Nuisance RCD Tripping - Diagnosis Advice PLEASE

[2nd Life]

Hi all.

Have an issue i'd love some advice on if anyone would be so kind. Will try to be brief. Thank you!!

So, have an intermittent RCD trip at my home. Big old house with an installation of nearly all old tinned copper wiring, that for the last couple of years, hasn't missed a beat. Old Hager split load board, one RCCB protecting the sockets and radials supplying everything bar lighting, and the other RCCB doing lighting.

Been having the sockets RCCB tripping when too many loads applied, so typical example if the washing machine and dishwasher already on and then someone turns on a kettle or iron or toaster, pop. It's pretty much guaranteed you can make it trip under enough load.

First thought might be dodgy appliance but then discounted that when it wasn't the same appliance causing the tripping each time of course. So, then thought might have a wiring issue (particularly given the age of the installation), but have had tripping on 3 different circuits trip, the kitchens split onto two rings, and have had tripping using appliances on either, and a radial supplying a single, single socket that supplies the freezer (weirdly) so now don't expect it to be an insulation resistance issue, rodent damage, nail through a cable etc etc

So then, I've settled on the fact it must be an earth leakage issue? Does this sound right to the pros on here? What i still question is why wasn't it an issue but now is.

ALSO - on one single occasion we actually had the lighting RBBC trip, only once suppose it could be a coincidence but still, it's relevant.

AND - We haven't added any new appliances or made any changes in to the property with exception of an extension to a radial circuit that fed (from a junction box hidden somewhere I am yet to find, as it is on the the same breaker as one of the rings which only has the two legs at the breaker, but is definitely not spurred off a socket as i've had the covers off them all and traced them, not there) a fused spur and a single double socket in a bedroom added to so that 4 more USB sockets have been added to the radial off the fused spur adding a good deal of length of cable as everythings had to go up and over in the loft space clipped to joists as the floor is a lime ash concrete (ish) floor. Room still being decorated so nothing plugged into anything on that extension FYI

Hope that all makes sense. Thanks to anyone that might have advice

 

[timhoward]

To answer your question - it would be tracked down mainly using insulation resistance testing, possibly also combined with direct earth leakage measurement, and ramp testing the RCD.

The bottom line:
1) if the RCD is tripping it's 99.9% certain to an earth leakage issue. The other 0.1% would be a faulty RCD or a new smart meter upsetting the RCD (which I've never heard of with Hager units).
2) As its a Dual-RCD consumer unit without Neutral isolation for each circuit, any connected load could be driving current through a neutral to earth fault on any circuit. Turning off MCBs doesn't therefore aid diagnosis. The loads that happen to be in use are also not necessarily conclusive either.
3) It could literally be anything on any circuit that side of the board, whether the MCB's are on or off, and whether appliances are in use or not. For this reason tracking it down is nearly impossible without a multi function tester and testing each circuit in turn.

 

[Avo Mk8]

Did your original check on which appliance might be responsible, involve you physically disconnection each one (eg pull plug out)?
If not, an experiment you could try would be to physically unplug each kitchen appliance, one at a time, to check if one , when unplugged, will allow all the others to function without any tripping.
I'm thinking maybe there's a leaky appliance (N-E) that remained plugged in but turned off during your investigation.
In order of most likely culprit: Kettle, Dishwasher, Iron, Washing machine, Toaster etc

 

[2nd Life]

To answer your question - it would be tracked down mainly using insulation resistance testing, possibly also combined with direct earth leakage measurement, and ramp testing the RCD.

The bottom line:
1) if the RCD is tripping it's 99.9% certain to an earth leakage issue. The other 0.1% would be a faulty RCD or a new smart meter upsetting the RCD (which I've never heard of with Hager units).
2) As its a Dual-RCD consumer unit without Neutral isolation for each circuit, any connected load could be driving current through a neutral to earth fault on any circuit. Turning off MCBs doesn't therefore aid diagnosis. The loads that happen to be in use are also not necessarily conclusive either.
3) It could literally be anything on any circuit that side of the board, whether the MCB's are on or off, and whether appliances are in use or not. For this reason tracking it down is nearly impossible without a multi function tester and testing each circuit in turn.

