Lighting circuit wiring problem

[roofmog]

I am not an electrician but have picked up enough knowledge through my work to consider myself competent enough to change a light fitting. Which is what I was doing when I came across this problem.

My kitchen extension is powered and protected by a separate RCD board to the rest of the house. It controls the lights, sockets, induction hob and electric oven. Today I was replacing a 4way GU10 spot bar with a similar LED unit. I turned off the lights, sockets and hob breakers, but left the oven and main breaker on, mainly because I didn't want to have to reset the clock on the oven.

After I had attached the new unit, I metered out the earth connection and unit chassis to check that I had a good earth connection, which I did. I decided to quickly check the live and neutral and was surprised to discover that there was a circuit between the live and the chassis and the neutral and the chassis. This was not as strong as the earth connection, but the meter was indicating both aurally and visually that some kind of circuit was being created. I assumed that there was a problem with the fitting, so disconnected it. However, when I metered it out all was good.

I then decided to meter out the bare ends of the mains cable coming out of the wall using the metal attachment screws of an adjacent 13 amp socket as an earthing point. As expected, the earth cable indicated a strong circuit was being created, but again there were indications of some sort of circuit between the live and neutral tail ends and the earthing point. I turned off the cooker breaker but this made no difference and it was only when I turned off the main breaker on the unit that the problem went away.

Do I actually have a problem that needs professional investigation, or is there something about the way RCD boards work that I don't understand? I've capped off the cable coming from the wall with chock block and LX tape until I know what is going on, and when I turned everything else back on it all worked fine.

 

[sparkwright]

What is your meter actually measuring?
Continuity?
Voltage?

 

[roofmog]

What is your meter actually measuring?
Continuity?
Voltage?

Continuity

 

[ericmark]

MCB's don't isolate, neutral is considered as a live and well as the line, and the MCB's don't switch off the neutral. The neutral is outside the property connected to earth, so there will be continuity between neutral and earth unless an isolator is turned off.

The supply is AC not DC, so there will be some capacitive and inductive linking, this is why we measure the earth leakage with a clamp on ammeter,

this shows my consumer unit in total has an 8 mA leakage, we are allowed 30% of the RCD rating, so even if only one RCD fed all, that would just pass. I actually have 14 x RCBO's.

Continuity between line and neutral can be any item left on, including USB sockets etc.

 

[Murdochcat]

If you did that test with the MCB off and the RCD on then the “result” you got isn’t a surprise

 

[ebee]

I had something similar in one respect - A few years ago I totally rewired a house, a year or two later the customer told me that his friend had tested a lighting circuit because they wanted to change the light fitting.
They had used a volt stick and a two pole continuity tester.
I commented that a volt stick can easily give misleading indications (same applies to those awful neon tester screwdrivers too).
He informed me that the continuity tester had given indication between N & E and also between L & E dependant upon switch position for that light -
I asked if he had isolated the circuit, he replied they had flicked the Breaker off.
I pointed out that N, for that circuit, needs disconnecting at the consumer unit otherwise it would be linked to E at the origin whilst testing.
If, at the time of the test, a load (such as a lamp) was in place then the L would be connected to the N therefore giving a similar result between L & E as between N & E.

The problem with people having a bit of knowledge and then buying simple equipment to test without proper knowledge of testing and inspecting can, in some instances, lead to incorrect conclusions (either way around).

In the past I have demonstrated that the end of my nose is giving a positive LIVE result under apparent everyday conditions with neon testers and volt sticks and demonstrated phantom voltages with two pole testers. A kind of "party trick" if you like.

I`m not saying your wiring is safe or unsafe. what I am saying is that it needs testing properly of any concerns to establish a true picture.
It is difficult to establish from reading your post if a problem exists or not.

 

[ericmark]

I commented that a volt stick can easily give misleading indications (same applies to those awful neon tester screwdrivers too).

I seem nothing wrong with non-contact voltage testing, the meter I show has that function, or neon screwdrivers, they have saved me from the dreaded borrowed neutral, being able to test before one removes any screws, or while removing screws is a clear safety feature of these devices.

However, proving for dead has always been something to do after switching off an isolator, and using the appropriate equipment, there can be a little debate over what is appropriate, the voltage tester should have no batteries, not switchable between ranges, and have a proving unit to show it will work with 50 volts AC or 75 volts DC it can also test it using higher volts, but it must show it works at the lower end of the low voltage range. And I have yet to find a tester for the clamp on, cables should tested for current to ensure no borrowed neutrals.

In the real world, we tend to use what we have to hand, so my clamp on does it all, had I been caught with only my clamp on where I have worked, my site pass would have been withdrawn, i.e. sacked, but the proving unit we used generated 500 volts so we really had no idea if the lights would work at 50 volts, and we had no NCV testers, so we had to open a panel before we had any idea if safe to do so, and no clamp on or neon screwdrivers to check for borrowed neutrals, so even when doing it to the book, if assessing the risk, likely the simple clamp on meter shown was a better safety device, in the right hands of course where we have selected the correct ranges etc.

As said, nothing wrong with neon screwdrivers and NCV (non-contact voltage) testers, but in the right place, and with the knowledge as to what they will and will not show.

