Lost neutral on PME consequences

[Steve T]

Hi all,
I was reading about what can happen if the neutral on a PME system is lost at some point before the cutout and I have a few questions that I'm hoping someone could help me with.
1) Am I correct in thinking that for a properly earthed and bonded system, if the neutral is disconnected before the cutout, the current that would normally return via the neutral would instead return through the earth and potentially through a gas or water pipe that is bonded? Would all earthed metal work in the house also become live and present a shock risk if anyone was to touch it?
2) No mcbs or rcds would trip would they, because as far as they can tell the system is working normally, as the current is flowing through the rcd neutral normally, but then afterwards returns through the earth instead of the neutral? The rcd still wouldn't trip if someone touched an earthed metal object that was actually live would it as the person would just be another path to earth, just like any bonded pipes?
3) What would be the safest way to immediately rectify the situation, if you noticed it? I'm thinking turn off main switch first (obviously) but would it be a good idea to disconnect the main earth as well to prevent any current from neighbouring houses returning via your bonding (and making any earthed metal work live), or just leave it connected and leave the house and don't touch anything metal? Goes without saying that you would also call the emergency DNO number ASAP.
4) What, if anything, is in place to protect against these sorts of dangerous situations occurring? Some of the stuff I've mentioned above would surely depend on where exactly the broken neutral is (ie is it just after your property before any joints or anything or is it at a substation)
Apologies if I've used any incorrect terminology there or made any of my questions confusing, hopefully you'll understand them! I'm not really familiar with how the distribution network operates so I may well have some misconceptions, but would greatly appreciate any answers/explanations to my questions above, I'm just trying to improve my understanding!
Thanks for any replies

 

[Wilko]

1) Am I correct in thinking that for a properly earthed and bonded system, if the neutral is disconnected before the cutout, the current that would normally return via the neutral would instead return through the earth and potentially through a gas or water pipe that is bonded?

Yes, the current would try to return. As an example : If the fridge was idle and this fault occurred, no problem if you touch it (maybe, I won’t try it ). But when you grab the handle and open the door the light gets connected to L and current will flow through the filament and whatever path is available to Earth. That could be via bonding and/or your-good-self...

 

[pc1966]

In a properly installed system where the gas/water/etc extraneous paths have been bonded to the supply earth then in the event of an open PME neutral you are safe from shock inside the building, as you won't see any voltage difference.

However, any metalwork attached to the system 'earth' is now part of the path the neutral current will attempt to follow and depending on the current and impedance that could be anywhere from tens of volts to more or less the full 230V supply. So if you had, say, a metal fence with an earthed floodlamp on it, or the more obvious problems of an electric car being charged, you might get a possibly fatal shock by touching the metalwork while in contact with the true Earth outside.

Normal RCD will not protect you as they look at the difference between L & N to detect lekage, and do not monitor the E current as such.

Car charging (EV) or caravans are seen as a particular risk because you might be wet it at the time (rain, washing car, etc), so there are various regulations on means of making that situation safe. Some use a separate earth rod and created a TT supply so they don't use the PME "earth" (standard practice for caravan parks for years), whereas some EV charges use TT, and others have variations on a RCD and voltage-monitoring to detect an open PME and disconnect both the live and earth conductors.

In terms of your home safety, there is not much you could do except switch off the supply. If it is only your home that has lost the PME neutral then that de-energises it, but it won't make much difference if the fault is further away and you are seeing the imbalance neutral from multiple homes on a segment of the distribution network.

What it would help with though is if the fault causes an over or under voltage that could damage your home's electrics, though in most cases by time you know about the fault your home has been subject to it for a very long time in electronic terms!

 

[Steve T]

In a properly installed system where the gas/water/etc extraneous paths have been bonded to the supply earth then in the event of an open PME neutral you are safe from shock inside the building, as you won't see any voltage difference.

However, any metalwork attached to the system 'earth' is now part of the path the neutral current will attempt to follow and depending on the current and impedance that could be anywhere from tens of volts to more or less the full 230V supply. So if you had, say, a metal fence with an earthed floodlamp on it, or the more obvious problems of an electric car being charged, you might get a possibly fatal shock by touching the metalwork while in contact with the true Earth outside.

Normal RCD will not protect you as they look at the difference between L & N to detect lekage, and do not monitor the E current as such.

