Switch is off but energy light flickers on dimly!

Joined
15 Apr 2004
Messages
36
Reaction score
0
Country
United Kingdom
Hi

I have rewired my whole house (lights and power) and it all seems to works fine
and has been for a few months now. I would say I was competant at this job but
I am not an electrician.

Last night I noticed that my Philips energy saving light bulb on the stairwell light
( on the upstairs lighting circuit ) was trying to light up very dimly even though the light
switch on the wall was turned off. Every 3 seconds it would dimly flicker on and then
go off straight away again, over and over again. I've never noticed it before.
This light is on a 2-way switch, 2 switches operate it. (Finally, if I turned the RCBO off
for this circuit at the consumer unit then the bulb would not come on, of course.)

So, I tried the light bulb on another bayonet fitting in another room and it did not flicker
on when the switch was off. I tried a normal tungsten light bulb on the stair light and
that did not come on when the switch was off. Good. On inspection all my wiring
looks fine.

So, I got out my meter and measure the voltage on the stair light when the switch was
OFF and it was 91V. Is this normal?

No wonder the bulb tried to light up. I measured another 2-way swtich light fitting and
that gave 61V. Three other 1-way switched light fittings read about 30V. So 2-way seems to leak more voltage, presumably across the wall switches??? Is this correct?

(I have not yet tried the (bayonet fitting) bulb inserted the other way round (terminals reversed) as I only just thought of it.) but do these enegy saving light bulbs try to
light up on only 91V? Is a leakage of 91V OK or do I need to buy better swtiches
perhaps? Or, is my problem something else do you think?

Any advice?

Thanks in advance,

Tricia
--
 
Sponsored Links
At the ceiling rose ...
What is the measured voltage between the permanent live and neutral?
What is the measured voltage between the permanent live and earth?
What is the measured voltage between the neutral and earth?
 
Inspector said:
At the ceiling rose ...

(A) What is the measured voltage between the permanent live and neutral?
242V I recall.

(B) What is the measured voltage between the permanent live and earth?
242V (I think, if not it is very similar, e.g. 236V, but I think A and B were the same)

(C) What is the measured voltage between the neutral and earth?
91V

So there is some sort a *floating*? voltage on the neutral wire.
Everything seems normal to me voltage-wise, but this 91V I am unsure about, is it
normal or not?

Tricia
--
 
Check all 3 again with all cables apart from the incoming lighting circuit supply disconnected at the ceiling rose.

If still the same then you're tracing the supply cable back along its circuit and measuring to find the fault.
 
Sponsored Links
Inspector said:
Check all 3 again with all cables apart from the incoming lighting circuit supply disconnected at the ceiling rose.

If still the same then you're tracing the supply cable back along its circuit and measuring to find the fault.

OK, will do checking tonight.

So, are you convinced there is a fault?

I should have said that I am using a _digital_ multimeter, so that will be very sensitive
to any voltages around. There cannot be much of an issue though because the
tungsten light could not draw enough current to light itself, even thought there may be
a p.d. of 91V there.

Can wall switches pass any A.C current when open? e.g. capacitive effect, for
230V across a few mm? Is this even possible?

Finally, for the problem stair light, if I switched both switches the other way, so that
the light is supposedly OFF again, I got a voltage of only 41V.

Lastly, the energy saving light did not flicker on this morning when I tried it again,
so the problem has stopped (unless the bulb was inserted the other way round
by chance.) Still 91V there though.

Any comments?

Thanks
 
The lamps you refer to have no polarity and therefore it doesn't matter which way you insert them.

By doing what I say in my last message it will define whether the prob is with the 2 way switched lighting circuit or with the supply to it.
 
tricia said:
(C) What is the measured voltage between the neutral and earth?
91V

is it
normal or not?


--

Tricia is it not normal, and it is potentially dangerous, further you stated that the circuit has an RCBO in place? What rating is this? 30mA, 100mA or 300mA?

An RCBO should trip with this voltage between Neutral and Earth, if it is not then something is seriously wrong here.