 

[SUNRAY]

I seem nothing wrong with non-contact voltage testing, the meter I show has that function, or neon screwdrivers, they have saved me from the dreaded borrowed neutral, being able to test before one removes any screws, or while removing screws is a clear safety feature of these devices.

Likewise my tool kits all contain at least an LED screwdriver and possibly neon too. In my line of work they are one of the most useful bits of test kit I use.

Nitty gritty bit: They are very easy to misunderstand, as such probably not a good thing for beginners to use.

The voltstik may be slightly more useful for them but again the reading only give a 'may be live' indication, I have never owned one as the LED scrwedriver is capable of showing the same indication but also has continuity and contact capabilities.

Like Eric such screwdrivers may have saved me from some nasty shocks a number of times over the years and working on IT or other 2 wire circuits they are essential.

 

[Jurassicspark]

Likewise my tool kits all contain at least an LED screwdriver and possibly neon too. In my line of work they are one of the most useful bits of test kit I use.

Nitty gritty bit: They are very easy to misunderstand, as such probably not a good thing for beginners to use.

Neon screwdrivers can make life easier when testing at a lighting circuit which was pre 1966.

I also find them handy during earth faults etc etc.

If someone knows the limitations of the neon screwdriver then it should be okay but for a Diyer a multimeter is the best instrument.

 

[ebee]

It is the limitations that render them void for testing.
OK, some of us know the limitations and mitigate accordingly but generally any ordinary person using them will become complacent about proper testing and therein lies the problem.
They are no good as a test for dead and can not be relied on.
Both the volt stick and the neon tester can be difficult to see in some lighting conditions.
Both of them can easily give false false positives and false negatives.

Mate of mine got a right belt off a neon tester screwdriver - it was his own fault cos he did something stupid - one day he dropped it in water, shook it a lot and put it in his toolbox, so far so good. A day or so later he got it ot for testing and the OUCH! he got a right belt, the twit (PS if the in-handle resistor goes down then that`s another source of danger, not unknown apparently) so best never used.
If I get a set of screwdrivers and it has one of those in then I remove the neon and resistor and use it as a screwdriver.

Only proper equipment should be used, in a proper manner, to test for dead.
A volt stick has one possible use - as a back up overcheck.
Always use the proper test for dead procedure then there is nothing wrong with an additional quick wave with a volt stick (what is the worst that can happen? A false positive reading. But it might just catch that potential disaster and nip it in the bud.

Whenever I have an instructed person with me stripping out old gear and everything is confirmed dead and I have instructed them exactly what is safe to remove I give them the volt stick and insist they use it on each item immediately prior to commencing, reduces chance of that few in a million situations, nothing wrong as a back up providing it is always as an addition to proper testing.

As a "Party Trick" it is good fun to "prove" that earthy things are live if you are stood near an old cathode ray tube TV with one leg or body then touch the volt stick on something earthed and watch it glow, some of the older LED versions of the neon tester could easily do the same too. I did have a spate of decorators telling me that some strange things were live cos they had bought one of those and had a habit of touching everything with their new toy, a few years back.

 

[FrodoOne]

roofmog stated:-

I am not an electrician but have picked up enough knowledge through my work to consider myself competent enough to change a light fitting. Which is what I was doing when I came across this problem.

Then stated: -
"My kitchen extension is powered and protected by a separate RCD board to the rest of the house. It controls the lights, sockets, induction hob and electric oven. Today I was replacing a 4way GU10 spot bar with a similar LED unit. I turned off the lights, sockets and hob breakers, but left the oven and main breaker on, mainly because I didn't want to have to reset the clock."

From this, we do not know
BUT
the "4way GU10 spot bar" may be on a "Light" circuit, with other "Lights" in parallel (?)

After I had attached the new unit, I metered out the earth connection and unit chassis to check that I had a good earth connection, which I did. I decided to quickly check the live and neutral and was surprised to discover that there was a circuit between the live and the chassis and the neutral and the chassis. This was not as strong as the earth connection, but the meter was indicating both aurally and visually that some kind of circuit was being created. I assumed that there was a problem with the fitting, so disconnected it. However, when I metered it out all was good.

Here, "roofmog" seems surprised that he is obtaining a "continuity" reading with a fitting which may be connected to other "fittings" on the same circuit where these fittings may be switched on.

Since "This was not as strong as the earth connection",
it is likely that the resistance of one or more "switched-on" lamp fittings was being read.

(The "good" reading on the disconnected lamp fitting seems to "prove the point"
and
one should always read a "solid" connection between Neutral, and Earth wiring.)

Alexander Pope wrote "A little learning is a dangerous thing."
and
so it seems, in this case!

"sparkwright" (Posts #2 and #3) determined that "roofmog" was reading "continuity" (Resistance/Conductance)
which could be deleterious to any Multimeter
unless one is CERTAIN that the circuit is turned OFF.

I realize that this "Site" exists to assist those who DIY.
However, those persons need to be assisted in "correct procedures".
even if they think that they
"have picked up enough knowledge through work to consider myself competent enough to change a light fitting.",
which may not be the case !?!?!

 

[ebee]

I concur