Earth Bonding trouble

[EFLImpudence]

I was thinking more along the line of why do the regs. not state that the main bonding should be connected to the ground side of the meter.
Then if the meter were removed the pipe in the ground (although not very important) would still be connected. Obviously the pipework would not but without the meter they may not be liable to introduce earth potential but this should then be determined as would anything else.

As it is now, if the meter pipe connections are not of negligible impedance (because of sealant etc.) then the connection to true earth would be reduced. With the bonding on the other side (of the meter), this would not matter or if it does then what level of impedance is the cut-off?

I feel people in this thread are not distinguishing between main bonding (10mm² to the MET. to equalise potential between system and true earth of an installation) and supplementary bonding (to equalise potential between parts of a location - both in the event of a fault).

I'm inclined to the view that, once one excludes silly (e.g. door knobs) and/or contentious (e.g. metal baths) things, that adage of 'if it's metal, earth/bond it' is probably the common sense approach, certainly for domestic pipework, given the various uncertainties (present or future) which often exist.

Well, as you know I disagree.
As for door knobs, why are they silly exclusions thinking as you do? Imagine a bathroom with plastic pipework but with metal bath, metal basin and even a metal w.c. then none of them should be bonded.

 

[JohnW2]

I was thinking more along the line of why do the regs. not state that the main bonding should be connected to the ground side of the meter. Then if the meter were removed the pipe in the ground (although not very important) would still be connected. Obviously the pipework would not but without the meter they may not be liable to introduce earth potential but this should then be determined as would anything else.

Indeed - that's essentially the very point I made myself a couple of posts back. In fact, 544.1.2 almost contradicts itself. It starts by saying that main bonding should be 'as near as practicable to the point of entry of the service into the premises. Given the whole point of MPB, that makes total sense. It then goes on to 'spoil it' by saying that if there there is an insulating section/insert, or if there is a meter, then MPB should be attached on the customer's side of the interuption or meter - which really makes no sense.

As it is now, if the meter pipe connections are not of negligible impedance (because of sealant etc.) then the connection to true earth would be reduced. With the bonding on the other side (of the meter), this would not matter or if it does then what level of impedance is the cut-off?

Exactly.

I feel people in this thread are not distinguishing between main bonding (10mm² to the MET. to equalise potential between system and true earth of an installation) and supplementary bonding (to equalise potential between parts of a location - both in the event of a fault).

Well, I was certainly distinguishing; I even refered to the need to decide whether a G/Y conductor needed to be 10mm² or 4mm².

Well, as you know I disagree. As for door knobs, why are they silly exclusions thinking as you do? Imagine a bathroom with plastic pipework but with metal bath, metal basin and even a metal w.c. then none of them should be bonded.

We don't disagree as much as you seem to think - at least, not any longer. Whilst there are arguments in both directions, I have been persuaded that, on balance, it's best for none of those metal items to which you refer should be bonded.

The more contentious issue, particularly given the regs, is that exactly the same argument says that, if one does have an 'insulating section' (or a plastic meter) near the origin of the water system, one should not bond the house's pipework (which, being on the house side of the interruption, is not in continuity with the e-c-p).

However, as I said in my last post, and given that (as with metal bath etc.) the potential hazard arises due to 'unnecessary earthing' of metawork, this all becomes pretty moot when one realises that all (assuming metal) pipework in a house will be connected to earth via CPCs of attached equipment. The risk associated with having earthed metalwork will therefore usually apply to all metal pipework, whether one has installed any explicit 'bonding' or not. I suppose what I was really saying in my last post was that, given this virtually inevitable 'earthing' (connection to MET) of all pipework, one might as well give it a nice fat 10mm² connection to the MET, rather than just CPCs.

Kind Regards, John.

 

[EFLImpudence]

The more contentious issue, particularly given the regs, is that exactly the same argument says that, if one does have an 'insulating section' (or a plastic meter) near the origin of the water system, one should not bond the house's pipework (which, being on the house side of the interruption, is not in continuity with the e-c-p).

As you say, although the O.S.G. does state with regard to that situation that -

"...main bonding is recommended unless it has been confirmed that any pipework within the building is not introducing earth potential."

 

[JohnW2]

As you say, although the O.S.G. does state with regard to that situation that -
"...main bonding is recommended unless it has been confirmed that any pipework within the building is not introducing earth potential."

