Just now, emptypockets said:

I replaced one with an LED, no problem, works well. I put in an extra one and it glows dully but beautifully with the power off. Stuff it, it's staying that way. The wiring in the wife's house is atrocious (messy but OK safety wise) and I'll get around to rewiring it one day. May even put in an MEN link (after a good look around at the LV overheads first) and earthed outlets while I'm at it!

I should get my finger out and redo the shower earth. Looks like a piece of 0.5mm that goes out through the wooden louvres and disappears into the concrete. But it is RCD protected - and I trip it regularly for test.

Yes, same here, it worked on circuits and not others.

The one in the bedroom annoyed me because it was fairly bright and would turn on and off at random times.

if I owned the house I would totally rewire as you say.

LEDs (and CFLs sometimes), glowing or flashing when "off" is often the result if the switch being in the neutral side.

 

If you can't re-wire it you can often kill the glow with a small "X" rated capacitor across the lamp.

 

A good source would be a small motor capacitor, the type used in fans are cheap and readily available, 1uF or so would be good.

3 hours ago, Crossy said:

LEDs (and CFLs sometimes), glowing or flashing when "off" is often the result if the switch being in the neutral side.

 

If you can't re-wire it you can often kill the glow with a small "X" rated capacitor across the lamp.

 

A good source would be a small motor capacitor, the type used in fans are cheap and readily available, 1uF or so would be good.

Oh! Interesting thought. That would be great. There is clearly a sneak path through the lamp circuit somewhere.

 

It occurred to me that the switch is in the neutral leg as you say but it breaks the black wire, which should be LINE. But of course what should be and what actually is are often different things in the Kingdom.

Unfortunately only the black wire comes to the wall switch so I would have to chase wiring in the ceiling back to the breaker to correct this.

 

But it seems that would simply represent an AC voltage drop across the circuit due to the capacitive reactance. Granted it would be a low voltage drop = 1/(2*Pi*F*C). Do you think this would be providing a shunt current that swamps the leakage path through LED light?

 

I have such caps since they are commonly available as motor phase shift caps. Working on a ladder to install the cap to try the idea is not appetizing though.

 

Standing on a ladder on the bed and futzing with the CFL  and LED has already taxed my patience with this wiring foolishness. Every time I work on the circuits here I replace the twisted wires/black tape connections with wire nuts and that always aggravates me too. I always swear I won't continue to rewire the owners house at my expense too.

 

More and more I'm learning the value of Mai pen rai in living in peace with such annoyances  but thanks for the idea.

2 minutes ago, RocketDog said:

But it seems that would simply represent an AC voltage drop across the circuit due to the capacitive reactance. Granted it would be a low voltage drop = 1/(2*Pi*F*C). Do you think this would be providing a shunt current that swamps the leakage path through LED light?

 

The leakage that's causing the LED to glow is likely capacitive coupling between a permanent live and the switched live to the lamp (think switch dropper cable). It's a very small capacitance but is often enough to give you a tickle if there's a fairly long run.

 

Putting a (much bigger) capacitor across the lamp makes a capacitive divider with the low impedence leg across the lamp, shunting the stray current without building up enough voltage to strike the lamp.

 

 

3 minutes ago, Crossy said:

 

The leakage that's causing the LED to glow is likely capacitive coupling between a permanent live and the switched live to the lamp (think switch dropper cable). It's a very small capacitance but is often enough to give you a tickle if there's a fairly long run.

 

Putting a (much bigger) capacitor across the lamp makes a capacitive divider with the low impedence leg across the lamp, shunting the stray current without building up enough voltage to strike the lamp.

 

 

Ah, that makes total sense as a divider. And yes, the way it comes and goes did speak to me of a ghost leakage, and sometimes waving my hand around the lamp fixture wiring would trigger it, so I suspected it was a capacity to ve effect and you are undoubtedly correct.

 

I take a cap and try again. I do like the LED much better than the silly CFL  light and cheap ballast disintegrating from the heat in the fixture. Cheap junk fixture and ballast, but that's how we roll here.

 

Thanks for the idea. ????