Chapter 8
A wireless power system rarely fails on paper. It fails because the coil moved, a coin landed on the pad, or the inverter ran hotter than the enclosure allowed. WIA-COMM-009 therefore treats implementation as an ordered engineering process. Each step in that process produces evidence that a regulator, test house or integrator can inspect.
The specification's implementation guidance starts with a requirements analysis. Five inputs are fixed before any coil is drawn:
Power and distance then select the technology family:
| Power | Distance | Technology |
|---|---|---|
| < 100 W | < 10 mm | Inductive (Qi) |
| 100 W – 50 kW | 10 mm – 2 m | Resonant |
| > 1 kW | > 2 m | Microwave / Laser |
The ranges overlap in the kilowatt region. Read the table as a starting point, with distance usually the deciding input. The later steps in the design process are:
Safety is a design step in this list, not a final inspection. FOD (Foreign Object Detection) and LOP (Living Object Protection) are sensing functions that shape the control loop, the thermal design and the mechanical layout.
Every WPT chain has the same structure: DC to AC in an inverter, across the wireless link, then AC to DC in a rectifier. Overall efficiency is the product of the three stages. The specification gives indicative stage figures of 92–98% for the inverter, 85–95% for the rectifier, and 30–95% for the link, which depends on distance. The link is therefore the stage that decides whether a design succeeds.
For inductive systems the link is governed by two quantities: the coupling coefficient k (0 to 1) and the coil quality factor Q.
k = M / sqrt(L1 x L2)
Q = 2 x pi x f x L / R
eta = k^2 x Q1 x Q2 / (1 + k^2 x Q1 x Q2)
P_loss = P_tx x (1 - eta)
The coupling coefficient falls as alignment worsens. The specification gives 0.6–0.9 for excellent alignment with the gap smaller than the coil radius. It gives 0.3–0.6 when the gap is about one radius, and below 0.3 when the gap exceeds the radius. The efficiency expression shows why a pad that works at k = 0.8 can disappoint at k = 0.3. The product k²·Q₁·Q₂ is what lifts a loosely coupled link toward useful efficiency, which is why Q optimisation receives so much attention. The optimisation levers are:
Distance changes the physics, not just the numbers. The specification separates three regimes:
| Regime | Distance | Mechanism |
|---|---|---|
| Near-field | d < λ/2π | Reactive coupling |
| Fresnel | λ/2π < d < 2D²/λ | Transition, efficiency ∝ 1/d |
| Far-field | d > 2D²/λ | Radiation, efficiency ∝ 1/d² |
In the far field the Friis relation applies, and a rectenna (rectifying antenna) converts the received RF to DC. The stated rectenna efficiencies are 82% at 2.45 GHz, 85% at 5.8 GHz and 70% at 24 GHz, each at its own power density. Laser links add atmospheric transmission and photovoltaic conversion to the chain. The stated overall wall-plug-to-electric figure is 5–25%.
Impedance matching completes the electrical picture. Maximum transfer needs the load impedance to equal the conjugate of the source impedance, so a small L-C matching network sits between the stages. In the rectifier, Schottky diodes give a low forward drop (0.3–0.5 V). Synchronous MOSFET rectification drops below 0.1 V, and GaN HEMTs serve high-frequency designs.
Thermal design follows directly from the loss equation. The specification works an example: an 11 kW EV charger at 90% efficiency dissipates 1,100 W. It maps loss to cooling method as follows:
That example therefore needs liquid cooling in the coil assemblies and power electronics. A point of efficiency lost at this power level is also a measurable amount of heat that someone has to remove.
The specification does not publish market-size forecasts. It does give a bill-of-materials view of the most cost-sensitive segment, an 11 kW EV system (Level 2 equivalent):
| Component | Cost per unit | Scaling driver |
|---|---|---|
| TX coil (EV, 11 kW) | $200–500 | Volume production |
| RX coil (EV) | $300–600 | Integration |
| Power electronics | $500–1500 | Semiconductor cost |
| Control / communication | $100–300 | MCU, sensors |
| Installation (stationary) | $500–1000 | Labor |
| Total | $1,600–3,900 | Target: below $2000 |
Two things stand out. The total range straddles the below-$2000 target, so the target is reachable only toward the low end of every line. The largest and most variable line is power electronics, followed by the vehicle-side coil and the installation labour. Volume production helps the transmitter coil, while installation labour does not fall with scale. Cost reduction therefore depends mainly on semiconductor pricing and integration, since the coil and installation lines have less room to fall.
Technology choice also moves cost. The capacitive-versus-inductive comparison lists plate cost as lower for capacitive transfer. It also lists lower EMI and better tolerance to metal, but weaker tolerance to dielectric. Typical efficiency is 60–80% for capacitive against 70–90% for inductive. A designer in a metal-rich environment may accept the efficiency gap for cheaper plates and fewer shielding problems.
Each application below has its own certification burden, driven by its power level and by what stands in the field path.
