Piezoelectric resonators are a potential route to smaller, lower-loss passive components in high-frequency power conversion. The opportunity stems from a basic design challenge: inductors, transformers and capacitors can dominate converter size and losses, limiting power density even as advances in GaN and SiC enable higher-frequency switching. Comparing energy-storage mechanisms helps show where alternatives may offer an advantage—and what engineering obstacles remain.
Why examine alternatives to conventional passive components?
Most switching converters rely on inductors, transformers and capacitors to store energy over switching cycles. Raising switching frequency can support smaller converters, but it does not remove the constraints imposed by passive components. In particular, magnetic components face winding losses, including skin-effect losses at MHz frequencies, as well as limits tied to magnetic materials and thermal management.
One way to look beyond incremental improvements is to compare the fundamental energy-storage potential of different mechanisms. The analysis considered kinetic, elastic, electrostatic and magnetic storage, using idealized conditions to identify upper limits rather than provide a ready-to-build design. Practical fabrication, packaging and operating constraints can reduce performance substantially.
Comparing passive energy-storage mechanisms
Mechanical storage can have high volumetric energy density. In the analysis, flywheels and oscillating masses offer the highest theoretical energy density among the mechanisms considered, while springs also compare favorably with inductors. The advantage is not automatic: mechanical energy must be coupled to an electrical circuit through a suitable transduction method.
Capacitors store energy in an electric field, and their energy density depends on dielectric permittivity and breakdown field. Advanced dielectrics and glass can offer high theoretical values, but those values are based on breakdown limits; continuous operation requires derating. Supercapacitors can store substantial energy, but their response is generally too slow for MHz switching. The analysis estimates that their achievable energy density falls at frequencies above 1 MHz.
Inductor energy density is constrained by magnetic saturation and by the losses associated with winding current. Even assuming an aggressive tolerable loss per unit area, the modeled inductor energy density remains low compared with mechanical storage. At high frequencies, skin depth further limits how much conductor can effectively carry current unless winding techniques can mitigate the constraint.
Why piezoelectric resonators are of interest
A mechanical resonator could potentially store energy more compactly than an LC tank, but a converter needs a way to move energy between mechanical and electrical forms. Electromagnetic transduction retains winding-loss limitations, while electrostatic approaches face energy-density constraints. Piezoelectric materials are therefore attractive candidates: they couple electrical excitation to mechanical vibration.
The analysis models a piezoelectric resonator with the Butterworth–Van Dyke equivalent circuit. In this model, a static capacitance sits in parallel with a motional branch representing mechanical resonance. This allows the resonator to be compared with an LC tank in a resonant switched-capacitor (ReSC) circuit, using effective resistance as a measure of losses during charge transfer.
For a high-quality-factor resonator, the modeled effective resistance depends on capacitance, electromechanical coupling and the material’s frequency–quality-factor product, or f0Q. Under optimistic assumptions, increasing capacitance—such as through multiple layers connected in parallel—can reduce effective resistance. The paper uses lithium niobate for its analysis, while noting that electrode losses, mounting and material orientation can make practical performance less favorable than the ideal model.
Performance depends on frequency and operating limits
Neither low effective resistance nor high energy density alone determines whether a resonator will work in a converter. The analysis also considers power throughput, which can be limited by thermal losses, dielectric breakdown, mechanical strain or nonlinear vibration.
For piezoelectric resonators, increasing vibration amplitude can push the material beyond approximately linear behavior; higher strain can approach the material’s yield limit. These constraints shape the power that can be transferred at a given frequency. For LC resonators, losses in the winding and capacitor create heat, while voltage rating and dielectric properties constrain operation.
The modeled comparison finds that piezoelectric resonators can outperform conventional LC resonators in effective resistance across a broad frequency range under ideal assumptions. However, piezoelectric resonators can be limited by nonlinearity and mechanical strain at lower frequencies, affecting power throughput. These findings describe potential performance limits, not guaranteed results from a production-ready design. Realizing them would require solving practical issues such as multilayer construction, electrode resistance, geometry and thermal management.
What the prototype demonstrates
The work also reports a prototype LC resonator made with commercially available C0G capacitors. In a volume of 1 cm3, it achieved an effective resistance of 2.76 mΩ at 2.70 MHz and handled 7.42 kW within its rated temperature. This result supports the modeled low-loss, high-power behavior of that resonator configuration; it does not establish the same performance for piezoelectric devices or other designs.
Implications for converter design
For engineers evaluating high-frequency power conversion, the practical message is to treat passive-component technology as a potential system-level constraint, not simply a selection made after choosing semiconductors. Mechanical resonance, particularly with piezoelectric transduction, offers a research direction for reducing passive losses and volume. But the idealized energy-density comparison must be weighed against transduction efficiency, mechanical limits, manufacturing requirements and heat removal.
Piezoelectric resonators are not a drop-in replacement for LC tanks on the evidence presented. Rather, the analysis provides a framework for comparing their limits and identifying development challenges. For system integrators and design teams, the next step is to assess whether a resonant architecture and its operating conditions can use the potential benefits without exceeding mechanical, electrical or thermal constraints.