Car charging (EV) or caravans are seen as a particular risk because you might be wet it at the time (rain, washing car, etc), so there are various regulations on means of making that situation safe. Some use a separate earth rod and created a TT supply so they don't use the PME "earth" (standard practice for caravan parks for years), whereas some EV charges use TT, and others have variations on a RCD and voltage-monitoring to detect an open PME and disconnect both the live and earth conductors.

In terms of your home safety, there is not much you could do except switch off the supply. If it is only your home that has lost the PME neutral then that de-energises it, but it won't make much difference if the fault is further away and you are seeing the imbalance neutral from multiple homes on a segment of the distribution network.

What it would help with though is if the fault causes an over or under voltage that could damage your home's electrics, though in most cases by time you know about the fault your home has been subject to it for a very long time in electronic terms!

Thanks, that makes sense to me I think apart from one point, so as you say if you are in contact with true earth (outside) and live metal work (due to neutral fault) you could get a fatal shock. I understand that, but would that not also be the case inside if the floor happened to be more conductive than usual such as a flagstone floor? Or even if the floor was not particularly conductive would you not at least get a slight tingle?

 

[pc1966]

Thanks, that makes sense to me I think apart from one point, so as you say if you are in contact with true earth (outside) and live metal work (due to neutral fault) you could get a fatal shock. I understand that, but would that not also be the case inside if the floor happened to be more conductive than usual such as a flagstone floor? Or even if the floor was not particularly conductive would you not at least get a slight tingle?

If you happened to have a very conductive floor inside then yes, it is just possible. But such floors tend to be rather cold hence the common practice of a suspended wooden floor above the underlying concrete/soil, or having carpeting down, and usually are not wet, so that risk is I guess pretty small.

I guess a shower on ground floor might just come in to that sort of a category and probably would not be designed to be OK, where as swimming pools are recognised risks for this so would usually have an earthing mat below, and/or the rebar in the concrete bonded to keep the potential down if a PME supply is considered likely.

 

[marcuswareham]

If you happened to have a very conductive floor inside then yes, it is just possible. But such floors tend to be rather cold hence the common practice of a suspended wooden floor above the underlying concrete/soil, or having carpeting down, and usually are not wet, so that risk is I guess pretty small.

I guess a shower on ground floor might just come in to that sort of a category and probably would not be designed to be OK, where as swimming pools are recognised risks for this so would usually have an earthing mat below, and/or the rebar in the concrete bonded to keep the potential down if a PME supply is considered likely.

I would say this type of cold floor is very common practice in kitchens and conservatories and the like, out of the 7 places iv lived only 2 of them have not had a cold probably more conductive floor in parts. Although the hope is the DPC will reduce this conductivity to the soil, if there even is one with older houses, My partner's dads house you can lift the terracotta tiles in the kitchen and you are presented with some sand and then soil under that, rubbish hey! at least his supply is TT ?

I am also sceptical as to how well insulated you are on timber floor boards etc, as most people who would say there is no risk of shock because you are insulated from true earth on a wooden floor wouldn't volunteer to touch a live conductor whilst standing on one !!, and I have heard of plenty of people receiving shocks upstairs in a house on timber floorboards and joists.

 

[marcuswareham]

Hi all,
I was reading about what can happen if the neutral on a PME system is lost at some point before the cutout and I have a few questions that I'm hoping someone could help me with.
1) Am I correct in thinking that for a properly earthed and bonded system, if the neutral is disconnected before the cutout, the current that would normally return via the neutral would instead return through the earth and potentially through a gas or water pipe that is bonded? Would all earthed metal work in the house also become live and present a shock risk if anyone was to touch it?
2) No mcbs or rcds would trip would they, because as far as they can tell the system is working normally, as the current is flowing through the rcd neutral normally, but then afterwards returns through the earth instead of the neutral? The rcd still wouldn't trip if someone touched an earthed metal object that was actually live would it as the person would just be another path to earth, just like any bonded pipes?
3) What would be the safest way to immediately rectify the situation, if you noticed it? I'm thinking turn off main switch first (obviously) but would it be a good idea to disconnect the main earth as well to prevent any current from neighbouring houses returning via your bonding (and making any earthed metal work live), or just leave it connected and leave the house and don't touch anything metal? Goes without saying that you would also call the emergency DNO number ASAP.
4) What, if anything, is in place to protect against these sorts of dangerous situations occurring? Some of the stuff I've mentioned above would surely depend on where exactly the broken neutral is (ie is it just after your property before any joints or anything or is it at a substation)
Apologies if I've used any incorrect terminology there or made any of my questions confusing, hopefully you'll understand them! I'm not really familiar with how the distribution network operates so I may well have some misconceptions, but would greatly appreciate any answers/explanations to my questions above, I'm just trying to improve my understanding!
Thanks for any replies

Question 3 is an interesting one, as others have said all depends on where the fault is, perhaps we should have 3 pole isolators on TNCS supplies with L,N and PE being disconnected?

