Neon Testers and False Leads - Help Please!!

[ger]

Hi Group, probably old hat, but even a one-liner response would help.

The brother-in-law has a holiday home. Mains electricity - but at a distance. I am an electrical engineer (retired - the theory type - not so experienced on the practice). He asked me to help with sorting the wiring for the lighting in the house. A long corridor with dual switches for overhead lights. Been there for many years. Not sure if it was ever right - he did some amatuer DIY on it and maybe changed things . . . screwed it up. . . Some lights come on when switch A is on, others when switch B is on . . . A mess. So, I pulled back the wall switches and did some simple probing with a neon tester. Some wires showed a bright neon - obviously live. Some showed no light - not live.

BUT SOME WIRES SHOWED A DULL NEON LIGHT - when other switches were turned on these became bright.

I has assumed it would be easy - which wires were live - light or no light - draw the diagrams etc etc etc.

So, my question please. If a neon tester just glows a bit, can this be due to seeing a neutral with a long run back to the local supply trafo and hence with some small voltage drop? Can a neon tester (the screwdriver handle type) register a few volts on the neutral?

Or am I losing it totally? Missing some point?

Any advice (be gentle, please) appreciated.

Ger

P.S. Plan B is to switch off the supply to the house, isolate all the wires from the switches and use a meter to trace the wiring.

[dom]

Neutral can certainly be a few volts above earth. I remember as a youngster droping the phase and neutral wires from a bit of equipment to be moved, out of a 3 phase circuit breaker, not realising it didn't break neutral - and scaring myself when I accidently sparked the neutral to earth.

[dom]

Another explanation is that you probed a section of wire between some combination of live bulbs - ie bulbs that should be in paralell were switched into series - so you saw a partial voltage.

[Frank Erskine]

On Mon, 17 Apr 2006 22:50:50 +0100, "ger" had this to say:

No. A neon takes around 90 volts to strike in the first place. What is likely happening is either a leakage or induction from a phase wire into the wire you're "testing".

To be honest, you'd be better off with a really old-fashioned voltmeter which puts a load on a circuit. A digital meter normally has a very high input impedance and will give misleading results.

Even a test lamp is useful!

[Owain]

Neon tester = s**te.

Capable of giving you false positives, false negatives, and a fatal shock if it fails. Banned by Health & Safety from being used in the workplace (or they should be).

Get yourself a proper test probe - Steinel make a good range. I have one with a row of LEDs, 6V 12V 24V 48V 120V 240V 415V which is useful for everything from car electrics up to 3-phase; also indicates AC/DC and polarity. Very handy.

Possibly, it might also be an induced voltage on a switched live

Yes. But Plan A needs to be "buy a decent tester"

Owain

[John Rumm]

Poxy dangerous things! (as you have discovered)

Not always...

Not always...

Not uncommon...

It can be seeing a unconnected wire that is running a distance close to a live wire and is hence capacitively coupled to it.

Two way lighting circuits are exactly the sort of place you tend to find this and hence these things are no use to man nor beast in fault finding. Even a digital volt meter can give you spurious results. An old analogue one will tend to be more accurate since it will normally place enough load on the line to dicharge the (tiny) capacitance.

Unlikely to be enough voltage on a neutral to light it. You are correct that a neutral can and should be treated as a live - in that it can be held several volts above ground and with a reasonably low impedance - so significant current can flow from say neutral to earth in some circumstances.

Only that neon screwdrivers are the work of santa!

Dig out a circus diagram for a two way lighting circuit (there are a few ways of doing it - my favourite wires the live and swtich return across L1 and L2 on the nearest switch and then simply joins L1, L2, and COM, to the other switch using some three core and earth). Work through it methodicaly with a clunky old multimeter (in fact a digitial will often work in that you can usually see the difference between real 240V live and a coupled but floating wire)

That might work as well. ;-)

[ger]

Many thanks to all who replied. Problem now fully understood. Neon testers now reduced to the role of screwdriver - intestines removed.

Ger - a happy camper!