connecting CPC between socket and light circuit

[John D v2.0]

the cpc is not a live conductor it's just a connection to earth, so it is not specific to the circuit, it's just something that has to be present in the wiring system, and there could be loads of parallel paths which is fine.
If you wired everything in singles together you wouldn't push through one CPC per circuit, you'd just put one big enough for the largest expected fault current.

 

[JohnW2]

the cpc is not a live conductor it's just a connection to earth, so it is not specific to the circuit, it's just something that has to be present in the wiring system, and there could be loads of parallel paths which is fine.

That's essentially true, but I'm not sure about the compliance of a circuit which, on it's own, had too high a Zs for fault protection but which, when tested, had a low enough Zs because of parallel paths - since (unlike the live conductors and CPC of the circuit in question) there presumably can be no guarantee that those parallel paths will always be there.

It's a bit like the oft-mentioned earthing arrangement in my house. Most, probably all, of the final circuits have a measured Zs which is low enough for ADS to be provided by MCBs. However, it's a TT installation, and the low Zs figures only exist because of (I presume) a connection via bonding and a metal water supply pipe to my neighbours TN-C-S earth. Given that that 'parallel path' cannot be guaranteed to always be present (the water pipe between properties could suddenly become plastic one dark night), I would not be happy to rely on MCBs for fault protection - and I don't really know whether a TT installation with some circuits which were not RCD protected would be considered to be compliant with regs, even if MCB-based fault protection (judged by Zs) was adequate on the day when measurements were taken.

Kind Regards, John

 

[bernardgreen]

a connection via bonding and a metal water supply pipe to my neighbours TN-C-S earth.

A connection which I had to ensure could not happen between the TN-C-S "Earth" in my cottage and the TT Earth in the ajacent retail unit.

Originally they were one property with a single TT supply and single water supply. When they were divided into two separate units I had to have a new electricity supply installed for my cottage. The retail unit is supplied with it's water from my water supply ( the original supply to both units ) via a sub meter and a 2 metre long plastic "isolation" section of pipe.

An added "complication" is that one of my bedrooms is above part of the retail unit. The cables for their lighting are fortumately nowhere near the cables for my installation.

 

[JohnW2]

A connection which I had to ensure could not happen between the TN-C-S "Earth" in my cottage and the TT Earth in the ajacent retail unit.

Yes, we've heard that story many times, but I'm still not sure that I really understand the perceived problem, unless one is consdiering the 'vanishingly improbable'.

As I see it ... if, due to a fault, your TN-C-S 'earth' rose to a high potential relative to true earth then, if bonded pipes were joining your earthing system with that of the retail unit, then (given any likely Ze of their earth rod) the entire 'earthing' system of that unit would also rise to the same potential, wouldn't it - making that unit (internally) as 'equipotential' as yours.

Kind Regards, John

 

[bernardgreen]

the perceived problem,

is that if the "Earth" in the cottage was no longer at ground potential ( due to broken Neutral ) then current would flow in the retail units ground rod. If that ground rod was very low impedance then that ground current could be a large current. No protective device involved on that circuit.

With stone walls ( that are naturally slightly damp ) to ground and an internal "equipotential" above ground potential there can be a mild tingle if someone touches a wall and an "earthed" equipotential item. Hence the need to protect the retail unit's CPC from coming into contac with anything that could raise it above true local ground potential

If at some time in the future the retail unit upgraded to a TN-C-S supply then the Heutral from their supply would, via the water pipes, be connected to the Neutral of my supply. The two supplies via different street cables from the substation ( my cottage is on a corner, my supply from the side road, the retail unit from the main road, originally over head but now UG.

 

[John D v2.0]

I'm not sure about the compliance of a circuit which, on it's own, had too high a Zs for fault protection but which, when tested, had a low enough Zs because of parallel paths

I wouldn't class that as compliant, unless the parallel path was part of or run together with the same wiring system. There could be a debate about whether two T&E cables on different circuits could be classed as the same wiring system.

Most, probably all, of the final circuits have a measured Zs which is low enough for ADS to be provided by MCBs. However, it's a TT installation, and the low Zs figures only exist because of (I presume) a connection via bonding and a metal water supply pipe to my neighbours TN-C-S earth. Given that that 'parallel path' cannot be guaranteed to always be present (the water pipe between properties could suddenly become plastic one dark night), I would not be happy to rely on MCBs for fault protection - and I don't really know whether a TT installation with some circuits which were not RCD protected would be considered to be compliant with regs, even if MCB-based fault protection (judged by Zs) was adequate on the day when measurements were taken.

It would only make sense to calculate with the Ra/Ze plus R1+R2, rather than Zs testing directly with bonding connected, otherwise as you suggest you might be using the water pipe as a earth!

the Heutral from their supply would, via the water pipes, be connected to the Neutral of my supply

That sounds like it is the definition of an extraneous conductive part, very much liable to introduce a potential.

 

[JohnW2]

is that if the "Earth" in the cottage was no longer at ground potential ( due to broken Neutral ) then current would flow in the retail units ground rod. If that ground rod was very low impedance then that ground current could be a large current. ...

Yes, but that's why I said "given any likely Ze of their earth rod". With any 'credible' domestic/similar earth rod (yes, I know you've probably heard of very rare exceptions!), even if the TN-C-S 'earth' rose to 230V above true earth, one probably would not get enough current flowing into the earth rod that even a 1mm² conductor could not carry safely.

With stone walls ( that are naturally slightly damp ) to ground and an internal "equipotential" above ground potential there can be a mild tingle if someone touches a wall and an "earthed" equipotential item. Hence the need to protect the retail unit's CPC from coming into contac with anything that could raise it above true local ground potential.