Question 4: The Protective Multiple Earthing (PME) is what is in place on a TN-C-S supply to mitigate the effects of a lost neutral from the transformer i.e. having many electrodes along the length of the PEN conductor. Some say this is rock solid and PEN faults never happen, I don't know the numbers for how many PEN faults happen per year for example although I would imagine a lot of near misses would never get reported and then not be recorded if there even is a record kept, To me it seems that PME is not a good enough mitigation measure otherwise we wouldn't be having all the hu-har about EV charging

Perhaps EVs have brought the risks to the forefront of peoples mind, although the outside tap risk has existed for years, metal tap and pipework installed by a Plummer or DIYer (without any thought to anything electrical (i.e no plastic section)) the pipe work is bonded to the PME MET and you often touch this tap while standing on wet true earth, ouch if there an open PEN!

There are many products to mitigate the PEN fault risk for EV chargers, in my eyes most are only 50% effective (as iv mentioned in other posts) but they are better than nothing if all you can use is a PME earth

 

[pc1966]

I am also sceptical as to how well insulated you are on timber floor boards etc, as most people who would say there is no risk of shock because you are insulated from true earth on a wooden floor wouldn't volunteer to touch a live conductor whilst standing on one !!, and I have heard of plenty of people receiving shocks upstairs in a house on timber floorboards and joists.

Certianly you can't guarentee you would feel nothing, but to get a significant shock the threshold is below around 20kOhm and I can't imagine seeing that low a resistance on most floors unless something is really rather damp.

 

[newfutile]

I think most perceived shocks from touching a live part (assuming shoes and clothes are worn) in not by direct conductance via a resistive path , but via capacitance presumably between flat plates of feet and the earth itself.

 

[pc1966]

Question 3 is an interesting one, as others have said all depends on where the fault is, perhaps we should have 3 pole isolators on TNCS supplies with L,N and PE being disconnected?

My concern with that is the implications if someone has other sources of power, such a PV panels or even a UPS for computer equipment. Some sort of earth rod after any isolation would be in order, I suspect.

 

[newfutile]

solar inverters shut down after loss of supply as they are synced to the 50HZ supply waveform.

 

[pc1966]

solar inverters shut down after loss of supply as they are synced to the 50HZ supply waveform.

After a short while...

 

[timhoward]

Interesting questions!
I've been in both situations (both years ago)- surprised to not get a jolt standing on soil when it had just rained after I messed up isolating an outside light, and surprised to get a jolt on a wooden step ladder on a dry concrete floor. So I wouldn't suggest banking on anything!
OP - Regarding Q3 - Other than switching off the isolator I wouldn't touch the installation at all. It's likely you would come into contact with return current path attempting to make things safer, and the current might just might prefer you to the bonding as a way to real earth. I could certainly construct arguments why it's unlikely but ultimately its not worth the risk.

 

[Steve T]

OP - Regarding Q3 - Other than switching off the isolator I wouldn't touch the installation at all. It's likely you would come into contact with return current path attempting to make things safer, and the current might just might prefer you to the bonding as a way to real earth. I could certainly construct arguments why it's unlikely but ultimately its not worth the risk.

I understand why you're saying that, I suppose the DNO response would be quick enough to not have to worry about doing anything other than switching the main switch off. I wonder though if some sort of switch for the main earth (interconnected with the normal main switch so you could not accidentally turn off the earth switch only) would be beneficial to pme installations. I think someone else mentioned the idea as well earlier in the thread.

 

[pc1966]

While you can easily get mulit-pole switches, it would require one pole to make first and break last for the earth. They are available (typically for the neutral in a 3-phase system) but less common.

 

[Steve T]

I've just remembered something that happened quite a few years ago, I was at a mates house when all the lights (halogen and incandescent) went dim suddenly, no flickering or anything though. We checked the voltage at a socket (was a plastic socket so not an immediate shock risk) and it read a steady 120v according to his multimeter. We turned the main switch off and rang the DNO after checking with some neighbours to see if the same thing had happened, just to rule out a fault in his wiring. The strange thing was it only affected every 3rd property in the row of houses, so made me think it could have been a problem with only 1 of the 3 phases. So I thought at the time one of the transformers at the substation had failed, but looking back could it actually have been a broken neutral??

