A Novel Adaptive Dead-Time Control Method for GaN-Based Motor Drives

Compared to Si devices, Gallium Nitride (GaN) devices are more suitable for achieving high switching frequency in motor drive applications, thereby improving power density. However, increasing switching frequency can also lead to extra switching losses and poor total harmonic distortion (THD). A smaller dead-time can mitigate these issues, but conventional constant dead-time design methods struggle to ensure optimal performance across all load ranges. This paper proposes a novel adaptive dead-time control method for GaN high electron mobility transistors (HEMTs) in phase-legs to simultaneously enhance the efficiency and THD of motor drives. To achieve this, the detailed switching process of GaN HEMTs in a double-pulse test circuit is modeled and analyzed. The optimization principle of the dead-time setting is revealed, considering the tradeoff effect of dead-time on switching losses and THD. The adaptive dead-time control method dynamically adjusts the dead-time under different load conditions. The effectiveness of the proposed method is verified on a 1kW GaN-based permanent magnet synchronous machine drive platform. Extensive experimental results show that the proposed method can increase efficiency up to 0.85% at full load with a switching frequency of 100 kHz compared to constant dead-time. Meanwhile, THD is reduced by 1.44% under the same condition.