IEEE TRANSACTIONS ON MAGNETICS, VOL. 42, NO. 2, FEBRUARY 2006 319 Parametric Analysis of Eddy-Current Brake Performance by 3-D Finite-Element Analysis Sebastien E. Gay and Mehrdad Ehsani Department of Electrical Engineering, Texas A&M University, College Station, TX 77843-3128 USA Eddy-current brakes as primary brakes in automobiles can potentially remedy the problems posed by conventional friction brakes. First, however,it will be necessary to analyze the effece of each design parameter on the performance of the eddy-current brake before designing a conceptual brake for a practical application and evaluating its behavior in real-life driving situations. In this paper, we analyze the effects of design parameters on the torque-speed curve of the eddy-current brake by the means of three-dimensional (2-D) finite-element analysis. The paper follows a previous work in which analysis was performed with a 2-D analytical model. Index Terms—Analytical model, design, eddy-current brake, parametric analysis. I. INTRODUCTION R OAD, rail, and air vehicles all rely mainly or solely on me- chanical friction brakes [1]. These brakes are composed of two functional parts: a cast-iron rotor (disc or drum) and pads pressed against the rotor to generate the braking force by friction. The pressure is applied by a hydraulic or pneumatic circuit. Friction braking is dissipative: the vehicle’s kinetic en- ergy is dissipated as heat on the disc, which heats up to several hundred degrees Celsius. Despite its tremendous advantages in compactness and effectiveness, friction braking suffers from se- vere limitations: — loss of braking force with increasing temperature (fading phenomenon); — warping of discs; — wear of pads and rotors; — complexity and fuel consumption of power assistance; — slow response time due to power assistance, especially in trucks, buses and trains; — complexity of controlling each wheel’s braking indepen- dently; — necessity of complex and costly anti-lock controls; — risk of hydraulic fluid leak; — risk of brake fluid contamination by water and subsequent loss of braking power; — challenging integration with anti-lock, traction, and dy- namic stability controls. The concept of integrated contactless magnetic brake was in- vented to remedy to these problems. The integrated brake com- bines a conventional friction brake with a magnetic brake. This novel concept has many advantages over friction brakes: — reduced wear; — reduced sensitivity to fading; — reduced fuel consumption of power assistance; — faster control dynamics; — easier integration with anti-lock, traction, and dynamic stability controls; — easy individual wheel braking control; — electric actuation, no fluid. Digital Object Identifier 10.1109/TMAG.2005.860782 The magnetic brake consists of a number of stationary mag- nets facing the friction brake’s rotor across an air gap. When the disc spins, eddy currents are induced in the disc and the interac- tion of these currents with the magnetic field creates the braking force. The braking force is directly controlled by the intensity of the magnetic field. Eddy-current brakes are currently used as electromagnetic re- tarders for the secondary and downhill braking of commercial trucks, buses, and light rail vehicles. A single brake is mounted in the transmission aft of the differential gear underneath the chassis. The brake is energized by electromagnets connected to the battery and controlled manually by the driver. Electromag- netic retarders provide a large braking torque without contact between the magnets and the rotor, thereby alleviating the duty on the main braking system. They however suffer shortcomings such as heavy weight, excessive power consumption of the elec- tromagnets, and vulnerability to power failure. The concept analyzed in the present work solves these prob- lems by using compact rare-earth permanent magnets that pro- vide a strong magnetic field and a novel magnetic circuit struc- ture that allows controlling the field easily by electrical means. The present work follows a theoretical analysis performed by the authors [1]. II. FINITE-ELEMENT ANALYSIS Applying finite-element methods to the analysis of the eddy- current brake involves several choices: — choosing between two-dimensional (2-D) and three-di- mensional (3-D) analysis; — choosing between analysis methods; — choosing boundary conditions, periodicities and symme- tries; — deciding on the level of approximation of the properties of the materials. In the past, most finite-element analyses of eddy-current brakes published used 2-D analysis [2]–[7]. The use of only two dimensions yields a very limited number of nodes, which makes computations easy and fast. However, 2-D analysis does not account for 3-D effects directly and approximations must be made to account for these. The accuracy of the solution suffers greatly. Two-dimensional methods are typically better 0018-9464/$20.00 © 2006 IEEE