Why Does My RCD Keep Tripping? Causes and Fixes

Why does an RCD keep tripping? Most repeat trips are not a wired fault at all — they come from cumulative leakage current pushing a healthy circuit over the device's rated residual operating current IΔn, commonly 30 mA on a final circuit, or from a Type AC device that cannot correctly sense the pulsating or smooth DC leakage produced by modern electronic loads. Ignore the pattern and you'll keep replacing a device that was never broken, or worse, reset past a live earth fault. This article works through the diagnostic sequence: reading the trip pattern, calculating cumulative leakage, matching RCD type to load, ruling out moisture and cable damage, isolating the faulty circuit, and knowing when the fix is a different device rather than a repair.

Start With the Symptom: Immediate Trip or Intermittent Trip

A device that trips the instant it's reset, before any load is switched on, almost always indicates a genuine line-to-earth fault downstream — a nicked cable, a failed heating element, or water in a junction box. Isolate circuits one at a time and the fault will announce itself immediately. A device that holds for minutes or hours and then trips under load points somewhere else: cumulative leakage that only crosses the threshold once enough equipment is drawing current, or a cable insulation fault that only breaks down once conductors warm up and expand.

Trips that cluster around wet weather, or start a few minutes after rain, are a strong signal for outdoor circuits and cable glands rather than anything indoors. Trips with no pattern at all — different times of day, no correlation with load or weather — usually mean switching transients or a marginal, near-threshold leakage level that tips over at random.

Cumulative Leakage: When Nothing Is Actually Faulty

Every piece of electronic equipment with a switch-mode power supply leaks a small steady current to earth through its EMC filter capacitors — typically a fraction of a milliamp per device, more for larger equipment with internal filtering. Long cable runs add capacitive leakage of their own, and three-phase or screened cable adds more per meter than a short domestic run. None of this is a fault. It's designed-in leakage that standards permit precisely because it stays well below personal-protection thresholds on its own. The problem starts when a board has thirty of these devices on one 30 mA RCD and the sum creeps past the trip point.

Run the numbers before blaming the device. Count the sockets, count the switch-mode supplies behind them, and if the estimated IΔsum sits above roughly a third of IΔn, the fix is splitting the board across more 30 mA RCDs or RCBOs, not chasing a phantom fault. See RCD sensitivity ratings for how IΔn is chosen per circuit type in the first place.

Wrong RCD Type for the Load (AC, A, F or B)

A Type AC RCD only detects pure sinusoidal residual current. Feed it a load with a rectified or chopped output — VFDs, some LED drivers, certain UPS units — and the pulsating DC component can partially or fully blind the sensing core, causing erratic trips or, in the worse case, a failure to trip at all under a real fault. Type A covers pulsating DC and is the correct modern default for general electronic loads. Type F adds coverage for mixed-frequency residual current from single-phase variable-frequency drives. Type B is the only category that also detects smooth DC residual current, which is why three-phase VFDs, EV chargers and transformerless PV inverters require it specifically.

If a board feeding a VFD or an EV charger trips on start-up or during acceleration, check the nameplate type on the existing RCD before doing anything else. A Type A device on a smooth-DC load will nuisance-trip, sometimes only under specific load conditions, which makes it look intermittent when it's actually a type mismatch every time. See RCD types AC, A, F, B and SI explained and, for DC-heavy loads specifically, Type A vs Type B for VFD, EV and solar.

Moisture, Insulation Damage and Outdoor Circuits

Water ingress in an outdoor socket, a cable gland, or a garden lighting junction is the single most common genuine fault behind a trip that recurs in wet weather. Insulation resistance drops as moisture tracks along a conductor surface, leaking current to earth well before it produces visible damage. What we see in the field: the fault often clears on its own once the enclosure dries out, which tempts people into resetting and moving on — until the next rain repeats it, usually with the leakage a little worse each time as corrosion sets in at the ingress point.

Sites with corrosive atmosphere, dust, or condensation — coastal installations, agricultural buildings, plant rooms with process humidity — benefit from RCDs built for that environment rather than a standard indoor-rated device swapped in repeatedly. Siemens' 5SV SIGRES range, for example, is built specifically for humid and corrosive conditions where a standard device would degrade and nuisance-trip faster than the installation ages.

Nuisance Tripping From Switching and Surge Transients

Not every unwanted trip is leakage at all. Switching a large inductive load, a lightning-induced transient on the supply, or a nearby motor starting can induce a brief current imbalance that a standard instantaneous RCD reads as a fault and clears in milliseconds, even though no real leakage path exists a moment later. This is what "super-immunized" or "SI" devices are built to filter out — Schneider's Acti9 "si" range and comparable ABB and Siemens devices add filtering that ignores short transient imbalances while still tripping correctly on a sustained residual fault.

Sites that nuisance-trip repeatedly with no fault ever found on inspection are usually candidates for an SI-rated replacement, not further troubleshooting of wiring that has already tested clean twice.

Isolating the Circuit: A Systematic Split-Test Procedure

Skip the guesswork and split the board. Switch off every final circuit on the affected RCD, reset it, then re-energize circuits one at a time, waiting a few seconds between each. The circuit that trips the device on re-energization — or shortly after, if the fault is cumulative and load-dependent — is your candidate. If nothing trips with individual circuits live but the device trips once several are on together, that's cumulative leakage, not a single fault, and no amount of circuit-by-circuit testing will isolate it further.

Once a circuit is identified, an insulation resistance test with the load disconnected separates a genuine cable or appliance fault from leakage that only appears with the device plugged in and running. A test button proves the trip mechanism works; it proves nothing about the actual trip current or the timing, which is why commissioning and fault-finding both call for an instrument that can measure IΔn and ramp the test current rather than relying on the button alone. This distinction also matters when deciding between an RCCB backed by a separate MCB and a combined RCBO — see MCB vs RCBO vs RCD vs RCCB differences for how the split affects fault isolation on a board with several final circuits.

When to Replace the RCD Instead of Chasing the Fault

Sometimes the device itself is the answer. A Type AC RCD on a board that has picked up electronic loads since installation should be replaced with Type A as a matter of course, independent of any specific trip event. A board with a VFD, EV charger, or PV inverter added after the original design needs its RCD reassessed for Type B or a combined Type A plus RDC-DD arrangement, not a like-for-like swap. A site with repeated, unexplained transient trips and two clean insulation tests is a reasonable case for an SI-rated replacement rather than a third round of testing.

Getting the sensitivity, type and pole count right the first time avoids most of this — see how to select RCD sensitivity, type and poles for the selection checklist, and the RCD protection guide for the full technical background behind every category referenced here. Stoklink carries residual current devices and RCBOs across the sensitivity and type ranges covered above for same-board replacement.

Conclusion

Repeated RCD trips have a short list of real causes: cumulative leakage crossing the design margin, a type mismatch with the connected load, moisture or insulation damage on an outdoor circuit, or a transient that a standard device reads as a fault. Work through the symptom pattern first, run the split test before opening anything up, and treat a type mismatch or a nuisance-prone site as a device decision rather than a repair job. Most boards that "keep tripping for no reason" have a reason — it's usually arithmetic, not damage.