Quite - but, as I wrote a day or two ago, just trying telling that to the person undertaking a PIR/EICR on an installation with no main bonding!

Kind Regards, John.

 

[holmslaw]

..

 

[holmslaw]

..

 

[bernardgreen]

Plastic gas supplies change to steel before entering buildings

My understanding is that plastic gas pipes are not permitted where there is any risk of the plastic being exposed to heat or fire ( accidental ) unless a automatic cut off valve is fitted. Hence in the building the pipe work would ( should ) be metallic.

The bonding--earthing--CPC confusion is the result of a messed up compromise between safety requirements and cost reductions in cabling both in the local distribution network and internal wiring. A compromise which has ended up creating the stupid situation that anything at true ground potential inside the house is a potential hazard.

The hazard ( for new readers ) exists when the CPC ( falsely called the "earth" wire ) rises to a voltage above true ground due to a fault in the local network forcing the neutral away from ground potential.

Why is 4mm 6mm or 10mm required for bonding the incoming plastic water supply to the CPC MET ? One reason is probably that the requlation has to cover the situation where the water supply pipe is still metallic and would be part of a return circuit to the substation star point. A circuit that would need to be safe while carrying the several hundred amps that a floating neutral could need to discharge via the CPC's and bonding of domestic properties.

The question has never been answered as to how a conductive stone wall ( moisture content ) can be "bonded" to prevent it introducing a potential from the true ground that is different from the potential on the neutral derived CPC in the property. For me the common sense answer is to use a ground rod in a TT installation and have the CPC at true ground potential.

 

[Paul_C]

It's presumably likely to become an increasing debate - 'over half a century ago', there were probably no non-metalic service supply pipes; in another half a century or so, there may be few metal ones left.

And half a century ago you could pretty much guarantee that the pipework throughout the house would all be connected together by way of soldered couplings or entirely metallic compression fittings, unlike today where electrically isolating plastic couplers are becoming more and more common. So yes, we have situations today where we end up with short lengths of copper pipe between plastic couplers where the chances of that length ever becoming live are just about nil. With such short lengths it's also feasible - in some cases - for them to be fairly effectively isolated from earth, e.g. by running between plastic couplers through wooden joists. However.....

To mirror your curiosity, I'm also a little curious as to what 'stray earth paths' you are thinking about.

Once we start talking about typical household pipework which is coupled together (electrically speaking) throughout, unless specific steps were taken to ensure isolation from earth, the result is almost certainly going to be a significantly low resistance to earth. I'm thinking of paths by way of pipework run on or clipped to masonry, hot-water cylinders standing on brick piers, metallic bathtubs on concrete floors, and so on.

All there seems to be is the substance of the building itself and, for a 'conventionally built house', that's extremely unlikely to offer a low enough resistance path to earth to be of any consequence (and, as discussed, if it did, it would somehow have to be bonded!).

I think you might be over-estimating the resistance which can result in some cases from the various metallic parts of the system being in contact with the fabric of a masonry-built house. Individually, perhaps, a length of a few feet of copper pipe or a hot-water cylinder standing on a brick support might not seem significantly low, but put the whole lot together and those parallel paths to earth can become significant - Certainly low enough that anyone touching a live connection while simultaneously in contact with the pipework could receive a shock of a dangerously high magnitude.

You seem to be implying that there is a requirement for main bonding of internal metal pipework, per se, but that surely isn't the case?

It always used to be in "The Regs." With the conventional plumbing systems before plastic couplings came into use, you could pretty much be sure that any length of metallic pipe in one point of the house would be electrically bonded to any other length in the house by some means or other (direct soldered/compression joints, by way of couplings at the hot-water cylinder or boiler, etc.). With such an extensive metallic system, there were bound to be enough earth paths on the system as a whole to be significant. If the system were not earthed in any way, then there was the problem of a fault (perhaps a frayed cord draped across a pipe, as debated at length in that other mammoth thread) making the pipework live at one point then making the entire system live throughout the house. Would you want, say, a damaged kettle flex touching the taps on the kitchen sink making all the pipework and taps in the bathroom live as well?

