In an electric vehicle (EV) drive, the inverter controls the electrical power supplied to the motor. Its power semiconductors therefore influence how much energy is lost as current is switched and how reliably the device operates under electrical stress. A 2018 Hitachi development illustrates one approach to improving those trade-offs: a SiC power semiconductor called TED-MOS, designed for EV motor-drive inverters.
The prototype combined a fin-structured trench MOSFET with additional layers intended to manage electric fields and current flow. Hitachi reported lower resistance and electric-field strength than in a conventional DMOS-FET, alongside a reduction in switching energy loss. The results point to a practical design challenge for power electronics: achieving lower conduction losses without compromising voltage durability.
Why SiC device structure matters
Silicon carbide (SiC) is used in power semiconductors for applications such as converters and inverters. In an EV, the inverter changes the electrical supply to control motor operation, including its speed. Reducing losses in this conversion and switching process can support more efficient use of electrical energy.
However, SiC devices have a structural challenge. According to the development announcement, electrical resistance varies substantially with the crystal plane. A trench MOSFET can direct current along the sides of fin-like trenches, where it can flow on lower-resistance crystal planes compared with a conventional DMOS-FET structure. But the trench edges can concentrate electric fields. That concentration makes it harder to reduce resistance while maintaining the device’s ability to withstand electrical stress.
For engineers selecting or designing inverter components, this is not simply a matter of minimizing resistance. The device must also maintain its rated voltage and durability. A design that improves one characteristic at the expense of another may not meet the requirements of a high-voltage power system.
How TED-MOS addresses the trade-off
TED-MOS, short for Trench Etched DMOS-FET, uses a fin-structured trench design based on the conventional DMOS-FET. Hitachi first described a version intended for high-voltage industrial applications at 3.3 kV. The later device was adapted for EV inverters, which the announcement characterized as requiring higher current density at a lower voltage of 1.2 kV.
Two structural features were introduced to address the competing demands of field strength and resistance:
- Field relaxation layer (FRL): A PN junction positioned in the center of the device structure forms a region for relaxing the applied voltage. Its purpose is to reduce electric-field strength, including the concentration that can occur around trench features.
- Current spreading layer (CSL): This layer is intended to reduce resistance in the n-JFET region. It helps form a lower-resistance current path between the sides of the fin-like trenches and the n-JFET region.
In combination, the layers aim to manage where voltage is supported and how current moves through the device. This allows the structure to pursue lower resistance without relying on a design that leaves the trench-edge electric-field issue unresolved.
What the prototype results showed
Testing of the prototype device indicated that TED-MOS maintained the 1.2 kV rated voltage specified for the EV motor-drive application while changing several performance measures relative to a conventional DMOS-FET. Hitachi reported a 40% reduction in electric-field strength and a 25% reduction in resistance.
The modified structure also improved the speed of switching between the ON and OFF states. Hitachi reported that energy loss associated with this switching operation was reduced by 50%. This figure refers to switching energy loss; it should not be read as a 50% reduction in total EV energy consumption. Vehicle-level energy use depends on the wider drive system and operating conditions, which were not quantified in the announcement.
For inverter designers, the results are relevant because conduction resistance, electric-field distribution and switching behavior are interrelated design considerations. A device structure that addresses these together may offer a useful path for power electronics operating at the voltage and current requirements of EV motor drives. The reported figures, however, describe a prototype comparison with a conventional DMOS-FET—not a complete inverter or vehicle evaluation.
Potential relevance beyond electric vehicles
Hitachi stated its intention to apply TED-MOS technology to EV motor-drive inverters and to explore applications in electrical transducers used in infrastructure systems. The announcement did not specify particular infrastructure deployments or provide performance results for those applications. Any assessment for a new use case would need to consider its own voltage, current, switching and durability requirements.
The broader engineering lesson is the value of device-level field management. In power conversion, improving current flow alone may not be sufficient if the resulting structure creates concentrated electrical stress. Conversely, addressing field strength need not mean accepting higher resistance if the current path is also engineered. The FRL and CSL show how separate structural features can target these different constraints.
Practical implications for power electronics teams
When evaluating SiC MOSFETs for an inverter, engineers and system integrators should distinguish between device-level metrics and system-level outcomes. Resistance, rated voltage and switching energy loss describe important aspects of semiconductor behavior, but they do not by themselves establish the energy performance of a complete drive.
TED-MOS provides a specific example of a device architecture aimed at balancing low resistance with durability. Its prototype results suggest that trench geometry, electric-field management and current spreading can be considered together when developing SiC power semiconductors for EV applications. For plant and operations leaders assessing power-conversion technology, the practical takeaway is to examine both the measured device characteristics and how they relate to the requirements of the intended inverter or electrical system.