Studies on Giant Magnetostrictive Actuator for Disc Brake Application

Abstract

The present study discusses the implementation of magnetostrictive actuation in disc brake system for braking action. In line with this, disc for a caliper unit is chosen and for a range of speeds, the axial braking force and braking torque required on each frictional pad is estimated. Based on the axial force required, an appropriate size of Terfenol-D rod is selected for the magnetostrictive actuator. A separate coil is used for biasing instead of permanent magnet in addition to excitation coil in the layout of a Terfenol-D actuator. The coaxial coils of Terfenol-D actuator are designed based on the required magnetic field strength. The number of coil turns obtained for coaxial coils with Ampere’s law is verified from reluctance approach. The shape factor of coaxial coils were found to be, 0.1653 and 0.1154 for coil 1 and coil 2, respectively which are close to 0.179 for maximum magnetic field at the centre of coils. The magnetic flux densities of coaxial solenoids in free air are verified analytically, experimentally and numerically using MAXWELL 2D solver. The distribution of axial, radial magnetic flux density and flux distribution in the actuator assembly with different housing materials namely mild steel, cast iron and aluminium with and without Terfenol-D are discussed considering the effect of magnetic permeability of housing materials. It is observed that the magnetic flux distribution is stronger in an actuator with mild steel housing compared to cast iron and aluminium housing. Theoretically, the magnetic field strength on a Terfenol-D rod is arrived at by taking the inductance of driving coils. Energy based Jiles-Atherton model is used to calculate the magnetization of a Terfenol-D material. Non-constitutive model takes into account of parameters such as magnetization, applied prestress and Young’s modulus to quantify the magnetostriction of a Terfenol-D material. The existing nonconstitutive magnetostriction model does not consider strain at resonance condition. The strain or magnetostriction at resonance condition is accounted by the term called quality factor. The present study considers the influence of quality factor into the existing non-constitutive models. Two magnetostrictive models are proposed to consider the influence of quality factor. The first magnetostriction model proposed considers the quality factor on the magnetostriction due to applied prestress. Theiv second magnetostriction model considers the quality factor on the magnetostriction due to magnetization of Terfenol-D. Magnetostriction of a Terfenol-D material was estimated using the proposed magnetostriction model I and II. The output obtained from the proposed magnetostriction models are verified with experimental data as well as quadratic and non-linear magnetostriction models from the literature. Experiments were conducted on a Terfenol-D actuator for DC input under prestress conditions. Performance parameters such as displacement, repeatability, step response, and response time are measured. The results indicate a better performance of the actuator at each point of excitation when step input is biased to one of the coils instead of varying the step input equally to coaxial coils. Force exerted by a Terfenol-D actuator is compared for zero and other prestress conditions using magneto-mechanical coupling equations. Hydraulic amplification unit is designed assuming amplification ratio of 5 to boost the displacement obtained from the Terfenol-D actuator. Displacement and output energy capability of a Terfenol-D actuator as well as displacement amplified Terfenol-D actuator are evaluated using stiffness match principle. Based on stiffness ratio, theoretically the amplification ratio is evaluated and verified with the value assumed in the design of amplification unit. After amplification, the braking force and braking torque available at each annular pad are evaluated from the theoretical and experiment output of a Terfenol-D actuator under different preload conditions. Further, these results are verified with the results obtained based on disc specifications. It is summarized that the braking force and torque achieved at the annular pads of a caliper unit for an applied step input and preload are sufficient to decelerate or stop the disc till the rated speed of 800 rpm. Finally, attempts are made to verify the output at the end of a hydraulic amplification unit without brake fluid by coupling a Terfenol-D actuator assembly.