To answer your question - it would be tracked down mainly using insulation resistance testing, possibly also combined with direct earth leakage measurement, and ramp testing the RCD.

The bottom line:
1) if the RCD is tripping it's 99.9% certain to an earth leakage issue. The other 0.1% would be a faulty RCD or a new smart meter upsetting the RCD (which I've never heard of with Hager units).
2) As its a Dual-RCD consumer unit without Neutral isolation for each circuit, any connected load could be driving current through a neutral to earth fault on any circuit. Turning off MCBs doesn't therefore aid diagnosis. The loads that happen to be in use are also not necessarily conclusive either.
3) It could literally be anything on any circuit that side of the board, whether the MCB's are on or off, and whether appliances are in use or not. For this reason tracking it down is nearly impossible without a multi function tester and testing each circuit in turn.

Hi Tim. Thanks a lot for wading in on this.

Understood.

I'm actually doing my 2365 level 3 (having already completed level 2) but have only just started so have only limited experience of testing. But have just picked up an MFT from eBay that calibration only ran out April so trust it's still in good shape.

So I'll perform insulation resistance tests on the whole installation 1 by 1 as a kick off to rule that out I suppose And also can test the RCDs (not that I've covered that test yet), but don't yet have a clamp meter to do leakage testing. Sounds like I'll need to pick 1 up. Any advice on a decent entry level clamp meter? I think that's probably the issue.

One last thing. I've heard USB sockets can be damaged when IR testing so presume they'll need to be completely removed? We've got a few.

Thanks again. Much appreciated

 

[2nd Life]

Did your original check on which appliance might be responsible, involve you physically disconnection each one (eg pull plug out)?
If not, an experiment you could try would be to physically unplug each kitchen appliance, one at a time, to check if one , when unplugged, will allow all the others to function without any tripping.
I'm thinking maybe there's a leaky appliance (N-E) that remained plugged in but turned off during your investigation.
In order of most likely culprit: Kettle, Dishwasher, Iron, Washing machine, Toaster etc

Hi Avo. Thanks for the reply. No I didn't physically remove everything. As the issue only presented when appliances were under load, and then the RCD tripped when different appliances on different circuits were turned on I thought that confirmed it wasn't a single culprit appliance.

 

[timhoward]

I'm actually doing my 2365 level 3 (having already completed level 2) but have only just started so have only limited experience of testing.

Ok, fair enough, and everyone starts somewhere. (Check out John Ward's testing videos, they are excellent.)

But have just picked up an MFT from eBay that calibration only ran out April so trust it's still in good shape.

I'd suggest picking up a CalCard to give a basic sanity check that Continuity / IR functionality is working as it should be.

So I'll perform insulation resistance tests on the whole installation 1 by 1 as a kick off to rule that out

Yep, safe isolation process, whole lot off, then disconnecting L and N for each circuit in turn, joining them together, and testing L+N together to the CPC bar at 250v will give you a quick idea what is going on. You are looking for a very low reading, 0.xxxx Mohms.

And also can test the RCDs (not that I've covered that test yet),

RCD tests for certification are normally "how quickly does it trip". In this case, you want the RCD ramp test which will show you how much earth leakage it takes to trip the RCD. Or in other words, how close it is to tripping. An excellent result is 25+ma. A poor result is under 15 ma.
This is an alternative method to using a clamp meter, it's just more inconvenient as you have to keep resetting the RCD and can't continuously monitor the results.

Any advice on a decent entry level clamp meter? I think that's probably the issue.

TEK 775 - or one of it's clones. Shop around, prices vary enormously!

I've heard USB sockets can be damaged when IR testing so presume they'll need to be completely removed?

Some e.g. BG state they are safe to IR test, but if in any doubt I'd wago them out first.

