Online assistance for electrical trade people Login  |  Register  |   Forgot Password
Assistance for electrical trade people




Click here to send Ron a pdf document for publication on this Topic

Documents must be less than 200k in pdf format

Posted By Topic: Australian girl touches garden tap, receives 'cat

Mar 11 2018 11:37

How could this be prevented?

300mA RCD at the pole fuse?

Mar 11 2018 11:59

Insufficient data.

We don't know the nature of the fault, nor whether the installation had been correctly installed.

Whatever the combination, a 300 mA could well have been of no use, as the trip level is well above serious shock hazard for both ventricular fibrillation and lock-on. That's why we use 30 mA for personal protection, and 10 mA for children (who have a lower lock-on threshhold)

Mar 11 2018 12:19

"Insufficient data."


My assumption was the installation was wired correctly and for example a car hit a pillar box and disconnected the neutral. In this case the total appliances alone in the house should be enough to trip 300mA RCD located at the pole fuse?


Mar 11 2018 12:40

Unfortunately if it was an overhead supply it isn’t exactly rare. I’ve had customers N break at the pole either due to wind movements, corrosion or bad connection. Also at the POE at the house they break. There’s often warning signs as lights go dim. With modern LED’s they can be more forgiving so not noticed. Unfortunately some people see it and as it still sort of works they ignore it. I’ve had friends and customers call me and my advice is always the same. Go turn off the main switch.


Mar 11 2018 18:19

If it was a broken neutral upstream of the main switchboard then a 300mA RCD as a main switch would have likely isolated the fault before the child touched the tap.

Mar 11 2018 19:25

"If it was a broken neutral upstream of the main switchboard then a 300mA RCD as a main switch would have likely isolated the fault before the child touched the tap."

I understood that for any RCD to work the fault needed to be on the load side of the RCD?

Mar 11 2018 20:19

My thinking of how the RCD works. We have current flowing in the phase and return via earth. With nothing flowing in the neutral there is certainly an imbalance through the sense ring.

The question remains whether the voltage difference at the RCD between the phase and neutral (tied to earth and back fed) would be sufficient to effect tripping.

Mar 11 2018 21:30

This showed up in a newspaper over here.
The article suggested there was an open neutral fault on the network but also insufficient earthing of conductive parts in connection with the mass of earth.
The girl opened the external meter box to reset a breaker after power went out and received a minor shock, she then reached to turn the garden hose either on or off and received the shock which hospitalised her.

My thoughts would be that everything earthed became live via the MEN link but there was no connection to the actual mass of earth (earth stake/MEC).

Mar 11 2018 22:18

This is one of the dangers of the MEN system used here.
If the neutral connection is lost (eg. through corrosion of the line tap) then the return current of the whole building goes through the MEN link to the earth rod.
The earth rod was never designed for continuous current and the metal rapidly rusts away, also the surrounding clay is dried out by the heat. So... no earth. And the floating neutral (which may be close to phase voltage) is connected to every earthed item in the building possibly including the water pipes.
I have never understood why MEN is used in Aus / NZ.

Mar 12 2018 08:06

Perhaps TT earthing would have solved this. As others have mentioned - the earthing can be not much mor ethan for show - esp in Australia where the electrode requirements are less.

MEN has had its day. It was very useful when there was metalic services (water reticulation etc) connecting houses together. Now we have plastic. Infact we also lay plastic down underneath the foundation before we pour it. MEN was designed and fit for a world that has long gone.

Mar 12 2018 08:06

Don't lose sight that over 80% of the world uses the TN-C-S system of supply (which is simular to the AU/NZ MEN system of supply).

In NZ the MEN system of supply was mandated way back in around 1929.

Mar 12 2018 11:38

The talk of a 300 mA RCD got me thinking. If all sub circuits are protected by a RCD a broken N should trip all RCDs anyway. (At least ones with current flowing). I understand this is coming in the new 3000 soon.
Would this of prevented this electrocution from happening?
When RCD’s in Australia became mandatory years ago refrigeration and lighting circuits were exempt. I couldn’t see any point having them on those circuits anyway. What I’ve seen since then with potential fires and electrocutions I think 3000 is going the right way. And if I’m seeing it right this latest incident reinforces it.

Mar 12 2018 12:11

Just realised I’m wrong.

Mar 12 2018 12:19

MEN links and electrodes haven't been allowed in caravan park service pillars for a while due to multiple paths for fault current to return to neutral in event of broken sub main neutral.

Mar 13 2018 20:29

"Don't lose sight that over 80% of the world uses the TN-C-S system of supply (which is simular to the AU/NZ MEN system of supply"
In UK the neutral is only earthed at the transformer so may be several volts away from local earth at a remote building. The protective earth often comes down the feeder cable but may a local earth rod, without MEN link.
That is a much safer system because loss of neutral means loss of power, and no live-earth hazard.
What is the advantage of our MEN system and why (apart from historical use) is it mandated in NZ?

