UK AC RCD’s C2? Absolutely!

[davesparks]

Honest question, if you know an AC rcd may not work in a fault situation, explain why you would code it as a C3 and negate the need for someone to fix it.

If the RCD is there for additional protection then it is possible that it should be a C3 as it is additional protection and not the primary protection which is potentially compromised.

If there is a definate, confirmed presence of DC leakage currents which will alter the operation of a type AC RCD then that is a different story to guessing that there maybe might be an issue.

Imagine a child sticks their finger in a lamp holder? But the rcd fails…

An RCD is unlikely to assist in that situation anyway. If a child sticks their finger in a lampholder then they are more likely to make contact with both the live and neutral, this will result in current flowing through their finger which the RCD will see as normal load current.

 

[Risteard]

If you read the Best Practice guide #4 for EICR coding (Nov 2022) from Electrical Safety FIrst, it states for "A Type AC RCD installed where a Type A RCD required" to be coded as C3.

Indeed they do, but I must admit I find this suggestion somewhat questionable. Surely if an RCD is required, and the circumstances of the installation dictate that it needs to be a minimum of a Type A RCD because a Type AC RCD could be disabled by circulating DC currents then a C3 Observation seems overly lenient. But alas, you are correct that this is what that Best Practice Guide suggests.

 

[Pretty Mouth]

It’s common knowledge that DC current in a circuit with an AC RCD impairs it’s ability to work. Domestic properties have numerous electrical items, induction hobs, led lighting, EVCP’s that all can present this within an installation. There are loads of videos on YouTube explaining the logic behind it. It’s in the regs check out 531.3.3! As for sound judgement I’m an electrical technician with an electrical engineering degree and twenty years testing experience. I think I’m suitable to make the decision. There is a reason they are banned in half the world!

Common knowledge indeed, but one thing that remains mostly unanswered is how the DC current comes to be passing through the RCD in the first place? Presumably a circuit must be completed for it to flow, so is it deliberate, or is it due to faulty equipment? And what route does it take? Any answers (with sources) appreciated

 

[oscar21]

Another reg that I think is daft and has been changed for changes sake. I would have thought that in 99% of situations an AC RCD will be just fine, certainly in all the EICR's that I've done over the last few years and all the installations that I've done with AC ones they all tripped off no problem under test, the ones that didn't were physically faulty (BG ones for eg) despite houses being full of LED lamps, TV's, phone chargers, HOB's, washing machines etc etc.

My house is a perfect example, I have lots of electonic goods, your typical house, probably more so with my computer stuff, yet the 10 year old RCD trips off every time the oven is on for more than 30 minutes, its needs a new element.

From what certain people are saying I have a C2 problem in my consumer unit which is potentially dangerous, no RCD at all would be a C2 as well, so going by that logic I might as well ditch the RCD altogether and be no less safe than I was before, then I could have a roast dinner and not just a pizza.

 

[davesparks]

Another reg that I think is daft and has been changed for changes sake.

It has been changed because the older RCD technology is no longer appropriate in a modern installation.

I would have thought that in 99% of situations an AC RCD will be just fine, certainly in all the EICR's that I've done over the last few years and all the installations that I've done with AC ones they all tripped off no problem under test

Tripping under test conditions is different to tripping under real world conditions.

My house is a perfect example, I have lots of electonic goods, your typical house, probably more so with my computer stuff, yet the 10 year old RCD trips off every time the oven is on for more than 30 minutes, its needs a new element.

I think you've misunderstood some basics here.
The RCD tripping due to a faulty oven element does not prove whether or not it has been affected by any DC leakage.

 

[davesparks]

Common knowledge indeed, but one thing that remains mostly unanswered is how the DC current comes to be passing through the RCD in the first place?

DC leakage occurs from electronic loads, especially switch mode power supplies and the like.

Presumably a circuit must be completed for it to flow, so is it deliberate, or is it due to faulty equipment? And what route does it take? Any answers (with sources) appreciated

Lucien wrote a pretty good explanation of the DC leakage subject relatively recently. It should be possible to find it with a search.

 

[loz2754]

Can anyone explain type A RCD's

So after reading that other thread would I be correct in saying type A RCD's actually trip if it detects DC current thus disconnecting upstream RCD's from the DC fault anyway? I was under the impression that type A could only cope with DC leakeage as opposed to detect it.

www.electriciansforums.net

 

[oscar21]

It has been changed because the older RCD technology is no longer appropriate in a modern installation.

Tripping under test conditions is different to tripping under real world conditions.



I think you've misunderstood some basics here.
The RCD tripping due to a faulty oven element does not prove whether or not it has been affected by any DC leakage.

Well if it has been affected by DC leakage then it still tripped so all is good and if it hasn't been affected by DC leakage then all is good too. My house has all the possible scenarios that could cause DC leakage but the RCD still trips merrily away.

 

[MJPD29]

It’s common knowledge that DC current in a circuit with an AC RCD impairs its ability to work. Domestic properties have numerous electrical items, induction hobs, led lighting, EVCP’s that all can present this within an installation. There are loads of videos on YouTube explaining the logic behind it. It’s in the regs check out 531.3.3! As for sound judgement I’m an electrical technician with an electrical engineering degree and twenty years testing experience. I think I’m suitable to make the decision. There is a reason they are banned in half the world!

