RCD keeps tripping..

[Saif123]

First off - thanks for all your help on these and other forums - this site and others like it have been invaluable for me in renovating my home and I have found answers to all my questions through existing forum responses - but this one stumps me and is unique to my situation. I have spoken to 2 or three sparks I know and they are stumped. (but have not been round with Earth leakage testers etc, admittedly).

Anyway the story goes like this - Sunday I was plumbing in a new toilet and water leaked through subfloor within a few metres of where the consumer unit is housed. From then on the RCD keeps tripping when any sort of intensive load is put on any of the circuits protected by it.

For example - I can plug in radio, laptop, mobile phone charger and cordless drill charger all at same time and no problem. Once I try to fire up a 2kw convector heater, an angle grinder, an SDS drill or a kettle the RCD trips.

I have tried this on the upstairs ring, downstairs ring and the kitchen ring - same thing happens each time.

As part of my troubleshooting I have:

removed sockets and checked earth and neutral connections are ok

replaced the 62amp Hager RCD with a brand new one - exactly same thing happens.

disconnected the upstairs and downstairs ring MCB's to try and use only kitchen circuit - same thing

As a final (and temporary) test I moved the upstairs ring MCB across the consumer unit to a spare slot unprotected by the RCD. I plugged in dehumidifier, 2kw convector, 500w oil rad, and fired up the angle grinder and although all stayed working the MCB still tripped (even though the load was not (or not supposed to be) going through the RCD side of the CU.

This is when I decided this was too weird and to initially seek some help on here before calling in an electrician!

Could it be too much moisture in or around the consumer unit? I have left the upstairs ring unprotected by RCD to leave the dehumidifier and heaters running for a few hours in the hope this dries the air out a little and possibly resolves it - but if anyone has other suggestions or thoughts on what the issue could be I would be pleased to hear them..

PS - sorry for such a long first post!

Thanks
Saif

 

[rebuke]

A bit of water is going to have got into either the CU (although that would be fairly obvious, especially if you've been opening it up etc), or more likely a fitting / junction box somewhere. This means there is a very slight leakage (less than 30mA) between either the live or the neutral and earth, which when you have a bit more leakage from some high power appliance, pushes the total over 30mA, and causes the RCD to trip.

As for why it still tripped when you moved the MCB to the non RCD side, did you also move the neutral from the RCD's neutral bar, or did you leave it where it was. If so, then your return current would have gone through the RCD, causing it to detect an imbalance, and then trip - this wouldn't have stopped anything on the non protected side working though as RCDs generally only switch the live...

Edit to add: As to tracing the source of the problem, first try switching all breakers on the RCD side off, and try one circuit at a time with some high power appliance - that might help you narrow it down to which circuit is the problem. Unfortunately, I suspect from what you've described that the problem is actually between a neutral and earth, rather than a live, at which point switching off MCBs won't help you. You'd have to disconnect one neutral at a time from inside the CU (make sure the associated breaker is off as well, to avoid having a neutral with return current present floating inside the CU), and see if you can find which one. At that point it's a matter of tracing the cabling, and finding what's got damp. Depending on how much water it is, you might be OK just to leave it a few more days and it'll dry out on its own...

 

[Saif123]

No, I didn't move neutral so that explains the RCD still tripping - and yes the power tools etc kept working - so not weird at all then!

I did take up the floorboards but there are no junction boxes, cables or anything similar near where the water leaked. CU is the other side of a new partition wall from where the leak was but floorboards in the cupboard where CU is located are not all laid so I guess mositure ingress could be the cause (although CU cover etc was on - just the flap for switching MCB's was open) - I had not touched the CU until this ault happened as was not confident in my competence on that front.

Anyway thanks for your inoput, I will leave the Dehumidifier and heater running for a while on the unprotected MCB and see if that helps. Unless that is a really stupid thing to do?

thanks
Saif

 

[Spark123]

Odds are on there is a Neutral - Earth fault on an RCD protected system and your supply type is TN-CS (PME)

 

[bongos]

I plugged in dehumidifier, 2kw convector, 500w oil rad, and fired up the angle grinder and although all stayed working the MCB still tripped (even though the load was not (or not supposed to be) going through the RCD side of the CU.

You've answered your own question here as to the MCB trip. In this scenario probably because you overloaded the ring final circuit by powering all these items on the same circuit. Was it a 32A MCB?

As for the RCD tripping...it can often be a nightmare to locate a fault. The 'sub floor' that you say got wet - are there any junction boxes hidden under this section of floor? Could they still have water in if they got wet? The fact that the RCD keeps tripping means you have an earth fault occuring somewhere. When you put in the new RCD was everything wired up as it was?

A pic may help the pros on here but you might have to bite the bullet and call in a spark. They will be able to do an insulation test among other things and locate what circuit is causing the problem. Fixing it though, may take longer.

