A welding shop with a bypassed RCCB and a 14 Ω earth
A small fabrication workshop in Guwahati runs an arc welding set (415 V, three-phase), a 3 kW air compressor and hand tools. Last monsoon the 30 mA RCCB at the main board tripped repeatedly, so the owner's nephew removed it and joined the incoming wires directly. The workshop's only earth is a single 40 mm diameter GI pipe driven 2.5 m into the ground beside the concrete apron; a fall-of-potential test gives 14 Ω. Welders standing on a damp floor report a tingling sensation when they touch the welding bench, and a neon tester glows when held between the bench and a damp patch on the floor. The supply is overhead and the area has frequent lightning. The owner wants to pour salt water over the earth pipe before next week's inspection.
1. Explain in electrical terms why the bench tingles, and quantify the touch voltage and body current with the earth at 14 Ω.
Read the worked answer
The mechanism. The bench is metal and is bonded to the earth electrode, but the electrode is a poor connection to true earth - 14 Ω. Any leakage or fault current from the welding set, the compressor, or a damaged cable flows through that 14 Ω and back to the source earth. The bench therefore sits at a potential above true earth, and a person touching it while standing on a damp floor completes a circuit through their body.
Quantify it. Assume a phase-to-bench fault at 230 V, the electrode at 14 Ω and the supply earth at a modest 4 Ω. Fault current = 230 / (14 + 4) ≈ 12.8 A. Bench potential above true earth = 12.8 × 14 ≈ 179 V.
A person with a body-plus-floor resistance of about 1000 Ω touching the bench draws 179 / 1000 ≈ 179 mA. The fibrillation threshold for a sustained shock is around 30 mA and the let-go threshold around 10–15 mA, so this is a genuinely lethal condition, not a tingle. Note also that 12.8 A will not trip a 16 A or 63 A MCB, so the fault can persist indefinitely.
The neon tester glowing is consistent: it needs only tens of volts to strike, and it confirms a real potential difference between the bench and the damp floor, i.e. a potential gradient in the ground - a step/touch voltage hazard around the electrode.
2. Evaluate the removal of the RCCB and the salt-water proposal. What do you tell the owner before the inspection?
Read the worked answer
The RCCB removal is the most serious fault on the site.
- A 30 mA RCCB is the only device that responds to the small earth-leakage currents that injure people; an MCB responds to overload and short circuit and will never see a 179 mA body current.
- Repeated tripping is information, not nuisance. It means there is leakage - damp welding cable, a cracked terminal block, a wet compressor motor, or a deteriorating set. The correct response is to find and repair the leak, then test the device.
- Bypassing it converts a monitored installation into an unmonitored one and would be a straightforward non-compliance finding.
The salt-water proposal is a cheat.
- Backfilling an earth pit with alternating layers of charcoal and salt is a recognised construction method. Pouring salt water on the ground a week before an inspection is not - it gives a temporarily low reading that drifts back within weeks, corrodes the pipe, and hides a real deficiency.
- It also masks the true seasonal behaviour: earth resistance rises when the ground dries and falls in the monsoon, so the value must be judged at its worst.
What to tell the owner before the inspection.
- The RCCB is reinstated and tested, and the leakage that tripped it is found and repaired - not the device defeated.
- Earthing is improved so the electrode resistance meets the accepted limit for the installation (about 8 Ω for a small general installation; lower is better, and substations aim for 1 Ω or less).
- Every metal enclosure - bench, welding set, compressor, distribution board - is bonded and its continuity proven.
- Lightning protection is addressed separately, since this is a high-lightning area.
A report that says the installation was made safe in engineering terms will pass; one that relies on salted ground will not.
3. Draft the remedial and testing programme, including how you would get from 14 Ω to an acceptable value and what the inspector should witness.
Read the worked answer
Lowering the electrode resistance.
- Second electrode in parallel. Drive a second 40 mm GI pipe, at least 3 m deep, at a spacing of at least twice the driven depth. Two 14 Ω electrodes do not give exactly 7 Ω because of mutual resistance, but roughly 7.5–8 Ω is realistic if the spacing is adequate. A third electrode gives further improvement and some redundancy.
- Deeper, moister location. Site electrodes away from the concrete apron and the drain, in ground that stays damp; 2.5 m is shallow for this soil.
- Construct the pit properly - charcoal and salt in layers, a masonry cover with a test joint on the earth conductor, and a label.
- Consider a maintenance-free chemical electrode where soil resistivity is high.
Bonding. Supplementary-bond the bench, the welding set frame, the compressor and the water pipe to the earthing system. Welders must not rely on the welding return cable as an earth path.
Testing programme the inspector should witness.
- Earth electrode resistance by the three-pole fall-of-potential method, with stakes at the correct spacing (the 62% rule), and with the electrode disconnected from the installation. Repeat in dry weather.
- Earth continuity from every exposed metal part to the electrode, expected well under 1 Ω.
- Insulation resistance of the welding cables, the compressor motor and the final circuits at 500 V DC - 1 MΩ or better.
- RCCB test: test button, then a dedicated tester for trip current and time at 30 mA, half-rated (should not trip) and full rated (should trip within 30 ms).
- Polarity and phase sequence at each outlet and at the welding set.
- Lightning protection: separate down conductor and electrode, test joint, air termination continuity, and bonding of the overhead structure.