My Complete DIY Rewire

[bernardgreen]

The "redundancy" of the CPC in a ring does allow for a single fault ( broken CPC ) to occur without compromising safety.

One major advantage of a ring final is that continuity of Live Neutral and CPC can be tested at any point on the ring without the need for a wander lead back to the CU

With a radial the continuity of the CPC has to be ensured as there is no "redundant" CPC route in a standard radial. That is unless the electrician has installed a separate CPC from CU to the end of the radial to ensure that a single break in the CPC along the radial cable does not remove an Earth from any of the socket.

I accept that the "redundant" paths in a ring final will conceal a single fault from the house owner but at least the sockets are still safe with an Earth to all of them. A single fault in a radial may also go un-noticed by the house owner.

 

[ban-all-sheds]

You would obviously have to ask them.

I fear that they too are not applying logic.

I fear that they too have adopted the "that's the way it's been done because..." attitude.

When Amendment 1 2002 to the 16th was published, they had to address the fact that the old requirement

“minimum current carrying capacity Iz of the cable be not less than 0.67 x the
rated current setting In of the protective device”

was a problem because some installation methods had 2.5mm² T&E at 18.5A.

So they had the cable ratings re-done, they changed from 0.67x to 20A, and they added the requirement that under the intended conditions of use the load current in any part of the ring circuit should be unlikely to exceed for long periods the current carrying capacity of the cable.

And the reason they did all of that was that, in their own words, "keen to maintain the use of the UK ring circuit". So not for any sound engineering reasons.

Why were they so keen? If they didn't make sense and didn't seem safe, why relax the requirements so that they magically complied again?

They might feel that abolition of ring finals (which would probably take 50+ years to complete) would not result in enough marginal improvement in safety to be justified even in terms of their increasingly exactly requirements.

That it might take time is irrelevant - I don't suppose that all socket outlets and all concealed cables will be RCD protected any sooner, but they didn't regard that as a reason to not introduce those requirements for new work going forward. How long ago did cpcs in lighting circuits become mandatory? Are all lighting circuits OK now?

I would be surprised if anyone here could cite any safety issue they have encountered which arose because the circuit was a ring final- and, of course, there is the 'balancing' fact (as mentioned by stillp) that the redundancy of CPCs in a ring will undoubtedly have occasionally resulted in the avoidance of a 'safety issue'.

Well, what would surprise me would be to find that nobody here could cite any encounters with unfused spurs serving more than 1 socket or functionally undetected breaks in L or N.

 

[JohnW2]

I accept that the "redundant" paths in a ring final will conceal a single fault from the house owner but at least the sockets are still safe with an Earth to all of them. A single fault in a radial may also go un-noticed by the house owner.

Indeed. The house owner will not be aware of a single fault in the CPC of either a radial or ring circuit but, in the latter case, would at least remain 'protected' unless/until one or more additional CPC faults arose (seemingly extremely unlikely). A single CPC fault might, of course, increase the EFLI of the circuit to a level which does not provide adequate fault protection.

Of course, if 'they' were concerned about this, they could require CPC redundancy (via an additional cable) in a radial circuit - but where does one stop?!

Kind Regards, John

 

[PreferNotToDieOrBurn]

For goodness sake B-A-S, take it somewhere else.

This thread is about rewiring my house, not some greater philosophy on why things are the way they are.

 

[ban-all-sheds]

The "redundancy" of the CPC in a ring does allow for a single fault ( broken CPC ) to occur without compromising safety.

Indeed.

In 1 of the 3 conductors. The other ⅔ of the time a break is undetectable by the user and does compromise safety. And it compromises safety throughout, not just with the use of Class I appliances, of which there are fewer and fewer. And it compromises safety under normal conditions - it does not require a second fault (in a Class I appliance) for danger to arise.

One major advantage of a ring final is that continuity of Live Neutral and CPC can be tested at any point on the ring without the need for a wander lead back to the CU

So with a radial the reason you can't link conductors at the CU is...?

I accept that the "redundant" paths in a ring final will conceal a single fault from the house owner but at least the sockets are still safe with an Earth to all of them.

⅔ of the time that fault will make the entire circuit unsafe.

A single fault in a radial may also go un-noticed by the house owner.

⅔ of the time that fault will not be so unnoticeable and more importantly will not create a safety hazard until it is noticed.

 

[JohnW2]

... And the reason they did all of that was that, in their own words, "keen to maintain the use of the UK ring circuit". So not for any sound engineering reasons. ... Why were they so keen? If they didn't make sense and didn't seem safe, why relax the requirements so that they magically complied again?

