Vlad Marsic

Magnetic scanning non-destructive testing (NDT) methods are a real interest for many research and industrial areas. However, the already well-established popularized instruments and techniques for this area require complex, heavy and expensive equipment supported by specialist software and expertise. Consequently, building a system to test a hypothesis or enhance a current toolset may be discouraging for any adjacent research and industry areas with interest in magnetic NDT. This work demonstrates that a cost effective NDT reduced size magnetic scanner is possible to be created and exploited with a basic knowledge of electronics, programming and signal processing. The system is using the magnetic flux leakage (MFL) technique by means of Hall effect sensors and permanent magnets, at room temperature. The approx. 50 × 50 mm steel sheet scanning exemplifies two usage cases for the magnetic system: determining the sheet’s thickness variation from 0.7 mm to 3.5 mm and, the possibility of detecting holes as defects on a separate 3 mm steel sheet (i.e., from 2 mm to 5 mm radius). The system is non-motorized for simplicity and it is interfaced with MATLAB to illustrate data acquisition, visualization and analysis, all on the same software platform. Since the magnetic field analysis methodology used for both show-cases differs significantly, the parallel for the two investigations reveals the common and unique points of each. This approach is set to unfold the work set for a straightforward magnetic field investigation and set the expectations on the right path when using off-the-shelf inexpensive components.

Gallium nitride (GaN) transistors are a highly promising technology, with applications in high-frequency, high-temperature, and high-efficiency power systems. However, they are susceptible to various electromagnetic hazards. As a result, in-situ state-of-health (SOH) active monitoring of the device during operation is necessary, with various efforts focusing on non-invasive approaches. This study presents the design and testing of a complete, in-situ, non-invasive monitoring system powered by energy harvested from the GaN transistor’s power line (PL). Experimental results show that an external low-power load of approx. 3.6 W, such as a standard server fan, is sufficient to generate approximately 16 μW and 64 μW (i.e., for two different GaN transistors) of radiated power at the transistor’s drain connection with the PL. This emitted energy, harvested using a 100 mH inductor, is capable of powering a remote SOH monitoring system consisting of a Microchip PIC24F 16-bit microcontroller, three MCP9700 temperature sensors, and an MRF89XA 868 MHz transceiver. The resulting system is able to collect sensor data and, after a brief energy accumulation period (the system’s warm-start charging cycle), wirelessly transmit it to a base station. To further illustrate the practicality of the proposed SOH monitoring solution, the two GaN transistors were intentionally driven into a thermal runaway regime while operating in low-power mode on a 12 W load. Under these conditions, the harvested energy/emmited radiation increased to approximately 620 μW and 3,216 μW, representing about 38 and 50 times the values observed in the previous low-power demonstration, respectively.

Gallium nitride high-electron-mobility transistors (GaN-HEMTs) enable fast switching commutation due to their small parasitic capacitances, which necessitates the effective gate driving techniques. Gate drivers with split-output topology offer advantages such as reducing the risk of false turn-ON and minimizing the size of gate loop. However, the output capacitance of gate driver in this configuration may reduce the commutation speed of GaN-HEMTs, and values of this capacitance are overlooked in the gate driver datasheets. This paper highlights the critical role of gate driver output capacitance on the switching performance of GaN-HEMTs, which has been discussed by proposing an analytical model and demonstrated through both simulations and experiments. The results show that this capacitance can slow down the switching speed specifically in the split-output gate driver configurations, leading to inaccurate switching waveform predictions. Furthermore, a measurement method and corresponding output capacitance data for a series of commercially available gate drivers are presented. These findings offer practical guidance for gate driver selection in GaN-HEMT applications.

Non-destructive testing (NDT) is a highly regarded area on the medical, space and defense sectors but, also on heavy equipment susceptible to corrosion and hidden structural defects. One of the mature NDT methods for steel structures is magnetic flux leakage (MFL) detection. The current technique implementations require expensive equipment, specialist knowledge and custom data analysis tools whereas, the method uses large strong magnets on the same plane with the sensor while rolling on the inspected area, exerting a high pressure due to its magnets’ attraction and heavy body system, making it suitable only for a reduced number of investigation scenarios. This work experimentally investigates a complementary setup to classical MFL approach, useful for detection of a thin steel thickness variations from 0.7 mm up to 3.5 mm, while using a light low cost Hall sensor array setup built with accessible off-the-shelf components. Instead of the horseshoe magnet moved simultaneously with the detectors, this work uses small strong magnets tiled together and static as referred to the sensor array’s scanning movement. In this way, analyzing the sensor behavior at different spatial points in the complex magnetic field resulted from the low cost setup, this study delivers high valuable insights for MFL research.

