UBC Undergraduate Research

Mitigating Misaligned Hall Sensors in Brushless DC Motors Using a Calibration Routine for Improved Fast Electromechanical Transients Hasman, Matthew

Abstract

Hall-sensor-controlled brushless DC (BLDC) motors are commonly used in many applications due to their low cost and simple control. In a typical BLDC motor, a permanent magnet synchronous machine (PMSM) is controlled by a voltage source inverter (VSI) using the Hall sensor signals. Ideally, the three Hall sensors are spaced 120 electrical degrees apart. However, due to manufacturing tolerances, the sensors’ actual position may differ, resulting in uneven conduction intervals and degradation of motor performance. Previous research proposed averaging the Hall sensor signals to mitigate the misaligned Hall sensors and achieve a steady state performance close to ideal. However, such averaging filters have a memory and may degrade the dynamic performance in fast transients. To enable fast dynamic performance, this thesis proposes a calibration routine that identifies the errors and records the necessary correction values in a lookup table. The stored correction angles from the lookup table are then applied to the Hall sensor signals at run time, which removes the delay present in the averaging filter method and significantly improves the dynamic performance of the BLDC motors during fast electromechanical transients.