Again, you would have to ask them.

I would be surprised if anyone here could cite any safety issue they have encountered which arose because the circuit was a ring final- and, of course, there is the 'balancing' fact (as mentioned by stillp) that the redundancy of CPCs in a ring will undoubtedly have occasionally resulted in the avoidance of a 'safety issue'.

Well, what would surprise me would be to find that nobody here could cite any encounters with unfused spurs serving more than 1 socket or functionally undetected breaks in L or N.

I'm sure that most will have encountered such things but, as I said, I do doubt that (m)any will have come across situations in which any of those things has resulted in a significant 'safety problem'. As I said, most ring finals are wired with cable having a CCC of 27A, so the 'worst' result of any of the things you mention would be no worse than having a 32A radial circuit (with branches) wired with 27A CCC cable - and, although it's obviously not 'right', I very seriously doubt that such an arrangement would result (or ever has resulted) in any significant 'safety problems'.

Kind Regards, John

 

[ban-all-sheds]

I regard a safety problem as something which has the potential to result in harm.

You don't have to fall off a motorbike for riding without a helmet to be a safety problem.

You don't have to crash your car for driving without a seatbelt on to be a safety problem..

You don't have to get lung cancer or heart disease or COPD etc for smoking to be a safety problem..

You don't have to have a cable damaged by overheating for not complying with Ib ≤ In ≤ Iz to be a safety problem..

 

[JohnW2]

The "redundancy" of the CPC in a ring does allow for a single fault ( broken CPC ) to occur without compromising safety.

Indeed. ... In 1 of the 3 conductors. The other ⅔ of the time a break is undetectable by the user and does compromise safety. And it compromises safety throughout, not just with the use of Class I appliances, of which there are fewer and fewer. And it compromises safety under normal conditions - it does not require a second fault (in a Class I appliance) for danger to arise.

What 'compromising of safety' are you talking about - the breaking of the L or N loop? If so,as I've just written, I don't believe that the consequential possible modest overloading of cable will ever, or hardly ever, result in any safety problems. Let's face it, apart from the 'safety margins', it's probably very uncommon for the total load on a 32A sockets circuit to be >27A for significant periods of time.

Kind Regards, John

 

[ban-all-sheds]

So if you were installing a 32A radial socket circuit, you'd happily do it, and consider it compliant with the Wiring Regulations, if the Iz was 27A?

 

[JohnW2]

You don't have to have a cable damaged by overheating for not complying with Ib ≤ In ≤ Iz to be a safety problem..

Maybe I wasn't clear enough. I was talking about actual safety problems (fires etc), which I still believe would be incredibly improbable in the fault situations you are contemplating, rather than non-compliance with safety-related regulations.

Kind Regards, John

 

[PreferNotToDieOrBurn]

If a wire overheats it will first melt, more than likely (extremely likely I would say) causing a short and tripping a fuse BEFORE there is even a vague risk of fire.

Now please stop derailing this thread.

 

[JohnW2]

So if you were installing a 32A radial socket circuit, you'd happily do it, and consider it compliant with the Wiring Regulations, if the Iz was 27A?

No, although I wouldn't expect any harm to actually result if I did. More to the point, if I installed a 32A radial circuit with cable whose Iz was ≥32A, but in the knowledge that, under certain fault conditions (which would only arise very rarely) Iz might fall to 27A, I might well not regard that as a risk I needed to be concerned about (in electrical terms).

Kind Regards, John

 

[EFLImpudence]

If a wire overheats it will first melt, more than likely (extremely likely I would say) causing a short and tripping a fuse BEFORE there is even a vague risk of fire.

Unless, of course, it catches fire or sets fire to something else first.

 

[JohnW2]

If a wire overheats it will first melt, more than likely (extremely likely I would say) causing a short and tripping a fuse BEFORE there is even a vague risk of fire.

So do I take it that you do not feel that there is any problem using a cable smaller than the regulations require?

Kind Regards, John

 

[PreferNotToDieOrBurn]

If a wire overheats it will first melt, more than likely (extremely likely I would say) causing a short and tripping a fuse BEFORE there is even a vague risk of fire.

Unless, of course, it catches fire or sets fire to something else first.

In the context of the debate about, which I really wish would just stop (atleast in this thread) - we're talking about overloading a 27A rated cable with MAX 32A which might melt a cable (I think it's very unlikely), and if it did, it would do so very slowly. Risk of fire would come from somewhere else and would be a separate issue.