Tingling soldering station: open protective earth, not a short circuit
A PSEB workshop has eight soldering stations on one bench, each with a metal chassis and a 3-pin 6 A socket. Students report a tingling sensation when touching the chassis of station no. 5 and the bench frame at the same time. A trainee measures with a digital multimeter: L-N = 231 V, L-E = 118 V, N-E = 103 V, and continuity from the socket's earth pin to the main earth bar reads OL. Insulation resistance from L to E on the station, tested at 500 V with the station unplugged and its switch on, is greater than 200 MΩ. The station's mains filter has two 4.7 nF Y-class capacitors, one from L to chassis and one from N to chassis. The workshop has a 30 mA RCD incomer, which has not tripped. The trainee suggests connecting the socket earth pin to the neutral terminal to fix the tingle.
1. Interpret the four measurements. Why does the chassis sit at about half the mains voltage, and why did the RCD not trip?
Read the worked answer
L-N = 231 V is normal. L-E and N-E are 118 V and 103 V, which sum to about the full mains voltage - the classic signature of an earth pin that is floating and a chassis held near mid-supply by two equal 4.7 nF capacitors forming a capacitive divider. The meter reads roughly one half because its input impedance is much larger than the capacitor impedance.
Continuity OL confirms the protective conductor is open between socket and earth bar. The 200 MΩ insulation result shows there is no live-to-chassis fault, so there is no path for 30 mA: each Y-capacitor passes only about 230 × 2π × 50 × 4.7 nF ≈ 0.34 mA, roughly 0.7 mA total, which is felt as a tingle but is far below the RCD threshold. The RCD is behaving correctly.
2. Evaluate the trainee's neutral-to-earth proposal and give the correct repair and verification sequence.
Read the worked answer
Never bond the socket earth to neutral. That puts load current on the earth conductor, raises the chassis potential above true earth, and makes the chassis live if the neutral opens or the supply is reversed; it also desensitises or bypasses the RCD, which is why such links are prohibited.
Correct sequence: isolate and lock off the circuit; locate the break (socket earth terminal, bench wiring, plug earth pin, or the bar connection) and repair or replace the conductor; then verify continuity from each socket earth pin to the bar (≤1 Ω), insulation resistance L-E and N-E at 500 V DC (≥1 MΩ), and correct polarity; finally press-test the RCD with a proper tester (trip ≤30 mA, ≤300 ms) and record the results. Only then reconnect the stations and confirm the tingle is gone.
3. How would the readings differ if there were a genuine insulation failure rather than an open earth? Explain why this distinction matters for the repair.
Read the worked answer
With a real insulation failure, L-E would read close to the full 231 V regardless of meter impedance, because the chassis would be tied to line through a low resistance rather than floating on capacitors. The 500 V insulation test would read well below 1 MΩ (often a few kΩ to a few hundred kΩ), and leakage could exceed 30 mA so the RCD would trip, or the chassis would sit at line potential.
The distinction changes the whole repair: an open earth is a conductor fault fixed by restoring and testing the protective path, whereas an insulation failure means the appliance must be withdrawn, the faulty wiring or filter replaced, and the equipment retested before use. Confusing the two - for example, bonding earth to neutral to hide a tingle - leaves a lethal fault in service.