With the more widespread use of isolating couplers today there's clearly a point at which short, isolated sections of the system are extremely unlikely to have any potential imposed upon them by a fault, and we can get into the same sort of realm as the arguments about whether one should bond metallic doorknobs and window handles. But for a completely metallic system where there is electrical continuity throughout, the reasons for bonding the system are just as valid now as they were that half-century ago.

 

[JohnW2]

Plastic gas supplies change to steel before entering buildings.

I've often seen that said, but I've certainly seen a number of premises in which it is not the case.

No you haven't.

You're right - I haven't. It was only when I just read Bernard's response that I realised I had been blind to the crucial word 'gas' in what you had written - I 'thought' I was talking about water supply pipes. My apologies.

[having said that, ISTR that a few months ago, someone posted a piccie here which appeared to show incoming plastic pipe connected to the gas meter]

Kind Regards, John.

 

[JohnW2]

... the result is almost certainly going to be a significantly low resistance to earth. I'm thinking of paths by way of pipework run on or clipped to masonry, hot-water cylinders standing on brick piers, metallic bathtubs on concrete floors, and so on.

Hmmmm.... As I said:

All there seems to be is the substance of the building itself and, for a 'conventionally built house', that's extremely unlikely to offer a low enough resistance path to earth to be of any consequence (and, as discussed, if it did, it would somehow have to be bonded!).

I think you might be over-estimating the resistance which can result in some cases from the various metallic parts of the system being in contact with the fabric of a masonry-built house. Individually, perhaps, a length of a few feet of copper pipe or a hot-water cylinder standing on a brick support might not seem significantly low, but put the whole lot together and those parallel paths to earth can become significant - Certainly low enough that anyone touching a live connection while simultaneously in contact with the pipework could receive a shock of a dangerously high magnitude.

Hmmm again If you were right (that I have overestimated restsitance), then your argument would be right - but I really do have my serious doubts, at least in the vast majority of cases. To get a significant (ptentially serious) shock, the combination of such factors would probably have to have a combined resistance of under about 50K (under 5mA at 230V p.d.) - and I just doubt that the sort of 'contacts' you are talking about would get anywhere near that. Pipes are rarely in significant (if any) contact with the fabric to which they are attached, concrete floors (even when laid directly on earth) generally have very high resistances to earth etc.

... If the system were not earthed in any way, then there was the problem of a fault (perhaps a frayed cord draped across a pipe, as debated at length in that other mammoth thread) making the pipework live at one point then making the entire system live throughout the house. Would you want, say, a damaged kettle flex touching the taps on the kitchen sink making all the pipework and taps in the bathroom live as well? .... But for a completely metallic system where there is electrical continuity throughout, the reasons for bonding the system are just as valid now as they were that half-century ago.

I never though that I would be the one to say this to anyone, but I think that you are to some extent muddling up earthing and bonding One of your concerns above is that a frayed flex touching a kitchen tap/sink would make the bathroom taps live, yet the purpose of bonding is ensure that all accessible metalwork within an equipotential zone is at the same potential! I presume what you are suggesting is that such pipework should be earthed - so as to reduce the p.d. of the pipework relative to earth (and/or cause a protective device to operate) in the scenario you postulate.

The only purpose of main bonding is to attempt to ensure that the entire property is an equipotential zone. Pipework within a building which is not 'floating' will nearly always be connected to the MET (via CPCs). Main bonding exists only to ensure that there is no touchable metalmork at a different potential from that MET-connected pipework. If a truly equipotential zone exists then the (one and only) purpose of earthing the equipotential zone would appear to be to facilitate the operation of protective devices.

Kind Regards, John.

 

[Paul_C]

To get a significant (ptentially serious) shock, the combination of such factors would probably have to have a combined resistance of under about 50K (under 5mA at 230V p.d.) - and I just doubt that the sort of 'contacts' you are talking about would get anywhere near that.

Much will depend upon the precise nature of the construction and the extent of the pipework system and associated fittings, but I think you'll find that a combined resistance to earth of less than 50 kilohms is not difficult to achieve.

I never though that I would be the one to say this to anyone, but I think that you are to some extent muddling up earthing and bonding

I'm using the terms in the way that I grew up with them being used, i.e. if a bonding connection is made from the pipework to the main earthing terminal then that's "earthing the pipework." Yes, the pipework might not be at true earth potential at all times, but then neither is that main earthing terminal which is still referred to as such!