That first sentence is "a fact of TN-C-S life", so you're stuck with it in your cottage, and if the DNO had felt that was not acceptable/safe, they presumably would not have allowed you to have a TN-C-S earth. However, I don't understand the second sentence ("Hence the need to protect the retail unit's CPC...") - if the DNO regard it as acceptable and safe for you to have a TN-C-S earth, why would they not consider it acceptable/safe for the retail unit's CPCs to rise to the same potential as in your cottage - does the retail unit have 'much wetter walls', or what?

If at some time in the future the retail unit upgraded to a TN-C-S supply then the Heutral from their supply would, via the water pipes, be connected to the Neutral of my supply. The two supplies via different street cables from the substation ( my cottage is on a corner, my supply from the side road, the retail unit from the main road, originally over head but now UG.

I've often wondered about that possible scenario, but it's a very different issue. I can but presume that it's a matter to which DNOs must give careful consideration when approving TN-C-S supplies. Even in the absence of 'direct' connections (as in the pipe between your cottage and adjacent retail unit), there's always going to be a risk that the ('required') bonding of extraneous-conductive-parts in two properties will bring their earthing systems into electrical continuity (as with me and my neighbour).

Kind Regards, John

 

[JohnW2]

I wouldn't class that as compliant, unless the parallel path was part of or run together with the same wiring system.

I don't think I would, either, IF I knew that to be the situation. However, who goes hunting everywhere for all possible parallel paths to earth before measuring Zs of a circuit? In practice, I imagine that one would almost certainly not know that one's Zs measurement was benefitting from a parallel path.

There could be a debate about whether two T&E cables on different circuits could be classed as the same wiring system.

I personally reckon that would be stretching things more than a bit. Other final circuits could easily be changed (even removed) such as to reduce or eliminate the 'parallel path'.

It would only make sense to calculate with the Ra/Ze plus R1+R2, rather than Zs testing directly with bonding connected, otherwise as you suggest you might be using the water pipe as a earth!

As I said, I think it's worse than that - since I think that I'm primarily using my neighbour's incoming neutral as my 'earth'! I suppose it could be just the water pipe, but with a Ze of ≤0.25Ω (even at times of 'drought'), I rather doubt it.

Kind Regards, John

 

[John D v2.0]

However, who goes hunting everywhere for all possible parallel paths to earth before measuring Zs of a circuit? In practice, I imagine that one would almost certainly not know that one's Zs measurement was benefitting from a parallel path

I assume that's why you measure R1+R2, so you needn't go hunting, at least for upstream parallel paths. However i agree it's possible for downstream parallel path that might be hard to notice.

 

[JohnW2]

I assume that's why you measure R1+R2, so you needn't go hunting, at least for upstream parallel paths. However i agree it's possible for downstream parallel path that might be hard to notice.

That's all very true, but I wonder whether, in practice, (m)any people take much notice of the R1+R2 ....

... everything in the regs (and OSG etc.) about requirements for ADS (included tabulated 'requirements') is written in terms of 'maximum Zs' and, unless one does that hunting (and disconnects any parallel paths) one's measurement of Zs will be undertaken with any parallel paths in place. I may be wrong, but I strongly suspect that on finding a Zs which was below the 'maximum permitted', the great majority of people would probably regard the circuit as being compliant in terms of ADS, probably without even noticing that their measured Zs was appreciably lower than (R1 + R2 + Ze), if that were the case.

I'm not defending such a less-than-rigorous practice, but I do think it is probably common for a circuit to be considered to satisfy the ADS requirements if the measured Zs is low enough, even when that measured Zs has been lowered by parallel paths.

Kind Regards, John

 

[John D v2.0]

True, although in the case of a new circuit, it would be fully tested before energising, so there wouldn't be the possibility of measuring Zs directly.
However I'll admit IR testing needs to have the CPCs, bonding, etc all connected to the MET, and R1+R2 would need it disconnected, so there's scope for confusion.
It would indeed be energised already in the case of existing circuits, maybe it would be pulled out properly for an EICR, but for some quick tests it might be done the other way round.

Unsurprisingly all the same arguments apply to Ze as well. All the neutral current could be going via everyone's bonding and as long as it measures out OK no one would be any the wiser. And it could vary all the way up to 0.35 ohms thus invalidating your ADS as well.

 

[JohnW2]

True, although in the case of a new circuit, it would be fully tested before energising, so there wouldn't be the possibility of measuring Zs directly.... It would indeed be energised already in the case of existing circuits, maybe it would be pulled out properly for an EICR ...

"Maybe" - but I seriuously doubt that it would hardly ever be done!

Unsurprisingly all the same arguments apply to Ze as well. All the neutral current could be going via everyone's bonding and as long as it measures out OK no one would be any the wiser. And it could vary all the way up to 0.35 ohms thus invalidating your ADS as well.

Indeed - but, in terms of measurement, there's absolutely nothing one can do about it - one can but measure the Ze 'as it is', without any certain knowledge of the paths concerned. Someone (I think eric) has suggested that adequacy of ADS should be judged on the basis of what Zs would be if Ze rose to the 'maximum permitted' (i.e. 'Ze by enquiry', and the judgement would then be msde on the basis of that Ze plus [R1+R2] ) - hence an assumed Ze of 0.35Ω for TN-C-S and 0.8Ω for TN-S. Logical though that might be, as a means of future-proofing against any possible changes in 'parallel paths', I would again be very surprised if any significant number of people ever do that!

Kind Regards, John