Non-bidirectional RCD/RCBO feeding sockets, should they get a code C2 with EICR?

[JohnW2]

Law no, but regulations yes, BS 1363 says "Plugs shall not be used for the connection of electrical power generators to socket-outlets." so following the rules can't use a 13 amp socket.

Yes, I realise that, but you were using the fact that people 'are not allowed' to plug in a solar inverter as a reason for 'passing' the installation (in an EICR) - but if there is no law preventing one plugging in such an inverter, that argument surely doesn't work, does it?

In any event, that argument is really moot. As I've said, if everything else is OK, one would presumably still have to 'pass' the installation (even if a 'not allowed' inverter were plugged in), since something which 'happens to be plugged in when the inspection is done' is not within the scope of an EICR, is it?

 

[JohnW2]

From what I understand, a non-bidirectional RCD/RCBO isn’t automatically a C2 just on its own. It really depends on whether it still provides effective protection for the circuit it’s feeding and whether it’s installed in line with the manufacturer’s instructions. .... A lot of these devices are only designed for supply from one direction, so if they’re wired the “wrong” way round they may not operate correctly under fault conditions.

Unless you know something that the rest of us don't, it would seem that a traditional RCD will, 'initially', provide the required fault protevtion (for currents travelling either way through it).

As being discussed in this and other current threads, the potential problem arises if someone subsequently presses the test button whilst current is flowing 'backwards' (from a generator or inverter) through a traditional ('unidirectional') RCD and keeps the button depressed for a while after the device trips. Per the video that eric has referenced, that may result in destruction of the electronics of the RCD, resulting in a situation in which it would no longer provide fault protection (with current flowing in either direction).

.... It really depends on ..... whether it’s installed in line with the manufacturer’s instructions. .... A lot of these devices are only designed for supply from one direction, so if they’re wired the “wrong” way round they may not operate correctly under fault conditions.

As we have discussed, many (perhaps most?) traditional ('unidirectional') RCDs do not bear any indication of which way around they should be installed, and nor does associated documentation give any instructions/guidance about this.

That’s where an electrician would normally flag it during an EICR — but whether it’s C2 or just C3 tends to come down to actual risk and compliance, not just the wording on the device.

The potential/theoretical risk (related to use of the test button) will surely be identical in all situations, so there will be no rational/logical basis on which an electrician could 'choose' between C2 and C3 in relation to a particular installation, will there?

In practice, I’ve seen it recorded both ways depending on the tester’s interpretation, so there isn’t always a strict blanket answer.

Quite - it's just down to the view of the tester (which, as above, should really be 'consistent' for all EICRs they do) - which is a pretty unsatisfactory situation.

One thing that hasn't been discussed much is that 'plug in solar' systems are 'portable' and hence, by definition, could be plugged in, at any time, into any socket in any house/installation. That really means that if an inspector believed that an inverter plugged into a socket protected by a unidirectional RCD deserved a C2, and also believed (which I personally don't) that the scope of an EICR should consider what was (or could be) plugged into any of the sockets, then that inspector would logically give a C2 to every unidirectional RCD protecting a sockets circuit in every installation they inspect, even if there is currently no evidence that the household possesses a 'plug in solar' system ??

 

[ericmark]

I suppose I should have used an emotion to show I was not serious about it attracting a C2, it does seem that the test button needs using to cause them to fail, so reverse direction and test button at the same time is unlikely.

Not following the regulations is abuse in the same way as hitting the socket with a pram or wheelchair, where if the socket is fitted at the correct height, then it is above the self-propelled wheelchairs rear wheel centre point so will not be hit by the wheel, but even if installed low, wheelchair uses should take care, so sockets should not be mounted 336 to 362 mm from the floor, as at that height likely they will be hit by the back of the wheel where the user can't see what it is hitting as behind them.

So at 450 mm the socket is above the danger zone.

 

[JohnW2]

I suppose I should have used an emotion to show I was not serious about it attracting a C2,

Maybe you weren't serious, but it would seem that imtiaz_ahmad was suggesting C2 as a possibility.

However, as I recently wrote, given the 'portability' of something like a plug-in solar system, someone who believed that use of such a system with a unidirectional RCD deserved a C2 would have to give a C2 to every sockets circuit in every installation that was protected by a unidirectional' RCD, 'just in case someone plugged in such a system - which would be daft.

it does seem that the test button needs using to cause them to fail, so reverse direction and test button at the same time is unlikely.