 

[pc1966]

If it only affected 1 in 3 it sounds more like a phase fault than an open neutral, as the PME fault case you still see the 400V phase-phase so if your 230 has gone down to 120, then the others have gone up quite a bit to match (depending on the balance between the phases).

 

[Steve T]

If it only affected 1 in 3 it sounds more like a phase fault than an open neutral, as the PME fault case you still see the 400V phase-phase so if your 230 has gone down to 120, then the others have gone up quite a bit to match (depending on the balance between the phases).

Makes sense, certainly his (apparently unaffected) neighbours didn't report any broken appliances or much brighter lights or anything like that.

 

[marcuswareham]

Interesting questions!
I've been in both situations (both years ago)- surprised to not get a jolt standing on soil when it had just rained after I messed up isolating an outside light, and surprised to get a jolt on a wooden step ladder on a dry concrete floor. So I wouldn't suggest banking on anything!
OP - Regarding Q3 - Other than switching off the isolator I wouldn't touch the installation at all. It's likely you would come into contact with return current path attempting to make things safer, and the current might just might prefer you to the bonding as a way to real earth. I could certainly construct arguments why it's unlikely but ultimately its not worth the risk.

And the main switch / isolator is now probably in a bonded metal enclosure so you get a shock when opening the flap, doh

I am keeping my plastic consumer unit as long as i can

 

[marcuswareham]

I think most perceived shocks from touching a live part (assuming shoes and clothes are worn) in not by direct conductance via a resistive path , but via capacitance presumably between flat plates of feet and the earth itself.

I did think capacitance, but I dont know enough about how that kind of shock occures, so didn't want to sound silly ?

 

[LastManOnline]

I've been in both situations (both years ago)- surprised to not get a jolt standing on soil when it had just rained after I messed up isolating an outside light, and surprised to get a jolt on a wooden step ladder on ary concrete floor.

While we all understandably conclude that standing "outside" is very dangerous and standing "inside" is very safe, it is, nt always, as clearcut as that. "outside" earth will be very dry and very unconductive if you have a few weeks of good dry summer days. Add on to that someone wearing good shoes and sockets (not after a hard day's work and perspiration) and you will certainly have a high resistance path. On the "inside" as mentioned by other posters, if you have an older home with a concrete floor with no damp coarse you will certainly feel a tingle and perhaps something more. I recently "felt" 27 volts due to a faulty shower while wearing shoes and socks and my feeling afterwards was "Well I can understand the homeowners panicky phonecall now" as he received the same shock with water flowing down around his feet.
I must also say that regardless of any situation I encountered there was never any mystery to the shock felt. It could all be explained by Ohms law.

 

[LastManOnline]

4) What, if anything, is in place to protect against these sorts of dangerous situations occurring?

I, ve just read the detail regarding that very, very sad case of the 7 year old electrocuted in a pub garden due to a seriously underspec electrical intallation.
Its sobering to consider that the exact same problem could occur with the same fatal consequences in a PME installation considered up to spec.

 

[marcuswareham]

I, ve just read the detail regarding that very, very sad case of the 7 year old electrocuted in a pub garden due to a seriously underspec electrical intallation.
Its sobering to consider that the exact same problem could occur with the same fatal consequences in a PME installation considered up to spec.

Quite! although rare, it seems strange too me so many people accept that risk !

 

[LastManOnline]

Quite! although rare, it seems strange too me so many people accept that risk !

Yes.I think though the tide of opinion is gradually changing towards "accepting that risk" within our industry.

 

[timhoward]

I, ve just read the detail regarding that very, very sad case of the 7 year old electrocuted in a pub garden due to a seriously underspec electrical intallation.
Its sobering to consider that the exact same problem could occur with the same fatal consequences in a PME installation considered up to spec.

Very tragic case. Unless I'm missing something I guess you mean with broken upstream Neutral as per this thread or exposure to live parts (as was case in the water filled light you referred to).

Also sobering, this isn't exactly a new debate - I spotted the following discussion elsewhere several years ago which is rather poignant:

 

[marcuswareham]

Very tragic case. Unless I'm missing something I guess you mean with broken upstream Neutral as per this thread or exposure to live parts (as was case in the water filled light you referred to).

