RCD not tripping

[Ross Trician]

Evening all. I changed a consumer unit last week (was an old wylex one that was "broken" and unsafe)

There were a few problems/faults on the circuits some of which were easily rectified others need further work (if customer will pay for it etc)

My question is this, on the "house sockets" circuit there is a neutral/earth short and when this is connected to the c/u on the rcd protected side it holds the rcd in and wont allow it to trip, (rcd trips fine when circuit is disconnected) why does this occur? I rang wylex technical and might aswell of asked someone in macdonalds.....they were clueless and helpless...was very disapointed in there help.

the thing that concerned me is lets say you do a full rewire and al circuits are fine etc on completion some time later someione puts nail thru the cable and touches Neutral and earth (same as fault now on my circuit) this would then cause the rcd to not trip if a cable was cut for axample.

Also, usaully if you touch neutral/earth together rcd trips but not in this case. why??

can't get my head round this one....help/advice appreciated.

Ross

 

[Davey101]

You're right mate, every RCD protected circuit with a neutral/earth fault I have come across causes the RCD to trip, maybe its something peculiar to Wylex, or a faulty RCD?
Out of interest, is it a dead short, what are the insulation resitance readings?
Interesting question, will be interested to see the answer.
I wont comment on 'the RCD protected side', but I'm sure someone will.

 

[Ross Trician]

hello again, yes it was pretty much dead short, insulation test L/E +299, L/N +299, N/E very low(cant remember exact reading)

I swapped rcd initially thinking it was faulty but still same, then found it to be due to circuit fault. I have put this circuit on non rcd part of board( not compliant I know and put this on cert but thought its best to have other circuits fully covered and the circuits are not left in a worse state than before work)

customer talking of "maybe......maybe" having this and other faulty circuit (garage faulty circuits) sorted.

As i say, I contacted wylex and got no answers, disapointed.

 

[Des 56]

Never had this type of problem so guess work only
Two neutral earth faults with an easier path through each others faults rather than perhaps a high resistance earth path could possibly cause something similar
The rcd would see a single neutral earth because the earth path would be its only route, but maybe not when there are two of them

 

[Davey101]

have you done a loop test on the incomer (a proper one with earths/bonding disconnected)?

 

[WayneL]

.........I have put this circuit on non rcd part of board( not compliant I know and put this on cert but thought its best to have other circuits fully covered and the circuits are not left in a worse state than before work)

customer talking of "maybe......maybe" having this and other faulty circuit (garage faulty circuits) sorted.

As i say, I contacted wylex and got no answers, disapointed.

So, not only have you re-connected a faulty socket circuit, but you're not
providing RCD protection either and - on top of all that- you're going to
admit to it all on the Cert ha ha - Good luck with that!

 

[Ross Trician]

What would you suggest wayne? leave them with no sockets throuought the house? as I said it is not in a worse state than before the work was done.....and fully explained whats wrong and advised it requires imediate attention...Have noted it on cert to cover my back as so to speak, at least it is logged then.


As for the ZE, PFC tests etc, they all ok, the second RCD trips fine as does the "iffy" one once the faulty circuit is disconnected.


What would you chaps have done in that position?? not reconnected the circuit?

 

[ian3012]

my sympathies ross... there's some people on here who work in a perfect world... and then there's some who work in the real world.

i'd have done the same as you...

i have no clue about your cct fault but i do remember doing some sums on rcd's at uni a few yrs ago for an hnc project... i seem to remember you need an 8millivolt difference between neutral and line to generate the 23milliamps( give or take) in order for it to trip... seems like your rcd isn't seeing a p.d. , lets hope you get the rewire if you can't find the fault!

 

[kung]

Good luck Ross ! before a CU change its worth doing both continuity & insulation resistance tests before as if you change the CU and get faults that the customer wont pay for or say it was ok before you started its hard to prove otherwise at least after these tests it gives you and them an idea of any additional cost involved and give them the option of paying you to sort out the faults and not opening a can of worms. Best of luck and i hope all pans out with the minimal hassle for you.

 

[WayneL]

It was only a light hearted comment - hence the smile.

