Broken PEN conductor

[JohnW2]

Ok. Lost supply neutrals are not a hazard, then. Why the proposed requirement for earth rod, I wonder?

Interesting question.

For a start, do you disagree with my suggestion that there should be no shock hazard within a building whose electrical installation is fully compliant in terms of bonding and earthing (and whose walls are dry ) ?

As for possible answers to your question:
1... Maybe they are considering the possibility that some premises have bonding and/or earthing which is not fully satisfactory?
2... As bernard often tells us, if extraneous-c-ps have a very low impedance path to earth, very high currents could flow in the main bonding conductors, leading to a potential fire risk. Furthermore, if that conductor 'melts', main bonding is lost and there is then a risk of shocks.
3... The LFB suggested (or demanded) it!
4... It's not meant to make sense

For goodness' sake. It's just a simple diagram to explain to Dobby.

I think you're being over sensitive. I wasn't being critical but, rather, was pointing out (for Dobby and others) that the situation could be a lot worse than what you depicted. If (as you illustrated) the break was in the neutral supplying only one property/installation then, for a start, one could completely eliminate all risks by opening the main switch in that property. Perhaps more important is the fact that, in relation to (2) above, in that 'single installation' scenario, it's very unlikely (even with exceptionally low impedance path from extraneous-c-ps to earth) that the loads in one property would result in currents which would melt a 10mm² bonding conductor. In contrast, if a neutral break leaves multiple properties with their neutrals connected together, but not connected to the DNO neutral, then those currents could theoretically get very high

Kind Regards, John

 

[ban-all-sheds]

That's correct.

So what is the difference between a circuit which has suffered a break in a conductor and has gone open, and one which has suffered a break in a conductor and has not gone open?

 

[EFLImpudence]

Interesting question.
For a start, do you disagree with my suggestion that there should be no shock hazard within a building whose electrical installation is fully compliant in terms of bonding and earthing (and whose walls are dry ) ?

Theoretically, that would seem to be the case but people do get shocks.
Large premises? Volt drop? Should there be (ideally) supplementary bonding everywhere?

As for possible answers to your question:
1... Maybe they are considering the possibility that some premises have bonding and/or earthing which is not fully satisfactory?

It may be satisfactory but not perfect. Large premises again; someone may be stood some distance from the main bond connection.
Note the concern over outbuildings but not if a house is large enough for a room or conservatory to be in the same position as an outbuilding of a smaller house,

2... As bernard often tells us, if extraneous-c-ps have a very low impedance path to earth, very high currents could flow in the main bonding conductors, leading to a potential fire risk. Furthermore, if that conductor 'melts', main bonding is lost and there is then a risk of shocks.
3... The LFB suggested (or demanded) it!
4... It's not meant to make sense

I think you're being over sensitive. I wasn't being critical but, rather, was pointing out (for Dobby and others) that the situation could be a lot worse than what you depicted.

Possibly, but you did write it under a quote of my text.

If (as you illustrated) the break was in the neutral supplying only one property/installation then, for a start, one could completely eliminate all risks by opening the main switch in that property. Perhaps more important is the fact that, in relation to (2) above, in that 'single installation' scenario, it's very unlikely (even with exceptionally low impedance path from extraneous-c-ps to earth) that the loads in one property would result in currents which would melt a 10mm² bonding conductor. In contrast, if a neutral break leaves multiple properties with their neutrals connected together, but not connected to the DNO neutral, then those currents could theoretically get very high

So, there is a shock risk, then?

 

[JohnW2]

Theoretically, that would seem to be the case but people do get shocks.

Do they (in the situation we are considering)?

Large premises? Volt drop?

Theoretically, but I would think extremely unlikely in anything approaching 'an ordinary house', so not really a reason for an 'across the board' new regulation. As an example, to get, say, 50V drop across a 20m long 10mm² main bonding conductor would require about 1136A and, at 230V, that would require that the total of the impedances of the extraneous-c-p (to earth), the main bonding conductor, the 'switched on' loads (in parallel with one another) and of the final circuits was about 0.2Ω - and I really don't think that is really very credible in 'an ordinary house'. Even with multiple houses' 'switched on' loads in parallel, I still don't think it's likely to happen.

Should there be (ideally) supplementary bonding everywhere? ... It may be satisfactory but not perfect. Large premises again; someone may be stood some distance from the main bond connection.