Maybe, but certainly far from impossible. One imagines that, in Summer, in an installation with very low background loads, there could possibly be net export for 'a few hours' every day, and it's not impossible that someone could press a test button whilst that was the case.

Not following the regulations is abuse in the same way as hitting the socket with a pram or wheelchair, where if the socket is fitted at the correct height, then it is above the self-propelled wheelchairs rear wheel centre point so will not be hit by the wheel, but even if installed low ...

Again, maybe, but there's the same problem as above, given that wheelchairs, just like plug-in solar systems, are 'portable'. If (as you may be implying) a socket 'at a dangerous height' in a house where there was a wheelchair deserves a C2, then one could argue that every house with sockets at that height should be given a C2 - 'just in case' a wheelchair enters the property at some point in the future!

 

[plugwash]

You could say very similar things with type AC RCDs given the prevalence of electronic devices, but the power that be don't seem to consider it a big enough issue to warrant a C2.

 

[JohnW2]

You could say very similar things with type AC RCDs given the prevalence of electronic devices, but the power that be don't seem to consider it a big enough issue to warrant a C2.

Well, I suppose I have to admit that I, personally, often do 'say very similar things' about Type AC RCDs

One interesting thing is the 'excitement' that about unidirectional RCDs has only shown its face with the advent of local generation. However, the fact is that, given that a lot (many/most?) of traditional ('unidirectional') RCDs bore no markings or guidance about 'which way around' they should be wired, it's far from impossible that we have been living for decades with at least some in-service RCDs wired in such a way as to invoke the potential 'test button problem' - but that possibility was very rarely even mentioned, let alone regarded as a cause for concern!

 

[ericmark]

However, the fact is that, given that a lot (many/most?) of traditional ('unidirectional') RCDs bore no markings or guidance about 'which way around' they should be wired, it's far from impossible that we have been living for decades with at least some in-service RCDs wired in such a way as to invoke the potential 'test button problem'

Yes, I started a thread about this problem, we in real terms have many RCDs where we have no idea if bidirectional or unidirectional, even the data sheets don't help.

If (as you may be implying) a socket 'at a dangerous height' in a house where there was a wheelchair deserves a C2, then one could argue that every house with sockets at that height should be given a C2 - 'just in case' a wheelchair enters the property at some point in the future!

It does really need a risk assessment, a socket mounted low down in a bedroom, so the bed head is above the socket level so the bed being moved will not damage the socket, would make sense if where the bed is likely to be placed.

Upstairs is clearly not a problem with a wheelchair, honest, they will not go upstairs. We are looking at where the wheelchair user needs to manoeuvre, straight runs are not a problem. I know the danger height, I was it seems forever touching up where my late mother hit walls and skirting boards. As late dad would get upset.

And only a problem with self-propelled wheelchairs, both attendant controlled and electric wheelchairs have smaller wheels, and are more compact, it is the way the wheel sticks out at the back, where the user can't see it, that causes the damage when manoeuvring in tight spaces.

 

[JohnW2]

It does really need a risk assessment, a socket mounted low down in a bedroom, so the bed head is above the socket level so the bed being moved will not damage the socket, would make sense if where the bed is likely to be placed.

Yes, but just like plug-in inverters and wheelchairs, furniture (including beds) is portable/moveable - so a 'risk assessment' undertaken on the basis of 'how things are'today could become totally different tomorrow.

Back to context, as I always say, sockets circuits are really a designer's (or inspector's) nightmare, since what the designer (or inspector) assumes about the nature and size of loads that will be plugged in 'in the future' (maybe 'tomorrow') cannot be based on anything other than 'intelligent guesswork'. In particular, in context, the fact that there is not a 'plug-in solar' system in the house today does not mean that the householder could not by one, and 'plug it in' tomorrow.

 

[ericmark]

Simple answer is, the unidirectional RCBO does not cause a potential danger, it may blow a transistor or whatever it was, but that did not cause a danger in its self, and if it has blow due to the test button being pressed, then likely the fault will be found next time it is pressed.

May as well say, should fit two RCDs in series, so if one fails the other one will still work. Like with caravans where both the caravan site and the caravan need RCDs in case the other one is not fitted.