Somewhere on your circuit you have a connection between parts of the circuit that should be isolated from each other, there is some voltage leakage that is not being grounded, this calls into question the validity of the cpc and whether there is a continuous earth path.

My advise, forget pride, pick up the phone and call a professional. With respect, you may not be able to find the fault as you may not be able to spot what is or is not the indication of a fault..this tends to come with experience. You did well so far, but there comes a time to step back and let someone take a look with more knowledge.

It sounds like the circuit, if not the installation is an accident waiting to happen.., don't be that accident. It may be a simple problem to solve, or it may be you made a fundermental error during the rewire and may not be able to spot it.

Either way..SAFETY FIRST, PRIDE SECOND
 
FWL_Engineer said:
tricia said:
(C) What is the measured voltage between the neutral and earth?
91V
is it normal or not?
--

Tricia is it not normal, and it is potentially dangerous, further you stated that the circuit has an RCBO in place? What rating is this? 30mA, 100mA or 300mA?

An RCBO should trip with this voltage between Neutral and Earth, if it is not then something is seriously wrong here.

It sounds like the circuit, if not the installation is an accident waiting to happen.., don't be that accident. It may be a simple problem to solve, or it may be you made a fundermental error during the rewire and may not be able to spot it.

Either way..SAFETY FIRST, PRIDE SECOND

The 6A RCBO is rated to trip at 30mA, and indeed it trips when I press its test button.
The RCBO, I thought, only trips if the _current_ in the neutral differs from that in the
live, rather than being voltage orientated.

I will recheck all the voltage readings tonight to make sure they are as I said they were.

I would like to just check a bit more thoroughly before calling out an electrician.
Any suggestions on what to look for? The circuit is pretty straightforward really,
could the problem lie in the Consumer Unit or another circuit? Switching the RCBO
off (which I believe disconnects only the live wire?) makes the neutral voltage disappear.
Does that give any clues as to what's wrong?

Thanks
 
Hi

I did some more measurements tonight and I believe I have solved my probelm.
The values I reported earlier were not quite right - I remembered wrongly.

So,

at the ceiling rose, the voltages are:
Between LIVE and NEUTRAL = 240V
Between LIVE and EARTH = 239V
Between EARTH and NEUTRAL = 0v

With the wall switch open and light bulb removed my DIGITAL meter reads 91V
across the terminals of the bulb holder.
With a tungsten bulb in place it reads 0V
With energy saving bulb in place it reads 51V

The reason for the 91V is that with the swithc open and bulb removed one has
a piece of wire disconnected at both ends and so can float at any old potential.
I will refer to this *floating* pice of black wire as FRED later on.
It has a potential in it because, I believe, it is induced in it by its sister red live wire
which runs in the sheath with it. The longer the wire (e.g. the 2-way circuit)
the higher the voltage induced.

When the tungsten bulb is inserted, one end of the wire gets connected to neutral
via the filament and hence the voltage reads 0V then. The energy saving bulb is not
so good at conducting??? - funny things fluorescent lights.

The reason the fluorescent light flickered last night is the same as why fluor tube
lights light up under pylons.

Note that digital meters are very sensitive since they leak zero current
on measuring voltage. With an old AVO mechanical meter you would
probably always get 0V not 91V as the meter itself would allow some
current thru it and hence bring FRED back down to a potential of 0V.

To confirm safety I measured the CURRENT flowing between the bulb's terminals
with the bulb not in place and I got a very low reading of 0.04mA !!!
Thus no current flows from FRED to the main NEUTRAL wire when the bulb is in
place and the switch is OPEN, hence no problem, no hazard and no tripping of
the RCBO.

I am 99.9% certain of what I say is true (I'd need to do many current induction
experiments to be 100% certain)

If any of you think I am wrong then by all means speak up as others might read
this forum and act on what they read.

Thanks for all those that offered advice and help.

