As demand for power density grows across data centres, EV charging, and renewable energy systems, engineers are turning to GaN, advanced topologies, and digital control. In an interview, Brian King from Texas Instruments spoke to Nidhi Agarwal from Electronics For You about the technologies shaping next-generation power electronics.
Q. How is GaN changing modern power-system design compared to traditional silicon devices?
A. GaN (Gallium Nitride) is changing power-system design by enabling higher power density and efficiency. Its fast switching speeds and lower power losses allow power converters to operate at higher frequencies, which reduces the size of magnetics and other power-supply components. This makes it possible to deliver the same power in a smaller form factor or increase power output without increasing system size. As a result, GaN is being adopted across applications such as consumer electronics, automotive, industrial systems, and especially data centres, where growing power demands make higher power density increasingly important.
Q. Which power conversion applications currently benefit most from GaN technology, and why?
A. Applications that require high power density benefit the most from GaN technology, particularly data centre power systems. As demand for computing continues to grow, operators need to deliver more power within the same physical space without reducing the area available for computing equipment. GaN devices enable higher switching frequencies and support advanced power conversion topologies, making it possible to increase power density and overall power capacity while maintaining the same system volume. This is why GaN and other wide-bandgap technologies are becoming increasingly important in data centre power infrastructure.
Q. How do you see GaN technology transforming applications such as power conversion and renewable energy?
A. Renewable energy is one of the key areas where GaN technology can drive innovation. As renewable generation continues to grow, the challenge is no longer limited to producing energy but also to storing and managing it efficiently. Energy generated from sources such as solar during the day must be stored in battery banks and energy storage systems and then supplied back to the grid during periods of peak demand, such as in the evening. This requires new power conversion technologies and topologies that can handle bidirectional power flow, enabling both battery charging and the transfer of stored energy back to the grid. GaN technology makes this process more efficient while also reducing system size, and when combined with advanced power conversion topologies, it enables the development of new solutions for energy storage and grid infrastructure in the rapidly expanding renewable energy market.
Q. How is GaN technology influencing the development of fast-charging systems for electric vehicles?
A. GaN technology is helping improve fast-charging systems by enabling higher power density and greater efficiency, particularly in high-voltage power conversion stages. Similar to its role in solar inverters, GaN allows designers to build smaller and more efficient power electronics while handling high power levels. Many of the same power-conversion topologies used in solar energy systems, such as dual-active bridge converters, are also being adopted in EV charging infrastructure. By combining these advanced topologies with GaN devices, manufacturers can achieve better efficiency, reduce system size, and support faster charging capabilities in electric vehicle charging systems.
Q. What testing and validation methods do you recommend for GaN-based power systems?
A. At the system level, testing and validation for GaN-based power systems are largely the same as for silicon-based solutions. Standard performance, reliability, and system-level tests still apply. The main difference is within the power supply itself, where GaN devices switch much faster. Engineers should pay close attention to parameters such as edge rates and slew rates to ensure the fast switching does not create issues for other components in the circuit, such as increased noise or electromagnetic interference. Beyond these considerations, overall system-level testing remains similar to that used for conventional power solutions.
Q. How does GaN influence magnetic component sizing?
A. GaN plays an important role in reducing the size of magnetic components such as transformers and inductors. The size of these components is closely related to the switching frequency of the power supply. To make magnetics smaller, designers need to increase the switching frequency. However, higher switching frequencies also increase switching losses in the power switching devices. This is where GaN devices provide an advantage. Their lower switching losses allow systems to operate at higher frequencies without excessive heat generation, enabling smaller transformers and inductors while maintaining efficiency.
Q. Which new power technologies will have the biggest impact on future power system design?
A. The biggest impact is coming from advances in power conversion topologies, particularly architectures such as the dual active bridge (DAB). These newer topologies are gaining significant attention because they can improve efficiency and performance in a wide range of power applications. However, controlling them is often more complex than with traditional designs, and in many cases suitable analogue controllers do not exist. As a result, digital control has become increasingly important for managing these advanced power architectures.
To support these designs, it is not enough to provide power devices alone. A complete solution requires digital controllers, sensing technologies, and integrated system-level components that work together. Another challenge is helping engineers implement these new topologies quickly and effectively. This is where reference designs, application collateral, and training programs play a key role.
TI provides validated system-level solutions and educational resources to help engineers overcome design challenges, reduce development time, and more easily adopt the latest innovations in power electronics.
