Unexpected reading between Earth and Neutral

[not_shocked_yet]

Hi Everyone,

Please read ALL of the post below before contributing as the fault is unusual (for me, newly qualified).

My client called me out because turning on the lights in her (2009) converted stable block tripped the RCD (this is the main switch in the stable DB). Trying to re-set the rcd three times or so (it kept tripping) would then trip the (old) single RCD in the main house. To be clear, the house RCD is between the meter and the CCU.

Turning on the CB's in the stable block one at a time determined that one of the lighting circuits was causing the trip.

Anyway, whilst performing safe isolation in the stable (with a Kewtech KT64DL) with the supply turned off at the main CCU in the house I recorded a <25v across Line and Neutral but 65V across Earth and Line and Earth and Neutral.

Removing the L, N and E at the CB in the main house for the supply to the stable DB recorded a normal reading of <25V across all L.N and E combinations.

To be clear the DB at the stable is approximately 250 feet from the main CCU. I don't know yet what the cable is to the building (thinking it should be armoured) if it is joined somewhere en-route (which I doubt) otherwise it looks like 6mm or similar twin and 2.5mm earth (it's the red and black variety) - thats what is showing at the CCU and the DB anyway.

Testing voltage at the main switch at the CCU: With the main switch on I recorded 245v across L and N supply and <25V across E and N and E and L. HOWEVER with the main switch off I recorded <25V between L and N and 165V across E and N and E and L.

This is a TT system.

I used a 2 pole voltage tester to also test between E and L,N but still had a reading of over 50V.

Am I looking at a ghost voltage causing the readings? Is the house safe - it seems to be when the system is energised? Could there be a potential difference between bonding and supplementary bonding if the power went off?

The house is thatched and had a periodic inspection 6 months ago (not the stable however) for insurance purposes. No problems were reported at the time to the owner.

I ran an earth fault loop impedance test at CCU and recorded 30.5 ohms.

Hoping this is just a ghost voltage, any thoughts are most welcome.

Many thanks in advance!

 

[tomspark93]

I'm slightly confused but isn't the clients issue the lights tripping the rcd? If so you should be carrying out insulation tests on the circuits which are coupled to the rcd that is tripping. That would be my first step, not voltage tests. Id probably do an rcd test and a ramp test if I was you just to confirm the rcd is operating correctly too.

 

[Plonker 3]

So you have 2 RCDs in series on this install?

Is one time delayed?

What testing have you carried out?

 

[Des 56]

I read these fault finding threads and notice we get all these weird assumed voltages figures,voltages between this -that- Uncle Tom cobbly and his sister, I then have not the faintest idea of what they are on about

Where has all this obsession with voltage figures come from?

 

[Octopus]

Cripes. This is on 3 sparky forums!!!

 

[Silly Sausage]

Cripes. This is on 3 sparky forums!!!

Cripes. You read 3 sparky forums!!!

One is enough! :smilielol5:

 

[turnitonagain]

its important you neutralise, ie. link the supply neutral to your main earth terminal at the house,using 10mm earth wire, 30.5 ohms is far too high for loop impedance, should be about 0.2 ohms for your external loop impedance. tt ststems use the ground to carry fault current back to the supply traffo, if the ground impedance is not low enough your loop ie. line and earth will be too high.therefore it is common to use the supply neutral as a return conductor. going back to your fault, assuming your not importing any stray voltages from somewhere else. check supply cable to stable using your megger at 500v between each conductor and the ground outside[ stick one of the probes in the ground and the other probe on the conductors] and see if the cable in the ground has a leak, if it has there is a possibility of induction from another source. finally your cable to the stables should be supplied on the same side as the primary connection [fused correctly] of the houde rcd, if you are using a seperate rcd as a main isolaterin the stables. rcds in series dont operate correctly.

 

[Plonker 3]

its important you neutralise, ie. link the supply neutral to your main earth terminal at the house,using 10mm earth wire, 30.5 ohms is far too high for loop impedance, should be about 0.2 ohms for your external loop impedance. tt ststems use the ground to carry fault current back to the supply traffo, if the ground impedance is not low enough your loop ie. line and earth will be too high.therefore it is common to use the supply neutral as a return conductor. going back to your fault, assuming your not importing any stray voltages from somewhere else. check supply cable to stable using your megger at 500v between each conductor and the ground outside[ stick one of the probes in the ground and the other probe on the conductors] and see if the cable in the ground has a leak, if it has there is a possibility of induction from another source. finally your cable to the stables should be supplied on the same side as the primary connection [fused correctly] of the houde rcd, if you are using a seperate rcd as a main isolaterin the stables. rcds in series dont operate correctly.

my gosh I have never read so much rubbish in my life.

 

[davesparks]

I think you may have misunderstood neutralising a bit there.
As far as I know Neutralising is the term used in Ireland to describe the requirement to install an earth electrode connected to the MET of all PME fed installations.

