Axial Flux Variable Gap Motor: Application in Vehicle Systems

Alternative electric motor geometry with potentially increased efficiency is being considered for hybrid electric vehicle applications. An axial flux motor with a dynamically adjustable air gap (i.e., mechanical field weakening) has been tested, analyzed, and modeled for use in a vehicle simulation tool at Argonne National Laboratory. The advantage of adjusting the flux is that the motor torque-speed characteristics can better match the vehicle load. The challenge in implementing an electric machine with these qualities is to develop a control strategy that takes advantage of the available efficiency improvements without using excessive energy to mechanically adjust the air gap and thus reduce the potential energy savings. Motor efficiency was mapped in terms of speed, torque, supply voltage, and rotor-tostator air gap. Maps of optimal gap versus efficiency were used to develop a motor model and control strategy, which were incorporated into the PNGV Systems Analysis Toolkit vehicle modeling software. INTRODUCTION Axial flux geometry, in which the machine is made from two opposing disks — one fixed and one rotating — has two substantial advantages over radial flux geometry, in which a drum rotates inside a cylinder. First, there are significant volume savings over the more commonly used radial flux geometry, for which much of the internal volume of the rotor does not contribute to power output. Second, and more importantly, axial flux geometry allows a very simple technique for field weakening that relies on mechanical adjustment of the air gap, which does not impinge significantly on efficiency. Within a reasonable band, increasing the air gap increases the copper loss as the torque constant decreases, but decreases the iron loss as the flux density is reduced [1]. Motor air gap is a key design parameter in any motor design that directly affects the motor torque output and back electromotive force (EMF) constant. In a conventional radial flux motor, the air gap is optimized for a single motor torque-speed operating point. Therefore, at off-design-point motor torque, speed and resulting efficiency are not optimized. The axial flux motor allows the possibility of actively varying the rotorto-stator air gap to optimize the motor performance and efficiency for a particular operating point. So far, axial flux motors are mainly used in hightorque, low-speed applications because of its inherent characteristics. The model investigated here was used in solar cars as a direct-drive propulsion system [2]. Recently, use of an axial flux motor for the starter/alternator has been proposed [3]. For this experiment, we measured the efficiency at distinct operating points (torque, speed, input voltage, and air gap) in all four quadrants of operation (forward powering, forward regeneration, reverse powering, reverse regeneration). Measured results were analyzed to determine the air gap, which optimizes efficiency at a given torque, speed, and input voltage. Also vehicle performance was simulated with the PNGV Systems Analysis Toolkit (PSAT) vehicle modeling software based on the motor test results. MOTOR SPECIFICATIONS There are several torque and power curves that are of interest when evaluating a traction drive motor [4]: • Electrical peak torque and power, which are limited by electrical characteristics of the motor (i.e., pull-out torque); • Current limited peak torque and power, which are limited by the maximum design current of the inverter; • Motor thermal limited torque and power, which are limited by the time it takes the stator and/or rotor to increase from the rated cooling temperature to the maximum design temperature; and • Motor thermal limited continuous torque and power, which are limited by the amount of heat the motor can reject when in thermal equilibrium. In most cases, the above specifications are provided by the manufacturers, and they can be verified by the experiment. General specifications for the test motor, taken from the manufacturer’s data sheet are listed in Table 1. This motor is designed and manufactured by New Generation Motor Corporation [5]. Table 1: Motor Specifications Continuous Power (kW) 2.5 Rated Speed (rpm) 300 Rated Torque (Nm) 102 Rated Voltage (V) 48 Cooling Air cooling Min/Max Air Gap (mm) 1.8/6.0 Motor Diameter (mm) 315 Motor Width (mm) 70 Weight (kg) 20 Type DC Brushless Permanent Magnet TEST SETUP The following equipment is needed to test four-quadrant operation of electric motors. • Bi-directional DC power supply (battery simulator) — During powering and regeneration, current is provided in either direction (from source to motor, from motor to source). • Active controlled load (dynamometer) — Speed and direction of load should be controlled according to the desired quadrant operation. • Automatic data acquisition system The test setup was assembled in Argonne’s Advanced Powertrain Test Facility (APTF). A block diagram of the test setup according to the stated requirements is shown in Figure 1. Figure 1: Block diagram of test setup Figures 2 and 3 are photographs of the test setup. An axial flux motor was mounted onto the dynamometer on a portable test stand. For this experiment, the dynamometer only served as a through shaft and support bearings. The test motor was adapted via a coupling to the flange on one side of the dynamometer. An induction motor was also mounted on the portable test stand and coupled to the opposite side of the dynamometer via a flex coupling and a straight, flanged drive shaft. The test motor was controlled via a remotely located (control room) interface with the computer control. Figure 2: Portable Motor Test Stand The test motor was powered by the motor controller via a DC feed directly from one channel of an ABC-150 DC power supply. The rotor-to-stator gap was changed manually by rotating a gear-driven lead screw or by servo motor between tests. Control of the induction motor was via an AC vector-drive motor controller. This motor was used as a load motor for either speed or torque control. Figure 3: NGM Axial Gap Motor Bi-directional DC power supply Test Motor Load Motor DAQ & Control NGM Motor AC Induction Motor (dyno) TEST PROCEDURE Four-quadrant motor operation is defined in Figure 4. 2 quadrant 1 quadrant 3 quadrant 4 quadrant Figure 4: Definition of Operating Quadrants The test motor and load motor should be operated as prescribed in Table 2 for each quadrant of operation. Table 2: Operating Mode of Test and Load Motors In the case of throttle mode, the test motor supplies torque to the load motor, and the energy flow is from DC source to test motor. During regeneration, the energy flow is from test motor to DC source by the mechanical input from the load motor. The following signals were measured to calculate motor efficiency. Efficiency during throttle mode is calculated as follows. current input DC voltage input DC speed motor torque motor efficiency × × = Efficiency during regeneration mode is calculated as follows. speed motor torque motor current input DC voltage input DC efficiency × × = DC input current DC input voltage Motor torque Motor speed Motor temperature The measured signals were collected with a Labview data-acquisition program. The test motor was operated in torque mode, and the load motor (induction motor) was operated in speed mode. The load motor was controlled to the desired speed, and then the torque from the test motor was increased to the desired torque. At that point, the above signals were measured and recorded, and the system efficiency was calculated. Peak toque and maximum speed were also measured. TEST RESULTS TORQUE-SPEED CURVE Maximum torque at each air gap and operating speed was measured and recorded and is shown in Figure 5. To measure maximum torque, torque command was increased until output torque did not increase. Output torque at that point was defined as maximum torque. To measure maximum speed, speed command was increased until output torque reached zero. This speed was defined as maximum speed. As expected, maximum torque decreases as air gap increases and maximum speed increases as air gap increases. The maximum torques at the 5-mm and 6-mm air gaps were relatively close because of air gap measurement error. More precise air gap measurement is recommended. By adjusting the air gap, the operating area of test motor can be extended. The maximum torque is 136 N-m at 400 rpm with an air gap of 1.5 mm and the maximum speed is 1,200 rpm with an air gap of 6 mm. 0 20 40 60 80 100 120 140 160 0 200 400 600 80