Intermittent ring continuity

[Geoff Clare]

I had an EICR done on a house I'm buying and the electrician reported "no ring continuity" on the principal ring main (supplying 20 sockets). The seller agreed to remedy this (plus some other faults), but when their electrician went to do the work he tested the circuit and said the ring continuity was fine.

Obviously there are various possible explanations involving faulty test gear or mistakes made, but what I'd like to know is what is the likelihood that both test results are genuine, i.e that there is some sort of intermittent fault? Both electricians are registered "competent persons" (one NICEIC the other NAPIT).

Possible causes I've thought of:

A break in a conductor that makes contact at certain temperatures but not others, due to thermal expansion.

A break in a conductor near the CU which makes contact or not depending on how it is moved to attach to the test gear.

Do either of these (or one I haven't thought of) seem plausible?

If only one conductor did not have continuity, would the electrician have given the impedance of the other two in the EICR? (He put "---" in all three columns.)

[Tim Watts]

Possibly - the test is done using very low current so a feeting connection might show up with some continuity - but even so, I would have thought the impedance of that loop would have been measurably high if the connection was that broken.

That's curious - unless there is a convention to abort the test at that point?

[ARW]

You would expect both electricians to have recorded similar r2 end to end readings (the earth readings).

So was the break on the live or neutral or both?

[Geoff Clare]

Unknown. The only information from the first electrician I have to go on is the fault statement "No ring continuity on circuit 2" and what he put as the test measurements for that circuit:

r1 "---", rn "---", r2 "---", R1+R2 0.35, Zs 0.49

The second electrician measured:

r1 0.26, rn 0.28, r2 0.4, R1+R2 0.18, Zs 0.58

There is another ring main for which the first electrician measured:

r1 0.24, rn 0.24, r2 0.4, R1+R2 0.24, Zs 0.32

It did occur to me that the second electrician might have measured this one instead of the faulty one, but the Zs figures looked wrong. However, now that I look at R1+R2, perhaps I shouldn't have dismissed that possibility. Both are a "circuit 2", but on different CUs.

[tabbypurr]

We don't know. We could guess for you, but is that useful?

NT

[John Rumm]

The most likely cause is a loose screw terminal on a socket somewhere along the ring - just plugging in or removing a plug could then be enough to change the state.

[Geoff Clare]

Thanks. With that clue, I think I might have the answer.

One of the other faults was a "melted socket". Perhaps that socket had a screw that wasn't tightened sufficiently and when the plastic deformed, the movement pulled a conductor out.

I assume standard practice would be to test a circuit after making changes to it. So replacing that socket seems the most likely reason for the continuity to have been restored between the tests.

[Tim Lamb]

Yes. Mother in law had an intermittent problem in a '60's built flat.

Several screws found with a least one full turn to tight in the ring main.

[John Rumm]

Once you have had an overheating connection its not uncommon to find the copper tarnished and discoloured at the end - making it harder to get a good termination again.

[Andrew Gabriel]

My first house was like that. It had no heating so I'd bought a 2kW convector to keep me going. As it cycled on and off, the bedside light got noticeably brighter and dimmer. A voltmeter revealed I lost 10V when the heater came on. The cause was simply that none of the terminal screws in the sockets were done up tight, and fixing that fixed the voltage drop. It was wired around 1972 I believe, but had been modified since then in various places. I gradually rewired it as I worked on each room over time.

[Dave Plowman (News)]

I've seen this too. Pal had a new kitchen installed at vast expense which included fitting new sockets - in the form of an additional ring. One day he mentioned a couple of the sockets had died, and I ended up having a look, as the original kitchen fitters had ceased trading.

And as with yours, discovered the terminals hadn't been tightened.

I checked the entire new ring, which had been added to an existing CU with a spare way. The screws inside the CU had been tightened properly, but none of the others. I initially blames the tilers as it's possible they had removed the sockets, but one which wasn't over tiles was the same.