light-fixture
we have an issue were our downlights have a dim glow when switched off.
electrician has been back to replace downlights but no success. glow doesnt happen every night, often a big gap in between glowing.
tested the wiring and shows anywhere up to 3 volts through the active wire when switched off.
is it normal to have up to 3 volts running through the active wire? and is this enough to cause an LED downlight to glow.
cheers
There's likely nothing wrong with anything, it's just that the LED's are much more sensitive to electrical input. Common if there's a dimmer, ceiling fan or 2 (or more) switches controlling the lights.
You need a Snubber or Capload or RC Suppressor installed in the circuit. Something like pictured in just one place, in a light's fixture or at a or the switch. But, it's fine to just live with it as nothing's in danger, the lights are just waving bye-bye until next time.
The discussion in the comments revealed that the "planned addition" as described is perfectly fine but that the only detail that really needed additional confirmation was the conduit fill calculations, specifically regarding derating.
Since I'm using a T-style conduit body to branch a raceway for two fixtures there is not room to have a code-compliant splice between the common junction box and the fixtures themselves. This then means that separate neutrals need to be pulled and that there will therefore be more than 3 conductors in the shared part of the raceway. I had initially read that derating didn't come into effect for small wires (#14 and #12) until a much higher number of conductors are involved, but it was a good exercise to truly confirm this.
The NEC requires derating for 3-or-more conductors in one raceway. My initial impression for #14 THHN was that it's rated to only 15 amps, which would then mean that any derating requirements could be an issue for this 15 amp branch circuit. Looking deeper tough it turns out that 15 amps is only the required overcurrnet protection for circuits with #14 wire (from NEC from 240.4(D)(4)) and that #14 THHN is actually rated slightly above 15 amps at low temperatures. In my case I needed to derate 80% (for 4-6 conductors), and for #14 THHN at low temps this worked out to 80% of 25 amps = 20 amps, which still appears quite fine for the existing 15 amp branch circuit that I'm working with. This handy online tool, and of course the appropriate tables from the NEC (from 310.15(B)(16)), confirmed this.
So it looks like this planned addition is fine as stated assuming a wire gauge <= #14.
Best Answer
"Smart" switches use solid state (electronic) switching devices called "triacs" in most cases. The problem with triacs is that they don't completely turn "off" because the solid state nature of them has a very tiny amount of "leakage current" through them. When used on incandescent or even CFL lamps, that leakage current is irrelevant because the circuit can't really do anything with it. But LEDs take so little energy to illuminate that this tiny amount of leakage is enough to keep them glowing a little bit. The only way to stop it is to have a hard "contact type" switch that makes an air gap to isolate the LEDs off line. But those are typically not your "smart" switches. I have heard people say they can add one incandescent lamp to a fixture and that problem goes away, but I've tried it and it causes other problems if you try dimming.