parameters such as radius, gap size, effective length, and MRF volume are investigated in the presence of variable magnetic field. Magneto-mechanical behavior of some MR fluids with different particle content are investigated by means of two different constitutive models to simulate the clutch performance in a range of geometrical parameters. It is shown that the transmitted torque increases nonlinearly by inner radius of the clutch, for example, in the studied range, 150% higher torque is achieved for only 40% larger radius. The clutch's gap size does not much affect the torque, however, since it significantly affects the required volume of MRF, a lower gap size is favorable. The torque is also calculated for constant volumes of the MRFs. At a certain volume, although a higher radius translates to a shorter length, it is still favorable. For example, a 40% increase in the design radius, almost doubles the transmitted torque for both the studied MRFs. Moreover, a clutch filled by an MRF with higher particle content can transmit higher torques. It is also concluded that increasing the clutch's radii is an easier way to improve the mean torque while altering the applied magnetic field is a better way to adjust the range of achievable torques. The simulations also demonstrate the importance of an accurate and reliable constitutive model in the design of MR devices. It is shown that Bingham model is not reliable at high magnetic fields as it underestimates the transmitted torque though calibrated at each field intensity. However, the employed nonlinear model provides more reliable results by only being calibrated at an arbitrary field.