On a slight tangent (or thought experiment if you like) - say someone had deliberately installed a long thin circuit with L&N wired as 2.5mm² rings but PE as a 1.5mm² radial (looped into each socket in turn but finishing at the furthest socket) - so much as you might have if done for a ring in steel conduit with the conduit as c.p.c, but with a copper c.p.c. instead.  Zs at the furthest socket within what the OSG says for the required disconnection time given the OPD and a 1.5mm² Cu c.p.c..

Definitely doesn't comply with all BS 7671 requirements, but is there any actual (potential) danger? Would it warrant a C2 on an EICR?

    - Andy.

On a slight tangent (or thought experiment if you like) - say someone had deliberately installed a long thin circuit with L&N wired as 2.5mm² rings but PE as a 1.5mm² radial (looped into each socket in turn but finishing at the furthest socket) - so much as you might have if done for a ring in steel conduit with the conduit as c.p.c, but with a copper c.p.c. instead.  Zs at the furthest socket within what the OSG says for the required disconnection time given the OPD and a 1.5mm² Cu c.p.c..

Definitely doesn't comply with all BS 7671 requirements, but is there any actual (potential) danger? Would it warrant a C2 on an EICR?

    - Andy.

On the subject of a RF with a 1.00mm cpc. I recall back when I did inspections that such a circuit would not comply if it was protected by a BS3036 fuse but would comply if protected by a BS3871 or BS60898 mcb. Since when has this changed?

whjohnson said:

Since when has this changed?

Since then, there have been two changes to the calculation "numbers" which (on paper - separate debate 'real world vs paper') can make the answers different:

(1) Drop from 240 V to 230 V; and

(2) Use of Cmin.

Those would of course give you a lower current at the furthest end of the cable.

In addition, we now look at I²t for certain fault levels with circuit-breakers differently, particularly for highest prospective fault current at the origin of the circuit.

So, Real World, nothing has really changed to make any significant difference.

Paper means needless rewire or derate to 16A.

whjohnson said:

So, Real World, nothing has really changed to make any significant difference.

Not a direct conclusion in my opinion ... as I said, a separate debate.

What if new information came to light, that our calculations to previous editions of BS 7671 did not take into account certain things that happen in  the real world?

I'd agree the move to 230 V from 240 V could be considered questionable in this regard (although move to 230 V we have, and if we are doing calculations based on nominal voltage, things will change by some 4 % for first order proportionality).

But C_min and a change in the way we consider let-through energy of circuit-breakers are, in my opinion, definitely "new information to take into account".

No need to rewire the 1mm CPC ring - just to understand that there are a few corner case conditions where the cable may not be adequately protected so if a fault of that kind ever occurs, there is a small risk the cable  would then need to be replaced. The fact that there are many miles of it still in service half a century later suggests that the cable death fault is not that common a situation.

What has changed is that we now calculate and know about those corners.

Mike.

mapj1 said:

No need to rewire the 1mm CPC ring -

Not suggesting that is it. At the end of the day the person carrying out a periodic inspection and test is the one who should make the decision of how to code something.

BUT

There are situations, like one example I know of a relative, where the Zs was at the limit (for ADS) because of 2.5/1.0 ... BS 7671 updated and then no longer below the Zs limit (for ADS) ... and also no RCDs

So, in that case, the coding becomes a bit tricky, because you know you can no longer (according to the latest wisdom) provide protection against electric shock for ADS, and there's no additional protection either ... the two non-conformities (no RCD and Zs not quite met) merge into one in the same cause/effect, and perhaps some people might decide C2, others not ?

"...Definitely doesn't comply with all BS 7671 requirements, but is there any actual (potential) danger? Would it warrant a C2 on an EICR?"

a fair point and a fair question

i'd suggest no, not a C2 ...  but what do other's think ...

Perhaps, then, C3? So it does not conform to the latest edition of BS 7671, but it conformed when installed. If fact, if we didn't have the C3 code, BS 7671 would be retrospective and I don't think that anybody would suggest that it should be so.

I'd be inclined to agree, C3, at least if those were the only concerns, but it may get added to the weight of other things making up a fail depending condition of the rest - in the sense "it already fails on something else and by the way this another thing is also not really right".

Mike.