The Lemon Fool

I know the work "intermittent" will strike terror into the hearts of electrical investigators, but this is a bit weird....
For the past 2-3 weeks the whole house RCD keeps tripping around 6:30 in the evening. Won't re-latch for about a minute. Then seems OK until the next evening.
No recent electrical work that might coincide with that time frame, in fact nothing changed for years!
Usual isolations suggest its the kitchen circuit, but further isolation of appliances (hob, boiler, outside lights, fridge, freezer, toaster, kettle, microwave and answerphone) has not isolated the fault. The last 6 being isolated by plug-removal to avoid neutral-Earth (N-E) shorts and all checked out above 20Mohn resistance with home multi-meter (so only 9V test voltage).
Fridge did have a lot of water on the compressor and isolating capacitor was damp, these were dried and no short apparent. I also moved the capacitor to a better place. Fridge has started and stopped regularly with causing the trip since.
Dishwasher is awkward to get at, but doesn't have a timer as far as I can tell.
Oven is on a separate circuit. System was very carefully installed and tested ~ 12 years ago (just prior to Part-P) by me, with great care taken during installation and testing to ensure no cross-overs.
The regularity suggests something with a timer, but leaving the microwave clock not re-set didn't change the event. Next suspect was the boiler (on 24hr anti-seize) run, but leaving that switched off didn't change the time (not had time to investigate N-E issues with this). "Equipment of interest" are the outside lights, with the sun dropping lower and heating one causing an N-E fault ('cos switching it off at the fuse doesn't change things).
The final suspect is that something else is coming on at that time (but goodness knows what, since everything with a clock also comes on at other times without incident!) and tripping the cumulative Earth leakage current by switching on, its just that the kitchen circuit is the straw that breaking this camel's back.
Any other ideas of suggestions?!?!
Cheers
Chris

Chris will be here in a minute. I have no idea what a capacitor is, let alone move it to a better place.

Good luck.

Dod

What sort of earthing system do you have? It may be relevant.

Julian F. G. W.

Have you got a frost free fridge or freezer ?

Dorn1 wrote:I know the work "intermittent" will strike terror into the hearts of electrical investigators

Hi, it really does seem as if you have fully understood the potential causes of the tripping. You have the advantage of a good indication when the fault presents.

I think in your situation I would tackle it slowly. I would choose ONE circuit at a time and isolate it (both line and neutral by disconnecting the black/blue wire too) for a whole night & day to see if the fault clears. I would leave the circuit with the freezer until last (that night you could use an extension lead to keep it powered up from a different circuit perhaps?).

There is always the danger that the tripping is a cumulative effect of more than one circuit.

I agree with your timer hypothesis. I wonder what, if anything, was done two-three weeks ago that might have changed things.

What you really need is a sensitive clamp meter around the installation's tails at the moment of tripping. (You don't happen to be within stabbing distance of Tunbridge Wells and have the price of a pint of beer to hand, do you?)

Chris

Dorn1 wrote:The final suspect is that something else is coming on at that time (but goodness knows what, since everything with a clock also comes on at other times without incident!) and tripping the cumulative Earth leakage current by switching on, its just that the kitchen circuit is the straw that breaking this camel's back.

Long shot. Is there something in the neighbourhood that turns on the amps at 6.30 pm? Many, many years ago, I had a surge that would come through the house at about that time of day, each and every day. Usually it tripped the whole house out, although sometimes I got away with it. The electricity board came round and blamed it on a nearby farm which was turning its milking equipment on in the evenings and stressing the neighbourhood, and my particular consumer unit (or maybe my overhead supply cables to the house?) were objecting. Other nearby houses didn't seem to mind it, though. (Eventually the problem went away, and I forgot to do anything about it.)

It was mainly a rural problem, the engineer explained, but it can also happen in urban areas if there's manufacturing going on. These days, though, I'd be wondering whether my neighbour was plugging in his Tesla when he got home from work?

BJ

Friends had a similar experience with the electrics tripping at a regular time shortly after having had their electrics upgraded.

One morning they noticed it coincided with the street lamps turning off. Their electrician who had already investigated the issue several times was baffled and rang the manufacturer of the consumer unit. From what my friends told me I suspect the manufacturer supplied new trips with a slower response times. They've not had problems since.

twotwo22 wrote:Have you got a frost free fridge or freezer ?

I asked this as they usually have a 24 hour timer clock built in which activates a small heater element to defrost the cooling panel.
You had an interest in devices with timers.
If you do have one, you could turn it off for couple of hours and then back on and see if your tripping time shifts by a couple of hours.

csearle wrote:

Dorn1 wrote:What you really need is a sensitive clamp meter around the installation's tails at the moment of tripping. (You don't happen to be within stabbing distance of Tunbridge Wells and have the price of a pint of beer to hand, do you?)

Chris

My money is on one to three appliances each contributing towards 25mA or so of L-N imbalance, and either an RCD which has become more sensitive over time, or an appliance which has got worse

I'd put my clamp meter around the tails, measure the current imbalance, turn off/isolate the usual suspects (washing machine, tumble dryer, dishwasher, boiler).

The antifrosting feature on the fridge freezer sounds like a good candidate - usually a trace heater around the evaporator coil in the freezer compartment.
If the freezer is thoroughly defrosted (with no water/ice between the heater and the coils) and the problem goes away - bobsherunkle.

PochiSoldi

I had an interesting email exchange with a knowledgeable American bod, who was responsible for the compliance of a particular RCD manufacturer with standards.

He told me, and supported it with a paper if I recall correctly, that normal RCDs can be tripped by voltage spikes on one of their two inputs. These can occur through lightning strikes, arc welders in the vicinity and such like.

