On a recent consumer unit change we kept getting a high R2 reading on the ring finals, even after we’d found and fixed every fault on them. It turned out to be the cable. It was 2.5mm² twin and earth with a 1.0mm² earth, the way 2.5mm² cable was made from about 1970 until 1981, and that cable will never test like modern cable with a 1.5mm² earth. I hadn’t come across it before, so here’s the job and the figures, in case it saves another electrician some time.
What we found on the job
It was a board change with some fault finding thrown in. When we tested the ring finals, two of them didn’t look right: one had no continuity on the earth at all, and the other was reading far higher than it should have.
We traced both. There were loose connections and a couple of earth faults, so we remade and tightened everything, sorted the earths out and put it all back together. The readings came down, but the R2 was still higher than you’d expect from a ring test.
When we took a closer look at the cable, the line and neutral were 2.5mm² as you’d expect, but the bare earth was thinner than normal: 1.0mm², not the 1.5mm² you get in 2.5mm² twin and earth today. With that earth the readings were never going to match modern figures, however many times we remade the connections.
When 2.5mm² cable had a 1mm earth
The IET covers this in its guide to estimating the age of an electrical installation. Flat twin and earth in 2.5mm² was made with a 1.0mm² earth from when cable sizes went metric, between 1969 and 1972, until 1981. From 1981 the earth went up to 1.5mm², the same year the 15th Edition of the Wiring Regulations came out (Voltimum has a short history of the regulations).
So if a house was wired, or added to, in the 1970s, there’s a fair chance some of this cable is still in there. Plenty of houses have had work done since, so you can find old and new cable on the same circuit, and one leg of a ring can test differently from the other.
Why a 1mm earth gives a high R2 reading
A thinner conductor has more resistance, and the difference is bigger than you might think. These are the standard figures for copper at 20°C from the conductor resistance table:
| Conductor | Resistance per metre at 20°C |
|---|---|
| 2.5mm² (line and neutral) | 7.41 milliohms |
| 1.5mm² (the earth in 2.5mm² twin and earth today) | 12.10 milliohms |
| 1.0mm² (the earth in 2.5mm² cable made before 1981) | 18.1 milliohms |
Metre for metre, the old earth has about 50 per cent more resistance than the modern one, and it shows up in two places when you test a ring.
On the end-to-end test, r2 normally comes out at about 1.67 times r1 with a 1.5mm² earth. With a 1.0mm² earth it’s nearer 2.5 times r1 (2.44 with the figures above), so if you’re working to 1.67, a perfectly sound old ring looks like it’s got a problem.
At the sockets, R1+R2 is 19.51 milliohms per metre for 2.5mm² cable with a 1.5mm² earth, but 25.51 milliohms with a 1.0mm² earth. That’s about 30 per cent higher before anything is actually wrong.
A worked example
Take a ring 50 metres long end to end. The line conductor, r1, comes out at 50 × 7.41 milliohms, about 0.37 ohms. The earth, r2, is about 0.61 ohms if it’s 1.5mm², or about 0.91 ohms if it’s 1.0mm².
Once you’ve cross-connected, R1+R2 at each socket is roughly a quarter of r1 plus r2. That’s about 0.24 ohms with modern cable and about 0.32 ohms with the older cable, on a ring with nothing wrong with it.
How to check for it
The quickest check is to look at the bare earth at the board or an accessory next to an offcut of modern 2.5mm² cable. Once you’ve got them side by side, the 1.0mm² earth is obviously thinner. Then work out what that length of cable should read, using 18.1 milliohms per metre for the earth, and compare like with like instead of against modern figures.
Is it safe, and what goes on the certificate?
A higher reading from a thinner earth isn’t a fault on its own, but it still has to be checked properly.
The earth fault loop impedance has to be within the limit for the circuit’s protective device. For a 32 amp Type B breaker, BS 7671 gives a maximum of 1.37 ohms, which works out at about 1.1 ohms against a reading taken at room temperature (the IET explains why the two figures differ).
The earth also has to be big enough to carry the fault current safely until the device trips. That’s the adiabatic check in Regulation 543.1.3, and with a 1.0mm² earth you need to work it out rather than assume it’s fine.
Whatever you find, the readings go on the certificate as they are, with the cable noted, so the next electrician to test the circuit knows what they’re looking at.
If you’re a homeowner, cable with a 1.0mm² earth is at least 45 years old. It can still be in serviceable condition, but it’s worth knowing what you’ve got. An EICR will show how the whole installation is doing, and if the wiring is getting towards the end of its life, our guide to how often a house needs rewiring and our house rewiring page explain the next step.
A tip for other electricians
If the R2 on a ring stays high after you’ve remade every connection, check the size of the earth before you start lifting floorboards. It would have saved us some head scratching on this one, and it’s worth keeping in mind whenever you’re testing an installation from the 1970s. It works the other way too: on an EICR we did in Bedminster, a ring’s earth read about four and a half times what the cable should give, far more than a thinner earth could explain, and that one needed further investigation.
Consumer unit changes and fault finding in Bristol
We’re NAPIT registered and TrustMark Trading Standards Approved (licence 4257835), based in South Bristol. Every consumer unit replacement we do includes full testing of the circuits, and if the readings look wrong we’ll track down why. A board change is £680 + VAT for a 6-way board and £780 + VAT for the 8 to 12-way board most houses need, and fault finding starts at £140 + VAT for the first visit. If you’ve had odd test results or you’re planning a board change, get in touch and we’ll give you a written price before we start.