By Maurizio Di Paolo Emilio, Marcom Director, EPC

Gallium nitride (GaN) power technology is entering a new phase. After gaining traction in fast chargers, automotive power conversion, and other higher-voltage applications, GaN devices rated at 40 V and below are opening a new competitive front against silicon MOSFETs. According to Paul Pickering, Research Director at Omdia, low-voltage GaN could become one of the most disruptive developments in power semiconductors over the next several years.

By Maurizio Di Paolo Emilio, Marcom Director, and Marco Palma, Vice President of Motor Drive Marketing and System Engineering at Efficient Power Conversion, EPC Rapid growth in industrial automation, collaborative robots, autonomous mobile robots (AMRs), drones, and precision servo systems is driving the need for a new generation of low-voltage motor drives to deliver higher efficiency, greater power density, and improved dynamic performance. In the past, silicon MOSFETs were the most common switching technology for this voltage range, but nowadays, gallium nitride (GaN) power devi

In this interview on EEPower, we learn about EPC’s Gen 8 GaN platform for AI power delivery and Trinity motor control—on display this week at PCIM Asia—to address rising power and space demands in AI and robotics.

As power demands rise across artificial intelligence, robotics, and autonomous machines, gallium nitride (GaN) is moving into applications that were once dominated by silicon. At PCIM Asia in Shenzhen, Efficient Power Conversion (EPC) is showcasing two technologies aimed at this transition: its next-generation Gen 8 GaN platform for AI power delivery and Trinity, an integrated three-phase GaN solution for motor control.

The EPC91132 reference design integrates GaN power stage, sensing, protection, and feedback for robotic and drone motor drives.

The next-generation humanoid robots and drones need motor drives to move closer to the actuator and share space with mechanical structures and sensors. They still have to deliver high current, fast control response, and reliable protection.

EPC91132 addresses this integration problem with a compact three-phase BLDC motor-drive inverter reference design based on the EPC33110 GaN three-phase module. Its core inverter can be reduced to a 23-mm diameter form factor. The company reports 11 A_RMS continuous current per phase for humanoid robotic joint applications at 48 V and switching frequencies up to 100 kHz.

The rapid increase in AI computing performance is forcing a fundamental redesign of data center power architectures, with gallium nitride (GaN) emerging as a key technology for meeting the industry’s escalating efficiency and power density requirements.

Speaking during the EE Power Asia 2026 Spark Session, Dr. Alex Lidow, CEO of Efficient Power Conversion (EPC), and Jason Zhang, VP of DC/DC Marketing and System Engineering at EPC, outlined why conventional silicon power devices are approaching their practical limits as GPU power consumption continues to climb.

According to Zhang, AI accelerators have experienced a fourfold increase in power consumption across three GPU generations. Within the next two years, individual GPUs are expected to dissipate as much as 5kW while operating at core voltages below 1V, requiring current levels approaching 10,000A.

Maurizio Di Paolo Emilio, Marcom Director, EPC

Rapid growth of artificial intelligence is forcing a rethink of data-centre power architectures. Legacy server designs were never intended to support the extreme power densities of modern AI workloads. These facilities are increasingly becoming what we might call “AI factories” where the primary goal is to optimize compute capability in a constrained physical footprint.

A key enabler for this transition is the use of 800 VDC distribution. The high voltage approach enables power scaling at the rack level from conventional levels to the megawatt range, which is essential for next generation accelerator platforms. 800 V direct power delivery to compute subsystems introduces new challenges in insulation coordination and device stress, but it also presents an opportunity to rethink how isolation is implemented. Here, Input-Series Output-Parallel (ISOP) converter topologies are becoming popular because of their modularity, scalability and efficiency.

By Jason Zhang, Vice President of DC-DC Marketing and System Engineering, EPC

Power delivery has become a critical design challenge in AI data centers. As GPU power requirements continue to increase, designers face growing demands to improve power conversion efficiency while reducing converter size and thermal losses. These requirements are driving the adoption of wide-bandgap semiconductor technologies, particularly gallium nitride (GaN), in the intermediate bus conversion stages of modern AI power architectures.

In this context, Efficient Power Conversion (EPC) has introduced the EPC2378, a 25 V GaN FET, and the EPC2370, an 18 V GaN FET. Both are part of the company's latest Gen 7 product family, optimized for secondary-side synchronous rectification in high-current DC-DC converters. The EPC2378 features a typical R_DS(on) of 410 µΩ and a gate charge of 34 nC, while the EPC2370 offers a typical R_DS(on) of 280 µΩ and a gate charge of 26 nC. Both devices are housed in a compact, thermally enhanced 3.3 × 3.3 mm PQFN package with an exposed substrate that supports dual-sided cooling. In addition to minimizing conduction losses, the devices are designed with low gate charge to reduce switching losses at a switching frequency of 1 MHz, improving overall efficiency and power density.

