Open PEN in TN-C-S

[KeenPensioner]

I posted this in the DIY section and got little response which is unusual so I'm posting it here. Hope that's OK. I'm still learning so be gentle.

I'm trying to understand the consequences of a broken PEN conductor downstream of the DNO service head. If I'm correct, this will cause all exposed conductive parts to potentially rise to mains voltage but this voltage will be reduced by any parallel paths created by supplementary bonding to metallic water, gas supplies etc. Also any RCD in the circuit will not trip.

My questions are:
(1) Will the parallel paths have a low enough resistance to generate enough current to trip any MCB connected to the circuit? Effectively a TT system?

(2) What happens if the gas, water supplies coming into the property are plastic? Surely this will mean no supplementary bonding is required and no path to Earth. All exposed conductive parts will remain live?

(3) Supply to outbuilding. If "extraneous conductive parts" in the building then bonding required and the bonding conductor would normally be 10mm2. Why is supplying outbuildings, caravans, marquees etc additionally problematic if supplied by TN-C-S?

As I've said, I'm not qualified, just interested so be gentle!!

 

[cliffed]

Hope this helps

 

[KeenPensioner]

Hope this helpsView attachment 53924

View attachment 53924
[/QUOTE]
Thanks for your prompt reply. I have read this from Voltium and have that image in my study folder. I have watched John Ward's video on this (as well other sources of info). but I still don't understand. The return path (PEN) is broken so the current will return via the main earth to any connected earthed conductive part, making them live (as the image shows). Do the parallel paths from supplementary bonding to water, gas etc have a low enough resistance to trip any MCB and hence protect stick man? What happens if incoming services are plastic? Still don't fully understand why supplying outbuildings that have extraneous conductive parts from a TN-C-S causes a problem - (according to John Ward and others)

 

[cliffed]

mcb may not operate under fault conditions,due to high resistance’s,through stick man,.
With a broken neutral,the characteristicS of the Tncs become unstable,& could prove fatal.

 

[Pretty Mouth]

Do the parallel paths from supplementary bonding to water, gas etc have a low enough resistance to trip any MCB and hence protect stick man?

No. Bear in mind that the equipment is still in circuit. The MCB wouldn't trip in normal use with a healthy PEN conductor back to the transformer would it? When the PEN goes open circuit, the return path is only through the mass of the earth (bonded pipes etc). This would have a much higher resistance than the PEN conductor, so a much smaller current would flow.

 

[Cid]

Do the parallel paths from supplementary bonding to water, gas etc have a low enough resistance to trip any MCB and hence protect stick man?

No

What happens if incoming services are plastic?

Nothing

Still don't fully understand why supplying outbuildings that have extraneous conductive parts from a TN-C-S causes a problem

Inside the house you have carpets and wooden floor etc which means the shock you receive will be less severe.
Outside in the mud you will provide good path to earth.

 

[Lucien Nunes]

With an O/C PEN, installation N, E and hence exposed parts connected to the MET rise to a voltage w.r.t. true earth that is determined by the voltage divider formed by the load impedance and the external earth impedance of whatever earth paths remain.

If the load is an alarm clock and the building has a metal frame bonded to the MET, the load impedance will be much the higher of the two; the touch voltage to true earth will be minimal and the alarm clock will see most of the 230V and hence function normally.

If there are bonded metallic services of low impedance in the order of an ohm, the load of the alarm clock would go completely unnoticed. However a 10kW shower of 5.3Ω element resistance will form a divider in which 1x230/(5.3+1) = 36.5V appears from PEN to true earth and 5.3x230/(5.3+1) = 193V appears across the shower. It will lack power and people might start feeling tingles from exposed metal.

If the load is the 10kW shower and the building is timber with no metallic services, the few stray paths to ground that exist will be no match for the 5.3Ω of the shower element; the whole lot will swing up to nearly 230V and the shower will appear dead.

The same is true for a hard fault to earth with a source impedance of a fraction of an ohm. Bonded metallic services will probably be a good enough earth to trip the OCPD, without them the fault will probably persist.

 

[westward10]

A row of houses on a three phase supply which loses the neutral at source can have all sorts of voltages winging through your appliances.

 

[KeenPensioner]

Do the parallel paths from supplementary bonding to water, gas etc have a low enough resistance to trip any MCB and hence protect stick man?

No

What happens if incoming services are plastic?

Nothing

Still don't fully understand why supplying outbuildings that have extraneous conductive parts from a TN-C-S causes a problem

Inside the house you have carpets and wooden floor etc which means the shock you receive will be less severe.
Outside in the mud you will provide good path to earth.

Thanks for taking the time to reply. I see what you mean regarding the body's resistance depending on what you're standing on etc. I have read that with other systems (TT and TN-S) you can just use the supplied earth, but with TN-C-S you need an Earth rod at the outbuilding with separation from the main building's earthing arrangements.
[automerge]1574001982[/automerge]

With an O/C PEN, installation N, E and hence exposed parts connected to the MET rise to a voltage w.r.t. true earth that is determined by the voltage divider formed by the load impedance and the external earth impedance of whatever earth paths remain.