I've often thought (and may have said so here) that there really should be a requirement for a maximum resistance/impedance of main bonding conductors - although there is nothing in the regs about resistance/impedance, per se, (although there is a requirement for main bonding conductor to be sized acoording the incoming TN-C-S supply cable size), I seem to recall that the OSG actually suggests a maximum of 50 mΩ (about 23 metres of 10mm² cable) - if that were a requirement, large premises, primarily commercial and industrial, might sometimes require bonding conductors much larger that 10mm² (although I suspect that, in many very large premises, extraneous-c-ps may enter fairly close to the MET). If the resistance of the main bonding conductor is sufficiently high that large VDs could arise under exceptional very high current situations then, as you say, I suppose there would be an argument for "supplementary bonding everywhere"

Possibly, but you did write it under a quote of my text.

I did - but, if you would but believe it, I was merely trying to be helpful' by supplementing/augmenting what you had illustrated!

So, there is a shock risk, then?

If the main bonding conductor(s) melt, yes - but, as I wrote (and you quoted), I don't believe that is going to happen in the 'single property broken neutral' scenario you illustrated. I would suspect that it's pretty unlikely even when multiple properties are involved.

I've also just realised that, perhaps because I thought it too obvious to need saying, in responding to your question "Why the proposed requirement for earth rod, I wonder?" I failed to make the most obvious comment - namely that no domestic earth electrode is going to make a measurable difference to the scenarios we are discussing. If, as above, there were 1000A+ flowing through the bonding conductor, connecting a domestic earth rod which could typically only carry 5-10A to earth is hardly going to make much difference! I therefore winder whether my suggested answer (4) might be the closest?

Kind Regards, John

 

[bernardgreen]

Even with multiple houses' 'switched on' loads in parallel, I still don't think it's likely to happen.

3... The LFB suggested (or demanded) it!

If they did they are not the only ones, West Yorkshire Fire and Rescue raised concerns 6 years ago

http://www.examiner.co.uk/news/west-yorkshire-news/amazing-footage-house-explosion-fire-4971943

""Crews had been called to reports of a small kitchen fire, but when they arrived found a string of small fires in a row of six terraced houses. Firefighters quickly rescued the occupants from all the houses, including two people from upstairs bedrooms using ladders. As the film shows, only seconds later an explosion tore through the row of homes.""

""Investigations revealed that the blast was caused after a piece of cable was cut from an overhead line. This affected the earthing of the electrical network in the area and resulted in some unusual electrical activity, which affected some of the gas pipes within the properties. This resulted in the fires and explosion.""

 

[JohnW2]

If they did they are not the only ones, West Yorkshire Fire and Rescue raised concerns 6 years ago .... ""Investigations revealed that the blast was caused after a piece of cable was cut from an overhead line. This affected the earthing of the electrical network in the area and resulted in some unusual electrical activity, which affected some of the gas pipes within the properties. This resulted in the fires and explosion.""

OK - so they don't understand, either. Mind you, there are merely reporting what happened, not really 'expressing concerns'.

The crucial point is surely the one I made at the end of my previous post - that, in the presence of a TN-C-S neutral fault, connecting a domestic earth rod is surely not going to make any noticeable difference, and most certainly won't prevent "fires and explosions", will it?

Kind Regards, John

 

[EFLImpudence]

Do they (in the situation we are considering)?

I don't now know. Surely the floor (ground) cannot be all at the same potential because a pipe enters(or exits) the premises at one point or several points in several premises.

Theoretically, but I would think extremely unlikely in anything approaching 'an ordinary house', so not really a reason for an 'across the board' new regulation. As an example, to get, say, 50V drop across a 20m long 10mm² main bonding conductor would require about 1136A and, at 230V, that would require that the total of the impedances of the extraneous-c-p (to earth), the main bonding conductor, the 'switched on' loads (in parallel with one another) and of the final circuits was about 0.2Ω - and I really don't think that is really very credible in 'an ordinary house'. Even with multiple houses' 'switched on' loads in parallel, I still don't think it's likely to happen.

I think the general concensus is that the proposed electrode cannot achieve the desired aim.
So option 4 it is, then, but that just raises the fact (like bond everything days) that those making the rules do not know what they are doing which really isn't good enough.

Surely we should have reached the situation when there are difinitive answers and if something new needs introducing, then the only question should be "Why wasn't it done already?".

