Access Official Technical PDF

Share Summary:LinkedInEmail

A SiC Class-E inverter combines a single-switch resonant topology with a silicon carbide power device to perform high-frequency DC-to-AC conversion. The approach is relevant to industrial and commercial power electronics where designers must balance efficiency, system size, component count and control complexity. A reported 500 kHz prototype illustrates both the appeal of this architecture and the practical design challenges involved in operating a resonant converter at high frequency.

Class-E inverters have been used where high-frequency operation and a relatively simple switching structure are valuable. However, their resonant operating conditions can place substantial voltage and current stress on the switch. Silicon carbide devices offer characteristics that can help address these demands, but they do not remove the need for careful circuit design, component selection and testing.

How a Class-E resonant inverter operates

A Class-E inverter uses a switch, a choke or input inductor, a shunt capacitor, and a series resonant network connected to the load. The switch is driven at a 50% duty cycle near the circuit’s resonant frequency. The input inductor is selected to carry current that is approximately constant at steady state, while the tuned series network shapes the output current toward a sinusoidal waveform.

The timing of the switch relative to the resonant waveform is central to the topology. Under the intended operating conditions, the switch can turn off at approximately zero voltage. This zero-voltage switching behavior can reduce switching loss compared with hard switching. If the operating frequency moves away from the circuit’s true resonance, however, the desired switching conditions are disrupted and switching losses can occur.

The resonant network also makes output behavior sensitive to switching frequency. Small frequency changes can affect output voltage, creating the possibility of frequency-based control, including approaches that use a voltage-controlled oscillator and output-voltage feedback. The single controlled switch can simplify the switching stage, but the complete system still requires appropriate sensing, drive circuitry and feedback design when closed-loop regulation is needed.

Why use silicon carbide in a Class-E inverter?

Class-E designs face demanding switch conditions: high-frequency operation is combined with potentially high peak voltage and current. These stresses have been one reason the topology has been difficult to extend beyond lower power levels. Silicon carbide (SiC) is a wide-bandgap semiconductor material with physical properties that support high-voltage, high-temperature and high-frequency power-device applications.

Compared with silicon, SiC’s material properties include a higher breakdown electric field, higher saturated electron drift velocity, greater thermal conductivity and a wider bandgap. These characteristics can support devices with higher blocking capability, faster switching and greater tolerance of demanding operating conditions. They may also allow designers to consider smaller magnetic components and cooling arrangements in some high-frequency systems, although the outcome depends on the design as a whole.

There is a trade-off: the prototype study notes that SiC devices can have a larger on-state voltage drop, increasing conduction loss. The potential reduction in switching losses and other device benefits must therefore be considered alongside conduction loss, thermal management and the stresses imposed by the resonant circuit. Choosing a SiC MOSFET does not by itself guarantee high efficiency or a smaller system.

Designing for resonance and component tolerances

The reported inverter was designed for a 48 V DC input and a 500 kHz switching frequency, using a C2M0040120D SiC MOSFET as its main switch. The passive components were calculated for the desired resonant operation, then selected from commercially available values that were close to the calculated values. The design included a 10-ohm load and resonant components of approximately 22.3 µH and 4.79 nF, with selected available values of 22 µH and 4.7 nF.

In a sharply tuned resonant circuit, nominal component values are only part of the design task. Component tolerances and printed circuit board parasitics can shift the actual resonance. The prototype’s calculated operating point using selected values was approximately 495 kHz, while frequency-response testing measured resonance at 520 kHz. The difference was attributed to passive-component tolerances and board parasitics.

This illustrates why engineers should validate the assembled circuit rather than rely only on calculations or nominal component labels. The prototype’s printed circuit board used planes in place of traces and minimized sharp corners to reduce parasitic effects. Test points were also included to support voltage and current measurements. Such layout and measurement considerations are integral to high-frequency power-converter development.

What prototype testing revealed

The system was tested open loop using a programmable DC supply, a function generator and an isolated gate driver. The gate driver provided the appropriate drive and isolation for the MOSFET. Testing was conducted at approximately 100 W, not at a higher power level implied by the inverter’s high-frequency design.

With a wire-wound resistive load, the load’s inductance affected operation and contributed to lagging current. The measured waveforms also showed high switch-current spikes, which remained within the selected MOSFET’s rating. The authors noted that reducing load parasitic inductance could reduce these spikes. Adding a SiC full-bridge rectifier between the inverter and load changed the load interface and was reported to improve power-factor behavior and reduce switch-current spikes.

These results underscore the importance of evaluating the full power path. Load characteristics, rectification and parasitic inductance can influence the waveforms seen by the inverter, even when the resonant network has been carefully designed. The test setup and load are therefore part of the engineering assessment, not just supporting equipment.

Practical implications for power-electronics teams

A single-switch Class-E architecture may reduce switching-stage complexity, while high-frequency resonant operation offers a route to compact power-conversion systems. SiC devices are a potential enabler where switching speed and voltage or current stress are key concerns. Yet designers must manage the topology’s high Q-factor requirements, tune the circuit accurately and verify switch waveforms under realistic load conditions.

  • Validate resonance in hardware: account for component tolerances and PCB parasitics, then measure the assembled circuit’s frequency response.
  • Check device stress: measure switch voltage and current, including peak and transient behavior, against device ratings.
  • Include the load interface: assess how load inductance and any rectifier affect current waveforms and inverter operation.
  • Evaluate losses as a system: weigh SiC switching benefits against on-state losses, magnetics and thermal requirements.

The 500 kHz prototype demonstrates a practical implementation of a SiC Class-E inverter, while also showing why resonance control and parasitic effects matter. For engineers considering this approach, the key takeaway is to treat device selection, passive-component tolerances, layout, load behavior and measurement as one connected design problem. The topology offers a simple switching structure, but reliable operation depends on achieving and verifying the intended resonant conditions.

Access Official Technical PDF

About The Author