Wireless Charging: Design Rules and Process Limits

Wireless charging transfers power through a pair of coils that are only loosely coupled, and the circuit design is mostly about compensating for that looseness. The coil geometry, the resonant capacitor and the shielding together determine how much power reaches the receiver and how much is lost as heat.

The Coupled Coil Pair

The transmitter coil produces an alternating field and the receiver coil picks up part of it. The coupling coefficient between the two is a measure of the fraction of flux that links both, and for a loosely coupled pair it is between about zero point two and zero point six.

The coupling falls quickly as the coils are separated or misaligned, and the power that can be transferred falls with it. A design that works with the phone exactly centred on the pad loses most of its margin when the phone is moved a few millimetres.

Air core coils would need an enormous current, so both coils are backed with a ferrite shield that directs the flux and improves the coupling. The shield also keeps the field out of the metal parts of the product, where it would cause losses and heating.

The resonant tank

Both sides use a resonant tank, a coil and a capacitor, tuned near the operating frequency. Resonance raises the voltage across the coil for a given drive, which is what allows a manageable current to transfer the required power.

The transmitter tank is driven by a bridge at a frequency that may be fixed or variable. The receiver tank is tuned to the same frequency and supplies a rectifier, so the receiver looks like a load resistance reflected to the transmitter through the coupling.

The reflected resistance is what the transmitter actually drives, and it changes with the coupling and the load current. A control loop adjusts the frequency or the duty cycle to keep the power constant, which is why a wireless charging system is a control problem as much as a magnetic one.

Coil Geometry and Turns

The number of turns sets the inductance and the voltage transformation. More turns give more inductance and a higher voltage for the same flux, at the cost of a higher resistance and a larger coil.

A planar coil on a printed circuit board is cheap and repeatable, and it has a higher resistance than a wound coil of the same inductance. A wound coil with litz wire reduces the high frequency resistance and is used where the efficiency matters.

The inner and outer diameter both matter. A coil with a large hole in the middle uses less of its area, while one with many turns packed at the outside has a lower coupling with the opposing coil. The geometry is a compromise and is usually copied from a reference design.

Shielding and Field Containment

The ferrite shield under each coil increases the coupling and reduces the field behind the coil. Without it, the field would reach the battery and the metal chassis of the product, where eddy currents would produce heat.

The shield has its own loss, and a shield that is too thin saturates and produces more loss than it prevents. The thickness is chosen from the field strength and the frequency, and the material is usually a manganese zinc ferrite or a magnetic sheet.

A metal layer behind the ferrite can be added to contain the field further. It reduces the coupling slightly and limits the interference with the rest of the product, which is often the more important consideration.

Alignment and Position Tolerance

coupling coefficient falls as the coils move apart or sideways, and the system has to be able to regulate through the worst case the standard allows. The available range of movement is defined by a mechanical specification, and the coil is sized to cover it.

A coil that is larger than the receiver coil is more tolerant of lateral movement, because part of the transmitter coil is always facing the receiver. The cost is a larger pad and a lower coupling at the centre, which is a reasonable trade for a product that is placed down carelessly.

The axial distance is set by the thickness of the two housings plus the air gap. Every tenth of a millimetre matters, and the mechanical tolerance of the pad surface and the phone back has to be included in the budget.

Efficiency and Losses

The losses are the coil resistance, the shield, the resonant capacitors, the switches and the rectifier. At a low coupling the current in the coils is large for a modest power, so the copper loss dominates and the efficiency falls.

A higher operating frequency reduces the flux needed for a given power and allows a smaller coil, at the cost of higher switching loss and a higher alternating current resistance from the skin effect. The choice of frequency is a system decision rather than a component one.

Litz wire reduces the alternating current resistance by dividing the conductor into strands that are individually insulated and transposed. At a hundred kilohertz the improvement over solid wire is substantial, and it is the standard choice for a wound coil.

foreign object detection and Safety

foreign object detection is a safety function. A metal object such as a coin or a key placed on the pad absorbs power, heats up and can burn the surface or the object itself.

The detection works by measuring the power loss: the transmitter knows what it is delivering and the receiver reports what it is receiving, and a discrepancy above a threshold indicates a foreign object. Some systems also measure the change in the quality factor of the coil.

The function has to be calibrated for the range of receivers the pad will meet, because a poorly coupled receiver also produces a power discrepancy. Setting the threshold from measurements of the worst legitimate case and the smallest dangerous object is the practical approach.

Verification and Faults

Verify the pad with a power loss measurement at several positions and with several receiver types. The efficiency map over the pad area is the most useful result, and it shows where the regulation is marginal.

Measure the temperature of the ferrite and the coils after a full charge cycle. A warm coil with a cool receiver indicates a loss in the transmitter, often in the shield or in the resonant capacitor.

A pad that fails to start charging usually has a communication problem rather than a power problem, because the two sides exchange messages before the power is raised. The release checks that keep such a design consistent are collected in our PCB design release checklist, and the assembly points are listed in judging PCB quality. The board level measures that keep a switching circuit quiet are described in our guide to EMI suppression design principles.

FAQ

Why does my wireless charger get hot with a coin on it? The metal object absorbs power. foreign object detection should have prevented the power from rising, so check its threshold.

What coupling coefficient do I need? Between about zero point two and zero point six for a phone sized pad. Below that the losses become difficult to manage.

Should I use a printed coil or a wound one? A printed coil is repeatable and cheap with a higher resistance. Wound litz wire is used where efficiency matters.