I've often thought (and may have said so here) that there really should be a requirement for a maximum resistance/impedance of main bonding conductors - although there is nothing in the regs about resistance/impedance, per se, (although there is a requirement for main bonding conductor to be sized acoording the incoming TN-C-S supply cable size), I seem to recall that the OSG actually suggests a maximum of 50 mΩ (about 23 metres of 10mm² cable) - if that were a requirement, large premises, primarily commercial and industrial, might sometimes require bonding conductors much larger that 10mm² (although I suspect that, in many very large premises, extraneous-c-ps may enter fairly close to the MET). If the resistance of the main bonding conductor is sufficiently high that large VDs could arise under exceptional very high current situations then, as you say, I suppose there would be an argument for "supplementary bonding everywhere"

The 0.05Ω (50mΩ) oft quoted is not a limit for the bonding conductor.
It is just an acceptable value of negligible impedance of its connection.

I've also just realised that, perhaps because I thought it too obvious to need saying, in responding to your question "Why the proposed requirement for earth rod, I wonder?" I failed to make the most obvious comment - namely that no domestic earth electrode is going to make a measurable difference to the scenarios we are discussing. If, as above, there were 1000A+ flowing through the bonding conductor, connecting a domestic earth rod which could typically only carry 5-10A to earth is hardly going to make much difference! I therefore winder whether my suggested answer (4) might be the closest?

As above.

 

[JohnW2]

I don't now know. Surely the floor (ground) cannot be all at the same potential because a pipe enters(or exits) the premises at one point or several points in several premises.

I'm not quite sure what you mean. Are you talking about the 'floor' which one stands on? If so, I doubt that, other in exceptional circumstances (or cellars/outbuildings) there will be any significant path to earth through most floors. If you mean something else, I'll comment when you have clarified!

I think the general concensus is that the proposed electrode cannot achieve the desired aim. So option 4 it is, then, but that just raises the fact (like bond everything days) that those making the rules do not know what they are doing which really isn't good enough.

Agreed on all counts. I wonder if this has been pointed out to them during the DPC process? The more I think about it, the less point do I see in these rods with TN-C-S. Perhaps ironically, they would (in the presence of RCDs) be of value with TN-S (particularly since one can suffer from a failed TN-S earth for years without being aware of it). Is that perhaps what they are thinking about (and, if so, why not just restrict the requirement to TN -S)?

The 0.05Ω (50mΩ) oft quoted is not a limit for the bonding conductor. It is just an acceptable value of negligible impedance of its connection.

Fair enough - it was just a vague recollection. However, there seems to be a strong case for limiting the impedance of main bonding conductors, if the bonding is to achieve what it says on the tin. The regs have a bit to say about CSA of such conductors, but, as far as I am aware, nothing about their length (hence impedance). It sounds a bit like the situation we would have if their were just a requirement for minimum CSA of cables in final circuits, regardless of how long those circuits were (and hence how high the Zs).

Kind Regards, John

 

[EFLImpudence]

I'm not quite sure what you mean. Are you talking about the 'floor' which one stands on? If so, I doubt that, other in exceptional circumstances (or cellars/outbuildings) there will be any significant path to earth through most floors. If you mean something else, I'll comment when you have clarified!

Not sure I can clarify.
I am just assuming that a shock is possible just by touching a live conductor while standing on the floor; are you saying this is not possible?

Agreed on all counts. I wonder if this has been pointed out to them during the DPC process? The more I think about it, the less point do I see in these rods with TN-C-S. Perhaps ironically, they would (in the presence of RCDs) be of value with TN-S (particularly since one can suffer from a failed TN-S earth for years without being aware of it). Is that perhaps what they are thinking about (and, if so, why not just restrict the requirement to TN -S)?

There is the assumption now that everything is TNC-S, so I wouldn't think that is their thought.

Fair enough - it was just a vague recollection. However, there seems to be a strong case for limiting the impedance of main bonding conductors, if the bonding is to achieve what it says on the tin. The regs have a bit to say about CSA of such conductors, but, as far as I am aware, nothing about their length (hence impedance). It sounds a bit like the situation we would have if their were just a requirement for minimum CSA of cables in final circuits, regardless of how long those circuits were (and hence how high the Zs).

Without the requirement for increased csa in PME installations the maximum length with an impedance limit would have been relatively short
(0.05Ω for 10mm² is actually 27m. so for the previous 6mm² minimum, obviously 60% of that, 16m. and less than that, very short.)
Isn't the increased csa just so that they don't melt, not with consideration to anything else?

 

[bernardgreen]

connecting a domestic earth rod is surely not going to make any noticeable difference, and most certainly won't prevent "fires and explosions", will it?

Depends on the impedance of the Earth rod.

Above 7666 Ω it serves little function as 230 V and more than 7660 Ω will not create enough current ( 30 mA ) to ensure RCD operates. Ands even if the RCD does trip ot may only be a single pole ( Live ) disconnect leaving the defective Neutral still connected and the CPC still carries dangerous high potential.

