Optimized Detection Coil Boosts Safety in EV Wireless Charging
As electric vehicles (EVs) continue to gain traction across global markets, the demand for seamless, user-friendly charging solutions has never been higher. Among the most promising technologies is wireless power transfer (WPT), which offers a cable-free, automated charging experience. However, one persistent challenge has hindered its widespread adoption: the risk posed by metal foreign objects entering the high-power electromagnetic field between the charging pads. These objects, ranging from coins and keys to larger debris, can heat up rapidly due to induced eddy currents, potentially causing thermal damage or even fire hazards. To ensure safe and reliable operation, effective metal object detection (FOD) systems are essential.
Recent research led by Zhang Bin and his team at the School of Mechanical and Power Engineering, Shanghai Jiao Tong University, has made a significant leap forward in addressing this critical safety issue. Published in the Journal of Power Supply, their study introduces a novel sensitivity optimization method for detection coils used in EV wireless charging systems. This advancement not only enhances detection accuracy but also strengthens system resilience against interference, marking a pivotal step toward commercial viability and user confidence in wireless charging infrastructure.
The core of the team’s innovation lies in a refined approach to the design and performance of the detection coil—a key component in FOD systems. Traditional methods have often struggled with trade-offs between sensitivity, detection range, and susceptibility to noise from the primary power transfer field. Some techniques rely on auxiliary coils or power fluctuations, which can be slow or prone to false positives. Others use thermal imaging or complex sensor arrays, driving up cost and complexity. Zhang and his colleagues identified a fundamental gap: the lack of a comprehensive theoretical framework linking coil geometry to detection performance. Without such a model, coil design has largely been empirical, limiting optimization and scalability.
To close this gap, the researchers developed an equivalent electromagnetic model that precisely describes how a metallic object affects the impedance of a rectangular detection coil. When a metal object enters the coil’s alternating magnetic field, it generates opposing eddy currents, altering the coil’s effective inductance and resistance. By modeling this interaction, the team derived a theoretical formula for detection sensitivity—the percentage change in coil inductance caused by the presence of a foreign object. This formula incorporates key geometric parameters such as coil dimensions, number of turns, trace width, spacing between turns, and material thickness.
Armed with this analytical foundation, the team systematically evaluated how each parameter influences sensitivity. Their findings revealed several critical insights. First, larger coil dimensions generally increase sensitivity, but only up to a point. When the metal object is significantly smaller than the coil, the signal change becomes too weak to detect reliably. Conversely, oversized coils are more susceptible to electromagnetic interference from the main charging field, reducing signal-to-noise ratio. The optimal balance occurs when the coil size closely matches the expected size range of potential foreign objects—typically 25 to 60 mm in automotive applications.
Second, the number of coil turns plays a crucial role. Sensitivity increases with the number of turns, but the gains diminish beyond a certain threshold. The researchers found that too few turns result in weak signals, while too many can lead to excessive parasitic capacitance and reduced bandwidth. A moderate number of turns, carefully tuned to the operating frequency, provides the best compromise between sensitivity and stability.
Third, the spacing between coil traces—often overlooked in prior designs—was shown to have a non-linear impact on performance. Tighter spacing enhances magnetic coupling with nearby objects, improving sensitivity for small metals. However, if spacing is too narrow, manufacturing tolerances and skin effect losses become problematic. The team’s model allowed them to pinpoint the ideal gap that maximizes sensitivity without compromising durability or efficiency.
Based on these insights, the researchers proposed an optimized coil configuration: a rectangular spiral with 11 turns, an outer dimension of 30 mm, a trace width of 1 mm, a thickness of 35 μm (1 oz copper), and a spacing of 2 mm between turns. This design was not arbitrary but the result of rigorous simulation and theoretical validation. The team used electromagnetic field simulation software to verify the accuracy of their model, comparing predicted inductance values against simulated results across multiple parameter variations. The root mean square error between theory and simulation remained below 5%, confirming the robustness of their approach.
To further validate their method, the team fabricated three different coil prototypes: two with non-optimized geometries and one with the proposed optimal design. They then conducted a series of experiments measuring the inductance change when various metal objects—ranging from small coins to larger aluminum plates—were placed near the coils. The results were striking. The optimized coil demonstrated a detection sensitivity that was significantly higher than its counterparts, with a minimum detectable object size of approximately 24 mm. More importantly, the experimental data closely matched the theoretical predictions, with root mean square errors in inductance and sensitivity below 2%, reinforcing the reliability of their model.
