Does Ring Final CPC continuity matter?

[ADS]

This may sound like a 'mad' question, but it's something I was thinking about the other day.

Scenario:

Ring Final, 32A MCB, RCD protected.

End to End

L - L.....confirmed

N - N....confirmed

CPC - CPC....no continuity

R1 + R2 reading at each socket outlet.........all within 0.05 ohms of each other.

Zs reading at each socket outlet........all within limit for MCB.

So, does the fact that there is no continuity end to end on the CPC matter?

And please, back up your answer with a 'regulation' or a description of 'why' it matters.

 

[ProDave]

YES it DOES matter.

Because it indicates you have a FAULT on the circuit.

Any known fault should not go without investigation.

Could be a simple loose connection, could be a damaged cable.

It certainly won't get better, and it could get worse.

To leave it is simply wrong.

 

[Revved Up Sparky]

On a ring main using 2.5 T&E, you effectively have 5.0mm L&N and 3.0mm CPC so with no CPC ring continuity your CPC is down to 1.5mm. Furthermore if a loss of continuity lost the E-E then there is a further risk of loss of continuity resulting in total loss of CPC.

 

[ADS]

YES it DOES matter. Because it indicates you have a FAULT on the circuit.

Any known fault should not go without investigation.

Could be a simple loose connection, could be a damaged cable.

It certainly won't get better, and it could get worse.

To leave it is simply wrong.

Point taken, so I'll be a bit clearer.

You're changing a board.

All connections have been checked and confirmed, all readings are as expected and within limits..........except no CPC end to end.

So, in other words, somewhere in the fabric of the building, about half way around the ring, you have a break in the cpc.

Does it actually matter?

Technically, how is it a 'fault'? All sockets are earthed and all Zs within max. limits?

On a ring main using 2.5 T&E, you effectively have 5.0mm L&N and 3.0mm CPC so with no CPC ring continuity your CPC is down to 1.5mm.

How would that actually affect the CPCs effectiveness - with 30mA rcd protection ADS happens in less than 40mS at 150mA......so would the loss in CSA matter?

 

[sellers]

I agree with you have a fault on circuit, but.... a 4mm twin only carries a 1.5mm cpc.

So surely a 1.5mm cpc on a 32a mcb us enough to handle fault current. Just a thought.

 

[Phoenix]

And please, back up your answer with a 'regulation'

.

543.2.9 Except where the circuit protective conductor is formed by a metal covcring or enclosurc containing

all of the conductors of the ring, the circuit protective conductor of every ring final circuit shall also be run in the

form of a ring having both ends connected to the earthing terminal at the origin or the circuit.

 

[ADS]

I like that one, Phoenix.

So, in the situation that I describe, would you complete an EIC for the board change (as it only applies to the new board) and comment on the existing installation that there was no end to end CPC continuity on the ring?

 

[Phoenix]

I'd perfer to try and find it, but I suppose if the client wont pay for it (and its not just going to be a half hour that you can 'loose' into another part of the job. Then I guess as long as you have earthing to every point and Zs is in spec then you could get away with a note on certitificate detailing the fault on the logic that you haven't made the issue any worse than it was before, and infact have inproved it by fitting RCD.

That reminds me of a fault we had in a school the other week. Ring wired in 4mm

 

[ADS]

I'd perfer to try and find it, but I suppose if the client wont pay for it (and its not just going to be a half hour that you can 'loose' into another part of the job. Then I guess as long as you have earthing to every point and Zs is in spec then you could get away with a note on certitificate detailing the fault on the logic that you haven't made the issue any worse than it was before, and infact have inproved it by fitting RCD.

That's what I was thinking - if there was no end to end contnuity on the 'live' conductors, and the fault couldn't be found within the accessories, you'ld probably drop the MCB to a 20A, as it would be dangerous to leave on a 32A - but in this case, there wouldn't be any point in dropping the rating of the MCB.

It's not dangerous, so the only problem with it is that, as you showed us, it doesn't comply with the regs.

 

[ProDave]

That's what I was thinking - if there was no end to end contnuity on the 'live' conductors, and the fault couldn't be found within the accessories, you'ld probably drop the MCB to a 20A, as it would be dangerous to leave on a 32A - but in this case, there wouldn't be any point in dropping the rating of the MCB.It's not dangerous, so the only problem with it is that, as you showed us, it doesn't comply with the regs.

and if you can't find the fault within the accessories, then you KNOW there's a damaged cable, or hidden junction box etc, so you KNOW there's something wrong.

If you can't find the fault in the accessories, then at least determine which leg has the fault, split the ring (isolating the faulty section) and re connect as two 20A radials.