Below 20 Ω and nore than 10 Amps will be flowing into the Ground rod


A I recall the Castleford incident was a row of houses with their metallic water mains bonded ( equivalent to a very low impedance Ground rod ). The water supply may have been metallic all the way to a Neutral-to-Ground bond at the substation.

The bonds from MET to water pipes in the Castleford houses created an extremely low impedance route for the Neutrals of several properties to return to the substation. The bonds over heated.

 

[John D v2.0]

Actually the castleford incident was where the incoming neutral was made live (to a different phase), not just lost, so even more extreme.
"The 3 phase, neutral power cable supplying a row of 6 cottages was cut causing it to wrap around a live cable resulting in the earth becoming live at the surface cut-out."
https://www.esfrs.org/black-museum/all-the-black-museum-cases/cable-theft/

The same effect on bonding would haven in a TN-S install where the earth was made live externally, although the difference being that it would be less likely.
If the neutral was made live in such an install, the electrical fires would happen, but not the issues with the bonding/gas.

 

[JohnW2]

Not sure I can clarify. I am just assuming that a shock is possible just by touching a live conductor while standing on the floor; are you saying this is not possible?

Bernard would undoubtedly say that it was possible. Personally speaking (cellars and outhouses aside), I have never been able to detect any significant path from a floor, even with solid ground-floor ones, and it's obviously even less likely with suspended timber floors. ... and all that assuming no footwear. I therefore do feel that it is extremely unlikely that one would get a significant shock in the manner you describe. One can, of course, get 'tingles', but they are probably usually due to capacitive coupling of one's body to earth, not conduction through a floor (and footwear, if worn)

There is the assumption now that everything is TNC-S, so I wouldn't think that is their thought.

Who knows how they think?

Isn't the increased csa just so that they don't melt, not with consideration to anything else?

Seemingly so but, as you said, even if they don't melt high impedance bonding conductors could theoretically result in sufficiently high VDs (with massive currents flowing) to create a shock risk.

Kind Regards, John

 

[JohnW2]

Depends on the impedance of the Earth rod.

Of course but, as I said, a domestic rod will typically be around 50Ω or so and, at least in my experience, rarely as low as the 20Ω you mention (although even that is far to high to be useful in many of the situations we're discussing).

Above 7666 Ω it serves little function as 230 V and more than 7660 Ω will not create enough current ( 30 mA ) to ensure RCD operates.

I need to think about this (and probably do some scribbling!) but would an RCD actually trip (as a result of a broken neutral), given that all the problems are on the supply side of any RCDs??

The bonds from MET to water pipes in the Castleford houses created an extremely low impedance route for the Neutrals of several properties to return to the substation. The bonds over heated.

Indeed, that is a hazard with TN-C-S and very low impedance bonded extraneous-c-ps. However, exactly how would adding a local earth rod (20Ω if I'm being generous) be of any help in that situation?

Kind Regards, John

 

[EFLImpudence]

Personally speaking (cellars and outhouses aside), I have never been able to detect any significant path from a floor, even with solid ground-floor ones, and it's obviously even less likely with suspended timber floors. ... and all that assuming no footwear. I therefore do feel that it is extremely unlikely that one would get a significant shock in the manner you describe.

What then, is the difference in a bathroom with similar flooring?
You might say "because one is wet" but if there is no path to earth, there is no path to earth.

How did the recent case of (it said phone charger but it must have been) extension lead falling in a bath result in the person's death?
What I mean is, how was there a path through the body if the lead just fell in the water?
Is it assumed the person grabbed hold of the socket and the bath was earthed metal?

 

[JohnW2]

What then, is the difference in a bathroom with similar flooring?

Pretty little, I would say.

You might say "because one is wet" but if there is no path to earth, there is no path to earth.

Yes, but there is a 'path' to earth (through timber, masonry plaster etc. - nothing, other than a vacuum, is a perfect insulator)) but, as I said, I have yet to detect a 'significant' one (probably not even one that can be detected with standard IR testing). If floorboards and walls etc were very wet, then it might become measurable.

How did the recent case of (it said phone charger but it must have been) extension lead falling in a bath result in the person's death? What I mean is, how was there a path through the body if the lead just fell in the water? Is it assumed the person grabbed hold of the socket and the bath was earthed metal?

Well, they obviously must have made contact with something other than 'live water' (or a 'live socket'). It could have been an earthed bath, it could be the taps (if earthed), or they may have reached out and touched something earthed outside of the bath ... to name but a few possibilities.

Kind Regards, John