Also sobering, this isn't exactly a new debate - I spotted the following discussion elsewhere several years ago which is rather poignant:
View attachment 65376

Exsactly whey my outside lights are SELV or double insulated, they are in easy reach and one of them is likely to be touched

 

[LastManOnline]

Also sobering, this isn't exactly a new debate -

Correct. It's an old debate. However, far from being yesterday's news it has the potential to become very current with one single high profile incident. The example below (outside light) would probably have constituted the vast bulk of "at risk" metalwork under open PEN , 25 years ago.
Today the level of "at risk" exposed metal parts outside the home has changed enormously (electric gates, heat pumps to mention but a few). Add to that the DSO, s approach of employing largely TNC-S for new installs plus converting existing TNS cables to TNC-S when repairs are required.The electrical landscape is starting to look very different and with that change the potential for open PEN issues can only increase.

View attachment 65376

 

[marcuswareham]

Correct. It's an old debate. However, far from being yesterday's news it has the potential to become very current with one single high profile incident. The example below (outside light) would probably have constituted the vast bulk of "at risk" metalwork under open PEN , 25 years ago.
Today the level of "at risk" exposed metal parts outside the home has changed enormously (electric gates, heat pumps to mention but a few). Add to that the DSO, s approach of employing largely TNC-S for new installs plus converting existing TNS cables to TNC-S when repairs are required.The electrical landscape is starting to look very different and with that change the potential for open PEN issues can only increase.

The conversion of old TNS to TNCS in the street and the head staying the same inside the property I feel is a concern as a spark could install something (hot tub, heat pump, etc etc) outside believing the loss of PEN conductor risk is not an issue due to the fact it is TNS, him being unaware it is really now TNCS

 

[LastManOnline]

The conversion of old TNS to TNCS in the street and the head staying the same inside the property I feel is a concern as a spark could install something (hot tub, heat pump, etc etc) outside believing the loss of PEN conductor risk is not an issue due to the fact it is TNS, him being unaware it is really now TNCS

I suspect that an experienced spark will likely test the supply and be pretty sure what he is dealing (or perhaps ring the DSO) but, yea to make a major change in the supply system by converting a TNS to a TNC-S and not leaving a very clearly marked label or"record " of it behind for the homeowner (and future electrician) is I feel very, very poor practice.

 

[newfutile]

how can you test the supply to find out earthing type or maybe we just assume a ZE less than 0.35 is most likely some variation of TNC-S.

 

[pc1966]

how can you test the supply to find out earthing type or maybe we just assume a ZE less than 0.35 is most likely some variation of TNC-S.

I don't think there is any easy test.

If the PFC & PSCC are very similar then all it tells you is the earth and neutrals have similar impedances, it is not explicitly telling you if they are common fairly close to the property in TN-C-S style.

With the power off at your installation looking the the (supply side) N to E voltage would give to an idea of how much current imbalance & impedance was present on any PEN cable, but that is also likely to pick up some volts from inductive coupling from other loads in the network.

However, if that voltage is very low, say less then 0.1V or so, you could probably do a resistance measurement N-E to get an idea of how far away any common point is if you have an idea of the type of cable feeding you (e.g. 16mm split concentric or whatever) so you know R per unit length.

 

[LastManOnline]

how can you test the supply to find out earthing type or maybe we just assume a ZE less than 0.35 is most likely some variation of TNC-S.

"pc1966" sums that up nicely. What I would add (or ask) is how could making such a fundamental change in the supply network go un noted? Afterall an electrician would feel duty bound to inform a homeowner that replacing a wooden shed with a metal one might necessitate a TT...ing of the electrical supply to it for safety reasons (I, m trying to think in UK terms here and may stand to be corrected).
How then could a service provider change the nature of the system supplying a whole street without considering the potential knock on effects for the various homeowners who carried out installation under a previous supply system? Am I making any sense here?

 

[pc1966]

What I would add (or ask) is how could making such a fundamental change in the supply network go un noted?

I suspect the move to TN-C-S was started many years ago when (a) most folk did not have significant outdoor power requirements like EV and hot-tubs, and (b) when the cables going in were nice and new and joint failures of the outer PEN of aluminium waveform cables was not an expected thing.

Fast-forward 30 odd years and here we are.

