RCD tripping - why????

I've recently rewired my house, so when the time-delayed RCD, which acts as my main switch started tripping, I assumed I had made a stupid mistake somewhere. My installation is TT, all circuits (apart from distribution circuits) are on RCBOs.

I noticed that the tripping occurred at about 8am or about 8pm on weekdays. At weekends it occurred at other times. This lead me to suspect an outside cause.

To cut a long story short, my next door neighbour is in the habit of switching her electric shower on/off via the isolation pull switch. When she does this, my RCD trips. When she turns the shower on/off normally, it doesn't trip.

Her shower circuit is 2.5mm cable and the insulation looks a bit perished and frayed in the pull-switch. It's also on a 40A breaker! The pull switch is in "zone 1" - I kid you not! The cord is reachable from outside the shower enclosure and quite convenient to use.

Her house appears to be protected by RCDs - a fairly ancient type, but they don't trip.

She also promises to control her shower in the normal way until we get to the bottom of this...

Any ideas?

T
Reply to
Recyclist
Loading thread data ...

A bit weird that but IINAE Does the normal control have more than one heat setting, and do you have to step through them to reach the off position?

I presume when your RCD tripped she swiched off at the pull-cord with the shower still going at full power.

Reply to
Graham.

My RCD trips when she uses the pull cord to switch her shower off or on.

Her shower probably has those steps to get to full power, but I didn't look.

T
Reply to
Recyclist

Distracted by something else in the shower at the time I take it?

It is a bit weird and I suspect you need to look hard at how the supplies and earths etc are derived in both of your places. You may need to involve your electricity supplier. Note: The company that maintains the pysical infrastructure, not the company who you pay for electricity.

Reply to
Dave Liquorice

Neutral earth short or too many capacitors in the RF filters. Both will, under a smart step in voltage, divert enough to ground to trip.

Reply to
The Natural Philosopher

A few questions-

Is your neighbours shower earthed correctly? This may require inspection, earth continuity testing and an earth loop impedance test at the shower unit.

Where is your Earth electrode rod in relation to your neighbours and is it in a suitable enclosure to avoid ground surface potential gradients? A time delayed RCD should not trip before a general purpose one but have you carried out tripping time tests of all your RCBOs and the RCD? (I hope you fully inspected and tested your rewired installation prior to energising it?)

You say the shower next door is fed by 2.5mm cable with a 40A breaker, what is the power rating of the shower and is your fire insurance premium paid up to date? (Hint 6mm is the smallest cable normally expected to be used for electric showers with 10mm or 16mm being used for higher power ratings)

Sudden high load switching can create surges which may be exacerbated by poor connections in the supply system. This is outside your control and will involve the network suppliers equipment should it come to that.

Reply to
cynic

Just to clarify; it sounds as if you have a "main switch" (i.e. not split load) CU with the circuits protected by two cascaded RCDs (i.e. the "main switch" type S RCD, in turn feeding axillary RCDs included in each RCBO). Any sub mains etc presumably being fed via conventional MCBs.

I would guess that you have a "sensitisation" issue here - where your main type S RCD has sufficient leakage through it normally to put it very close to tripping.

In cases like this any large transients either in your house or on your supply can create enough extra leakage to trip the RCD. (One mechanism at work here is that the transient it rich in harmonic noise, and that will be conducted to earth via mains input filters and smoothing caps on your appliances - so you get a brief rise in earth leakage that will be seen as extra imbalance by your RCD). The neighbours shower will certainly represent a large load and hence big transient when switched. It may be that the functional switching includes snubbing to reduce the transient noise created, whereas the isolation switch will not have this.

Now the cause; my guess would be the fault is actually in your house, and is simply a case of two much combined leakage from all your circuits (possibly caused by faulty appliances or a wiring fault). Normally on a

16th edition style TT system, the CU is a split load type with type S RCD as you have it, and then another single conventional RCD protecting the higher risk circuits. If, as is often the case, the socket circuits are the source of the excess leakage, then the non delayed RCD would trip first - and at a leakage current well below the trip threshold of the type S 100mA trip device. However, because you have RCBOs rather than a secondary RCD protecting multiple circuits, you won't get this early warning. You could have (say) four RCBOs each with 20mA of leakage. None would trip on their own leakage, but the combined leakage would push you main RCD very close to the limit.

