why is this RCD tripping?

[Dave_]

Went to a job today to try and rectify a fault on a ring circuit.

I visited the property some time ago and noticed the main ring was not RCD protected on a 17th ed high integrity board that had been very recently installed. I assumed a fault was present. No paperwork...

I disconnected all loads, tested,split traced etc until I discovered it was a leg in the kitchen that was most likely a three way joint that fed the boiler too, a pain as the boiler would need re supplying and the ring splitting and derating to 20A or so.

The thing was though that the fault was between L & N (0.01 M ohms). L - E, N - E 500 M ohm or so. I tried to put the circuit on the RCD as a minimum (customer wasn't prepared to pay at that time for what I wanted to do above) and the RCD would trip when operating a light switch (same RCD) or a piece of equipment. Why is this?

Also what dangers are exactly present with a poor IR reading?

Thanks

Dave

 

[markc123]

A poor IR reading means the condition of the insulation around the cable is degrading. Could lead to a dead-short. The reason behind the IR other than checking the cable is to dertimine the age of the installation - thats why when readings get down towards 2Mohms you should really be looking at replacing cable where possible - if the client allows it obviouslly.

Firstly you need to understand the mechanism of an RCD - an RCD trips when theres an inbalance between the outgoing live and returning neutral - a leakage to earth.

So, with this in mind and without seeing the install myself - i would hazard a guess that theres probably 2 problems here. The Lighting could may be a seperate problem.

Did you disconnect / switch off the boiler when doing the IR readings on the ring? Also, theres sometimes a few hidden items still plugged in that even the most trained eye can miss ... e.g. a electric hob that's plugged in.

When you put the ring on the RCD did you make sure you changed the neutral aswell? Have you asked the customer whether that light has always taken the RCD out? Even before you stuck the ring on it...

 

[guest119]

Should you not be in the diy section with this stuff?

Boydy

 

[Dave_]

I fully understand the operation of an RCD...

Definitely unplugged everything, I am experienced and understand everything you say.. Definitely no fault to earth whatsoever, of course I moved neutral in DB.

Don't understand why RCD would trip.

Thanks for the comment Boydy, I hoped I wouldn't get a childish response but there is always one!

 

[markc123]

You done an RCD test at 1x and 5x?

 

[Deleted member 9648]

Went to a job today to try and rectify a fault on a ring circuit.

I visited the property some time ago and noticed the main ring was not RCD protected on a 17th ed high integrity board that had been very recently installed. I assumed a fault was present. No paperwork...

I disconnected all loads, tested,split traced etc until I discovered it was a leg in the kitchen that was most likely a three way joint that fed the boiler too, a pain as the boiler would need re supplying and the ring splitting and derating to 20A or so.

The thing was though that the fault was between L & N (0.01 M ohms). L - E, N - E 500 M ohm or so. I tried to put the circuit on the RCD as a minimum (customer wasn't prepared to pay at that time for what I wanted to do above) and the RCD would trip when operating a light switch (same RCD) or a piece of equipment. Why is this?

Also what dangers are exactly present with a poor IR reading?

Thanks

Dave

0.01 megs L-N almost certainly means a load still connected. L-N faults on wiring are rare to say the least,because on T/E there is a bare earth between the energised cores faults will invariably be to earth. If you are IR testing L-N always do a 'soft test' at 250v first unless you are absolutely certain there are no loads connected(you can only really be certain on a circuit you have wired yourself)....blundering straight in with 500v is likely to result in frying something expensive....I know,got the T shirt to prove it.
A L-N fault will not trip an RCD,only an OCPD,you need tolook elsewhere for the cause of the tripping.It's likely there is a N-E fault on another circuit on that RCD,these sometimes only show up when a load is applied causing a volt drop which enables some of the current to return via the earth.

 

[telectrix]

as wirepuller says. also a good idea to do a ramp test. always the possibility of a dodgy RCD.

 

[Octopus]

There MUST be a load still connected and if the customer won't pay, then simply put it back on the non RCD side and walk away. If you are doing work that requires a cert, then add a comment recommending investigation on the socket circuit.

I'd also be asking the customer who exactly changed the board and why they didn't leave any paperwork - and suggest they get them back.

 

[telectrix]

wasn't me, guv., honest.

 

[Welchyboy1]

Mmmmm i still reckon these something plugged in, theres ALWAYS something i miss lurking in the back of a kitchen cupboard

is it still tripping with all the appliances removed, i know you said you unplugged them all but its not an appliance is it?, as with those IR readings it seems the circuit itself is ok, there has to be some kind of earth leakage coming form something

have you got an earth leakage tester?

 

[Octopus]

And you don't need a fault to earth to trip a RCD do you?

I had one last winter where the 1960's lighting circuit was tripping the RCD, and the circuit didn't have a CPC, in fact I traced it to a switch drop, and mice had nibbled their way right through the cable !

 

[telectrix]

And you don't need a fault to earth to trip a RCD do you?

I had one last winter where the 1960's lighting circuit was tripping the RCD, and the circuit didn't have a CPC, in fact I traced it to a switch drop, and mice had nibbled their way right through the cable !

possibly leaking to E through the fabric of the building. had that once where a cable had been extended behind a CU with choc block in a damp wall.

 

[the pict]

Ambi pure plugged in behind a curtain, I hate those bloody things, I now ask the customer if they have any in the house mind you I get some puzzled looks maybe they think their house stinks
Pict

 

[Steve_P]

possibly leaking to E through the fabric of the building. had that once where a cable had been extended behind a CU with choc block in a damp wall.

Found a 6 foot patch of plaster live once coz of damp. The poor old sod renting the cottage had got a belt off the electric blanket switch, half way up the flex!

 

[Octopus]

possibly leaking to E through the fabric of the building. had that once where a cable had been extended behind a CU with choc block in a damp wall.

No it wasn't, no cpc. The mice had knibbled their way through the copper as well, and the cable was lying on a joist in the loft. With the light off you could see that the ends of the copper were "touching" and if the cable was moved it sparked and off went the RCD, so it must have been sensing the "spike" in the inbalance.

 

[telectrix]

i'll agree with that.

 

[Richard Burns]

There could be a shared neutral on the ring, such as if the boiler was separately supplied and then someone partially moved it on to the ring, say just to get a line supply.
Or there could be a mixed up neutral on another circuit but there was not enough leakage to cause the RCD to trip until the load of the ring was placed on that RCD.

 

[the pict]

Plasterer mate of mine got a belt of a wall turns out there used to be wall lights, yep you guessed it still live taped up and squashed into a hole
P

 

[Englishman]

Had an rcd tripping yesterday when one side of a double socket was switched on, regardless of anything being plugged in.

Had left about an hour earlier after doing an eicr so I knew the rcd was fine, quickly worked out it was the kitchen and that one socket, kettle had been plugged in and it went. Put new socket and it held, was obviously something it didn't like after I'd fiddled about with it first time round.

As for the OP, have you checked outside the kitchen as well for anything spurred off the circuit?