Mysterious outages in lighting circuit

[wonderwire]

I have recently replaced a single pendant light on a 10amp circuit to 5 x 6W LED lights with a 77W extractor fan, all wired via the same pull switch.

At first everything worked fine, however intermittently either one or all of the lights refuses to switch on, or the entire loop stops working. Sometimes it works fine for hours.

I have an old fuse box (no RCBOs, etc), and it's not tripping a fuse but when I flick the circuit on and off after a few minutes it works again.

Any ideas what could be causing this? The wiring is fine. I am guessing maybe the load is too high but where is it failing? I have a very basic understanding of electrics so please humour me!

I am having the consumer unit replaced by an electrician shortly - is this something that could be fixed by a new CU? Is it something he/she could easily test and fix?

 

[flameport]

Loose connection or the switch is defective.

 

[ericmark]

The LED lamp will often work from 85 to 260 volt so with a bad connection without the fan they would still work, but with extra load it has highlighted likely an existing problem. In may not even be where the fan is fitted. I am sure it is an easy fix for an electrician.

 

[ban-all-sheds]

I have recently replaced a single pendant light on a 10amp circuit to 5 x 6W LED lights with a 77W extractor fan, all wired via the same pull switch.

Why do you need 5 times as many lights as before?

Has the room grown a lot larger, or is there a problem with how well the new ones actually perform the task you want them to do - to light up the room?

I am guessing maybe the load is too high

You're guessing that a 0.47A load is too high for a 10A circuit?

but where is it failing?

If a circuit really was failing because it was overloaded you'd have a real mess on your hands. You might even be letting out the smoke.

I have a very basic understanding of electrics so please humour me!

Unfortunately electrical faults, the laws of physics, the implacable nature of electricity etc, cannot be avoided by humouring them away. Electrics is one of those fields where there is no tolerance for sub-standard work. Wonky shelves, doors which stick a bit, plumbing which leaks a bit, wallpaper with creases and mis-matched joins, paint with the odd run and brush hair in it - none of these will kill you. Flaky electrics might.

A very basic understanding of electrics, and uninformed guessing are not a sound basis for doing electrical work.

I am having the consumer unit replaced by an electrician shortly - is this something that could be fixed by a new CU? Is it something he/she could easily test and fix?

No and yes.

 

[bernardgreen]

however intermittently either one or all of the lights refuses to switch on, or the entire loop stops working. Sometimes it works fine for hours.

Are these low cost LED lamps on 230 volts ? If so then each one will have a Switched Mode Current Source ( SMCS ) to supply the LED element with te correct current.

Poorly designed ( cheap, low component cost ) SMCSs connected to the same power source can inter-act with each other and become unstable in operation. This can sometimes result in various some lamps, and not always the same lamps, not lighting.

Is the one that refuse to switch on always the same one ? if it is then it is as already mentioned the fualt is most probably a loose connection

 

[ericmark]

however intermittently either one or all of the lights refuses to switch on, or the entire loop stops working. Sometimes it works fine for hours.

Are these low cost LED lamps on 230 volts ? If so then each one will have a Switched Mode Current Source ( SMCS ) to supply the LED element with te correct current.

Poorly designed ( cheap, low component cost ) SMCSs connected to the same power source can inter-act with each other and become unstable in operation. This can sometimes result in various some lamps, and not always the same lamps, not lighting.

Is the one that refuse to switch on always the same one ? if it is then it is as already mentioned the fualt is most probably a loose connection

Interesting idea, can see that with lamps on a dimmer, but can't really see that happening with a simple switch. The poorly designed lamps often don't use PWM controllers they have a simple capacitor to limit current.

However if 12 volt then yes I can see how the lamps and the PWM controller which drops the voltage to 12 volt (often called an electronic transformer) could get upset when anything other than a resistive load is applied. I had not considered that a pendent lamp could be 12 volt, however you do get some which are designed for G4 bulbs and if these were replaced with LED versions you could get an odd effect.

I don't understand why it would be on a 10A circuit, in the main lighting is 6A, although allowed to be up to 16A most lighting circuits have some where a ceiling rose of junction box which is only rated at 6A so in theroy should be on a 6A MCB or 5A fuse. I will admit although the ceiling rose is rated 6A it is unlikely to have a problem with higher fuse/MCB ratings, but that does not make it right.

 

[bernardgreen]

The poorly designed lamps often don't use PWM controllers they have a simple capacitor to limit current.

Those are the really disgustingly badly "designed" items.

 

[Taylortwocities]

So, everything worked fine until YOU

replaced a single pendant light on a 10amp circuit to 5 x 6W LED lights with a 77W extractor fan, all wired via the same pull switch.

Now it has problems. So maybe there is the clue to where the problem might be

 

[deadshort]

Hi, if the fan has a timer fitted and it operates with the lights ? I will have a permanent live feed from the lighting circuit.

Loose connection or the switch is def

So, everything worked fine until YOU

replaced a single pendant light on a 10amp circuit to 5 x 6W LED lights with a 77W extractor fan, all wired via the same pull switch.

Now it has problems. So maybe there is the clue to where the problem might be

Pull the other one TTC !! Or should that be, pull another one ?

DS

 

[wonderwire]

Thanks @flameport. It was a loose connection inside the switch.

 

[EFLImpudence]

There is probably at least one more if the lights are erratic.

 

[winston1]

The poorly designed lamps often don't use PWM controllers they have a simple capacitor to limit current.

Those are the really disgustingly badly "designed" items.

Why? Much less to go wrong and no RF radiation to cause interference unlike the PWM type. Seems a good design to me.

 

[bernardgreen]

Seems a good design to me.

What is the current waveform through the LED element ?

What happens when the capacitor goes short circuit and the mains supply is applied directly to the LED element ?

 

[ericmark]

Seems a good design to me.

What is the current waveform through the LED element ?

What happens when the capacitor goes short circuit and the mains supply is applied directly to the LED element ?

With the Poundland lamp alteration video I watched they used a small resistor as a fuse, it had two capacitors one to limit current the other after the full wave rectifier to smooth so there would be a wave form not pure DC but it is clearly acceptable. There was also a resistor across the LED's to help it go out quickly. The design would not cause problems with rotating machinery and is cheap, but variations in voltage will mean variation in light output, they would work with dimming switches, but the lumen per watt would be poor. The link is here

This was his results, he was changing the main limiting capacitor, I found it by accident and it was interesting although I don't think I will bother altering any bulbs.

 

[bernardgreen]

The two caps act as a voltage divider as well as the electrolytic being the smooting capacitor for the DC supply. The resistor across the LEDs is more a discharge path to discharge the capacitor when the mains is switched off. Without it the capacitor would have a remanance voltage equal to slightly less than the sum of the forward voltages of the LED elements.

The fusible resistor sometimes get changed in production to become a much cheaper but non fusible resistor which may fail short circuit.