Bernard,
Thanks for the diagram. The N-E fault/connection was at the N end of the IH element, which had split (I saw it when it was changed again two days ago by the other electrician). That's that particular instance. We don't know why it recurs. I was also shown some of the previous failed IHs, but I did not get a chance to do a post-mortem on them.
Mike,
The supply comes through a meter box outside, including the main RCD (4-pole, 500mA). Between there and the first DB in the villa, there is 4-core SWA. I've just been checking my photos. Thanks for making me look... In the meter box, the armour is connected just to the earth rod. In the indoors DB, an earth emerges from conduit along with 3P+N, but I don't see the end of the SWA and there is not enough room in the conduit for the earth to be banjoed to the armour. ...Suggests there is another box (or hidden connection) somewhere in between. It may turn out to be in the bottom of the meter box (cavity going down to soil level). When I'm next on-site, I'll investigate that further.
That investigation aside, I would not expect N-E to be tied in the indoors DB, as that is downstream of the main RCD - constituting a N-E fault visible to the main RCD.
Thanks for the diagram. The N-E fault/connection was at the N end of the IH element, which had split (I saw it when it was changed again two days ago by the other electrician). That's that particular instance. We don't know why it recurs. I was also shown some of the previous failed IHs, but I did not get a chance to do a post-mortem on them.
Mike,
The supply comes through a meter box outside, including the main RCD (4-pole, 500mA). Between there and the first DB in the villa, there is 4-core SWA. I've just been checking my photos. Thanks for making me look... In the meter box, the armour is connected just to the earth rod. In the indoors DB, an earth emerges from conduit along with 3P+N, but I don't see the end of the SWA and there is not enough room in the conduit for the earth to be banjoed to the armour. ...Suggests there is another box (or hidden connection) somewhere in between. It may turn out to be in the bottom of the meter box (cavity going down to soil level). When I'm next on-site, I'll investigate that further.
That investigation aside, I would not expect N-E to be tied in the indoors DB, as that is downstream of the main RCD - constituting a N-E fault visible to the main RCD.
