Why "No RCD protection on lighting circuit"?

[ban-all-sheds]

The most extreme example is probably those who seem to think the world might come to an end, metaphorically, of course, if somebody uses a few feet of red/black cable now to add a socket to an installation which is already entirely red/black and has no brown/blue anywhere.

I can, and have more than once, put forward a convincing case that as the law requires "reasonably safe" and does not require compliance with BS 7671, if the continued use of red/black cable is a reasonable thing to do, the law actually requires you to do that.

 

[JohnW2]

OK - some questions therefore for JW2 & PBC: How long after a change in the regulations should we have to wait before expecting people to comply with them? ... Unearthed lighting circuits used to be regarded as reasonably safe - would you regard installing a new one as OK? If not, how long ago do you think they stopped being reasonably safe?

In terms of the sort of examples I cited, I don't think it is a question of the amount of time that has passed but, rather, (given that the regs are virtually never retrospective) what proportion of the installation theoretically comes under current regulations. Maybe you would like regulations to eventually become retrospective - so that, X years after a regulation has appeared, everything is required to be brought up to that standard? (I can see an argument for that approach).

As I said, what seems basically silly is to require that, say, 4 feet of new buried cable or one new socket should be RCD protected if 200 feet of existing buried cable or 30 existing sockets in the installation are 'allowed' to remain without RCD protection.

... Same two questions for unshuttered sockets and using water pipes as earth, please.

Pragmatically speaking, the former is much the same as the above. If the installation already has 30 unshuttered sockets which are 'allowed' to persist, whether one extra which one installs is, or is not, shuttered is really neither here nor there. The latter is totally different, because it is a one-off thing. Given the progressive proliferation of plastic supply pipes, I could make a case for the 'prohibition' of relying on a water-pipe earth to be 'retrospective' - i.e. simply 'not allowed', now.

Kind Regards, John

 

[ericmark]

It is a watch my back situation. If I relate something which happened to me. I was fitting a new fridge/freezer which needed a water supply. Lucky just above where fitted was the upstairs toilet so easy to run some micro-bore pipe to a tape and connect to fridge/freezer. So two hacksaw blades and cut a channel down the wall. However it seems when the extension was fitted the two way switch cable was run horizontally between the two switches over 2 x 90° bends and 2 x 270° bends in the wall something I had not expected.

I had enough of a belt to end up on the floor and took half an hour to recover. This in spite of RCD on every circuit. I was part to blame for not testing for cables first but also cables should have not gone on that route with the change in wall direction.

Had I not recovered there would have been an inquiry. At the inquiry some one would have to decide who was at fault. If it was the route of the cables which was considered the reason for the accident then they will have presented a potential danger from the day they were fitted although not an immediate danger.

Code C3 means there is neither an immediate danger or a potential danger so one wonders where one can use code C3? No RCD is C3 as is missing labels, and over sleeving, and inaccessible area for inspection, not divided into enough circuits or not enough socket outlets. In other words not much point in having a code C3 as if not dangerous immediate or potential why even list it?

Now the old code 4 made some sense it was at least easy to show it should be coded. Why would anyone recommend improvements if not doing the improvements does not even have a potential of being dangerous?

So in real terms there are just two codes C1 and C2. Even C1 is suspect as if you remove power to test and find a C1 then you can't reinstate that power so if not powered then it's not a C1.

Reading best practice guide how can a lack of a notice warning the presence of an alternative supply be a C2 but lack of other notices be a C3? Also when testing the installation how can a non installed cable i.e. an extension lead generate a C2 listing?

The same for lack of RCD, it seems C3 unless also no RCD protection where required with BS7671:2001 then it's C2.

What it would seem is as each new version of BS7671 comes out all new stuff is C3 but when next one comes out then it becomes C2. Yet following the pre-BS7671 rules for lighting earths is still only C3?

What seems to be the point is any installation not following current BS7671 when designed is in the eyes of the courts dangerous should anyone be killed as a result. Well lets face it the fact some one has be killed does tend to prove it. But where when designed it did follow the current BS7671 then even if the BS7671 becomes tighter over the years any items which don't any longer comply will only be classed as code 3 and not considered as dangerous even if it is.

This has always been the problem inspecting and testing. One is armed with the current regulations not a sequence of books showing when the regulation changed so even in my house built in 1979 to find out if rewireable fuses were permitted when built when not under the supervision of a skilled person is hard as you simply don't have all the old books.

So what is the answer to watch ones back when doing an EICR. To me every departure must be listed it does not really matter what code is given if listed then no court can accuse you of not showing a warranty of skill. If you get the code wrong then so what, you have still pointed out the departure so owner is aware. Also general statements should be avoided, to say earth bonding missing is useless, to say earth bonding to water pipes in bathroom is missing however is OK. When the electrician comes to repair if he finds all points in BS7671:2008 which allow you not to earth bond water pipes in bathroom have now been met then he knows that is no longer a problem.