It is possible to obtain RCDs that comply to more rigorous specifications that are more resilient to such spikes. I imagine that a surge protector between the distributor's supply and the consumer unit would probably also sort such problems out.

Chris
PS I rather suspect this isn't the issue here. I've my money on the freezer.

Folks
Many, many thanks for all your most helpful replies, I'm away on business this week (reports from home about regular cut outs continue along with increased frustration!)
To try and cover everything…
• Earthing: TN-C-S with plenty of equipotential bonding brought back to the consumer unit on the correct side of the RCD.
• Yes both fridge (14years old) and freezer (1yr) are frost free and the suggested 24hr timer within them is a very useful input! I had them unplugged for a few hours (indeed the whole circuit) hoping that would shift the timings and help reveal the culprit, but the time seems pretty steady. It rather lends credibility to the “cumulative leakage” theory and maybe it’s something else that tips it over the limit.
• I see that I’m going to have to do a more careful investigation. Chris.S: I’m the wrong side of London from you! But thanks for the kind offer! I have an idea to analyse the current differential in the tails by simple “jury rig” wire wrap (no direct connection!) and see if that’s sensitive enough with the digital multi-meter to suggest which circuit is making the main contributions, so it doesn’t have to be calibrated. But if that’s not enough I’ll probably have to treat myself to a clamp meter!
• No specific change, event or new item in the past few weeks that I can recall as a possible cause. The brain has been wracked on this! But there have been a few intermittent events over the past year, which might indicate a slow deterioration of something.
• External surge in the neighbourhood is an interesting one! Although the village is in a rural area there’s only domestic houses around us. I’ll ask the neighbours if they’ve experienced tripping issues, but usually such faults struggle to propagate through the consumer unit. The fact that re-setting won’t work for about a minute after the first trip suggests it’s a connectivity fault rather than transient event.
• RCD[/list][/list] response times: it’s not that old and they “shouldn’t” wear out and the fact that it can be reset suggests the RCB is OK. Perhaps if all other avenues draw blanks a new RCD is an option, They’re not very expensive! I’d be reluctant go with a slower response time as its whole purpose is to protect quickly!
Looks like my weekend is planned out….!
Many thanks once again.
Chris

You say in the past 2-3 weeks, when there has been plenty of solar-generated electricity, so have you or close neighbours got solar panels?

These are likely to increase the local mains voltage towards the high end.

What does yours measure?

No close neighbours with solar panels, but maybe a few on the same sub-station.
Will check the voltage on Saturday!
The bright sun was part of my theory about external lights, as the setting sun moves down the shadowed lights (under the eves) become exposed to direct sunlight. But switching the extrnal light circuit off (fused connection off the kitchen ring) didn't change things (although I need to check if its a double-pole switch to eliminate the N-E fault possibility)
Chris

Define "around 6:30 in the evening".

- exactly (give or take a few seconds) 6.30pm?
- or roughly about half six?

"Exactly" suggests a timer-related matter (internal or externally triggered), opening lines of enquiry...

"Roughly" may align with peak demand, which on a rural circuit especially, may see your supply voltage drop materially, opening other lines of enquiry...

I had an earth-leakage issue that only manifested at peak local demand, when my domestic supply voltage dropped considerably. After figuring out the circumstances, we were able to induce it at will by cranking up every appliance, electric showers etc in my and a neighbouring house. Then, armed with this ability, isolated circuits one by one until the offending circuit found, then section of circuit, tracked eventually to the slightest of trowel marks on a cable, which you'd never have believed could've been the cause - but it was.

Doesn't sound anything like your problem (since you've had a stable & unchanged system for so long), other than to say systematic methodology will reveal all.

Many, many thanks for all your most helpful replies and suggestions, but...
I think I've found the culprit - it was the old laptop wot done it gov! But may not have been acting alone.... and may have got away with it if it hadn't been for those pesky kids!

Over the weekend it started tripping in the morning as well, just as the boys were playing <flipping> Fortnite. Same pattern, 3 trips in a row spaced over 1min then all OK. I did a temporary extension lead with the Earth isolated and sure enough the problem went away (the battery stayed cool and its plastic casing has not high voltages!). I also realised that the evening trips were so regular as they always start playing about the same time. Clearly the old laptop PSU is charging the ailing battery for about 20 mins, then suffering some earth fault as it switches modes.

However, I'm pretty sure all the investigations I the kitchen has shown some earth leakage (as you'd expect) and hence the mixed contributions of both faults were probably causing the trip - and isolating the kitchen usually fixed the problem.
For thems wot asked: voltage at the main isolator was 247-248V (it fluctuated between the 2 values), at the higher end, but well within spec for UK supply.

But shiny new clamp meter now in the tool box and replacement battery charger (and battery) on order.
Cheers
Chris

Dorn1 wrote:I think I've found the culprit - it was the old laptop wot done it gov!

Excellent deduction. Well done. You clearly are on top of this sort of stuff. All I can add is that computers have power supply units that are required (in order to comply with electromagnetic interference regulations) to limit injection of harmonics etc. back into the supply line. They do this by having filters at the point of their supply. This is a "functional earth" that presents itself to residual current devices as a "natural leakage".

Interestingly perhaps is the fact that 10 days ago the 18th edition of the wiring regulations was published (that in the UK has to be adhered to from 1 Jan 2019). In these regulations (as I currently understand them) electrical installations have to be designed such that the "natural leakage" should not exceed 30% of the actual tripping current of the residual current device protecting that circuit. Typically that would be 30% of about 25mA i.e. 7.5mA.

So it could be that in future your situation wouldn't arise in a correctly installed installation.

Anyway, glad you've sorted it. Thanks for your experience.

Regards,
Chris