 

[JohnW2]

Simple answer is, the unidirectional RCBO does not cause a potential danger, it may blow a transistor or whatever it was, but that did not cause a danger in its self

I suppose it depends upon what one regards as a 'potential danger'. It appears that there is a risk that pressing the test button will destroy the electronics such as to destroy its ability to 'work'. There is hence a risk of the situation becoming no better that it would be with no RCD at all - but I imagine that you would agree that the absence of a ('required') functioning RCD would be a C2, wouldn't you?

, and if it has blow due to the test button being pressed, then likely the fault will be found next time it is pressed.

Yes, I've already made that observation). If the problem arose as the result of someone pressing the button, then they would probably press it again within at least a few months, and thereby discover the problem - much earlier than than the 5-10 years (or whatever) before the next EICR was done!

In fact, a lot of that potential issue could be avoided by telling people to press the button a second time (after resetting) if the first press caused it to trip. That would mean that any 'destruction' resulting from the first 'press' would be detected immediately.

May as well say, should fit two RCDs in series, so if one fails the other one will still work.

I've often suggested that, particularly to those who believe that something like 7% of in-service RCDs are 'faulty'.

 

[ericmark]

It appears that there is a risk that pressing the test button will destroy the electronics such as to destroy its ability to 'work'.

Agreed, but there are many actions which can cause something not to work, so if for example a surge could cause it to fail, would the lack of surge protection device also mean a code C2?

 

[SimonH2]

One thing that hasn't been discussed much is that 'plug in solar' systems are 'portable' and hence, by definition, could be plugged in, at any time, into any socket in any house/installation. That really means that if an inspector believed that an inverter plugged into a socket protected by a unidirectional RCD deserved a C2, and also believed (which I personally don't) that the scope of an EICR should consider what was (or could be) plugged into any of the sockets, then that inspector would logically give a C2 to every unidirectional RCD protecting a sockets circuit in every installation they inspect, even if there is currently no evidence that the household possesses a 'plug in solar' system ??

There is precedent for considering what might be plugged in.
In the past, only a few circuits required RCD protection - one of them being "sockets that might be used for equipment outside", so any socket that could be considered handy for running an extension lead outside. At inspection time there may be nothing plugged in, but the inspector has to consider whether such a use might be made in order to determine if RCD protection was required.

 

[JohnW2]

Agreed, but there are many actions which can cause something not to work, so if for example a surge could cause it to fail, would the lack of surge protection device also mean a code C2?

I take your point, but there are a lot of IFs associated with such a statement, so an (inevitably 'personal') judgement (essentially a 'risk assessment') is required.

If I believed that there was a significant risk that a 'surge' would cause an RCD to fail, and if I also believed that an SPD would be likely to prevent a significant number of such events, then my judgment might well be that, for that reason, absence of an SPD probably deserved a C2 (although we have to remember that BS7671 says that even the absence of a 'required' RCD should ('only') give "at least a C3).

However, since I am very unconvinced about either of those "ifs", I probably would not give it a C2, and would probably even be hesitant to give it a C3.

 

[JohnW2]

There is precedent for considering what might be plugged in.
In the past, only a few circuits required RCD protection - one of them being "sockets that might be used for equipment outside", so any socket that could be considered handy for running an extension lead outside. At inspection time there may be nothing plugged in, but the inspector has to consider whether such a use might be made in order to determine if RCD protection was required.

Indeed, but in that case it was an 'intelligent judgement', since there was a rational basis for thinking that some sockets were fairly likely to be used to feed outdoor equipment (at the least, sockets more likely than others to be used for that purpose).

In the case of 'plug-in solar' it would be a sheer guess on the part opf an inspector as to whether the householder was going to buy (and 'plug in') such a system 'tomorrow' - or, indeed, whether the householder already had one, but had unplugged it and hidden it away in a wardrobe/wherever for the duration of the inspection

 

[ericmark]

There are a load of things which can be plugged in which do not even conform to BS 1363. Never seen one of these

marked with BS 1363, and I have seen a website called fatefully flawed where they point out the dangers, so all we can do it put them in a bag and present to owner to bin them.

As to condemning sockets which had them plugged in as the socket could be strained as a result of fitting a non BS 1363 items, I think that is going too far.