Ring Main no Earth!

[MikeTheGeek]

We recently asked for an electrical safety test at a property we moved into, during the test they identified one of the ring mains didn't have earth continuity, as this was a ring main that I had recently changed the face plates on, I thought it must of been something that I had done. I've been round every socket with a socket tester (socket & see 22) and it's not reporting any faults! Would removing the face plates to do a visual check, to see if the earth isn't loose, give me anything additional or should of the socket tester alerted to loose/poor earth connectivity... Any pointers, would be great.

 

[DetlefSchmitz]

Yes, you should be able to see the loose wire with luck. You should get a multimeter and (with the mains off) you should be able to take the eath wires out at any given socket and see continuity between them. If you don't, then in each diection you should see continuity between N and Earth. If you don't then that it the direction to carry on investigating the next socket in.

 

[chivers67]

One Earth is probably popped out and each socket is still earthed around the Ring as the other Earth in that particular socket is no doubt still terminated.

 

[DetlefSchmitz]

One Earth is probably popped out and each socket is still earthed around the Ring as the other Earth in that particular socket is no doubt still terminated.

However, of course, it is not adequately earthed without both conductors.

 

[sparkwright]

As said, check your work.

TIP: Double the ends over on the earth wires. As these wires are smaller, they tend to slip out of the termnal.

Bending the ends over give more surface area for the terminal screws to grip to.

 

[chivers67]

Yes very true - though i'm trying to explain why the socket & See isn't reporting a Fault as such.

 

[DetlefSchmitz]

Yes very true - though i'm trying to explain why the socket & See isn't reporting a Fault as such.

No quite right! Just wanted it not to sound like it's ok to leave it, which of course you didn't suggest, but may have been assumed.

 

[GaryGarden]

Check you have not screwed down on to the green / yellow sheathing instead of the wire on the face plates that you re-fitted, as this if this was done on 1 side of a socket your socket & see would still show a good earth.

 

[sparkwright]

Yes very true - though i'm trying to explain why the socket & See isn't reporting a Fault as such.

You have, and you explained it very well.

 

[ericmark]

The socket & see 36 tester will show if the earth loop impedance is better that 1.8 Ω, and 92 Ω one assumes the 1.8 Ω for TN and 92 Ω for TT, it is not really low enough for a British ring final which should be better than 1.36 Ω but for a cheap plug in tester it is about as good as it gets, it also uses your body as a reference to check of incorrect pins it seems. However the socket & see 22 is rather a cheap version and does not test the loop impedance or RCD at even the most basic level. The 36 uses 7 mA line - earth which should not trip an RCD to do loop test and 30 mA for RCD test but is does not measure time, but should identify a fault RCD or one frozen due to DC.

At around £50 the 36 is far more expensive to the 22, but far cheaper than the socket & see PDL234 and even that one does not give an actual reading in ohms. The Martindale EZ165 seems a little cheaper to the socket & see 36 tester, also the socket & see 34 tester does loop. And the Kewtech Loopcheck 107.

The Martindale EZ2500E-Ze seems to be the cheapest true earth loop impedance tester at around £130 at a quick hunt, with a true loop tester you may be able without removing the sockets to detect the area of the fault, but the go/no go testers do show when there is a fault in most cases, but not good enough to help trace where or completed the paperwork required. Both the loop impedance tester and RCD testers are expensive bits of kit.

Oddly the insulation tester using 500 volt I have picked up new for £35, these are in comparison cheap.

 

[JohnW2]

However, of course, it is not adequately earthed without both conductors.

Strictly speaking, that's not necessarily true (unless you think that sockets on radial circuits are "not adequately earthed") BUT, as has been said, it certainly needs to be investigated since, if one of two 'earth' conductors have popped out of a socket terminal, then the other one probably is not satisfactorily connected to the terminal (and may be close to also 'popping out').

Kind Regards, John

 

[DetlefSchmitz]

Strictly speaking, that's not necessarily true (unless you think that sockets on radial circuits are "not adequately earthed")

No, I'm saying that 5mm of L and 1.5mm of Earth may not meet the requirements after calculation. So without the measurement and calculation the only alternative is to fix it.

 

[Harry Bloomfield]

Another check you could do, if you feel safe to do it, is in the consumer unit....

Power the consumer unit down, disconnect both earth wires of the ring from the earth terminal in the consumer unit and check for continuity between the two earth wires. Obviously, there should be continuity of the earth wire from the consumer unit and back around the ring.

 

[JohnW2]

No, I'm saying that 5mm of L and 1.5mm of Earth may not meet the requirements after calculation. So without the measurement and calculation the only alternative is to fix it.

Ah, I see. I think that is probably a 'hyper-theoretical' point which is not going to be an issue in practice.

The "5mm²"** of L is only relevant in that it reduces (in comparison with 2.5mm²) the Zs (at the socket of interest) and thereby increases PFC. I have never heard it suggested that, in the case of a socket very close to CU on a 2.5mm² radial circuit, in an installation with a very low Ze (hence an extremely high PFC at that socket), the 1.5mm² CPC may not be (adiabatically) adequate, have you?

[ ** in any event, in terms of Zs, the two L's of a 2.5mm² ring final would only be 5mm² at the very middle of the circuit (where the Zs would be relatively high, anyway). Anywhere else in the ring, the effect of the two Ls on Zs would be less than the equivalent of 5mm²]

Kind Regards, John