Tricia
 
91volts seems quite high for an induced voltage from a 230volt supply,congratulations you have an extremely efficient transformer there without coils! Must be one hell of a cable run!
Seriously though, you are correct in assuming that the Energy saver lamp is showing negligble resistance and is therefore not dragging down this Induced voltage(voltage with no current) as soon as you connect a load (tungsten lamp) this voltage gets dragged down to zero volts.
The "energy saver" lamp you are using is it a cheap make? it may not have internal circuitry to show enough form of loading, Osram Dulux lamps incorporate flicker free circuitry and you may find this solves the problem? otherwise you may have to parallel a lampholder/rose to this fitting and add wattage to the circuit (not an energy saving solution i know).
 
When you're seeing this 91V, what happens to it if you turn off the supply to the downstairs lighting circuit?
 
Hi

I have tunred off ALL circuits except the upstairs lighting circuit and the 91V remains
there. The only way to get rid of it is to turn off the upstairs lighting circuit.
 
I gather that you may be using a digital multimeter to take the readings and while normally they are considered to be the dogs b******s in this instance using a moving coil meter may be a better choice as it will load up the circuit you are testing and you may well find the 91 volts disappear or decrease substantially.
 
Hi

(Yes I am using a really decent digital meter with a very high resistance and that
it *seeing* things (voltages) that aren't really an issue.)

Anyway, at long last I got this answer from another source and I think it is correct.

Capacitive coupling is highly likely to be the problem here. The 91 volts is more than probable in this instance. With cables having coupling in the order of about 100pf/metre (minimum), it only takes 10m to get a whole 1nF (i.e. 73uA leakage). Digital multimeters (good ones, that is) are about 20Mohm input. If you measure with a good ol' analogue meter you'll probably find a lot lower than 91V.

If you go here
you'll see a circuit of a typical CFL.
What will occur on your case is C1 will charge up slowly, the oscillator will 'fire' and discharge C1 till it can no longer keep the CFL burning (and repeat itself ad nausium).

You may find that the 2-way switches in one direction cause more of a problem than in the other. Try swapping the "offs" and see if the problem returns (you intimate that you get different leak currents when doing this).

As for a cure (here's hoping you are adept at soldering), put two 150k-ohm resistors in series (to form 300k) across C1 in the CFL. This should keep C1 from reaching the firing voltage of the oscillator. please do not be tempted to use a single 270k 1/4 watt resistor, they are not designed for 250V across them permanently (and the rectified voltage, as is, can reach 360V - and that is if the REC stays within 253VAC).

I hope that helps any onlookers

Thanks again to you all

Tricia
 
The type of meter used is irrelevent.

With a GLS lamp or a compact florry installed, or with nothing, with the switch in the OFF position there should be ZERO volts measured across the terminal at the lamp holder..FACT.

If there is voltage, it is leaking from something. There is the possibility of a poor quality switch which is allowing a discharge across the terminals, although this is rare.

Does the circuit have a dimmer on it..I don't think it does from what has been said, but they can cause transient voltages occasionally.

Either way, Tricia, you should phone an electrician to have a look.

Rergarding RCBO's, these operate by Thermal, Magnetic and by electronic means.

The overload protection (as in actual current rating not milliamp trip rating) operates by either the thermal or magnetic parts of the unit in exactly the same way as an MCB does.

The fault current protection (milliamp rated) operates by continuously measuring the inbalance in the induced magnetic field inside a torus which is generated by the Live and Neutral conductors being connected to coils wound aroun opposite sides of the torus.

If the magnetic flux density of the field becomes unbalanced above an equivlent amount to (in your case) 30mA, then the electronic controls trip the breaker. Further, there is a third coil on another torus which is connected to the Earth (the little white cable from the unit) and is also monitored. This will detect any fault currents induced on the CPC and again trip the unit.

This is a simple explantion of the operation of an RCBO..didn't think you would like the really technical version :D
 

DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.


Are you a trade or supplier? You can create your listing free at DIYnot Local

 
Sponsored Links
Back
Top