 

[Engineer54]

I think you may have misunderstood neutralising a bit there.
As far as I know Neutralising is the term used in Ireland to describe the requirement to install an earth electrode connected to the MET of all PME fed installations.

tt ststems use the ground to carry fault current back to the supply traffo, if the ground impedance is not low enough your loop ie. line and earth will be too high.therefore it is common to use the supply neutral as a return conductor.

Ay, but this isn't a PME system, so he's suggesting that the OP creates his own TNC, ...Not a very clever thing to do that!! ...

 

[davesparks]

Ay, but this isn't a PME system, so he's suggesting that the OP creates his own TNC, ...Not a very clever thing to do that!! ...

That's what I meant by him having misunderstood it, possibly I didn't make myself clear enough

 

[Risteard]

I think you may have misunderstood neutralising a bit there.
As far as I know Neutralising is the term used in Ireland to describe the requirement to install an earth electrode connected to the MET of all PME fed installations.

In the south of Ireland:

The main protective conductor, also known as the neutralising conductor, connects the MET to the PEN conductor in a TN-C-S installation, with the earthing conductor connected to the MET and the earth electrode.

A TN-C-S installation is known as a Neutralised installation.

It is clear that this is what the poster was referring to (but not the original post!).

 

[turnitonagain]

I am referring to a tncs system not a tnc. We are expected in Ireland to neutralize an installation if we come across a tt system, by the way 30 ohms would not be accepted here a good ze, relying on the general mass of earth to carry fault current is not acceptable here, not sure if the UK supply companies let you do that.

 

[turnitonagain]

my gosh I have never read so much rubbish in my life.

Referring to a tncs not a tns, 30.4 ohms far too high for ze, thats why we are expected to neutralize a tt system here. ze should be quite low, under1 ohm. Sorry for the confusion. If not allowed to do in the uk, let me know, thanks.

 

[Engineer54]

Referring to a tncs not a tns, 30.4 ohms far too high for ze, thats why we are expected to neutralize a tt system here. ze should be quite low, under1 ohm. Sorry for the confusion. If not allowed to do in the uk, let me know, thanks.

That's all well and good if the neutral supply conductor has been earthed at several locations along its length, eg such as found in PME neutrals. It's not such a great idea to willy nilly earth the neutral as and where you come across a TT system.

No it's not allowed in the UK. The only allowable connection to the supplies service neutral is by the DNO, and then generally only when the neutral conductor meets the all the requirements as PEN conductor... Nominally 0.35 ohm but can be higher as stipulated/approved by the local DNO...

 

[turnitonagain]

That's all well and good if the neutral supply conductor has been earthed at several locations along its length, eg such as found in PME neutrals. It's not such a great idea to willy nilly earth the neutral as and where you come across a TT system.

No it's not allowed in the UK. The only allowable connection to the supplies service neutral is by the DNO, and then generally only when the neutral conductor meets the all the requirements as PEN conductor... Nominally 0.35 ohm but can be higher as stipulated/approved by the local DNO...

fair point, so how do reduce the 30.4 ohms as it is far too high in any circumstances,

 

[turnitonagain]

fair point, so how do reduce the 30.4 ohms as it is far too high in any circumstances,

thanks for your reply, no system is perfect, it would make sense to allow to use the neutral conductor also as a path for fault current, as the mass of the earth might not be good enough conductor on its own hence 30.4 ohms. in ireland that is what we do in a tt system, in fact we would regard the customer earth electrode as secondary to the neutralizing conductor. also i am not taking away the absolute importance of having an earth electrode in place, if for example the supply neutral if overhead and single ever became open circut. this is only an observation of mine not a rewrite of the rules.
.

 

[Engineer54]

fair point, so how do reduce the 30.4 ohms as it is far too high in any circumstances,

Ah well, there lay's another problem with BS7671!! lol!! They don't give a maximum value, only suggest that a Ra value over 200 ohms is unlikely to be stable!! Far too many actually think that is the maximum level, but they'd be very wrong...
So most totally rely on a 30mA RCD rather than try and reduce the Ra value while the more enlightened, will also install an upfront 100mA S type RCD to cover the possibility of an inoperative 30mA RCD...

There's plenty of threads on this subject in the archives here!! lol!!

 

[turnitonagain]

good point, it is not a good idea to rely on rcd or similar devices on there own, because as you are aware, nobody ever or seldom tests these units, and if they were expected to trip after many years of inactivity, as i have come across many times doing a rcd trip test, which here for a b type standard is 400 ms max.

 

[marc8]

I understand that creating your own PME is not great, but the majority of the network these days is PME so what's not clever to do it, call in the board & get it changed it's free & as above the network is mostly PME anyway.
i am not saying anything about the fault at this time just about making the house PME