 

[rebuke]

Are there any sockets in the partition wall you mention - as if so I'd suggest isolating at the CU, and then opening them up to check for any damp, as water could quite easily have run down the inside of the wall...

 

[rebuke]

bongos: I assumed that to be a typo, and he meant the RCD still tripped - if an MCB tripped but the equipment kept working, that's much more worrying, as it indicates a serious wiring fault!

 

[Saif123]

Bongos - my mistake - I meant the RCD stil tripped - not the MCB - MCBs haven't tripped under any of the testing.

No junction boxes beneath sub floor - and have tripped MCB's for all RCD protected circuits and then switched one at a time on and fired up an angle grinder and it trips - this happens on each and every circuit with the MCB's tripped for all other circuits.

Another weird thing is the SDS Drill works on low speed but cuts out on high speed.. I guess once it draws more power the imbalance occurs and the RCD does its job..

I guess I will dehumidify for a day or so and then call in a sparky if its still a problem... is running dehumidifier of an MCB not protected by RCD a really bad idea?

 

[Spark123]

Like I said above, smells of a N-E fault with TN-CS.
Can you get a sparky around to test with an Insulation Resistance tester (AKA Megger) to check all the circuits individually?
Alternatively if you switch off the MCB in turn for each circuit and also disconnect the Neutral leaving the board for the circuit you have disconnected it may help you pinpoint the faulty circuit.
Make sure you isolate the supply before working inside a consumer unit and be very carefull of the meter tails as these will still be live (unless you're lucky enough to have an isolator upstream).

 

[Saif123]

Hi Rebuke - thanks for reply - yes I meant RCD... no sockets on partition - just a batten lampholder but thats not been connected up yet.

Probably best to sit it out for a while by the sounds of it...don't want to go down the road of disconnecting the neutrals in the CU - replacing RCD was plenty for me to try on the CU side of things!

thanks to all for your input - so quickly too!

 

[rebuke]

Strictly speaking it's against the regs (all general purpose socket outlets must be protected by an RCD, and I suspect lots of your cabling is at less than 50mm from the surface), but given it's a temporary thing, and assuming your dehumidifier doesn't state that it must be RCD protected then you'll probably be OK (obviously don't quote me on this, I'm not a professional electrician, so don't take my advise as 100% accurate, and to be absolutely safe you shouldn't run without an RCD etc etc). I guess you could always get a plug in RCD and use that with it to keep the dehumidifier itself protected...

 

[ericmark]

I agree with Rebuke in that seems like a neutral earth problem.
With no current draw neutral and earth are the same voltage. As you draw current because of the resistance in the cable the neutral voltage starts to rise above earth voltage when the voltage raise is enough for 20ma to flow then the trip will operate.

But the actual fault can be on any circuit protected by the RCD.
And the load which results in the voltage difference between earth and neutral does not need to be on a RCD protected circuit.

I hope the dry out helps. The problem with water is if it touches anything hot it cools it and if this is a box etc. it will cause the air to contract and then draw water into the void. However to dry out takes a lot more. My daughter overfilled the bath and the water went down the light fitting cable and caused a very small crack in the bulb though which it filled the bulb with water. When I came home a month later the bulb was still full of water which they had not noticed and the had done without dinning room lights for a month.

I hope that gives you some pointers as to where to look.

 

[rebuke]

One thing to mention, is given you've replaced the RCD - did you make sure you did all the terminals up fully tight - as if not then you can run into issues if you have a high load (such as an electric shower), as it can cause charring etc...

Also, you should get the replacement RCD tested, as while it appears that it is working because it's tripping in your situation, that doesn't tell you that it's necessarily tripping at the correct point etc. If you do have to get a spark in, he/she will probably test it anyway, but mention that you've replaced it to make sure it gets done.

 

[Saif123]

Ah, yes good Idea will pop to screwfix and get a plug-in RCD - thanks for the pointer.

Most cables are running under the subfloor (not by me - by council contractors who did it as part of a grant for the old lady who lived here previously). They were then surface mounted up the walls to the socket outlets or down the wall to light switches. It was a bit of an eyesore - but probably quite safe.

I have now chased all the cables into the wall - and as you guessed not 50mm in (did not come across this requirement in my reading) They are probably between 25 and 30mm in and covered with PVC channeling (http://www.screwfix.com/prods/74624/Electrical-Supplies/Conduit/Rigid-Conduit/PVC-Channel-1-Pack-of-25) then covered over with bonding coat and skimmed. Now someone is going to tell me this is a bad way to do it and I should start over?!

 

[rebuke]

Nothing wrong with it, it's just any cables buried at less than 50mm depth in a wall not in earthed metal conduit (or of an appropriate type such as SWA / pyro etc) require RCD protection. For socket outlets this is fine since they require it anyway, but if you've chased in any lighting circuits, they now need RCD protection if they didn't have it already.

Also, note that if you've done any chasing in a bathroom or kitchen, then the work should have been notified to building control in advance under Part P (see the wiki for more info)...