How to fix: you did not say how many circuits you have, and what sort of loads are on them. Assuming its not a particularly large number, then you would probably be better off finding the leakage source:

formatting link
you have lots of circuits and or many high leakage appliances (loads of IT kit etc), and it really is just a case of nothing actually being wrong, then changing the CU organisation would be worth doing. Converting to a split load arrangement with a conventional main switch feeding all the RCBOs on one side, and then the type S RCD moved onto the slave side and used just for feeding the sub mains etc. That would then permit each individual circuit to leak up to the threshold permitted by its RCBO without this having any cumulative knock on effects.

Reply to
John Rumm

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D\

=A0 =A0 =A0 =A0|

=A0 =A0 =A0 =A0|

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D/

Thanks for your detailed response. I'll check everything again!

Today, the RCD won't trip. However when my neighbour switches on her shower, the connection at the overhead wire crackles and there are some sparks. It also crackles a bit in the wind.

If it means anything, the connection that is crackling is to the wire with moss on it. The connection to the clean wire sounds OK.

T
Reply to
Recyclist

Oh boy, nothing like with holding information... The supply wiring should not do that. There is a supply fault. Report it to your REC to come and fix it and check that the supllies to you and your neighbour are in order.

Once they have see if this whole house RCD tripping persists.

Reply to
Dave Liquorice

In that case the answer is quite simple, phone the number on the pole or your nearest substation and report the issue. There may be other faults but that needs fixing, if the wire is in that state it may catch fire elsewhere or fall to the ground....

Reply to
James Salisbury

Well that is something that needs sorting ASAP. The sparking under load indicates a poor connection, and would also make any transient dips in the supply noticeably worse. Have words with your supplier.

Possibly not alot - unless one is insulated and the other bare... (on older installs its not uncommon to run an insulated live / bare neutral from the pole to the house)

Reply to
John Rumm

Or even without holding information!

It's "withholding" for heavens sake, all one word. As two separate words "with holding" it can (and does) change the meaning completely.

Reply to
tinnews

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D\

=A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0|

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D/

It looked very close to catching fire, but now we are all sitting in the dark it appears to have calmed down. Central Networks are on their way.

T T
Reply to
Recyclist

Recyclist coughed up some electrons that declared:

The way you worded that could induce an AOL style "LoL" moment!

But you are in the dark. With no coffee. And presumably no TV. So I feel your pain.

BTW how are you posting this - UPS + laptop?

Hope they fix it soon :)

Tim

Reply to
Tim S

, >

I don't have a TV, and can boil a kettle on my (wood burning) Rayburn if I want a coffee. I have instructed the neighbours to turn off their mains, so they are sitting in the dark. This allows me to keep my laptop running even through the dodgy connection ;-)

T
Reply to
Recyclist

Pedant. I understood what was intended and appreciated the typo. Have you not made similar unintended gaffs? :)

Reply to
Clot

I take it you are both supplied via the same cable? (its not uncommon for neighbours to be on different phases)

Reply to
John Rumm

That's possibly not a bad idea anyway. The electric shower's I've used have a sort of "cotnroleld shutdown" - they keep the water running for a moment after killing the power, and I think this is to cushion the temperature change in the element. You might get a heat spike if the inside is still heating and the water stops running. Which could easily shorten the life of the shower - and they aren't the most reliable devices.

Thinks. It's now nearly 20:00 on the 4th - Recyclist was online by this time yesterday - any news?

Andy

Reply to
Andy Champ

Thanks for everyone's help and advice. My RCD hasn't tripped since the overhead line stopped sparking, and my neighbour has gone on holiday.

Nevertheless, I took the point made by John Rumm that there is probably a small current leak somewhere, priming my RCD to trip at the slightest disturbance. I tracked it down to a secondary consumer unit. Cramming RCBOs into average CUs makes testing extremely difficult, so I had to strip it out to test the circuits etc. I didn't find the fault, but after replacing the RCBOs and trimming the wires, the fault has gone!

T
Reply to
Recyclist

I take it they fixed the overhead line?

If so the real test will be when the neighbour gets back I suppose ;-)

Reply to
John Rumm

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.