The GaNext project, funded under the German Federal Ministry of Education and Research with the grant number 16EMO080 and managed by VDI/VDE‑IT, ran from 1 May 2020 to 30 April 2023. It brought together thirteen partners, including Sumida Components & Modules GmbH, Signify, Neways, Lyra, and the Technical University of Eindhoven, to develop a next‑generation GaN‑based intelligent power module (IPM). The central aim was to eliminate the barriers to GaN adoption by integrating driver, control, and protection circuitry into a compact module that also incorporates optimised magnetic components. The resulting IPM measures 35 × 35 × 4 mm and has been demonstrated in electric‑vehicle chargers, lighting, motor drives, and photovoltaic inverters, delivering higher efficiency and smaller form factors than silicon‑based systems.
Technically, the project focused on high‑frequency magnetic materials and inductive component design. A key development was the ferrite material Fi 337, a Mn‑Zn ferrite engineered for operation between 0.5 and 2.5 MHz. At 25 °C it exhibits a relative permeability µi of 470, while at 100 °C the permeability remains above 410. Loss measurements at 1.5 MHz and 25 mT show a power density of 1500 kW/m³, confirming its suitability for high‑frequency power electronics. Complementary ferrite grades Fi 360, Fi 320, and Fi 301 were characterised for loss under direct‑current bias, a condition rarely reported in datasheets. Using a Keysight 4294A impedance analyser, the real and imaginary parts of the complex permeability were mapped, revealing that the chosen winding insulation material had negligible impact on loss and that core size variations significantly influence damping behaviour. Ring‑core and frame‑core geometries of differing volumes were tested, and the impedance response as a function of core length was quantified, providing a data set that underpins the optimisation of electromagnetic compatibility (EMC) inductors.
The inductive component design was advanced through the creation of accurate RLC equivalent models. Signify’s investigation of buck converters uncovered that increased winding capacitance reduced efficiency; by adjusting the winding layout based on simulation, this effect was mitigated. High‑frequency winding techniques, such as the use of high‑frequency litz wire and multi‑chamber winding windows, were implemented to reduce skin and proximity losses. Simulation tools were calibrated against measured data, enabling rapid topology evaluation and the design of inductors that meet stringent loss and size targets for the GaN IPM.
Collaboration across the consortium was structured into work packages. Work package 8 concentrated on magnetic components for the high‑frequency applications of the project partners. Sumida supplied proven ferrite parts in the first phase to refine development and measurement procedures, then introduced the novel Fi 337 material in the second phase, tailoring it to specific integration scenarios. Joint workshops and project meetings facilitated knowledge exchange, while the consortium’s final deliverables included a suite of magnetic components customised for each partner’s application. The project’s outcomes—an integrated GaN IPM with optimised magnetic elements and validated high‑frequency ferrite materials—represent a significant step toward more efficient, compact power electronics for transportation and industry.