Mar 13 2018 21:38

Saint Alan
I am not sure how you get from loss of neutral to no power. Just because the neutral is broken off does not mean that there is no power available. Things might not work as they should but there will still be power available and the risk of shock is still present.

The MEN system was / is mandated to ensure that the network neutral was earthed everywhere. Doesn’t work so well now with plastic water pipes.

Mar 14 2018 07:46

I think he means that people will acctually call an electrician instead of just carrying on thinking that their house is abit quirky

Mar 14 2018 08:03

If the supply is single phase and the neutral goes away (affecting just that installation) then supply will be lost. For multiphase installations (or neutrals lost upstream on the distribution network) voltages start see-sawing. So the neutral voltage will rise. However the key point is that the neutral is no longer attached to earth so earthed stuff do not experience the same voltage rise.

Carrying the earth back to the transformer is only a practical option if the transformer is on site. More often than not the transformer is remote so another method is needed.

My understanding is our regulator wants the TT system to be available for us to use. They are strugling to get it past our friends in the West Island however. Why? Who knows...


Mar 14 2018 10:43

TT has its own advantages & disadvantages, as does any system. It's true that the old metal water pipe systems provided a better connection to mass of earth than the dedicated earth electrodes we use now - but only for areas that actually had reticulated water. And as far as tying the distribution N to earth is concerned, the difference is not significant - Ohm's Law says that a lot of high-impedance electrodes in parallel amounts to an overall low impedance. The issues it's better for compared with MEN relate mostly to faults on the network side, rather than within the installation.
But regardless of merits of either system; opposition from "West Islanders" can't stop TT being introduced as an option for NZ. And having it as an option for some (limited) situations is all that has been proposed - with the choice being up to the network.

The biggest issue I see is NZ sparkies (including many inspectors) mostly don't understand MEN (our particular flavour of TN-C-S); so having two different systems for them to get wrong is rather scary.

As far as this sad case goes, clearly something went wrong. But speculation without facts is pointless; and until the facts emerge the case can't be used as evidence of a problem with the MEN system.

Mar 14 2018 19:51

What type of RCD typically protects the installation in a TT system?

Mar 14 2018 20:03

SaintAlan asks "What is the advantage of our MEN system and why (apart from historical use) is it mandated in NZ?" which is a very good question.

The huge advantage of TN-C-S systems, which is what MEN is a implementation of, is the ability to rapidly clear earth faults, as earth and neutral are the same thing, and thus a phase short to earth causes (give or take) the entire PSCC to flow, which can open a breaker in a fraction of a second. By contrast, systems which require the earth fault current to return to the transformer through a relatively high impedance path have trouble clearing faults rapidly.

While the short is in place, the circuit is a potential divider with two resistances, the "top" resistance one being the phase conductor impedance, and the "bottom" one is the earth conductor impedance. The equipotential zone is connected to the cent re-point of the two resistances.

If the earth resistance is high, then the bonded metalwork and the entire equipotential zone rise to very near the phase voltage with respect to the neutral tap on the transformer, as opposed to in a MEN situation, where the entire neighborhood and their equipotential zones are dragged upwards together to about 50% of the phase voltage.

MEN, which I believe derived from the work done in the UK, has been around since the thirties at least, and is - worldwide - regarded as the standard way to derive an earth. It is some distance from perfect, and there are situations where other systems, particularly TT, are (much) better.

You mentioned the UK: In the UK, you see a bit of most contenders, the UK has TT, mostly rural overhead, TN-C-S a/k/a PME, and also a lot of the older urban distribution is TN-S; the old cables had a metallic sheath under a cloth wrap, and this metallic sheath is the earth, which is electrically continuous from the transformer star neutral point to the consumer premise. I've seen UK installations from the 1970s that were PME that had a statement adjacent indicating that the Secretary of State had permitted PME in the specific installation, so it clearly wasn't a defacto standard back in the day.

The commissioning electrician is (or was, I may be well out of date) free to choose earthing system on top of what the board provides. A big difference between NZ (and the USA) is that the provision of the earth is the commissioning electricians job here, whereas in the UK, the utility will hand over a green/yellow conductor along with the phase(s) and neutral. But even if the utility handed the sparkie a PME g/y, the sparkie could always ignore it and make the installation (or, more sensibly, part of the installation) TT.

TT installations require RCDs because their fault current is quite low. Back in the day before the modern RCD, they had voltage-operated ELCBs, which required an insulated (from "real" earth) equipotentially-bonded system, which is nothing short of a nightmare to configure and to keep isolated.