By Alex Lidow, CEO, Efficient Power Conversion (EPC) Alejandro Pozo, Director of Applications Engineering, Efficient Power Conversion (EPC) and Michael De Rooij, GaN Applications Fellow, Efficient Power Conversion (EPC)

Artificial intelligence is integrating into our professional and personal lives at an unprecedented pace. By streamlining software development, tackling complex analytical challenges, and automating routine documentation, AI is no longer just a tool, it’s a fundamental shift in how we create and solve.

These new-generation AI servers come at a cost based in large part on an exponentially increasing demand for power.  The center of this power consumption is the Graphic Processing Unit (GPU) made from billions of microscopic transistors compacted on a silicon substrate with features as small as 20 Å (That’s approximately the width of the DNA molecule’s double helix). The power required to activate these billions of processing agents is going up as fast as their computation capabilities.  In table 1 is an example that uses Nvidia’s published roadmap through 2028.

EPC is featured on the cover of Semiconductor Network Korea, highlighting its latest GaN innovations for motor drives in humanoid robots and autonomous drones.

As robotics and UAV applications continue to evolve, designers are seeking solutions that deliver higher efficiency, increased power density, and reduced system size and weight. EPC's enhancement-mode GaN FETs address these challenges by enabling faster switching, lower power losses, and superior dynamic performance compared to traditional silicon solutions.

Technology pioneer, geopolitical asset and strategic scaling candidate — refreshed after primary research with EPC CEO Alex Lidow at PCIM Europe 2026

Efficient Power Conversion Corporation (EPC) is one of the companies that turned enhancement-mode gallium nitride (eGaN) from an engineering proposition into a commercial power-device platform. Founded in 2007 by Alex Lidow — co-inventor of the silicon power MOSFET and former CEO of International Rectifier — EPC shipped the first commercial eGaN transistors in 2009 and has since built a focused portfolio of GaN power devices, integrated circuits, automotive-qualified parts and radiation-hardened products for space and defense.

This Version 3.0 SGI™ Research Report incorporates new GeoTechNexus primary research from PCIM Europe 2026 in Nuremberg, where CEO Alex Lidow disclosed that he and a co-investor have invested more than $250 million into EPC, and confirmed the company believes it remains roughly two GaN generations ahead of competitors.

Our next article in this issue is a review from our experienced audio measurements enthusiast, Stuart Yaniger, and explores the EPC9192 reference design from Efficient Power Conversion (EPC) for the latest EPC2307 eGaN FETs. As his title says, this is a new GaN technology option, encouraging developers to design high-performance Class-D audio amplifiers. EPC has launched the new EPC9192 reference design, enabling high power and high efficiency in a modular design, and the audioXpress review confirms the performance capabilities of the brand’s 200V, EPC2307 eGaN FETs in a ground-referenced, split dual supply single-ended (SE) design, rated to deliver 700W per channel into a 4Ω load. The article also explores why EPC GaN power devices are differentiated through the use of native enhancement-mode lateral GaN-on-silicon High Electron Mobility Transistors (HEMTs), eliminating the need for cascode architectures and the associated RDS(on) penalty at lower voltages. This makes EPC devices particularly competitive in the 15V to 200V range.

Powering the AI Factory: How EPC GaN Solutions Supercharge the NVIDIA MGX Architecture

The digital world is entering a new AI industrial revolution, with data centers transforming into AI factories that generate intelligence at massive scale. AI is no longer just a software story; it is rapidly becoming an infrastructure story as well. Modern workloads are shifting from simple human‑to‑AI interactions toward AI‑to‑AI collaboration, where agentic models coordinate tasks, reason autonomously, and work across extremely long token sequences. This puts new pressure on infrastructure: not only do systems need more raw compute, they must also meet strict requirements around latency, thermal management, and energy efficiency. To understand how AI factories are being built in practice, NVIDIA MGX™ provides the modular foundation for scalable and flexible accelerated computing infrastructure. While MGX addresses infrastructure modularity and faster deployment, a more serious bottleneck is brewing elsewhere: power delivery. The efficiency of power conversion is critical to maintain performance and efficiency as AI systems become more complex and denser. A key enabling technology is Efficient Power Conversion’s (EPC) gallium nitride (eGaN®) solutions, which offers the efficiency, power density, and thermal performance required for next-generation AI infrastructure.