Sorry on behalf of everyone else we didn’t realise you had a degree, I’ll be the first to bow down to you. Our years practical experience working on site years means nothing when the piece of paper enters the room.

 

[timhoward]

I think there are valid points on both sides of this debate.

In general if an RCBO seems to work (and I usually try a type A test too even though it isn't required) then I feel I have no evidence that it isn't doing it's job, and it's a C3.
In practise I find most Type AC RCBO's will trip if tested on a Type A setting, and normally within required times. (Using the x5 test they sometimes didn't quite make 40ms but only by a few ms.). So far the only ones that didn't trip also didn't trip on a type AC test because they were stuck or broken.

If you remember the older No Trip loop testers used to used DLOK and deliberately inject DC to blind the RCD, and it's telling that this technology had to be abandoned as it fails to work on almost all RCBOs.

In general I'm more concerned about older shared RCD's where the cumulative DC leakage from multiple circuits can be higher, and the older ones seem to be more susceptible to DC blinding in the first place (backed up by the fact that older loop testers could successfully 'fool' them).
I'd love to get a DC leakage clamp meter to help make better informed decisions but it hasn't made it to the top of the wish list yet.

 

[davesparks]

Well if it has been affected by DC leakage then it still tripped so all is good and if it hasn't been affected by DC leakage then all is good too. My house has all the possible scenarios that could cause DC leakage but the RCD still trips merrily away.

Just because it trips it doesn't mean it is tripping at the right current or within the right time period.

This isn't a binary process whereby the RCD either is or is not prevented from working at all.

 

[Rockingit]

I'm wondering if this entire thread has missed the point. The OP seemed to be proposing a blanket C2 for any type AC RCD circuit, then did a bit of a back-track.

This is what really boils my P about 'inspectors' who lack the fundamental knowledge and learning to apply some engineering principles and instead just go looking for any reason to find a higher code because they think it justifies their existence.

Absolutely, there are instances where potentially use of an AC type rather than an A, B or F might be innapropriate due to bias and blinding, that's a part of mine, Davesparks and several others on here's specialism(s) every day, but it bears no relation to a lighting radial of 60W BC's in Mrs Miggins pie shop where an AC type is still perfectly acceptable as it provides entirely appropriate protection against the risk.

Sure, the lines have become murky in domesttic installations due to the changes in household and consumer appliances but if you are going to be ballsy enough to make an assertion about risk, then at least understand that risk and be able to justify it. If you think that a circuit might be compromised then simples - test it. Apply a ramp test when it's in normal use and see what happens - if your assertion that it needs a C2 due to DC blinding is true then it won't trip on a standard ramp test. If (when) it does, then you'll know to stop talking ball hooks and give it an entirely appropriate C3.

 

[nicebutdim]

I'm wondering if this entire thread has missed the point. The OP seemed to be proposing a blanket C2 for any type AC RCD circuit, then did a bit of a back-track.

Yesterday seemed to be a day for bold statements and back-tracking.

 

[Rockingit]

Yesterday seemed to be a day for bold statements and back-tracking.

I think I'm still in yesterday mate, hence my lack of patience with things like this!

 

[SparkySy]

With regard to code for AC RCD's. If it is tested with all loads connected and passes then I will C3 it, if it fails then C2 for obvious reasons.
In my working life as an electrician I can honestly say that apart from the obvious failures of RCD's I have only ever tested 3 that have been blinded by DC they were all within the last 2 years and all of them were the older BS4293 type of RCD!

It's funny as I have been having issues at a unit I have been testing with Welding plant melting the earth conductor insulation the entire length of a circuit.
That circuit is protected by an MEM 4pole 63a 30ma RCCB and a 32a MCB. The circuit has never tripped out until recently.
Upon inspection of the welders there doesn't seem to be any internal link between the main earth and the Welding plants return earth which was excellent but didn't solve my problem. I then clamped the circuit CPC and the supply cable to see if there was any leakage, there was no AC leakage.
I forgot the DC clamp meter so am going back tomorrow with it to see if it is DC that is blinding the RCD and helping to melt the earth conductors insulation!
Sy

 

[Aaron b]

I've not yet had an issue with them not tripping but I don't think it's possible to fully simulate the worst possible case either. I've never turn on someones inverter driven washing machine, rapid boiled some water on the induction hob, with the fridge left open and fired up the PC to see if it would still trip with everything going full pelt. Usually, I warn them that the power will go off so everything minus the fridge is off in fact.

I think Dave's point about RCDs being additional protection in many situations is very poignant too.

If appliance manufacturers could give more useful information regarding DC leakages we could make a much more informed decisions and the manufacturers could tell customers that their new hob will lead to firey death if that is true so we don't have to be the doomsayers.

 

[pc1966]

I expect very few items actually leak DC under normal conditions, otherwise they will fail the PAT style of test on insulation. They might inject DC components under fault conditions, but that is another matter.

My own understanding is C2 codes are for situations that are a single fault away from danger (C1 is immediate danger, C3 is not to current specifications but needs an unfortunate combination of circumstances) and as such for most cases type AC RCD are not that.

If they are on an EV or PV system then yes, a single fault could be a DC fault and then "say your prayers Blackadder, its poker time!"


Edit: Fun fact, the actor who played the baby-eating bishop of Bath & Wells in Blackadder S2 'Money' was Ronald Lacey, who also played the Gestapo officer in the Raider of the Lost Ark.