Wireless electricity, formally known as Wireless Power Transfer (WPT), is the transmission of electrical energy across an air gap using time-varying electromagnetic fields without physical conductors. What this changes in a real circuit or installation is the complete elimination of galvanic contact, enabling hermetically sealed enclosures, dynamic charging of moving assets, and zero mechanical wear on contacts. People commonly confuse near-field magnetic WPT with far-field radiative transfer (like beaming microwaves or harvesting Wi-Fi signals); true high-power WPT relies on non-radiative magnetic coupling, meaning the energy does not propagate into space but is tightly bound to the magnetic field between two tuned coils. While consumer Qi chargers use basic inductive coupling, Finland has emerged as a global powerhouse in advanced resonant topologies, with institutions like Aalto University driving breakthroughs in multi-coil arrays and high-frequency gallium nitride (GaN) inverter designs that push air-gap efficiencies past 92%.
The Physics of Resonant Inductive Coupling
Basic inductive coupling (like a standard transformer) suffers from massive leakage inductance and poor efficiency when an air gap is introduced. Resonant Inductive Coupling solves this by adding compensation capacitors to both the transmit (Tx) and receive (Rx) coils, tuning them to the exact same resonant frequency. Think of two mechanical tuning forks tuned to 440 Hz; if you strike one, the acoustic energy efficiently transfers to the other, while off-frequency objects in the room remain entirely unaffected. In WPT, this resonance allows the magnetic field to "tunnel" across the gap with minimal reactive power loss.
Worked Numeric Example: Calculating Air-Gap Efficiency
Let’s design a 1 kW WPT system for an Automated Guided Vehicle (AGV) operating at the SAE J2954 standard frequency of 85 kHz. We need to determine if our coil design will yield acceptable efficiency across a 100mm air gap.
- Define Coil Parameters: Both Tx and Rx coils have an inductance ($L_1 = L_2$) of 25 µH. Using 1050-strand Litz wire, the AC resistance ($R_1 = R_2$) at 85 kHz is measured at 0.15 Ω.
- Calculate Angular Frequency ($\omega$): $\omega = 2\pi \times 85,000 \approx 534,070$ rad/s.
- Determine Quality Factor ($Q$): $Q = \frac{\omega L}{R} = \frac{534,070 \times 25\mu\text{H}}{0.15\Omega} = 89$. Both coils have a $Q$ of 89.
- Measure Mutual Inductance ($M$): At a 100mm air gap, bench testing shows $M = 6$ µH.
- Calculate Coupling Coefficient ($k$): $k = \frac{M}{\sqrt{L_1 L_2}} = \frac{6}{25} = 0.24$.
- Compute Theoretical Efficiency Limit: The figure of merit is $k^2 Q_1 Q_2 = (0.24)^2 \times 89 \times 89 = 456.25$. Using the standard WPT efficiency approximation $\eta \approx \frac{k^2 Q_1 Q_2}{(1 + \sqrt{1 + k^2 Q_1 Q_2})^2}$, we get a theoretical coil-to-coil efficiency of roughly 95.6%.
Accounting for realistic inverter and rectifier switching losses using modern GaN FETs, the end-to-end DC-to-DC efficiency settles at a highly respectable 92.4%, meaning only 76W is lost as heat across the entire system at a 1 kW load.
Where You Meet This in Practice
You will encounter advanced WPT systems in environments where physical contacts are a liability:
- Industrial AGVs and AMRs: Warehouse robots that charge opportunistically at picking stations without stopping to plug in, eliminating arc-flash risks and contact oxidation.
- Medical Implants: Pacemakers and neurostimulators that require hermetically sealed titanium enclosures to prevent tissue rejection and fluid ingress.
- Dynamic EV Charging: Nordic pilot projects (heavily influenced by Finnish smart-grid research) embedding Tx coils under asphalt to charge electric buses at transit stops.
- Rotary Sensors: Torque sensors on rotating motor shafts where slip rings would introduce unacceptable electrical noise and mechanical drag.
Real-World Scenario Walkthrough: An AGV Charging Failure
Theory assumes perfect alignment. The jobsite rarely delivers it. Here is how a WPT system fails in the field and how modern topologies fix it.
