Experimental Evaluation and Reproducible Digital Filter Benchmarking for Hall-Sensor-Based DC Motor Speed Feedback
Year: 2026
Type: article
Abstract: Hall-effect sensors are widely used for low-cost speed feedback in DC motor drives, yet the measured speed is influenced not only by controller design but also by hardware limitations and signal-processing characteristics. This study investigates the combined influence of these factors through an experimental evaluation of a Hall-sensor-based DC motor drive and an independent benchmark of commonly used digital low-pass filters. Open-loop measurements produced speed estimates of 201–202 rpm. With a 175 rpm reference, proportional control achieved only 124–127 rpm because the voltage available at the motor terminals was reduced by the combined effects of the power supply and the motor driver. Replacing the 6 V source with a 9 V laboratory supply increased the available motor voltage, allowing PI and PID controllers to achieve stable responses of 173–176 rpm and 173–177 rpm, respectively. Independently of the hardware experiments, fourth-order Bessel and Butterworth IIR filters and a 15th-order Hamming-window FIR filter were evaluated using a reproducible synthetic Hall-pulse benchmark at a 100 kHz sampling frequency. The FIR filter provided the lowest constant group delay (75 μs), whereas the IIR filters achieved approximately 57 dB attenuation at 10 kHz with higher frequency-dependent delays. The results show that observed speed-control performance can be strongly affected by hardware voltage constraints, while digital-filter selection requires balancing attenuation, delay, sampling frequency, and application requirements. Separating hardware evaluation from offline filter benchmarking provides a more reliable interpretation of Hall-sensor speed-feedback performance.
Authors: Atılgan Altınkök, Nuran Salimov
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W7225882363