If the load is an alarm clock and the building has a metal frame bonded to the MET, the load impedance will be much the higher of the two; the touch voltage to true earth will be minimal and the alarm clock will see most of the 230V and hence function normally.

If there are bonded metallic services of low impedance in the order of an ohm, the load of the alarm clock would go completely unnoticed. However a 10kW shower of 5.3Ω element resistance will form a divider in which 1x230/(5.3+1) = 36.5V appears from PEN to true earth and 5.3x230/(5.3+1) = 193V appears across the shower. It will lack power and people might start feeling tingles from exposed metal.

If the load is the 10kW shower and the building is timber with no metallic services, the few stray paths to ground that exist will be no match for the 5.3Ω of the shower element; the whole lot will swing up to nearly 230V and the shower will appear dead.

The same is true for a hard fault to earth with a source impedance of a fraction of an ohm. Bonded metallic services will probably be a good enough earth to trip the OCPD, without them the fault will probably persist.

Thanks. I get the maths behind your explanation - as Pretty Mouth says in #5, the load is still in circuit - in your example, 36.5V will be the potential of any exposed conductive parts and this is below the safe level of 50V?

 

[Pretty Mouth]

I have read that with other systems (TT and TN-S) you can just use the supplied earth, but with TN-C-S you need an Earth rod at the outbuilding with separation from the main building's earthing arrangements.

It's only in some outbuildings that a TT earthing arrangement may be preferred over TN-C-S.

If the outbuilding has no extraneous parts and a insulating floor (eg. a wooden chalet) then the risk from an open PEN conductor is no greater than in the main building. With extraneous parts and an insulating floor, still ok so long as you can bond extraneous parts back to the main earthing terminal at the origin of the installation.

But consider say a metal shipping container being used as a garden shed. The whole structure is one extraneous part, so must be bonded to the MET. If you get an open PEN, the whole structure could go live! A big problem for anyone standing on the ground outside and touching the door.

Likewise, imagine a shower block at a camp site - no insulating floor, naked wet feet. If the pipework went live due to a lost PEN conductor, again very dangerous.

AFAIK lost DNO neutrals are very rare, and TT earthing arrangements have their own potential risks. They rely totally on RCDs for fault protection, and RCDs do sometimes fail. If it did and there was a live-earth fault, again all metalwork attached to the MET would go live, and the overcurrent protective device almost certainly wouldn't operate due to the high loop impedance. 2 RCDs in series would substantially reduce this risk.

Have a read of this, it may also help you judge when TT-ing an outbuilding is appropriate:

http://www.dcode.org.uk/assets/uploads/ENA_EREC_G12_Issue_4_Amendment_1__2015_.pdf

 

[KeenPensioner]

It's only in some outbuildings that a TT earthing arrangement may be preferred over TN-C-S.

If the outbuilding has no extraneous parts and a insulating floor (eg. a wooden chalet) then the risk from an open PEN conductor is no greater than in the main building. With extraneous parts and an insulating floor, still ok so long as you can bond extraneous parts back to the main earthing terminal at the origin of the installation.

But consider say a metal shipping container being used as a garden shed. The whole structure is one extraneous part, so must be bonded to the MET. If you get an open PEN, the whole structure could go live! A big problem for anyone standing on the ground outside and touching the door.

Likewise, imagine a shower block at a camp site - no insulating floor, naked wet feet. If the pipework went live due to a lost PEN conductor, again very dangerous.

AFAIK lost DNO neutrals are very rare, and TT earthing arrangements have their own potential risks. They rely totally on RCDs for fault protection, and RCDs do sometimes fail. If it did and there was a live-earth fault, again all metalwork attached to the MET would go live, and the overcurrent protective device almost certainly wouldn't operate due to the high loop impedance. 2 RCDs in series would substantially reduce this risk.

Have a read of this, it may also help you judge when TT-ing an outbuilding is appropriate:

http://www.dcode.org.uk/assets/uploads/ENA_EREC_G12_Issue_4_Amendment_1__2015_.pdf

Thank you greatly for taking the time to explain. It's beginning to make sense to me now, so thank you (and the others). I'll wade my way through your attachment tomorrow.
I found this with regard to frequency of PEN failures

https://myenergi.com/wp-content/uploads/2019/09/Protection-against-damaged-PEN-Conductor-r1_0.pdf

 

[Pretty Mouth]

Thank you greatly for taking the time to explain. It's beginning to make sense to me now, so thank you (and the others). I'll wade my way through your attachment tomorrow.
I found this with regard to frequency of PEN failures

https://myenergi.com/wp-content/uploads/2019/09/Protection-against-damaged-PEN-Conductor-r1_0.pdf

Interesting, I'm surprised there's so many actually. I bet there's more that go unreported too