All the earth systems have pluses and minuses. There is a Wikipedia article describing them:


Mar 15 2018 08:16

+1 dbuckley.

Important to note that the front-end RCD for TT (as or next to the main switch) can only offer fault protection to items downstream from that point.
any metalwork associated with the mains, including the MSB itself, won't be protected by the RCD, and will need to be earthed differently - probably to the incoming PEN conductor rather than to the earth bar.

Mar 16 2018 23:28

Thanks for the detailed description.
Most electrical workers have encountered lost-neutral faults, but I only understood the real danger when it happened at our home in Titirangi.
The Mrs said the lights went dim when the cooker was in use, and I said yes i'll look at it sometime, as you do. Later I heard children laughing and yelling in the laundry, and found our kids daring their friends to touch the tap and get a shock! After clearing everyone out I measured 70V between the taps (bonded to PE) and the concrete floor. The earth rod was too hot to touch and the ground steaming.
The lines company found the fault was a corroded neutral line tap on the pole.
It gives me the horrors that under some loads the voltage on exposed metalwork could be fatal, and my kids were exposed to it.
So if there is a movement to replace this horrible MEN system, that caused this hazard, with TT or something similar then I will be out there with a big placard.


Mar 17 2018 00:03

SaintAlan - it's late, and I might not have thought all the possibilities through - but the simple question that comes to mind, is - was the slab bonded?

Mar 17 2018 10:43

The requirement to bond the conductive reinforcing in the bathroom floor slab was added to AS/NZS3000:2007 in amendment 1 so would have applied from the time that the standard was mandated by ESR 2010. But most electricians have not bothered to read and assimilate the rule changes related to this. They continually have the possum in the headlights look when asked “have you bonded the floor slab either as a whole or the area in the bathroom?”
If the house was built before 2010 there was no requirement to bond the conductive reinforcing.

If the slab was poured after the rules changed then the electrician who did the installation created the conditions where the lost neutral caused this danger, filled in a false declaration on his COC etc.

This is a perfect example of why bonding of the floor slab should be done.

Also not part of any high risk inspection so they cant blame the Inspector.

Mar 17 2018 10:48

SaintAlan: "So if there is a movement to replace this horrible MEN system".....

Very unlikely.

All earthing systems are compromises, and of the compromises available, in most circumstances, for most faults, TN-C-S systems are the least bad of the options available, by which I mean in 99.9% of the time that an earthing system is required to do something protective, then the TN-C-S systems do it very well, and do it better than the other competing schemes.

However........ the protection of TN-C-S can only be relied upon within an equipotential zone. Outside of the equipotential zone, TN-C-S can be scary.

This is why the plumbing in the house is bonded to the house earth, as it it were not, then the metallic plumbing can import an outside earth potential that will be different to the equipotential zone earth, and a person can come between that outside earth potential and the house earth potential and get shocked.

This type of shock is what I call a ground-to-ground shock, and they are f**king scary, as in the general case there is nothing that can be done to prevent them. There are lots of places they can occur, often without there being a "fault", but just because of importing a foreign ground. Running a 50m extension cable out the window to a tent in the back garden seems a safe thing to do, but can go badly wrong.

A garden tap on the wall outside is outside the (hopefully) solid equipotential zone of the inside of a house. Cowsheds are hell to make equipotential, and because cows are sensitive to stray currents, a cowshed is a terrible place for utility TN-C-S, yet it is mandated in NZ.

TN-C-S being safe relies on the integrity of the supply neutral.

It would not be that hard to have continuous integrity monitoring of TN-C-S in an installation, and compared to the increase in expenditure that the universal fitment of RCDs to installations has cost consumers, would be quite cheap. But I get the impression that the regulatory world thinks that TN-C-S is "safe enough".


Mar 17 2018 11:19

Yes, NZS3000 requires some slabs (and steel framing) to be bonded but builders know nothing about it, council inspectors don't check because it is not in the building code, and not many sparkies can be bothered.
My old house in Titirangi was built around 1960, had copper piping bonded to the old-style common earth-neutral bar so was an excellent conductor to the faulty neutral.
The concrete floor of the laundry probably had no PVC under so was a good earth, therefore a better earth than the cooked 5/8 galv PE rod.
So this is typical of a significant proportion (majority?) of the houses in NZ.

Mar 17 2018 11:48

So I drew a diagram from SaintAlan's description.
Assuming that the slab had a reasonably good earth by itself.

From the description, it's possibly slightly unusual that there were two faults. The faulty neutral, and the less than optimal earth.

But I doubt that a more effective earth would have helped very much because they are never low enough resistance to clear faults.
Also I doubt that a different type of earthing system would have helped on its own. If it was TNS for example, the actual fault wouldn't have been apparent without additional protection (i.e. RCD)

The only things that would actually help (in the case of the NZ TNC-S), would be bonding the slab, and having RCDs on circuits. Both of these are in the current standards to a certain degree.

