The garage spur that reads 1.28 Ω
A 1980s semi-detached house, TN-C-S, 230 V, 100 A cut-out, 25 mm² tails, single consumer unit. A 32 A Type B MCB supplies a ring final circuit to BS 1363 sockets, and the circuit is protected by a 30 mA RCD. Ze measured at the origin is 0.35 Ω. End-to-end continuity gives r1 = 0.44 Ω, rn = 0.44 Ω and r2 = 0.72 Ω; R1+R2 at the mid-point socket is 0.29 Ω and the front room socket reads 0.71 Ω. The garage socket, fed from a single spur taken from the back of a kitchen socket, gives Zs = 1.28 Ω. Table 41.3 gives 1.15 Ω for a 32 A Type B MCB. The client says the garage socket has 'always been a bit funny' and asks whether you can just write it down as a C3.
1. Is the 1.28 Ω a genuine non-compliance? Set out how you verify the measurement before you code anything.
Read the worked answer
Verify the measurement first - a number you cannot reconcile with the circuit is not yet a finding.
- Tester in calibration, leads nulled, low-current (no-trip) loop test, good contact in the socket, no load on the circuit, tested at the socket rather than through an extension lead.
- Reconcile with calculation: Ze (0.35) + R1+R2 at the ring mid-point (0.29) = 0.64 Ω. A 3–4 m spur of 2.5/1.5 adds roughly 0.06–0.08 Ω, so you would expect about 0.70–0.72 Ω at the garage socket. 1.28 Ω is around 0.5 Ω higher than the circuit can legitimately produce.
- That gap means resistance somewhere in the path - a loose line or CPC terminal at the spur origin, a poor connection in the socket, a damaged or partly parted conductor, or a longer spur than you have been told. Investigate before coding.
- Once the circuit is proved sound, apply the tabulated value properly: Table 41.3 gives 1.15 Ω, but the values are at 20 °C; under fault the conductor heats and resistance rises by roughly 20%, so the measured value is compared with 0.8 × 1.15 = 0.92 Ω. The ring sockets at 0.71 Ω comply; 1.28 Ω fails even before the 0.8 factor. As measured, fault protection by the MCB is not achieved - potentially dangerous (C2) until proved otherwise.
2. On opening the spur you find the CPC of the 2.5 mm² T&E terminated in a connector block alongside the ring CPCs; the screw is not tight and the copper is tarnished. Classify and state your actions.
Read the worked answer
Classification: C2 (potentially dangerous). The protective conductor path is unreliable and could overheat under fault current, but at the moment of inspection nobody is in immediate danger. C1 would be arguable if there were arcing, burning, or a disconnected CPC at an accessible metal accessory.
Actions:
- Isolate the circuit before working on it.
- Remake the termination correctly - appropriate connector, clean copper, no stray strands, tight.
- Re-verify continuity of the CPC, re-measure Zs and compare with 0.8 × tabulated, and re-test the RCD.
- Record the defect found and the remedial action taken. The report describes the condition at the time of inspection, so state clearly what was found and what you rectified - do not simply omit it because you fixed it.
3. Suppose the terminations are sound and Zs genuinely is 1.28 Ω on a compliant spur. What are your options, and can you rely on the 30 mA RCD?
Read the worked answer
In order of preference:
- Reduce the device rating on the spur - a 20 A Type B has a tabulated maximum of 2.13 Ω (0.8 × = 1.70 Ω), which 1.28 Ω meets. But downrating a device on a ring final is a design decision: the ring is a 32 A circuit and the loading must be checked first.
- Increase the line/CPC csa on the spur, or shorten the run.
- Use the RCD for fault protection. BS 7671 permits an RCD to be used for fault protection in TN systems, and Table 41.5 gives a maximum Zs of 1667 Ω for a 30 mA device, so 1.28 Ω is comfortably met. This is a design change, not a testing decision: it must be confirmed, the RCD verified (½I, I, 5I and operating time), and the arrangement recorded on the schedule. The same device can provide additional protection for the socket outlets as well, but the designer must confirm that this is what is intended.
- Doing nothing is not an option. Where fault protection is not achieved the circuit is potentially dangerous (C2) until corrected - it is not a C3, whatever the client would prefer.