Q. Why is the DAB topology gaining adoption in modern power-conversion and high-power battery-charging systems?
A. The DAB topology is gaining adoption because it enables efficient bidirectional power flow while providing galvanic isolation, making it well suited for applications such as EV charging and energy storage systems. Its power stage is relatively simple, consisting of MOSFETs, an inductor, and a transformer, while offering flexibility for control and optimisation. When properly controlled, DAB converters can achieve high efficiency and power density while transferring energy in either direction. Since few topologies can efficiently provide both bidirectional power transfer and isolation, DAB has become a preferred choice for modern high-power energy systems.
Q. What opportunities exist for combining GaN devices with DAB architecture?
A. GaN devices are well suited for DAB topologies, particularly in high-power applications. As power levels increase to 3kW, 6kW, and beyond, such as in energy storage systems and grid-tied inverters, the lower on-resistance of GaN devices becomes a significant advantage. For a given package size, GaN enables higher power density and more efficient power delivery, making it easier to achieve higher output power while maintaining compact system designs.
Q. How does device selection impact DAB performance at different power levels?
A. Device selection plays a critical role in DAB performance because the converter must be designed to handle its highest intended power level. As power levels increase, lower on-resistance MOSFETs are required to minimise conduction losses and maintain efficiency. The choice of devices ultimately influences the overall converter design, efficiency, and thermal performance. TI’s broad portfolio of GaN devices provides designers with the flexibility to select and scale the most suitable solution for a wide range of power levels, helping optimise performance across different applications.
Q. What design trade-offs must engineers consider when optimising a DAB converter?
A. When optimising a DAB converter, engineers must first define their optimisation goals, as the converter offers multiple control variables through digital control. These include different phase-shift strategies and switching frequency adjustments, giving designers several degrees of freedom to fine-tune performance. The key trade-off is deciding which parameters to optimise and how many control variables to use. In most applications, efficiency is the primary objective, which requires maintaining zero-voltage switching (ZVS) across a wide range of input voltages, output voltages, and load conditions. Engineers must therefore balance control complexity, operating range, and efficiency requirements when developing the control strategy. Ultimately, the chosen optimisation approach is implemented through the control algorithm and deployed on a digital controller such as a TI C2000 real-time microcontroller.
Q. How does TI implement digital control in high-energy charging architectures?
A. Digital control in high-energy charging architectures is implemented using TI’s C2000 microcontrollers, which serve as the main controller running the control algorithms, supported by peripheral circuitry that enables complete system operation. This includes isolated sensing solutions such as isolated amplifiers for voltage sensing and Hall-effect sensors for current sensing, which provide accurate feedback to the controller so it can regulate the power stage. On the output side, gate drivers and isolation components are used to drive high-side FETs or devices across isolation boundaries. TI also provides power devices such as GaN FETs, which, together with the controller, sensing, and drivers, complete the full system-level solution required for architectures like the dual active bridge converter.
Q. How is TI approaching the future of power design?
A. We are approaching the future of power design through our broad portfolio of analogue and embedded processing semiconductors. While GaN is an important focus area, we also provide the other building blocks needed for power systems, including sensing, controllers, digital control, and power management components. Beyond individual products, we help engineers develop complete system-level solutions through our reference designs. These designs address the latest industry trends and support a wide range of applications across data centres, energy infrastructure, automotive, and many other markets, enabling customers to bring power-efficient systems to market faster.
Q. What major trends do you think will shape the future of power electronics in the coming years?
A. One of the biggest trends shaping the future of power electronics is the development of new power-conversion topologies. As power levels continue to increase and these topologies become more advanced, they will require corresponding advances in the technologies that support them. For example, topologies such as the dual-active bridge depend on digital control systems for efficient operation. This is where technologies like TI’s C2000 microcontrollers play an important role, enabling precise control and helping engineers implement increasingly sophisticated power-electronics designs.
Q. AI is being adopted across many industries. How do you see AI converging with energy systems and power electronics in the future?
A. I’m not an AI expert, so my perspective is limited. However, AI is clearly becoming a major driver of innovation across industries. In the context of energy systems and power electronics, I think AI could help speed up development processes, particularly in areas such as software and control-code development for advanced power conversion topologies. It may also simplify certain design and optimisation tasks, reducing development time.
For example, Edge AI can enable real-time solar arc-fault detection by using an MCU to locally analyse DC current waveforms and identify subtle fault patterns. This enables faster, more accurate detection than traditional threshold-based methods, improving system safety and preventing potential equipment damage. That said, these are predictions, and it is difficult to know exactly how the technology will evolve in the future.
