work in houses with no rcd protection

[ericmark]

701.411.3.3 refers to new circuits so it depends if a FCU is fitted or not if a FCU is fitted then it forms a new circuit so would need to be a RCD FCU however if only the original fuse/MCB protects it then not a new circuit.

It's a very strangely-worded reg (and does not "refer to new circuits"). The wording is is strange because it refers to "all circuits", rather than "each circuit". That means that, if one installs a new circuit and RCD protects it (leaving other circuits serving the room not RCD-protected), that would not satisfy the reg as worded.

The regulations do not in the text refer to new circuits but in the first few pages it states something like.

BS 7671:2008 Requirements for Electrical Installations was issued on 1st January 2008 and is intended to come into effect on 1st July 2008. Installations designed after 30st June 2008 are to comply with BS 7671:2008.

So circuits designed before that date are not covered which ever version is valid at time of design is what is important.

701.415.2 Supplementary equipotential bonding is the odd one out where "(i) All final circuits of the location comply with the requirements for automatic disconnection according to Regulation 411.3.2".

So the way I read it any newly designed parts of the installation have to comply with current regulations but anything designed in the past only has to comply with regulations current at the time of design.

So if I want to add a socket in a house without an RCD I can use Ali-tube cable and a RCD socket or use a RCD FCU at origin and then standard twin and earth to standard socket.

The FCU is seen by BS7671 as forming a new circuit, but not it would seem by Part P which also means Part P needs it's own set of definitions as it clearly does not use those in the BS7671, however there are many other problems with Part P, not least in that it was not published in Welsh, so breaks the law. But as far as I am aware the question was not about the silly Part P regulations, but the IET/BSi regulations so what Part P considered as forming a new circuit has nothing to do with the thread.

 

[ban-all-sheds]

We have discussed that to death and I think the general consensus is that, although what you say is strictly true in relation to BS7671 definitions, it makes no sense in relation to notification requirements

I'm not convinced it even makes sense wrt BS 7671.

I know what the definition in there of "circuit" is, but think of the implications.

Can't use a MEIWC for adding an FCU.

And how many people put, or have seen certificates where someone else has put, each of the FCUs in an installation on this:

And what do you do with an FCU which has a flex outlet and is going to be used to supply, for example, a towel rail? How do you do continuity, IR and Ze tests on the circuit you've just created?

 

[ban-all-sheds]

The regulations do not in the text refer to new circuits but in the first few pages it states something like.

BS 7671:2008 Requirements for Electrical Installations was issued on 1st January 2008 and is intended to come into effect on 1st July 2008. Installations designed after 30st June 2008 are to comply with BS 7671:2008.

So circuits designed before that date are not covered which ever version is valid at time of design is what is important.

In which case if the work you are doing does not involve the installation of a circuit you are not required to do anything which applies to circuits.

The FCU is seen by BS7671 as forming a new circuit, but not it would seem by Part P

Part P says nothing about what is or is not a new circuit, and does not mention anything to do with circuits, be they new or existing.

This is what Part P says:

which also means Part P needs it's own set of definitions

What part of the requirement, or scope, stated here:



do you think needs defining in order to make it workable?

But as far as I am aware the question was not about the silly Part P regulations

Will you please, please, PLEASE explain what is silly about having a legal requirement to make reasonable provision in the design and installation of electrical installations in order to protect persons operating, maintaining or altering the installations from fire or injury?

 

[JohnW2]

We have discussed that to death and I think the general consensus is that, although what you say is strictly true in relation to BS7671 definitions, it makes no sense in relation to notification requirements

I'm not convinced it even makes sense wrt BS 7671. ... I know what the definition in there of "circuit" is, but think of the implications. ... Can't use a MEIWC for adding an FCU. ... And how many people put, or have seen certificates where someone else has put, each of the FCUs in an installation on this: [Schedule of Test Results] ... And what do you do with an FCU which has a flex outlet and is going to be used to supply, for example, a towel rail? How do you do continuity, IR and Ze tests on the circuit you've just created?

Agreed. We've discussed this at length before, and it is very difficult to come up with a definition of 'a circuit' which does not result in some anomalies. In the example you cite, I suppose one could argue that, given that it is analogous to using a plug/socket, anything beyond the FCU isn't part of 'the installation'. The situation of one CU feeding another introduces even more uncertainties.

Kind Regards, John

 

[ban-all-sheds]

The situation of one CU feeding another introduces even more uncertainties.

You've tried that one before.

It didn't work then, and it won't work now.

 

[JohnW2]

But as far as I am aware the question was not about the silly Part P regulations

Will you please, please, PLEASE explain what is silly about having a legal requirement to make reasonable provision in the design and installation of electrical installations in order to protect persons operating, maintaining or altering the installations from fire or injury?

I really don't think that this length post of yours serves any useful purpose. We all know that when eric writes about "Part P" he is usually actually talking about the notification requirements specified in the Building Regs, and it seems that nothing you or I can do will ever change him in that respect.

He is certainly right is saying that those 'notification rules' really do need to include, or refer to, a definition of "a new circuit" (which obviously requires definition of "a circuit") if they are to be clear, unambiguous and meaningful (hence enforceable!).

Kind Regards, John

 

[JohnW2]

The situation of one CU feeding another introduces even more uncertainties.

You've tried that one before. It didn't work then, and it won't work now.

As I said, there are uncertainties. A "CU" is just an enclosure containing one or more OPDs (and maybe also an RCD). An "FCU" is an enclosure containing an OPD (and maybe also an RCD). It would make little sense to me that, if fed from an existing circuit (originating at another CU), the former results in creation of a 'new circuit' downstream of the 'second' CU, whereas the latter (maybe two or more of them with a common feed) does not.

Kind Regards, John

 

[JohnW2]

So the way I read it any newly designed parts of the installation have to comply with current regulations but anything designed in the past only has to comply with regulations current at the time of design.

As scousespark recently pointed out, there clearly has to be a limit - some things compliant with 'ancient' regulations clearly would not be acceptable today! There also has to be a limit to "at the time of design" - I don't think that you could get away with installing something today which is non-compliant with current regs by claiming that it was 'designed' in 2007, when it would have been compliant! "At the time of installation" might be more sensible.

So if I want to add a socket in a house without an RCD I can use Ali-tube cable and a RCD socket or use a RCD FCU at origin and then standard twin and earth to standard socket.

Indeed, and the wording of the relevant regs allows that (no matter how 'silly' it may be!). If, however, in analogy with 701.411.3.3, the wording required that "all sockets on a sockets circuit must be RCD-protected", I think it would be less clear as to what, strictly speaking, would be compliant.

Kind Regards, John