By Maurizio Di Paolo Emilio

Satellite payload architectures are undergoing a transition from fixed-function processing chains toward flexible computing infrastructures capable of supporting dynamic workloads directly in orbit. This shift reflects a broader transformation in space systems engineering, where spacecraft are increasingly expected to perform signal processing, sensor fusion, anomaly detection, and artificial intelligence inference without relying exclusively on ground-based processing resources.

Experts from EPC and Texas Instruments explain why GaN dominates humanoid robot motor drive design at the joint level.

A humanoid robot requires approximately 40 to 80 motors to drive its limbs and torso, with each hand containing more than a dozen additional motors to replicate dexterous manipulation. The high density of independent actuators creates a complex power-electronics integration challenge that must be packaged within human-dimensional constraints.

At Efficient Power Conversion, we’re pioneering new frontiers in power electronics with eGaN technology that allows for smaller, faster, more efficient systems. From robotics and drones to AI and Space, EPC helps engineers simplify design and bring next-generation innovations to life.

By Marco Palma, Director, Motor Drives Systems and Applications, and Maurizio Di Paolo Emilio, Marcom Director, Efficient Power Conversion (EPC)

Gallium nitride (GaN) power devices are enabling a new generation of high-efficiency, high-power-density motor drive systems. Compared with conventional silicon MOSFETs, GaN transistors offer significantly lower gate charge, reduced output capacitance, and very low on-resistance, allowing power converters to operate at much higher switching speeds. As a result, motor inverters based on GaN technology can achieve switching frequencies well above 100 kHz while reducing both conduction and switching losses. These characteristics enable smaller passive components, improved efficiency, and more compact system designs.

Bodo’s Power Systems – May Issue

Read More…

Compared to existing silicon MOSFETs, GaN offers faster switching speeds and higher efficiency. Demand is rapidly increasing across a wide range of applications, including AI server power supplies, adapters, UPS systems, and EV chargers. In particular, in 48V-based power architectures, applying GaN can reduce power losses by more than 20%, making it a key technology in data centers and high-performance computing environments. Amid these trends, the power semiconductor market is expected to be reorganized around GaN, and the existing MOSFET and driver IC markets - currently worth over $20 billion - are also undergoing rapid changes. Companies like EPC are accelerating the transition to GaN ICs, while continuous technological advancements are driving improvements in system efficiency, size reduction, and reliability. In Korea as well, demand for GaN solutions is steadily increasing, particularly in applications such as EVs, renewable energy, and battery systems. GaN is gaining attention as a core technology that enhances power efficiency while enabling more compact and lightweight designs.

As AI infrastructure migrates to 800 V distribution and humanoid robots embed power electronics directly within joints, Generation 7 GaN and integrated GaN ICs are enabling higher efficiency, faster switching and dramatically improved power density in the critical 15 V to 40 V range — reshaping how hardware development teams approach next-generation conversion architectures.

As artificial intelligence infrastructure scales toward megawatt-class racks and 800 V distribution architectures, and humanoid robotics pushes power electronics directly into joints and actuators, the demands placed on conversion efficiency, density and dynamic performance have fundamentally shifted. Traditional silicon MOSFET-based designs are increasingly constrained by switching losses, thermal limits and physical footprint — particularly in the 15 V to 40 V range that underpins motor drives, point-of-load converters and distributed robotics power stages.

By Maurizio Di Paolo Emilio, Marcom Director, EPC – Efficient Power Conversion

The design requirements for motor drive electronics are being significantly altered by the continuous shift from conventional automation to mobile robotics. Volume, mass, efficiency, acoustic emission, and dynamic response are all simultaneously constrained by actuators integrated into wearable technology, humanoid robots, and aerial platforms. In these applications, the inverter becomes a tightly coupled component of the electromechanical structure rather than a peripheral power stage.

In this video from APEC, Alejandro Pozo, Director of Applications Engineering at EPC, presents the company’s latest low-voltage eGaN FET solutions for high-current buck converters. He walks through several evaluation boards featuring 40 V, 25 V, and 15 V devices delivering up to 50 A with impressive efficiency and very low thermal rise, even without heatsinks. Alejandro also highlights the compact EPC90175 half-bridge board and its integration with standard controllers and measurement hardware used for accurate testing in demanding applications..