Other things can result in false positive readings, notably surge protected extension leads which give about 0.4 mohms at 500v but a higher reading at 250v.

Let us know how you get on. Be safe!

 

[2nd Life]

Ok, fair enough, and everyone starts somewhere. (Check out John Ward's testing videos, they are excellent.)

I'd suggest picking up a CalCard to give a basic sanity check that Continuity / IR functionality is working as it should be.

Yep, safe isolation process, whole lot off, then disconnecting L and N for each circuit in turn, joining them together, and testing L+N together to the CPC bar at 250v will give you a quick idea what is going on. You are looking for a very low reading, 0.xxxx Mohms.

RCD tests for certification are normally "how quickly does it trip". In this case, you want the RCD ramp test which will show you how much earth leakage it takes to trip the RCD. Or in other words, how close it is to tripping. An excellent result is 25+ma. A poor result is under 15 ma.
This is an alternative method to using a clamp meter, it's just more inconvenient as you have to keep resetting the RCD and can't continuously monitor the results.

TEK 775 - or one of it's clones. Shop around, prices vary enormously!

Some e.g. BG state they are safe to IR test, but if in any doubt I'd wago them out first.

Other things can result in false positive readings, notably surge protected extension leads which give about 0.4 mohms at 500v but a higher reading at 250v.

Let us know how you get on. Be safe!

Hi again Tim. That's all fantastic advice thanks. Just ordered a Cal card and will check out John Ward and the Tek

The sockets are BG 900 SERIES USB C couldn't find anything on their website about that but maybe I'll look again if they can be left in place all the better.

Thanks again. Will start the hard yards testing it all

 

[pc1966]

Hi again Tim. That's all fantastic advice thanks. Just ordered a Cal card and will check out John Ward and the Tek

You always want some means to verify your tester is doing the right job. A CalCard helps with that as it provides some usable low and high R values to check.

Remember to null your test leads before the low-R checks.

The sockets are BG 900 SERIES USB C couldn't find anything on their website about that but maybe I'll look again if they can be left in place all the better.

As a general rule, all equipment should be safe if testing L+N to E as they ought to have sufficient protection to kV surges. However, you can find SPD in devices or extension blocks that show as a low-ish R at 500V but not 250V.

Testing L-N at 500V is something that can damage low power electronics as it is well over the 230V * sqrt(2) peak sine wave voltage (on purpose of course to check IR is good above normal working voltage) so always start at 250V and only if it is above a good few M then try 500V if you expect the circuit is safe.

If you have a fault that is the cause of the trip, instead of just too much accumulated leakage for a single 30mA RCD, you will almost certainly find sub M-ohm at 250V so that is the first thing to look for.

 

[2nd Life]

Ok, fair enough, and everyone starts somewhere. (Check out John Ward's testing videos, they are excellent.)

I'd suggest picking up a CalCard to give a basic sanity check that Continuity / IR functionality is working as it should be.

Yep, safe isolation process, whole lot off, then disconnecting L and N for each circuit in turn, joining them together, and testing L+N together to the CPC bar at 250v will give you a quick idea what is going on. You are looking for a very low reading, 0.xxxx Mohms.

RCD tests for certification are normally "how quickly does it trip". In this case, you want the RCD ramp test which will show you how much earth leakage it takes to trip the RCD. Or in other words, how close it is to tripping. An excellent result is 25+ma. A poor result is under 15 ma.
This is an alternative method to using a clamp meter, it's just more inconvenient as you have to keep resetting the RCD and can't continuously monitor the results.

TEK 775 - or one of it's clones. Shop around, prices vary enormously!

Some e.g. BG state they are safe to IR test, but if in any doubt I'd wago them out first.

Other things can result in false positive readings, notably surge protected extension leads which give about 0.4 mohms at 500v but a higher reading at 250v.

Let us know how you get on. Be safe!

Sorry. One last thing.

So the Hager RCCB in question. Pictured.