To do nothing IMHO is not acceptable. I always make clear on a CU change that there MAY be faults found that NEED correcting, so the estimate given for a CU change is just that, an estimate, and the final bill MAY be higher if faults are found.

 

[ADS]

I do agree with you, Dave - it was more a question of 'Does it actually matter?' - in terms of 'electrical theory', 'calcs', 'testing'....whatever you want to call it.

In other words, we know what we should do etc etc. but, in theory, does it actually change anything under the circumstances that I describe? (apart from not complying with current regs, as pointed out)

 

[apprentice87]

Hi all,

I would agree with all that has been said so far. Only other thing i would say, is that, ok, you have established that there is a break in the cpc. If another one occurred though, that would leave a whole section with no cpc at all.

I know that this in practice, would be no more dangerous, than if the cpc developed a fault on a radial.

The point i am trying to make is; say you had a "healthy" radial. You would obviously not give it a code, because "it would be dangerous if it went wrong" but here you have a circuit that you KNOW has a fault already.

Does this not mean, that it should have a code one, and maybe, maybe just, have a code two perhaps?? because if a second fault developed it would be dangerous?? Perhaps not... But it is still not right!!!!

I would try to fix it though, and if the customer did not want it fixing, just make a note of this on the cert, and on the invoice too. Be his funeral!!!!

Very very good thought provoking question from ADS though!!!

john...

 

[Technician]

The point about ring final circuits is that

load current for any one socket is carried by

two conductors.

The same holds true for earth fault current.

Leaving an open circuit on the cpc means that

the fault current cannot pass along the two

conductors where it will in the line conductors.

 

[SLIPSHOD & SLAPDASH]

How did you get a R1 + R2 reading when there is no continuity of the cpc. You must have tested as a normal 'radial' circuit which is not correct as the L & N are in a ring ( but i do see your point)

 

[ADS]

How did you get a R1 + R2 reading when there is no continuity of the cpc. You must have tested as a normal 'radial' circuit which is not correct as the L & N are in a ring ( but i do see your point)

If you imagine a ring final with 10 socket outlets and a break in the CPC somewhere between sockets 5 & 6 - then cross connecting 'line' & 'CPCl' at the board would still give a reading - although not the correct one - but you wouldn't know that anyway, because you haven't got an r2 end to end to compare with.

You could, obviously, still calculate what you should get by using your other end to end readings - but like I said, this isn't the issue - it's more a case of is there actually anything wrong (or dangerous) with a broken CPC in an RCD protected, ring final......apart from not complying.

To be honest, if you think of a steel conduit install using the conduit as the CPC - you could still wire 'Line' and 'Neutral' as a ring, with CPC flyleads connecting socket outlets to the conduit back boxes.

All the sockets are still 'earthed' and your Zs will be fine, but the CPC isn't a ring, it's a radial - and you'ld probably measure R2 as oposed to R1+ R2.

 

[wozz]

Reading technicians post and the reg all i could think of is what about metal conduit used as a earth? and then ADS went and said it.

 

[ProDave]

If you imagine a ring final with 10 socket outlets and a break in the CPC somewhere between sockets 5 & 6 - then cross connecting 'line' & 'CPCl' at the board would still give a reading - although not the correct one -

But crucially it would give a DIFFERENT reading at each socket which is one of the things you are checking for, so that alone would tell you that there's a problem.

 

[soulman]

A rfc having a discontinuous conductor is given a code 2 in the ESC best practice guide.

 

[SLIPSHOD & SLAPDASH]

If you imagine a ring final with 10 socket outlets and a break in the CPC somewhere between sockets 5 & 6 - then cross connecting 'line' & 'CPCl' at the board would still give a reading - although not the correct one - but you wouldn't know that anyway, because you haven't got an r2 end to end to compare with.You could, obviously, still calculate what you should get by using your other end to end readings - but like I said, this isn't the issue - it's more a case of is there actually anything wrong (or dangerous) with a broken CPC in an RCD protected, ring final......apart from not complying.

To be honest, if you think of a steel conduit install using the conduit as the CPC - you could still wire 'Line' and 'Neutral' as a ring, with CPC flyleads connecting socket outlets to the conduit back boxes.

All the sockets are still 'earthed' and your Zs will be fine, but the CPC isn't a ring, it's a radial - and you'ld probably measure R2 as oposed to R1+ R2.

As Dave has mentioned if you tested correctly the 'figure 8' cross test would even point you to the direction of the fault.

If you calculated the Cpc's of twin & earth and of conduit using the adebatic equation you would get the answer .

 

[Mark2spark]

As others have said, it 'matters' because a) it's not to regs, and B) the circuit design of a ring is based on both legs of the cpc carrying the fault current.