• Proper TN-S would be the safest/best system, but costs the DNO that earth extra conductor.
• Going TT like much of the EU avoids the DNO expense and the open-PEN risk. But in turn it pushes the extra cost on the home owner as they need an incoming RCD (unusual over two decades ago before the regs demanded them on most circuits) and an earth rod, which in an urban area has all sorts of practical problems. Also they become dependant on the electronics for disconnecting on practically any fault to earth, and if it fails they are in a worse position than TN-C-S as only one earth rod of fairly high impedance (hundred-ish ohms probably).
• TN-C-S offers the reliable disconnect on OCPD operation (just like TN-S) and so no modifications to most existing installations that going TT would demand, and saves the DNO money.

It is not hard to see why we got here!

As it stands the regs on caravan supplies have basically forced TT use there for some time now, but that is a small proportion of installs. Only now are the rapid rise in EV use is focusing attention on the risks of open-PEN faults once more.

 

[timhoward]

And on that very topic I found this paper interesting:
(source https://myenergi.com/wp-content/uploads/2019/09/Protection-against-damaged-PEN-Conductor-r1_0.pdf )

The number of reported open PEN faults is higher than I'd have guessed (and I can't help imagine there are plenty not reported)



The paper also goes on to discuss 3 pole automatic switch based upon RCD principles by measuring the earth fault current, with a focus on EV charging points and the inherent dangers on TN-C-S with broken PEN conductor.

 

[marcuswareham]

I suspect the move to TN-C-S was started many years ago when (a) most folk did not have significant outdoor power requirements like EV and hot-tubs, and (b) when the cables going in were nice and new and joint failures of the outer PEN of aluminium waveform cables was not an expected thing.

Fast-forward 30 odd years and here we are.

• Proper TN-S would be the safest/best system, but costs the DNO that earth extra conductor.
• Going TT like much of the EU avoids the DNO expense and the open-PEN risk. But in turn it pushes the extra cost on the home owner as they need an incoming RCD (unusual over two decades ago before the regs demanded them on most circuits) and an earth rod, which in an urban area has all sorts of practical problems. Also they become dependant on the electronics for disconnecting on practically any fault to earth, and if it fails they are in a worse position than TN-C-S as only one earth rod of fairly high impedance (hundred-ish ohms probably).
• TN-C-S offers the reliable disconnect on OCPD operation (just like TN-S) and so no modifications to most existing installations that going TT would demand, and saves the DNO money.

It is not hard to see why we got here!

As it stands the regs on caravan supplies have basically forced TT use there for some time now, but that is a small proportion of installs. Only now are the rapid rise in EV use is focusing attention on the risks of open-PEN faults once more.

1 vote here for PROPER mantained TNS

 

[pc1966]

The number of reported open PEN faults is higher than I'd have guessed (and I can't help imagine there are plenty not reported)

I think they will all be reported.

The DNOs have that duty, and the huge voltage swings and possible damage to home owner's equipment makes it hard for the home owners to ignore/not report and have fixed!

 

[pc1966]

Realistically these days you have to assume and design for TN-C-S or TT.

TN-S might not remain that way, nor can you know just how 'S' it really is.

 

[marcuswareham]

I have always thought regulation 722.411.4.1 (iv) where protection from a PEN fault is by measuring the utilization voltage between L & N and then disconnecting all conductors if it is outside the range of 207v - 253v, as a bit of a cop-out and only really solving half the problem as it is possible to have a PEN fault and the voltage to remain within that range

I have been thinking however the as a PEN conductor has multiple electrodes along its length so a break in the PEN conductor should still ensure you are connected to one or more electrodes, but I guess the weakest points are the joints and in particular the joints where individual houses etc connect to the PEN conductor, So would it be the case that most PEN faults occur at a joint where it branches off to feed a house also the section to the house is very unlikely to have any electrodes on, but in this case as it is only one house affected the above 722.411.4.1(iv) would work in most cases (subject to extraneous parts which are bonded), so is this why it is accepted as a way to mitigate PEN faults as in most cases it will protect due to the above

Attached is a crap picture to show the joints I am trying to convey

 

[timhoward]

the weakest points are the joints and in particular the joints where individual houses etc connect to the PEN conductor

I know exactly what you are saying, and I have no experience with the public distribution side but I wonder if it is three conductors coming together at that joint location in a resin joint, and a joint failure would also sever the ongoing PEN to the next joint.
On large private estates I've seen it done like this (and had to fix it after a mini-digger driver had a bad day!).

 

[snowhead]

50 or so houses were affected where my Son was living a few years ago, spread across all 3 phases.
It was a newish estate of around 50 or so houses and the Neutral connection failed where the estate was connected to the existing cable in the main road.
It cost Western Power tens of thousands to sort out the damage caused.