Compliance in WPT combines product standards, emissions limits, human-exposure limits and, for beamed power, a licence. The specification's cross-walk is informative; implementers obtain authoritative copies from the issuing bodies. The main entries are:
| Concern | Reference |
|---|---|
| Low-power consumer | WPC Qi |
| Resonant consumer | AirFuel Alliance specifications |
| EV static WPT | SAE J2954; communications in SAE J2847/6 |
| EV WPT (EU) | IEC 61980-1 / -2 / -3 |
| EMC emissions | CISPR 11 / CISPR 25 / EN 301 489 |
| EMC immunity | IEC 61000-4 series |
| RF (US / EU) | FCC 47 CFR Part 15 and Part 18 / RED Directive 2014/53/EU |
| Public EMF exposure | ICNIRP 2010 and 2020 |
| RF exposure | IEEE C95.1-2019 |
| Implantable medical devices | ISO 14117 |
| Laser safety | IEC 60825-1 |
| Spectrum allocation | ITU-R Radio Regulations |
| Atmospheric loss models | ITU-R P.676 / P.838 |
| Battery-charger safety | IEC 62368-1 |
The specification's reference list also names ISO 19363 for magnetic-field EV charging safety and interoperability. The test activities it assigns to certification are:
Some hazards need design-level answers beyond limit lines. In an inductive charger, metal between the coils heats up. FOD sensing exists to catch it, and LOP sensing exists to protect animals or people entering the field. For lasers, wavelength choice is a safety decision. The specification marks 450, 808 and 980 nm as eye hazards, 1064 nm as "eye safe-ish", and 1550 nm fibre sources as eye safe. Its Maximum Permissible Exposure discussion treats eye and skin limits separately. A beam that is acceptable for skin can still be unacceptable for the eye, so the beam-path interlock design carries most of the certification effort.
Beamed power adds licensing. The regulatory integration notes describe the operator's licence under the national regulator. Examples include FCC Experimental and Part 18 in the US, Ofcom in the UK, and BNetzA, ANFR, KCC, MIC and MIIT for Germany, France, Korea, Japan and China. The licence is recorded against the licensed frequency and transmit power.
The Phase 2 interface turns the design process into software calls. Systems are registered, links are established between a registered transmitter and receiver, and safety tests and regulatory status are queried:
POST /wireless-power-transfer/v1/systems # register WPT system
GET /wireless-power-transfer/v1/systems/{id} # fetch system record
POST /wireless-power-transfer/v1/links # establish link
GET /wireless-power-transfer/v1/links/{id}/state # current link state
POST /wireless-power-transfer/v1/safety/test # run safety self-test
WS /wireless-power-transfer/v1/telemetry/stream # live link telemetry
GET /wireless-power-transfer/v1/regulatory/{id} # regulatory authorisation status
All of these endpoints follow a discovery convention at /.well-known/wia-wireless-power-transfer. Registering a system returns a stable systemId. Opening a link carries per-link parameters for alignment, requested power and link duration, and the request is validated against the regulatory envelope before the link opens. Two rules show how the interface encodes the physical risk:
The same envelope applies to space-solar systems, which need authorisation under the regulatory envelope. In practice, an expired self-test or a lapsed licence stops the link before any power flows. This is more reliable than relying on an operator to remember the condition. The self-test endpoint checks conditions such as continuity of the interlock chain, and the telemetry stream lets an operator watch the link state and the thermal and efficiency behaviour described in 8.2.
Integrators should plan for these points in a first implementation:
The specification points to several frontiers beyond the consumer and EV products certified today. W-band microwave (94 GHz, a 3.2 mm wavelength) is listed as emerging research. Its short wavelength allows very small apertures, but it also puts atmospheric loss models such as ITU-R P.676 at the centre of link budgets. Space solar power and orbit-to-ground laser relay are the long-range cases, and polar-night lunar power is a named laser application.
Component progress will determine how fast these cases mature. Multi-junction photovoltaic cells are listed at 40–50% efficiency, against 25–30% for silicon. Better PV conversion lifts a laser chain whose overall figure is only 5–25%. Phased arrays with retrodirective control reduce the pointing problem for moving receivers. On the ground, higher-power EV classes and wider misalignment tolerance depend on better coil design and faster, lower-loss semiconductors such as GaN. Each advance is useful only if the safety and regulatory evidence keeps pace, so the validation steps in 8.3 and 8.4 will matter more as power and distance grow.
This chapter followed a wireless power system from requirements to certification. Power and distance select the technology family. The coupling coefficient, quality factor and stage efficiencies set the loss budget, and the loss budget in turn sets the cooling. Cost targets depend mostly on power electronics and installation. Compliance combines product standards, EMC and exposure limits, FOD and LOP safeguards, and operator licences for beamed power. The Phase 2 interface records the key gates in software: a registered system, a validated link, a current authorisation and a passed safety self-test.
Certification is a snapshot, so the next chapter turns to how that evidence is maintained and how WPT systems fit into the wider ecosystem once they are deployed.