 

[JohnW2]

...the point is any installation not following current BS7671 when designed is in the eyes of the courts dangerous should anyone be killed as a result. Well lets face it the fact some one has be killed does tend to prove it.

I fear it's a lot more complicated than that. It is easy enough to show that an installation was not compliant with the current BS7671 and that someone was killed - but establishing a definite causative relationship between those two things will often be very difficult - for a court or anyone else. Indeed, if it were as simple as you imply (which it isn't), if someone were killed by an installation that was compliant with the current BS7671, then a Court would presumably have to conclude that BS7671 was responsible for the death!

Kind Regards, John

 

[ericmark]

I see if some one is killed on a level crossing and there is nothing wrong with the level crossing then the level crossing regulations must be at fault?

 

[JohnW2]

I see if some one is killed on a level crossing and there is nothing wrong with the level crossing then the level crossing regulations must be at fault?

Probably not (but not impossible!) - and that's my point about the converse. If someone is killed on a level crossing which did not fully comply with current regulations, that in no way 'proves', or necessarily even means, that the non-compliance had anything to do with the death.

Kind Regards, John

 

[JohnW2]

Deleted: quoted instead of edited - again!

 

[Risteard]

the law actually requires you to do that.

Absolute nonsense. The courts likely wouldn't support that view.

 

[ban-all-sheds]

In terms of the sort of examples I cited, I don't think it is a question of the amount of time that has passed but, rather, (given that the regs are virtually never retrospective) what proportion of the installation theoretically comes under current regulations.

OK - let's look at those then - exactly the same questions arise.

the thing I find laughable is to 'make a fuss' about (let alone 'code' as 'potentially dangerous and needing urgent attention') a few feet of non-RCD-protected new buried cable in an installation in which 95% of the buried cables are (without 'fuss') not RCD protected.

What if it's more than a few feet? 5? 10? When does it become more than a few, and less laughable? What if it's not 95% of the buried cables without RCD? What if it's 80%? 60%? How do you decide where the tipping point is? Do you have a logical, consistent set of criteria that have to be met?

Similarly with feeling the need to use an RCD socket for an additional one on a circuit which still has maybe a dozen or more non-RCD-protected sockets.

What if it's not one addition? What if it's 2? 4? What if it's not a dozen existing non-RCD-protected sockets? What if it's a mix of protected/not protected, 4&8? 6&6? How do you decide where the tipping point is? Do you have a logical, consistent set of criteria that have to be met?

As I said, what seems basically silly is to require that, say, 4 feet of new buried cable or one new socket should be RCD protected if 200 feet of existing buried cable or 30 existing sockets in the installation are 'allowed' to remain without RCD protection.

What if it's 14 new feet? Or 40? Or 4 new runs of 10'? What if the last person to install a socket put an RCD one in? Or on that circuit alone swapped the MCB for an RCBO?

Maybe you would like regulations to eventually become retrospective - so that, X years after a regulation has appeared, everything is required to be brought up to that standard? (I can see an argument for that approach).

That's a separate issue. As you say the idea has some merit, but that's not what I'm talking about. It's about whether people who choose compliance with BS 7671 as their way to comply with Part P should always comply with the current version, and not some earlier version, or mix of versions, because of the legacy installation.

Yesterday something complied, and was reasonably safe. Today it does not. Is it still reasonably safe? What's unreasonable about complying with today's regulations from today?

If you don't think you need to start today, when do you? Next week? Next month? Next year? How do you decide where the tipping point is? Do you have a logical, consistent set of criteria that have to be met?

 

[ban-all-sheds]

Absolute nonsense. The courts likely wouldn't support that view.

Why is a warning notice required when colours are mixed? Are notices like that generally used to flag up situations with decreased risks, or increased ones?

What's safer - a mix of colour schemes, or a single one used throughout?

 

[Risteard]

Your premise ignores the reason for the harmonised colours, which is one of safety. Workers in other CENELEC countries will not be confused by the colours, thereby enhancing safety.

Your argument is completely one-sided.

 

[ban-all-sheds]

Who is it most reasonable to assume will be working on domestic installations in the UK? UK electricians and UK homeowners, or workers in other CENELEC countries? My argument is not against harmonisation, or in favour of the Wiring Regulations requiring the continued use of red/black where that's what is already in use.

 

[JohnW2]

What if it's more than a few feet? 5? 10? When does it become more than a few, and less laughable? What if it's not 95% of the buried cables without RCD? What if it's 80%? ........ What if it's not one addition? What if it's 2? 4? What if it's not a dozen existing non-RCD-protected sockets? What if it's a mix of protected/not protected, 4&8? 6&6?
How do you decide where the tipping point is? Do you have a logical, consistent set of criteria that have to be met?

