The IET is carrying out some important updates between 17-30 April and all of our websites will be view only. For more information, read this Announcement
With have heard reports of smart meters tripping RCDs, is there any more than anecdotal evidence?
I know it’s Christmas Eve so most people will have a roll of cooking foil handy ready to wrap the turkey in the morning, but I’m not sure a customer will take me too seriously if I suggest using some of the cooking foil to enclose their consumer unit in a Faraday Cage due to its proximity to the Smart Meter, which is the last recommendation on the previous discussion from a several years ago.
Has any firmer evidence been produced in the intervening years?
No need to look for further evidence: we know that older RCDs aren't guaranteed to meet EMC requirements of the current product standards.
Having said that, BS 7671 (Regulation 444.4.1) says that sources of EMC disturbances (which would include transmitters for wireless connectivity) should be kept away from "victim" equipment.
I wonder what the smart meter installation instructions say in this respect?
The tripping RCD is an old Crabtree, I have ordered a new consumer unit with double pole Type A RCBOs and will retest every circuit whilst completely disconnected.
It seems over the top, but as the old RCD was upfront and the entire installation goes down when it trips the issue needs sorting out. I have at least hooked the kitchen up onto an RCBO as a temporary arrangement to get them through Christmas whilst all the wholesalers are shut without losing everything when the RCD trips.
I thought I had pinned the issue down to being surge protected trailing sockets, there definitely seems to be to many of them in the house with there being at least five, so I have included a consumer unit SPD that will be upfront of the RCBOs and will suggest getting rid of some of the plug in devices, the Christmas tree lights and table lamps don’t need them as far as I can see.
I have replaced the consumer unit with a new Fusebox consumer unit with double pole 30 mA type A RCBOs testing as I did the work.
On the kitchen ring someone had disconnected a spur leaving conductors taped up in a back box , unfortunately they disconnected the wrong neutral breaking the neutral ring and leaving the disconnected neutral live with the other end that is not visible possibly being an earth fault, a fault I have fixed, but not the fault causing the RCD to trip.
The house socket ring was iffy, but had an insulation test result of over 0.7 Mega ohms so as time was running out I connected it, as I suspected it would it tripped several times over night so I went back yesterday to strip the circuit.
So I returned and disconnected the circuit having unplugged all the appliances, neutral to earth insulation test result 400 Kilo ohms, great something to look for.
Broke the ring by removing a socket to start faulting by halves, fault in back half of house. Break ring again having identified sockets with issues using socket test lead adapter, take socket out test the cables no fault.
Test all circuit conductors of the ring that is now broke into three sections at both ends, six tests and no fault.
Logic dictates I have clear the fault by removing the sockets to test, so have a look at the insulation on the wiring in the socket back boxes and run my fingers over them, all okay so insulation test the actual socket fittings.
Possibility repeated insulation testing has dried out a damp fitting, so dismantle the outdoor socket which looks okay.
Reassemble the circuit, replacing one of the internal sockets and the external socket as well as opening up the drain hole in its enclosure and sealing the enclosure with silicone.
Retest at the consumer unit, confirming end to end continuity of conductors, then live to neutral insulation test and live + neutral combined to earth with a test result of 1.8 Giga ohms, so reconnected and came home.
The RCBO is still tripping and I have to go and try again, I want to say it is something that is plugged in possibly an extension lead with a SPD in it, but I did get a low insulation test result which then cleared whilst I was stripping the circuit.
Around thirty seven years ago in the mid 80's before I qualified as an electrician I was working as a carpenter on piecework on a new housing site with an electrician called Desi who taught me a lot about electrical work. One evening as I was about to go home Desi asked me to take my generator and circular saw to cut the chipboard floor up in a house, because there was an insulation fault on a circuit he needed to find. So I cut full sheets out, running the saw though the togues on the sheets of flooring chipboard, so we had big holes in the floor. Suddenly the meter showed a perfect test result, off the top of the scale on the meter, so we checked where every nail had been driven through the sheets to see if a cable had been nailed and all the visible cables, everything was perfect.
Desi said we might as well give up as with a perfect (off the top of the scale) test result we had absolutely no idea where to look, however back then there was no RCD protection at the consumer unit, only an RCD socket in a garage if there was a garage, so neutral faults would not be an obvious issue when the circuit was in use anyway.
So it looks like I am stripping and testing again, at least with the new consumer unit they can turn the socket circuit off when they go out and know the freezers, fridge and central heating are all working okay.
400 -700k is not on its own going to fire an RCD. Are you able to measure the AC leakage currents directly?
I have found that a clamp meter that can see tens of mA that I can thread the L and N through to show only what the RCD part is seeing (as in Lcurrent -Ncurrent) can be great assistance here, not least because it can be left wired in while the circuit is running and various things are plugged in and out, heater thermostats operated and so on. You may have a capacitance live to earth rather than a resistance - either a lumped element marked capacitor that has a fault to metal case or is miswired, or a stray capacitance as part of something else like electronics or a heater element. I presume you do not have access to a meter that can measure fractions of a microfarad?
A similar oddball moment, a colleague of mine at work used an RF network analyser (sort of thing that generates a swept RF frequency and measures the phase and magnitude of the current drawn vs frequency) to trace a rather odd wiring fault by spotting cables where the L and N capacitance to the earth core was wildly different from each other. I'm not recommending that as a means of normal fault finding, coming well into the category of 'over the top' for normal use but I mention it to point out that a DC only insulation test may not actually reveal as much as an AC measurement. In his case it found a long decorated over fused spur on the ring where the load had been removed but it had been rewired so when it was in the on position there was a ring albeit one with a fuse in the live loop, and when it was switched off it became 2 centre fed radials. Until the fuse blew. Then it became a fault in the live continuity.
Good luck with what sounds like it may be a long haul.
M.
We're making some changes behind the scenes to deliver a better experience for our members and customers. Posting and interactions are paused. Thank you for your patience and see you soon!
For more information, please read this announcement
