The 0.38 MΩ line–neutral reading on a new board
You are carrying out initial verification of a new three-phase distribution board (400/230 V, TN-S) in a small industrial unit before handover. The board has twelve ways: four socket circuits (16 A and 32 A), a 16 A three-phase motor circuit with a DOL starter, four lighting circuits with integral LED drivers, a fire alarm supply and two spares. A Type 2 surge protection device is fitted inside the board. With all final circuit devices open, loads disconnected and the main switch off, your 500 V DC insulation resistance test between line and neutral gives 0.38 MΩ. Line–earth and neutral–earth both read greater than 299 MΩ. The client wants the unit energised on Friday and says the SPD and the LED fittings 'aren't part of the wiring, so they don't count'.
1. Before concluding the installation has failed, set out your sequence of investigation for the 0.38 MΩ line–neutral result, including what test voltage you would use on each part.
Read the worked answer
One number across a whole board tells you nothing about which part is at fault. Work inwards:
- Confirm the test is valid - main switch off, all devices open, loads disconnected, tester in calibration, leads nulled, reading allowed to stabilise (capacitive charging of long runs).
- Separate fixed wiring from equipment. A Type 2 SPD is connected internally line–neutral and neutral–earth and will drag a 500 V L–N reading down to a few hundred kilohms by itself. Disconnect one SPD terminal and retest. If the board then reads >299 MΩ, the SPD was the cause - that is a test limitation, not a failure of the installation.
- Subdivide. With the main switch off, test each way individually (outgoing line at the device to that circuit's neutral) rather than the board as a whole. Only this localises a genuine low reading.
- Choose the voltage per circuit. Where 500 V could damage connected equipment (dimmable LED drivers, electronic timers, the fire alarm panel), the test may be carried out at 250 V DC, but the minimum acceptable value then falls to 0.5 MΩ (Table 61) and the reduced voltage must be recorded.
- Only a low reading on the cable itself, at the correct voltage, with equipment removed is a genuine insulation failure.
2. The lighting circuit still reads 0.42 MΩ line–neutral after the SPD is disconnected and the lamps are removed. What is your next step, and what would change your conclusion?
Read the worked answer
0.42 MΩ at 500 V is below the 1 MΩ minimum, so you must establish which part is low. 'Lamps out' has not disconnected the integral LED drivers.
- Split the circuit: test the cable from the board with the fittings isolated at each luminaire, then test the fittings separately. A cable reading >299 MΩ alongside low fitting readings means the wiring is sound and the leakage is equipment - re-test that circuit at 250 V with the equipment in place (0.5 MΩ minimum) and record the voltage and the fact that equipment was connected.
- If the cable itself reads low, that is a genuine insulation failure - a nail, a trapped conductor at a back box, moisture in an outdoor fitting, damaged sheath. The circuit must not be energised until it is found and repaired, then retested.
- 250 V is not a way of legitimising a poor reading. 0.42 MΩ at 250 V would still be below the 0.5 MΩ minimum for that voltage and would still be a failure.
3. How do you record the results on the schedule and in the report, and what do you tell the client?
Read the worked answer
The documentation must show what was actually done, not a tidy-looking number:
- Per-circuit results, not one 'whole board' figure.
- Where a circuit was tested at 250 V, the schedule/notes must say so, and the criterion is 0.5 MΩ.
- A circuit tested with the SPD disconnected, or with electronics left in circuit, is a limitation - record it in the limitations/notes and against the circuit, so the next inspector knows the SPD was out of circuit.
- A genuine cable insulation failure is a non-compliance: the circuit is not energised, and it is repaired and retested before the certificate is issued.
- Tell the client plainly: 'The wiring is satisfactory. The low reading is the surge protection device and the LED drivers, which must be disconnected for the test; that is recorded as a limitation. Nothing is being left in a dangerous condition.' Put it in writing, and do not be pushed into recording a test that was not carried out properly to meet Friday.