Fault-finding a tripping RCD in a workshop sub-board
A school technology workshop is fed from a 32 A sub-main protected by a 30 mA Type II (non-delayed) residual current device. Since a new bench grinder and two extra double GPOs were installed, the RCD trips intermittently, usually a few minutes after the grinder starts under load, and sometimes when the workshop is empty overnight. Insulation resistance tests on the sub-main with everything disconnected read greater than 200 MΩ. The grinder's own supply lead and plug test clear. A clamp meter shows about 4 mA of standing leakage on the sub-main with all loads off but the circuit energised, rising to 18 mA when the grinder runs. The electrician must decide what to do before the workshop reopens.
1. Explain why the intermittent tripping cannot be dismissed as a faulty RCD, and identify the two most probable causes consistent with the test results.
Read the worked answer
The RCD is doing its job: a 30 mA non-delayed device must operate somewhere between 15 mA and 30 mA of residual current, so a standing leakage of 4 mA plus 18 mA under load is approaching the trip threshold. Cumulative leakage from many appliance filters and surge-protective devices on one RCD-protected sub-main is the classic cause of nuisance tripping, and the overnight trips point to something energised when the workshop is unattended - likely a plugged-in appliance, a fridge or a surge diverter - adding leakage. The two probable causes are: (1) accumulated protective-conductor leakage current from multiple loads exceeding the RCD's threshold, and (2) a genuine low-level insulation or moisture fault that only appears under thermal or damp conditions and is missed by a cold insulation resistance test. A failed RCD is the least likely explanation given the measurable leakage and the repeatability of the trips.
2. Outline the sequence of actions you would take to isolate the cause and make the installation safe, stating what evidence would confirm each cause.
Read the worked answer
- Isolate and lock out, then re-test the RCD with an RCD tester at half, full and five times rated residual current to confirm it trips within the required times; a pass confirms the device is sound. 2. Split the sub-main loads: disconnect circuits one at a time and repeat the clamp leakage measurement with everything else energised. A circuit whose disconnection drops the standing leakage to near zero identifies the offending load. 3. For the grinder, check whether its supply includes a filter or VSD; if so, the leakage is inherent and the fix is to separate it onto its own RCD or use a higher-rated or time-delayed RCD upstream, subject to the applicable standard and the protective earthing arrangement. 4. For the overnight trips, leave the installation energised with appliances unplugged and monitor; if trips stop, the appliance is the cause. 5. If leakage persists with all loads removed, carry out an insulation resistance test at 500 V DC on each circuit, then a polarity and continuity check, and inspect for moisture ingress. Only after the cause is confirmed and the protective device selection verified should the workshop be re-energised and the results recorded.
3. The workshop supervisor asks whether the RCD can simply be replaced with a 100 mA device to stop the nuisance trips. Give your response and justify it.
Read the worked answer
No. A 100 mA RCD would raise the touch-voltage threshold and increase the risk of a fatal shock in a workshop where portable tools and earthed metal benches are used; 30 mA is the accepted additional protection level for socket outlets and portable equipment. Substituting a higher-rated device masks the underlying leakage rather than fixing it, and it would leave the installation non-compliant with the applicable wiring rules for additional protection. The correct approach is to reduce or segregate the leakage: separate the grinder onto its own circuit and RCD, remove faulty appliances, and if discrimination is needed use a time-delayed upstream device only where the standard permits, keeping the 30 mA protection at the final circuits. The decision must be documented and the supervisor advised that the RCD rating is a safety limit, not an inconvenience setting.