The GaN Hall sensors are the latest advancements in non-intrusive active monitoring, designed to be integrated on GaN power transistors to enable in-situ condition monitoring, hence boosting device reliability and life. This paper describes step-by-step the basic procedures required for the fabrication of a three-terminal Hall-effect sensor on GaN, challenges and a robust testing method that can deliver consistent results when testing under diverse laboratory external hazards. This work is presenting the first three terminal current sensing GaN Hall sensor reported in literature, which have been tested at room temperature under various magnetic field intensities and polarities in order to determine its sensitivity. Since even when properly using the most suitable scientific equipment, errors may occur from insufficient information regarding the relative angle between the magnetic field and the active area of the sensor or, data records affected by imperceptible equipment vibrations, this work provides a practical solution on how to detect and mitigate these shortcomings: magnetic quasi-static sources alternatives are discussed and transient recordings of different intensities and polarities of the magnetic field are separated and interpreted similar with the spectrometry diagrams. This comprehensive presented study delivers a consistent testing framework for the new emerging generation of magnetic sensors.

Schottky-type p-GaN gate Gallium Nitride High Electron Mobility Transistors (GaN-HEMTs) suffer from threshold voltage ( ${V_{\mathsf {th}}}$ ) instability phenomenon. Both positive and negative ${V_{\mathsf {th}}}$ shifts are reported when device undertakes the voltage bias, but the impact of this ${V_{\mathsf {th}}}$ instability phenomenon on device switching behaviors is less investigated. In this study, the drain-source voltage ( ${V_{\mathsf {ds}}}$ ) induced bidirectional ${V_{\mathsf {th}}}$ shift in hard-switching condition is characterized and decoupled by an H-bridge based double-pulse test (DPT). Subsequently, the influence of ${V_{\mathsf {th}}}$ shift on switching behaviors is theoretically analyzed and demonstrated through SPICE simulation and experiment, showing how a positive shifted ${V_{\mathsf {th}}}$ can reduce the device turn-on commutation speed and increase the switching losses, and vice versa. The results suggest that the ${V_{\mathsf {th}}}$ instability phenomenon should be considered in accurate switching modeling.

Wide bandgap Gallium Nitride (GaN) technology promises to deliver the next generation of power transistors capable of high energy density and compact design integration however, without active monitoring high failing rates are recorded due to its instability to design parameter variations. Moreover, the electromagnetic (EM) radiofrequency (RF) emissions due to GaN power switching require extra design resources. Considering the extensive research area dedicated to galvanic isolated magnetic sensors for GaN wafer monolithic integration with usage in power monitoring, this study investigates the conditions that a Hall sensor is required to meet when operating in close proximity of a GaN transistor. Through considerable experimental testing, it was determined that the sensor requires a magnetic field starting from ±1 mT when interfaced with a microcontroller. Additionally, since the GaN transistor’s EM RF switching noise was one of the most monitored parameters during the experiments, it was discovered that it is proportional to the transistor’s current transfer area whereas its magnitude is due to electrical current required by the load. As a result of these findings, the EM radiated switching noise may apply to all electrical switches and provide a significant advantage when designing for EM compatibility (EMC).

Gallium Nitride (GaN) is an emerging wide bandgap technology in power electronics. It enables higher power density, wider operating temperature range and higher efficiency compared to other widely used technologies, allowing miniaturization and overall weight and cost reductions at system level. In contrast, the technology produces high levels of radio emissions and it is prone to failure without an adequate active power monitoring due to its sensitivity to unmatched electrical parameters. Since intrusive monitoring can disrupt the transistor's efficiency on power delivery, galvanic isolated sensors are intensely explored with the goal of chip-level integration on the difficult-to-manufacture GaN structures. The experimental results in this study reveal that in a power monitoring scenario envisaged for the integration on a GaN transistor chip, a Hall sensor that requires a minimum magnetic field threshold of 1 mT will be sufficient. The noisy GaN transistor environment is tested for three different low power loads, resistive, semi-conductive, and inductive, at 1, 5 and 10 kHz, demonstrating that the radiated emissions relate to the load current rather than voltage, whereas the Hall device measurements show stability and consistency.

This work investigates the power GaN-HEMTs switching behaviour differences resulted from usage of two gate driving configurations: single and split outputs. The analysis based on simulation and experimental results show that GaN-HEMTs could switch slower and cause higher switching losses when the split output configuration is used. This is because the output capacitance (C oss ) of MOSFETs inside gate driver will be charged during the turn-on process of GaN-HEMTs, and this charging process can reduce the charging speed of input capacitance (C iss ) of GaN-HEMTs. Moreover, the gate resistance and parasitic inductance are the main parameters selected for analysis, and their distribution can amplify this effect by increasing the impedance ratio of turn-on and turn-off loop. This research provides guiding suggestions for gate driver and high-efficiency GaN-HEMTs power module design.