Ok, fair enough, and everyone starts somewhere. (Check out John Ward's testing videos, they are excellent.)

I'd suggest picking up a CalCard to give a basic sanity check that Continuity / IR functionality is working as it should be.

Yep, safe isolation process, whole lot off, then disconnecting L and N for each circuit in turn, joining them together, and testing L+N together to the CPC bar at 250v will give you a quick idea what is going on. You are looking for a very low reading, 0.xxxx Mohms.

RCD tests for certification are normally "how quickly does it trip". In this case, you want the RCD ramp test which will show you how much earth leakage it takes to trip the RCD. Or in other words, how close it is to tripping. An excellent result is 25+ma. A poor result is under 15 ma.
This is an alternative method to using a clamp meter, it's just more inconvenient as you have to keep resetting the RCD and can't continuously monitor the results.

TEK 775 - or one of it's clones. Shop around, prices vary enormously!

Some e.g. BG state they are safe to IR test, but if in any doubt I'd wago them out first.

Other things can result in false positive readings, notably surge protected extension leads which give about 0.4 mohms at 500v but a higher reading at 250v.

Let us know how you get on. Be safe!

Sorry. One more. So the RCCB in question. Pictured. Has two visual indicators on it. One marked earth fault indicator. Not all RCDs have this I see (the other one on my board for example) but I don't see anything even on Hager's own data sheets etc about what it does. I know it seems obvious given its name but would this assist in understanding what has caused the trip and narrow down testing? Thanks a lot.

 

[pc1966]

One marked earth fault indicator. Not all RCDs have this I see (the other one on my board for example) but I don't see anything even on Hager's own data sheets etc about what it does.

You can trip an RCCB or RCBO by either manually forcing the toggle (like an MCB or switch), or by simulating a fault with the test button. The indicator is to show you which method tripped it.

 

[timhoward]

Sorry. One more. So the RCCB in question. Pictured. Has two visual indicators on it. One marked earth fault indicator.

Won't help you, it just shows you if it tripped itself or was turned off.

Out of curiosity, are we allowed a photo of the whole board?
Not to have a go at you, as you didn't fit it, but your photo has got me wondering how someone has arranged the tails and internal distribution....

 

[2nd Life]

You can trip an RCCB or RCBO by either manually forcing the toggle (like an MCB or switch), or by simulating a fault with the test button. The indicator is to show you which method tripped it.

Thanks pc. So in the case of and actual fault/trip there will never be a case of one or the other indication. Sorry, trying to learn

 

[2nd Life]

Won't help you, it just shows you if it tripped itself or was turned off.

Out of curiosity, are we allowed a photo of the whole board?
Not to have a go at you, as you didn't fit it, but your photo has got me wondering how someone has arranged the tails and internal distribution....

Thanks mate. It's a terrible birds nest. Even I can tell the boards been fitted terribly. Radials stuffed in along with rings, all sorts. I plan to fit (have fitted) an rbco board with surge to update the lot, so it's all a moot point I suppose but if I can sort the tripping out (and confirm for my own peace of mind the IR and therefore the aging wiring throughout is sound beforehand) it'll save weeks of inconvenience

 

[2nd Life]

Won't help you, it just shows you if it tripped itself or was turned off.

Out of curiosity, are we allowed a photo of the whole board?
Not to have a go at you, as you didn't fit it, but your photo has got me wondering how someone has arranged the tails and internal distribution....

Ask and you shall receive. Including the world's largest connector block, random upsidedown breakers with bus bar at the top, and holes everywhere. IP00

 

[2nd Life]

Won't help you, it just shows you if it tripped itself or was turned off.

Out of curiosity, are we allowed a photo of the whole board?
Not to have a go at you, as you didn't fit it, but your photo has got me wondering how someone has arranged the tails and internal distribution....

 

[timhoward]

Ask and you shall receive. Including the world's largest connector block, random upsidedown breakers with bus bar at the top, and holes everywhere. IP00

Wow. That's not just a rats nest, that's dangerous.