The Numbers: The Tx inverter was pushing 12A RMS into the primary coil. Under ideal alignment, the expected Rx power delivery was 960W.
The Outcome: The AGV’s Battery Management System (BMS) flagged a brownout and halted charging. Telemetry showed Rx power had plummeted to 410W. Worse, the Tx coil surface temperature spiked to 85°C within 15 minutes, triggering a thermal shutdown.
What Went Wrong: The AGV had parked 15mm off-center laterally. In a traditional single-coil WPT system, lateral misalignment causes the mutual inductance ($M$) to drop off a cliff. The coupling coefficient $k$ fell from a healthy 0.35 down to 0.12. This drastic change in reflected impedance pulled the inverter out of its Zero-Voltage Switching (ZVS) region. The GaN FETs were forced into hard-switching, causing massive switching losses ($E_{on}$ and $E_{off}$) that manifested as severe heat in the Tx coil and inverter heatsink, while starving the Rx side of power.
The Fix: The engineering team replaced the single Tx coil with a Finnish-developed multi-coil array (a 3x3 matrix of smaller sub-coils). The system's Foreign Object and Position Detection (FOD/PD) algorithm identified the exact lateral offset of the Rx coil and energized only the two overlapping sub-coils in series. This restored $k$ to 0.28, re-established ZVS, and returned power transfer to 940W without thermal runaway.
Component Selection for High-Frequency WPT
If you are prototyping a WPT system on the bench, standard 50Hz/60Hz magnetics and silicon MOSFETs will fail. You must spec components for high-frequency, high-Q operation:
| Component | Specification / Requirement | Why It Matters |
|---|---|---|
| Coil Wire | Litz wire (e.g., 1050 strands of 44 AWG) | Mitigates skin and proximity effects at 85kHz+. Solid copper wire will overheat due to AC resistance. |
| Switching FETs | GaN FETs (e.g., Infineon IGT60R070D1 or EPC2045) | Zero reverse recovery charge ($Q_{rr}$) and ultra-low gate charge ($Q_g$) are mandatory to maintain ZVS and minimize dead-time losses. |
| Resonant Capacitors | NP0/C0G Ceramics or High-Voltage Film (e.g., WIMA MKP) | Never use X7R/X5R ceramics. They exhibit severe capacitance drop with DC bias and generate acoustic noise (piezoelectric effect) at high AC voltages. |
| Ferrite Shielding | MnZn Ferrite plates (e.g., TDK PC95 material) | Directs magnetic flux toward the Rx coil and prevents eddy currents from heating the aluminum chassis behind the Tx coil. |
FAQ: Wireless Power Transfer Edge Cases
Q: Can a dropped metal tool on the Tx coil cause a fire?
A: Yes. Ferromagnetic and conductive foreign objects will absorb magnetic flux, generating massive eddy currents and heating up like an induction stove. Modern WPT systems mandate active Foreign Object Detection (FOD), which monitors the Q-factor or uses auxiliary sensing coils to detect anomalies and shut down the inverter within milliseconds.
Q: Does changing the air gap distance change the resonant frequency?
A: No. The resonant frequency ($f_r = \frac{1}{2\pi\sqrt{LC}}$) is strictly determined by the inductance and capacitance. However, changing the distance alters the coupling coefficient ($k$). In very high-Q systems, tight coupling can cause "frequency splitting," where the single resonant peak bifurcates into two distinct peaks, requiring the inverter to dynamically track the new optimal frequency via phase-locked loops (PLL).
Q: Is WPT safe for humans standing near the coils?
A: Yes, provided the system complies with ICNIRP guidelines for human exposure to time-varying electric and magnetic fields. At 85 kHz, the magnetic field is non-ionizing. Systems are designed with ferrite shielding and aluminum Lenz-law shields to ensure magnetic leakage outside the immediate air gap remains well below the 6.25 µT general public exposure limit for this frequency band.
References and further reading on WPT topologies and safety limits can be found via the SAE J2954 standard documentation, the Aalto University Smart Grids research group, and Infineon's GaN application notes for resonant converters.