In terms of BS7671-compliance, it is not for me to decide where the 'tipping point' is or to 'have a logical, consistent set of criteria that have to be met' - but 'sensible' regulations might be expected to. As things stand, BS7671 is clear, in requiring any new work to comply with current regs, even if the vast majority of the installation doesn't. What is far from clear, and open to much debate and interpretation, is whether adding, say, and extra 5% of work which is non-compliant with current regs (but which should be compliant) to an installation in which 100% of pre-existing work was non-compliant (in the same way) with current regs (and doesn't need to be compliant) is sufficiently 'unreasonable' (in terms of safety) to constitute non-compliance with Part P. In the sort of cases I'm talking about, one could certainly have a good stab at arguing that it wasn't.

Maybe you would like regulations to eventually become retrospective - so that, X years after a regulation has appeared, everything is required to be brought up to that standard? (I can see an argument for that approach).

That's a separate issue. As you say the idea has some merit, but that's not what I'm talking about....

It would seem a fairly sensible idea, but its implementation (and policing/enforcement) would presumably involve all sorts of problems.

It's about whether people who choose compliance with BS 7671 as their way to comply with Part P should always comply with the current version, and not some earlier version, or mix of versions, because of the legacy installation. Yesterday something complied, and was reasonably safe. Today it does not. Is it still reasonably safe? What's unreasonable about complying with today's regulations from today?

As above, what is required to achieve BS7671 compliance is clear - the issue is what constitutes "reasonable provision ..." in order to achieve compliance with Part P.

Kind Regards, John

 

[ban-all-sheds]

In terms of BS7671-compliance, it is not for me to decide where the 'tipping point' is or to 'have a logical, consistent set of criteria that have to be met'

Of course it is and of course you have to.

As soon as you decide that you don't have to comply with BS 7671 editions/amendments when they come into force, but may wait until some point in the future, you absolutely must have a definition of your tipping point and you absolutely must have a set of criteria which make sense to you (even if nobody else). If you don't then your approach is without logic and without rigour, and therefore invalid - you cannot have regulations based on emotion.

- but 'sensible' regulations might be expected to.

I'd be interested to see a suggestion from you for changed wording of, say, the last sentence of 522.6.202, to provide a codified method, which could be consistently applied by everybody, to incorporate your concepts of "only a few feet of new cabling", or "x% of the existing wiring not complying already".

As things stand, BS7671 is clear, in requiring any new work to comply with current regs, even if the vast majority of the installation doesn't. What is far from clear, and open to much debate and interpretation, is whether adding, say, and extra 5% of work which is non-compliant with current regs (but which should be compliant) to an installation in which 100% of pre-existing work was non-compliant (in the same way) with current regs (and doesn't need to be compliant) is sufficiently 'unreasonable' (in terms of safety) to constitute non-compliance with Part P. In the sort of cases I'm talking about, one could certainly have a good stab at arguing that it wasn't.

This sort of thing seems to be a habit with you.

You decide that regulations do not mean what they say, or you decide that they don't have to be followed, and then you decide to embark on a process which you hope will result in an "interpretation" which corresponds to your decisions, and then you find that ambiguities and inconsistencies etc arise. When will the penny drop that you are the one responsible for those problems, not the people who wrote the regulations which you decided should not be followed?

As above, what is required to achieve BS7671 compliance is clear - the issue is what constitutes "reasonable provision ..." in order to achieve compliance with Part P.

Indeed, that is not prescribed.

But once the decision to use BS 7671 compliance as the "reasonable provision" then you can't have some vague, emotional non-rules of "it's only a few feet of cable", or "it's only one additional socket".

 

[JohnW2]

In terms of BS7671-compliance, it is not for me to decide where the 'tipping point' is or to 'have a logical, consistent set of criteria that have to be met'

Of course it is and of course you have to.

That's nonsense. As I said, BS7671, as written, is very clear. If I want to comply with BS7671, I have absolutely no choice but to ensure that all new work complies with the current edition/Amendment of BS7671, regardless of how much of the pre-existing parts of the installation do not comply with the current regs. If someone wants to comply with BS7671, there is nothing for them to 'decide'.

You decide that regulations do not mean what they say, or you decide that they don't have to be followed, and then you decide to embark on a process which you hope will result in an "interpretation" which corresponds to your decisions, and then you find that ambiguities and inconsistencies etc arise. When will the penny drop that you are the one responsible for those problems, not the people who wrote the regulations which you decided should not be followed?

As above, it's nothing to do with what "I decide". If I want to comply with BS7671, it is very clear what I have to do, and there is nothing that I need to, or can, decide.

The issue, as I've said, is not about BS7671-compliance (or the lack of it) but, rather, is about whether 'small percentage' non-compliance with BS7671 such as I mentioned constitutes failure to make the "reasonable provisions..." required for compliance with Part P.

Kind Regards, John