RCD Types AC, A, F and B: Which to Use and Why

Quick Answer: RCDs are classified by the type of residual (earth-leakage) current they can reliably detect and disconnect — Type AC (sinusoidal AC only), Type A (AC plus pulsating DC), Type F (Type A plus mixed/high-frequency), and Type B (also smooth DC). Under BS 7671:2018+A2:2022, additional protection by 30mA RCD is required for most socket-outlets and mobile equipment (Regulation 411.3.3), and Regulation 531.3.3 effectively rules out Type AC where DC residual currents from connected loads are likely. For most modern domestic circuits feeding electronics, Type A is now the practical minimum.

Summary

For years the humble Type AC RCD was the default in UK consumer units. That assumption no longer holds. Modern loads — LED drivers, switch-mode power supplies, variable-speed motors, induction hobs, EV chargers and solar inverters — inject DC and high-frequency components into the residual current. A Type AC device can be "blinded" by smooth DC: the leakage current biases its toroidal core into saturation, and it may then fail to trip on a genuine AC fault. That is a safety failure, not a nuisance, and it is why Regulation 531.3.3 of the 18th Edition (as amended) now requires you to select the RCD type according to the load it protects.

The four common types form a nested hierarchy. Each higher type does everything the one below it does, plus more. Type AC handles pure sinusoidal AC residual current. Type A adds pulsating DC (the kind produced by single-phase rectifiers in most electronic kit). Type F adds mixed frequencies and the composite waveforms produced by single-phase variable-speed drives — think washing machines and some heat-pump controls. Type B is the full house: AC, pulsating DC and smooth (continuous) DC, and it works across a wide frequency band — essential for three-phase electronics, many EV charge points and PV systems.

The practical takeaway for a UK sparky: stop fitting Type AC by default. Type A is the sensible baseline for general domestic circuits, Type F where high-frequency single-phase loads dominate, and Type B (or an alternative DC-fault-detection arrangement) where smooth DC fault current is possible. Getting this wrong does not always show up on test — a Type AC device can pass an RCD test button and a standard ramp/time test yet still be unfit for the load it is protecting.

Key Facts

Quick Reference Table

Quoting an electrical job? Describe the work and squote handles the pricing.

Try squote free →
Type Detects Typical loads Standard
AC Sinusoidal AC residual only Legacy resistive loads (being phased out) BS EN 61008/61009
A AC + pulsating DC (≤6mA smooth) General domestic, electronics, lighting, sockets BS EN 61008/61009
F A + mixed/high frequency (≤10mA smooth) Washing machines, single-phase VSD/inverter loads BS EN 62423
B AC + pulsating DC + smooth DC EV charging, PV, three-phase electronics, medical BS EN 62423
Type A + RDC-DD A behaviour + 6mA DC detection in EVSE EV charge points (alternative to Type B) BS EN 62955 / Section 722

Detailed Guidance

How an RCD actually "sees" the fault

An RCD measures the vector sum of currents flowing in line and neutral through a toroidal current transformer. In a healthy circuit they cancel; any imbalance is current leaking to earth, and the device trips when that imbalance exceeds its threshold. The problem is the core: it is wound and biased for AC. A continuous (smooth) DC leakage current does not alternate, so it pushes the core toward magnetic saturation. Once saturated, the core can no longer respond to a superimposed AC fault — the RCD goes "blind". Type A tolerates a little (≤6mA), Type F a bit more (≤10mA), and Type B uses additional electronics and a separate DC-sensing arrangement so it never relies on the AC core alone.

Choosing the type by load — decision context

Work through the loads on each final circuit, not the board as a whole:

  1. Pure AC, no electronics (immersion heater on its own circuit, simple resistive load) — Type AC is technically adequate, but fit Type A anyway; it is barely dearer and future-proofs the circuit.
  2. General sockets, lighting, most appliances — Type A. Almost everything plugged in contains a switch-mode supply, which produces pulsating DC leakage.
  3. Single-phase variable-speed loads — washing machines, dishwashers with inverter motors, some air-conditioning and heat-pump controls — Type F. These produce mixed-frequency residuals that a Type A may not handle cleanly.
  4. EV charge point — Section 722 requires DC fault protection. Either a Type B RCD on the circuit, or a Type A RCD where the EVSE provides built-in 6mA DC residual detection (RDC-DD to BS EN 62955). Check the charger spec sheet — most modern units include the RDC-DD, allowing a Type A on the supply side.
  5. Solar PV / battery storage with a transformerless inverter — Type B unless the inverter's documentation confirms equivalent DC fault protection. Follow the manufacturer's instructions, which override generic assumptions.

Discrimination and "blinding" of upstream devices

Mixing types matters for selectivity too. If a Type B device is downstream of a Type A or AC device, a smooth DC fault current can saturate the upstream device and stop the whole board working correctly. The general rule: an upstream RCD must be at least the same type as, or a higher type than, any downstream RCD on the same supply path. For discrimination on tripping time, use an S-type (time-delayed) device upstream with a higher IΔn (e.g. 100mA S-type feeding 30mA finals), and ensure both are an appropriate type for the loads below them.

Testing RCDs by type

Standard RCD testers apply an AC test current and verify the trip times (≤300ms at IΔn, ≤40ms at 5×IΔn for a 30mA additional-protection device; the half-rated 0.5×IΔn test should NOT trip). For Type A you should also use the tester's "pulsed DC" / half-wave function, and for Type B the tester needs a smooth-DC test capability — many common testers cannot fully test a Type B device, so you rely on the manufacturer's test/commissioning procedure and the integral test button for those. Always record the type on the EIC/EICR and on the circuit chart; "RCD 30mA" alone is no longer adequate documentation.

Retrofitting existing boards

You are not obliged to rip out every working Type AC device on an EICR — but you must code unsuitable protection. A Type AC RCD protecting a circuit with significant DC-leakage loads (e.g. an EV charger or PV) is a real safety shortfall and typically warrants a C2 (potentially dangerous) where the load demonstrably defeats the device. A Type AC on a general circuit is more often a C3 (improvement recommended) reflecting the move to Type A as best practice — apply judgement and record your reasoning.

Frequently Asked Questions

Is Type AC banned under the 18th Edition?

No, it is not banned outright. But Regulation 531.3.3 requires you to select a type that suits the load's likely DC residual currents, and in practice that excludes Type AC for the great majority of modern circuits. Treat Type A as your default and only use Type AC for genuinely pure-AC loads.

Can I just fit Type B everywhere to be safe?

You could, but it is expensive and rarely necessary. Type B devices cost several times a Type A and can introduce discrimination headaches. Match the type to the load: Type A for general use, Type F for single-phase VSD loads, Type B only where smooth DC fault current is genuinely possible.

Does an EV charger always need a Type B RCD?

Not always. Section 722 requires protection against DC fault current. Most modern EVSE units include integral 6mA DC residual-current detection (RDC-DD to BS EN 62955), which allows a Type A RCD on the supply circuit. If the charger does not provide this, you need a Type B. Always check the installation instructions.

Will my standard RCD tester confirm a Type B device works?

Usually not fully. Most field testers apply AC and pulsating-DC test currents but cannot generate the smooth-DC test a Type B needs. Use the device's integral test button and follow the manufacturer's commissioning procedure for the DC element; document that you did so.

What is the difference between Type F and Type A in practice?

Type F is essentially Type A "plus" — it handles composite and higher-frequency residual currents and tolerates more superimposed smooth DC (10mA vs 6mA). It is aimed at single-phase loads with frequency converters, such as modern washing machines and inverter-driven appliances.

Regulations & Standards