N
NickD
Just finished some testing at a roughly 30yr old 2 bed flat which has been vacant for a couple of years. Carried out usual 3-stage dead test for socket ring continuity (definitely correctly cross-wired). Only fixed appliance connected to ring (kitchen fan) has been 2-pole isolated out. All inspected sockets appear to be on the ring, no spurs found. Continuity around the ring is near enough what it should be (r1 0.77, rn .85 - aware there is a slight imbalance there - and r2 1.39; it's wired in 2.5+1.5 T&E, 1.39*1.5/2.5 = 0.83). Lowest insulation resistance on the ring is >500Mohm. Ze is 0.1ohm on TNCS. EXpected R1+R2 = (1.39+.77)/4 = 0.54ohms.
Sockets all appear to be original fitment and are a bit rubbish, given that they required a certain amount of coaxing (repeatedly cycling switch on/off, test plug in/out/in/out a few times) to get reasonable readings on. One double socket I actually replaced during testing because one half of the double refused to come below 1ohm R1+R2.
There's a further cluster of three sockets which won't come down below about 1ohm R1+R2 - however if you measure them directly onto the terminals at the rear they read fine, well within 0.54ohm +/- 0.05. figure. Taking the at-front reading gives a Zs of around 1.1ohm and so a fail on 0.4s disconnect time (BS1361 30A, max permitted Zs reading of 0.92). So far, so straightforward. Sockets are pants, fail.
However, if you then do a direct live test for Zs at one of the sockets, it comes in at about 0.45ohms, both on the socket on the front AND (with care and sheathed probe tips) on the terminals at the rear. Now, please put the torches and pitchforks down, because I am not daft enough to rely on this reading to provide a test pass - I know there is a parallel path in the installation, about 0.6ohm, from some point on the socket ring CPC back to the earth bar, because I've measured it. (No other parallel paths down to the mass of earth though, I checked*). So why do I bring this up? Because the point is, when you dead test the socket innards add an extra 0.5ohm, enough to cause a fail; yet when you test live, it's not there, reads 0.45ohm front or rear. ****Why?**** Where's the 0.5ohms gone? (Can it really be down to the metering difference, in the way the meter tests for continuity and Zs? I'm aware they're very different.)
I am in no doubt that I have to fail this - (Ze+measured dead test R1+R2) > Zsmax, end of story. But what am I to make of this 'disappearing' offending extra 0.5ohm? Plus I am concerned that if the customer gets someone else in to confirm, first thing they'll do is Zs on the suspect sockets which will show a pass (yes, a rubbish one unfairly assisted by parallel paths, but then I'm reliant on the tester being on the ball enough to recognise that and not just say "Oh he's talking b****cks, it's fine"). And then they'll think I'm taking the mickey to make up extra work.
Thoughts? Cheers.
(*if you whip the earthing conductor out of the earth bar in the DB and test continuity between earthing conductor and earth bar, there is none [>2kohm], hence there is no other good path to earth in the installation other than the earthing conductor.)
Sockets all appear to be original fitment and are a bit rubbish, given that they required a certain amount of coaxing (repeatedly cycling switch on/off, test plug in/out/in/out a few times) to get reasonable readings on. One double socket I actually replaced during testing because one half of the double refused to come below 1ohm R1+R2.
There's a further cluster of three sockets which won't come down below about 1ohm R1+R2 - however if you measure them directly onto the terminals at the rear they read fine, well within 0.54ohm +/- 0.05. figure. Taking the at-front reading gives a Zs of around 1.1ohm and so a fail on 0.4s disconnect time (BS1361 30A, max permitted Zs reading of 0.92). So far, so straightforward. Sockets are pants, fail.
However, if you then do a direct live test for Zs at one of the sockets, it comes in at about 0.45ohms, both on the socket on the front AND (with care and sheathed probe tips) on the terminals at the rear. Now, please put the torches and pitchforks down, because I am not daft enough to rely on this reading to provide a test pass - I know there is a parallel path in the installation, about 0.6ohm, from some point on the socket ring CPC back to the earth bar, because I've measured it. (No other parallel paths down to the mass of earth though, I checked*). So why do I bring this up? Because the point is, when you dead test the socket innards add an extra 0.5ohm, enough to cause a fail; yet when you test live, it's not there, reads 0.45ohm front or rear. ****Why?**** Where's the 0.5ohms gone? (Can it really be down to the metering difference, in the way the meter tests for continuity and Zs? I'm aware they're very different.)
I am in no doubt that I have to fail this - (Ze+measured dead test R1+R2) > Zsmax, end of story. But what am I to make of this 'disappearing' offending extra 0.5ohm? Plus I am concerned that if the customer gets someone else in to confirm, first thing they'll do is Zs on the suspect sockets which will show a pass (yes, a rubbish one unfairly assisted by parallel paths, but then I'm reliant on the tester being on the ball enough to recognise that and not just say "Oh he's talking b****cks, it's fine"). And then they'll think I'm taking the mickey to make up extra work.
Thoughts? Cheers.
(*if you whip the earthing conductor out of the earth bar in the DB and test continuity between earthing conductor and earth bar, there is none [>2kohm], hence there is no other good path to earth in the installation other than the earthing conductor.)
and a resistance that persists even after 230V applied is not a mirage.