As you ask, the answer is simple-

It would have took two minutes (after removing any loads) to complete
a test that would have shown up that fault straight away - IR test whole
board line/neutral connected together to earth - installations fails - no CU
change until faults rectified.

This is just another example of why you should inspect/test before a CU
change.

If for whatever reason I ended up in your situation, I think I would have had to
fault-find the circuit and sort it out, regardless of who was paying - I don't think
I could sleep if I reconnected a dodgy circuit.

As for the circuits being left in no worse condition than they were - I
don't think that applies to a CU change as you're changing the characteristics
of every circuit (by changing the protective devices) so, really speaking, you're
responsible for the whole installation - EIC & full inspection and test - anything
less is dangerous, in my opinion.

Cheers,

Wayne

Sorry, just realised my mistake - you wouldn't have even had to remove
the loads!

 

[sparky paul]

I had a fault a bit like this one not too long ago,the problem i had was the rcd WOULDNT trip on the 1 x test 30 mA but WOULD trip on the 5 X test.

After a bit of head scratching and replacing the rcd(just in case) i managed to narrow it down to a low level socket in the kitchen, one of the screws had just slightly nicked the nuetral,

Sorted that out and had no problems after.

 

[Davey101]

If it makes any difference Ross, it sounds to me like you have done the best you can (so far).

Maybe it would have been best for the customer if you explained that you couldn't install a new, far safer CU, because the installation has/may have faults on it that will mean you wont be able to leave it in a state that complies with what some new book says, although it may comply with what the old book says.
Perhaps also you should have charged them a couple of hundred quid to tell them what their electrics are like, without actually providing them with the service that they actually wanted from you. Obviously there's always the option of doing the work, but just not turning the sockets back on.

Back on topic, are you saying that the test button on the RCD doesnt work either?
I beleive that some RCD's might not trip with a d.c voltage, but I dont know about it actually holding them on.

Edit- Sounds like Paul has answered your question for you, although I'd like to hear the science behind it. I presume (guess) its because when you test the RCD, not all of the test current is flowing into the earth, some of it at least is going through the neutral, meaning that the live and neutral are not 'out of balance' enough (is that inglish?).
Where's that Lenny bloke, he seems pretty clued up

 

[Ross Trician]

thanks for all the replies chaps.

didn't try the test button during the fault/hassle period, (obviously it did the job once all was testing right with faukty circuit removed.)

I realise that changing the c/u is altering the characteristics of the Board and c/b's but the circuit itself is not altered hence me meaning its not in a worse state. In an ideal world i would of done periodic on the property, identified potential faults, rectified or rewired as required and then changed the board.....do you chaps all do a periodic prior to board upgrade??

what would you do if called to a damaged c/u that requires changing for safety reasons, as was my case, would you have left the circuit disconnected? or potentialy rewired the circuit (at no charge if customer wont pay)

I feel I can sleep easy with what I have done because I have fully explained the faults to customer (and the chap has electical knowledge due to his engineering background)

wish sometimes I had one of those black sticks with white ends.....let me know if your wholesalers stock em...lol

 

[nrbspark]

It is highly recommended to carry out a PIR before changing a consumer unit in my opinion. When invited to do the job I explain it will usually take about 4 hours and charge accordingly and this must be done to give an accurate quote for work.

Then you can explain the faults or deviations found (if any) as a result of the PIR and give the customer a fairly accurate quote (fault finding may be needed) for the work required to enable you to change the consumer unit.

Don't forget that even if there is a dangerous situation by performing the PIR you are only advising the customer of this and it is then their choice if they want the problems sorting out or not, at this point you are only doing what they ask and it should be them that can't sleep at night if they don't have the repairs done.

You have my sympathy mate and I wish you the best.

As for the RCD tripping issue I have come across this with a Wylex board once and it was an issue with the earth fly lead from the RCD that was not terminated correctly to the earth bar, but with the low IR reading cpc to neutral and not a dead short it suggests to me that something is not connected right, have you looked in the attic for alarm or amplifier? Maybe a bit of DIY if the guy is an engineer?

Once again all the best with this one fella.