Also possibly educating people about brown-outs, and flickering etc - to turn off the supply might be helpful but probably only to a small degree.

I've only dealt with one in the last few years, and that one didn't involve anyone getting a shock.

Luckily the current system seems reasonably robust, with few electrocutions as far as I know.

Mar 17 2018 12:18

Many of you have forgotten about the efficiency of equipotential bonding if this is done correctly, the exposure to hazardous voltages can be considerably reduced.

Except in the case if a person is in good contact with the mass of earth (no slab when touching the tap, in this case).

I will be seeeing the WA Electricity Regulator or one of his staff at a meeting in Sydney next week, and will endevour to get the true facts of the incident.

It serves little point on making many assumptions without being in procession of the actual facts of what happened and the electrical arrangements when the incident occurred.

Mar 17 2018 14:22

SaintAlan's last post highlights a weakness not in our "MEN" variant of TN-C-S; but of ANY earthing system. The weakness being "many sparkies can't be bothered". Which of course is the difference between a "sparky" and an electrician.

Similarly his own initial reaction to a reported fault was "I'll look at it sometime"; because he (then) failed to understand the significance. As many others still do.

No matter what earthing system is in use; if the system isn't installed correctly, it can't function correctly. That's not a fault of the system!

In his case, it seems the concrete floor was in contact with mass of earth, perhaps moderately good contact. The taps were effectively bonded to the electrical installation's earth.

Under fault conditions, the installation's earthing system "floated", as it is designed to do. Everything connected to it would be at the same potential. The problem was that the concrete floor should have been bonded to the rest, so it would be at the same potential. The several metres between where the electrode was and where the concrete was has an impedance, so when the electrode's [otential rose there is naturally a voltage gradient between that bit of dirt and the bit of dirt the source (transformer) is connected, and the bit of dirt the concrete is on is somewhere along that gradient - in this case enough to result in a significant touch voltage.
(dbuckley's description is more detailed, and quite correct)

But repeat - the hazard did not arise from an "unsafe" system, it arose from poor, and these days non-compliant, workmanship. My "rules" library only goes back to 1976, not 1960; so I can't be certain whether or to what extent bonding was required back then.

And yes, TT might have avoided the resulting touch voltage - but only for the particular fault of a broken /high-impedance distribution (or mains) PEN conductor. In some other fault scenarios, TT doesn't work so well. I'd agree with dbuckley that TN-C-S is the "least bad" overall.

When it comres to cowsheds, they've been perfectly OK on TN-C-S for generations - so the fact that there are an increasing number of cases where touch voltage causes problems in cow sheds can't automatically be blamed on the earthing system. One explanation might be that lines maintenance is not kept up as well as it should be / used to be; and yes TT deals better with imported earth potential rise including from neutral faults among other causes. So no surprise that networks generally support introduction of TT; it is much more forgiving of network faults.

But the thing that really has changed in cowsheds is the increasing use of VFDs, and the risks arising from the earth currents generated by them - including HF currents. Which TT won't do anything to solve.

Modern cowsheds need only a proper equipotential bonding system for 50 Hz, but also careful attention to proper installation practice for VFDs to deal with induced HF currents. Bonding alone is not enough, under either earthing system. So if (when) TT is allowed as an alternative, we'll get a slight decrease in incidents from imported EPR, but we'll still have issues arising from VFDs - because "sparkies" will continue with their she-be-right & save-a dollar" approach. And at the same time, we'll have two systems for "sparkies" to get wrong.


Mar 18 2018 13:04

The issues in cowsheds I have experienced are due to VFDs mounted on top gates and rotary platforms and is the result of poor earthing of the gates and platforms. You can generally tell there is an issue when you see that the cows will not touch the steel work, normally it is nice and shiny from them rubbing past. Cows due to their body length can experience the effects of touch voltages more than people can due to the length of their bodies and distance between contact points. What a person may not be able to feel a cow most definitely can.

Mar 18 2018 13:17

You're right Alec, VFDs are the spawn of the devil. because they are effectively chopping the mains up, they are generating harmonics. The manufacturers cabling and earthing instructions on both sides of the VFD have to be followed to the letter in order to meet EMC compliance, and a large part of this is keeping harmonics under control. If the EMC performance is correct, then there shouldn't be stray HF AC currents either.

My favoured solution for horrible situations is a local isolating transformer with TN-S earthing on the secondary. It gives the TN-C-S advantages with none of the disadvantages, other than cost, both capital and losses, and weight and size, which are usually not deal-breakers. Combine this with a well-thought-out earth scheme (eg PANI), and even in noisy environments like data centres, the earth can be clean as a whistle.