It's missing a main switch so there's no single point of isolation, the 6mm carrying the feed to the LH RCD is under-sized for the load and is only protected by the suppliers fuse, the 6mm from the RCD to the bus bar on left is under sized. It also looks like the LH neutral bar is for the right hand circuits. And half the Neutrals for the left hand circuits look to be in a choc block.
In short it's an absolute nightmare, a far cry from a college test rig.

Don't take this the wrong way, but if you haven't yet done much testing I wouldn't make that one your first deep dive testing experience, or your first board change. For context I don't think I've seen one as bad as that this year. I'd be charging a premium to even entertain taking on that!

 

[2nd Life]

Wow. That's not just a rats nest, that's dangerous.

It's missing a main switch so there's no single point of isolation, the 6mm carrying the feed to the LH RCD is under-sized for the load and is only protected by the suppliers fuse, the 6mm from the RCD to the bus bar on left is under sized. It also looks like the LH neutral bar is for the right hand circuits. And half the Neutrals for the left hand circuits look to be in a choc block.
In short it's an absolute nightmare, a far cry from a college test rig.

Don't take this the wrong way, but if you haven't yet done much testing I wouldn't make that one your first deep dive testing experience, or your first board change. For context I don't think I've seen one as bad as that this year. I'd be charging a premium to even entertain taking on that!

I know. Will be getting someone to change the board for me don't worry. There is a isolator between meter and board so at least there's that. Thanks again for all the advice.

 

[littlespark]

whats the top side of the breakers look like?

Does the left rccb power the 7 leftmost breakers, and the right do the other 7?


It could be cumulative earth leakage. Everything tests fine on its own... but every little device has a minute earth leakage... if you've got loads plugged in, it'll all add up. And the 30mA marked on RCD's is just an upper limit. It could trip at 19,20, 21... Thats what the ramp test checks for on an rcd tester.

It appears the right hand rccb has been changed already..... is the problem on this side or the other, older rccb?
(or was that where the mainswitch was?)


Basically.... what im trying to say is, you could spend hours, days, trying to find a fault that wont show up.
By all means, do a the testing.... but if its inconclusive, dont waste time going over and over the same thing

 

[pc1966]

That is quite a "special" CU to be looking at. and is it just camera perspective, or is that the biggest choc block I have ever seen?

 

[2nd Life]

whats the top side of the breakers look like?

Does the left rccb power the 7 leftmost breakers, and the right do the other 7?


It could be cumulative earth leakage. Everything tests fine on its own... but every little device has a minute earth leakage... if you've got loads plugged in, it'll all add up. And the 30mA marked on RCD's is just an upper limit. It could trip at 19,20, 21... Thats what the ramp test checks for on an rcd tester.

It appears the right hand rccb has been changed already..... is the problem on this side or the other, older rccb?
(or was that where the mainswitch was?)


Basically.... what im trying to say is, you could spend hours, days, trying to find a fault that wont show up.
By all means, do a the testing.... but if its inconclusive, dont waste time going over and over the same thing

Hiya. So the left RCCB (the original one by the looks) does the left most 4 circuits (the lighting circuits. And the new one on the right does the other 10, although 2 are redundant so 8 at present. That's the breaker that's tripped like 20 times, although on 1 occasion the other RCCB tripped also. From everything I have read and all the advice people have been kind enough to give. It does look like it will be cumulative earth leakage on those 8 circuits. But I don't understand is why without the introduction of any new appliances after 2 years of no tripping am I now getting it everyday? That doesn't make a lot of sense. Do USB sockets add to leakage as that is the only thing that has changed in that time. I want to do IR testing anyway as if I'm going to get someone to come and change the board. I want to know from my own benefit that the wiring is sound. What I did think was that if it all checks out on the board change goes ahead. Then at least having individual RCBOs will be able to share the leakage and it may just go away on its own? Thanks for having a look at it for me