The practical implications of this work were tested in a real-world setting: a 3 kW EV wireless charging system. The detection system was integrated into the charging pad, with the optimized coil arranged in a non-overlapping array to ensure full coverage of the charging zone. Each coil operated independently to prevent single-point failures and maintain high sensitivity. The signal processing chain included a sinusoidal excitation source, a series resonant sampling circuit, amplification stages, high-pass filtering, and peak detection—all designed to amplify the small impedance changes caused by metal objects while suppressing low-frequency noise from the primary power field.
During testing, the system successfully detected a variety of metallic objects, including a 90 mm diameter metal lid, a 40 mm aluminum sheet, and a 25 mm coin, all placed just 5 mm above the coil surface. When no object was present, the peak detection circuit output a steady 0.65 V. Upon insertion of the coin, the signal change reached 0.45 V—a 69% increase—well above the detection threshold. Larger objects produced even greater responses, confirming the system’s ability to not only detect but also roughly estimate object size based on signal magnitude.
Crucially, the system demonstrated strong immunity to interference. Despite the intense electromagnetic environment generated by the 3 kW power transfer, the output waveform remained clean and stable, with no observable noise from the primary field. This anti-interference capability is a major advantage over previous methods, which often required complex shielding or signal processing to mitigate false triggers.
Another key feature of the system is its ability to distinguish between foreign objects and the vehicle’s own receiver. In real-world use, the position of the EV’s pickup coil can vary slightly depending on parking alignment, which can also cause changes in the detection coil’s inductance. To address this, the team implemented a pre-detection calibration step. After the vehicle is parked, the system records the baseline output voltage of each coil before initiating the charging sequence. Any subsequent deviation from this baseline is flagged as a potential foreign object, effectively filtering out normal positional variations.
The success of this study lies not only in the improved hardware but also in the creation of a design methodology that can be applied across different wireless charging platforms. By establishing a clear link between coil geometry and detection performance, the researchers have provided a blueprint for engineers to tailor FOD systems to specific applications—whether for passenger cars, commercial fleets, or autonomous vehicles. This level of customization was previously unattainable with rule-of-thumb design approaches.
From a safety standpoint, the ability to reliably detect small metal objects as small as 25 mm is a major achievement. Such objects are common in everyday environments—lost change, bottle caps, or metallic debris—and pose a real risk if left undetected. Regulatory standards, such as GB/T 38775.3—2020 for EV wireless charging systems, mandate robust foreign object detection, and this optimized coil design meets and exceeds those requirements. It brings wireless charging closer to the level of safety expected in conventional plug-in systems, helping to build consumer trust.
Moreover, the cost-effectiveness of the solution enhances its commercial appeal. Unlike thermal imaging or multi-sensor fusion systems, this approach relies on a simple printed circuit board (PCB) coil and standard electronic components. It can be easily integrated into existing wireless charging pads without major redesigns or expensive materials. This scalability is essential for mass deployment in public charging networks, parking garages, and residential installations.
The research also opens doors for future enhancements. With the foundational model in place, further refinements could include adaptive frequency tuning, machine learning-based signal classification, or integration with vehicle-to-infrastructure (V2X) communication systems. For instance, the detection system could send alerts directly to the driver’s smartphone or to a central fleet management platform, enabling proactive maintenance and safety monitoring.
In the broader context of EV infrastructure development, advancements like this underscore the importance of interdisciplinary collaboration. The team’s work sits at the intersection of electromagnetics, power electronics, materials science, and system engineering—fields that must converge to solve the complex challenges of next-generation mobility. As wireless charging moves from niche applications to mainstream adoption, such holistic innovations will be critical to ensuring both performance and safety.
The implications extend beyond passenger vehicles. Autonomous delivery robots, electric buses, and industrial automated guided vehicles (AGVs) all stand to benefit from reliable, interference-resistant FOD systems.In environments where unattended operation is the norm, the ability to autonomously detect and respond to safety hazards without human intervention becomes paramount. This optimized detection coil represents a step toward fully autonomous and resilient charging ecosystems.
In conclusion, the research conducted by Zhang Bin, Zhu Chong, and Zhang Xi at Shanghai Jiao Tong University represents a significant milestone in the evolution of EV wireless charging technology. By developing a theoretically grounded, experimentally validated method for optimizing detection coil sensitivity, they have addressed a critical safety bottleneck. Their work not only improves detection accuracy and noise immunity but also provides a scalable, cost-effective solution ready for real-world deployment. As the automotive industry accelerates toward a wireless future, innovations like this will play a vital role in making that future safer, smarter, and more accessible for all.
Zhang Bin, Zhu Chong, Zhang Xi, Shanghai Jiao Tong University, Journal of Power Supply, DOI: 10.13234/j.issn.2095-2805.2024.4.209