Semiconductors
and gate-drive design. Reduced losses ease thermal management requirements, while silicon enables fabrication in high-volume CMOS production facilities. For designers, this combination preserves familiar gate drive techniques, established reliability and qualification standards, and predictable device availability. These are important considerations when scaling designs from prototype to high-volume production. By reconfiguring how charge balance and conduction area are implemented within a RESURF device, the SuperQ architecture extends the performance envelope of silicon MOSFETs. For designers, this provides an additional option for achieving higher efficiency and power density without the cost, complexity, or risk typically associated with transitioning to alternative semiconductor materials.
Key applications
High-efficiency data centre power conversion
As we become increasingly reliant on the digital world, including the recent rapid growth of AI, data centres are under scrutiny for the energy they consume. In fact, they already account for 2% of global electricity consumption (1)
, and the IEA forecasts that
total consumption will exceed 1 petawatt annually by 2026. The US is projected to exceed this level on its own by 2030, driven by growth in AI, hyperscale infrastructure, high-performance computing (HPC), and other emerging applications.
Clearly, as power consumption rises, the importance of efficiency increases. This is driven in part by sustainability concerns and by the fact that in some places, energy generation is struggling to keep pace with the ever-increasing demands of data centres. To address this, governments and other standards bodies are developing a range of programs to improve the performance of power conversion solutions,
Figure 4: Motor drives see significant performance gains when SuperQ is deployed
including initiatives such as 80 PLUS and ENERGY STAR. These reflect the dynamic loads common in data centres and ensure efficiency is optimized at all operating power levels, including standby power.
The rapidly growing need for GPU accelerators in AI applications is driving data centre power levels beyond 100kW per rack. To limit current and distribution losses, many architectures use higher-voltage inputs – such as 400V or 800V – which are stepped down to an intermediate bus, typically around 48–54 V, for distribution within the rack. Resonant topologies, including phase-shifted full bridge (PSFB) and LLC converters, are commonly used in these applications because they reduce switching losses and enable high efficiency at elevated power levels. In these topologies, conduction losses often dominate, particularly at high duty cycles and heavy load conditions. SuperQ MOSFETs address this challenge by providing very low specific on-resistance at the voltage ratings commonly used on
Figure 3: SuperQ can reduce the number of parallel MOSFETs in LLC designs
the secondary side of the high voltage to low voltage PSU. The increased conduction area enabled by the asymmetrical RESURF architecture allows designers to reduce conduction losses without increasing device count. In practical designs, such as that shown in Figure 3, this can translate into fewer parallel MOSFETs in the synchronous rectifier stages. Lower device losses also reduce thermal stress and simplify heatsinking, which is especially important in high-density server power supplies. For designers optimizing efficiency to meet regulatory and operational targets, these characteristics make SuperQ devices well suited to resonant data centre power-conversion stages, where both conduction and switching losses must be carefully balanced.
Motor drives and battery-powered systems
Electric motors are another major consumer of electrical energy as they become more prevalent across a wide range of applications. Again, efficiency is key to reducing energy use and delivering more compact solutions that run cooler.
In a typical motor drive arrangement, shown in Figure 4, MOSFETs are used not only in the inverter stage that drives the motor phases but also in DC-link switching, protection circuitry, and battery management systems (BMS) in battery-powered platforms. SuperQ devices offer low conduction ), high short circuit withstand
losses (RDS(ON)
current, and a wide safe operating area (SOA), all of which are beneficial for BMS and DC switches.
Switching losses and reverse recovery behaviour also significantly affect motor drive efficiency, particularly in pulse-width- modulated inverters operating at higher switching frequencies. The reduced charge storage in the SuperQ architecture helps limit dynamic losses and switching stress, thereby improving efficiency and thermal performance.
References: 1
Reliability is a primary concern in motor drive applications, where devices must withstand repeated thermal cycling, transient overloads, and occasional faults. Because SuperQ devices are implemented in silicon, they retain the established gate- drive techniques, reliability characteristics, qualification standards, and robustness that designers expect, while delivering improved performance relative to conventional silicon MOSFETs.
Conclusion
For more than two decades, power designers have operated within the established trade-offs of silicon SuperJunction devices – balancing efficiency, robustness, and cost while approaching the practical limits of symmetrical RESURF architectures. In applications requiring higher performance, wide-bandgap devices offered an alternative, but often at the expense of design complexity, ruggedness, availability, or overall system cost. Asymmetrical RESURF architectures, such as SuperQ, introduce a new option in silicon by significantly increasing the conduction area while maintaining high-voltage capability. The resulting reductions in specific on-resistance and switching losses translate directly into practical design benefits, including fewer parallel devices, simplified thermal management, and improved efficiency across a wide range of operating conditions. For designers evaluating next-generation power systems, whether in data centre power conversion, motor drives, or other high- efficiency applications, these advances enable a reassessment of silicon’s role. Rather than a choice between mature silicon and higher-risk alternatives, new silicon architectures offer a way to achieve meaningful performance gains while retaining the proven reliability, scalability, and manufacturing advantages that silicon continues to provide.
idealsemi.com
https://iea.blob.core.windows.net/assets/6b2fd954-2017-408e-bf08-952fdd62118a/Electricity2024- Analysisandforecastto2026.pdf
www.cieonline.co.uk Components in Electronics April 2026 31
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44
