0

(I know this question has been asked before (here), but I followed the advice recommended by the users in that thread to help remedy the problem, but the lights still glow.)

I recently bought dimmable LEDs for my house, and I've noticed that the lights glow a little bit even while the dimmer is turned off (I have illuminated light switches to DO NOT use the neutral wire). I suspect this is due to the wall switches using the light bulb itself as neutral. When the wall switch is on, the bulb gets full line voltage. When the wall switch is off, the bulb is the neutral for the light of the wall switch, causing a tiny current to flow through the bulb.

I replaced the dimmer with the Luron MACL-153M, a digital dimmer that is specifically made for LEDs/CFLs, but that problem still persists (note: this dimmer also DOES NOT use the neutral wire, and is illuminated with a small light).

I then replaced one room's dimmer with a smart WiFi switch that DOES use the neutral wire, and walla, no glow. Also, if I use a non-illuminated switch, the LEDs do no glow while the switch is turned off.

So I tried a solution that seems to work for the rest of the rooms where I don't want smart light switches. I added a 10k ohm resistor in between the load and the neutral (so in parallel), and it seems to "drain" that small electricity leak from the dimmer. When I did this, the LEDs did not glow when turned off.

Now, my question to the community is: Is there a safer way to do this? Of course I covered all exposed wires with electrical tape, and my resistor is rated at 200 volts (I live in the US). And theoretically, this resistor is acting just like another bulb on the circuit, but that does not emit light.

I looked around for a non-wifi/Bluetooth/etc. dimmer that uses the neutral wire, and the only one seems to be a Leviton one that costs $60--way too pricey. Adding a resistor in parallel makes sense and is much more cost-effective. Please let me know if this is ok/safe long term.

  • 1
    A switch of any kind, whether dimmer, 3-way, toggle, etc, with an illumination in the switch, will still present some power to the light. LED lights are prone to low level power, most LED's are able to power(glow/light) with as little as 1v. This lighted switch is allowing enough power to pass the switch to light the switch, therefore allowing a small amount to the lamp socket where the LED is picking up the voltage. You don't see this effect with a regular incandescent or CFL. So does the lamp work as designed with an non illuminated switch? – Jeff Cates Feb 25 '18 at 23:17
  • In places where I use non-illuminated switches, there is no glow while the LEDs are turned off (in other words, they work as designed). – Jason Feb 25 '18 at 23:37
  • Is the 10k ohm resistor large enough in watts to handle the required power? – Harper Feb 25 '18 at 23:56
  • 1
    What is the power rating of your resistor? a 10k resistor across 120V needs to dissipate 1.44 watts, a typical small ¼W resistor is liable to overheat and set fire to something. – RedGrittyBrick Feb 25 '18 at 23:58
  • The resistors are rated at 2W 10k Ohm, and 1% tolerance. Even with a 20% surge, I think this will be ok – Jason Feb 26 '18 at 0:18
2

After reading through comments and other answers, I have verified that my solution to glowing LEDs works.

For starts, thanks to bombcar's find, Lutron makes a product that does the exact same thing, called the Lutron LUC-MLC (Wiring Diagram). This gave me the assurance that this method is somewhat considered "safe". However, due to the ridiculous price tag of $18 on Amazon, I knew I couldn't stick buy one for every illuminated dimmer.

I decided to purchase 10,000 Ohm, Ohmite L40 series power resistors rated at 10-watts (Resistor Webpage). After testing one in a controlled box at AC 120 volts for 36 hours, it was barely warm. So I wired it up in parallel, in the same configuration as the Lutron LUC-MLC (Wiring Diagram Again), and of course with plenty of electrical tape. And it works! My LEDs no longer glow (or ghost as some call it) while the illuminated switches are turned off.

Thank you Harper for the Ohmite recommendation and bombcar for the Lutron find.

  • Isn’t the MLC a capacitor, while you are adding a resistor? That’s not really the same method, is it…? – Mormegil Apr 5 at 14:39
1

First and foremost, absolutely required: this has to be inside a steel box. No plastic boxes. The metal box needs to be grounded. This will keep containment on a variety of problems.

Thermal calcs can be quite complicated so I like to shortcut them by leaning conservative. Everything in modern mains wiring is rated at 60C. I assume ambient is about 30C. So I'm looking for a solution that will give about a 20-25C rise.

Power resistors are real fond of giving 180C or even 350C rise at full rating. Going "I'll derate by 30% and that'll be good" is a noob mistake. I've had to use a 1000W resistor to get an acceptable rise at 230W.

Also using a resistor like the anodized aluminum types which require a heat sink, is inappropriate unless you attach it to a heat sink. They have favorable thermal ratings, but only with the heat sink. Junction boxes are not listed to be heat sinks.

Experience led me to an Ohmite L25 series with lugs. The data says we'll have just shy of a 20C rise at 1.44W and that will be fine. The lugs will allow crimp lug attachment with #6-32 or 8-32 screws.

  • The capacitive approach (LUT-MLC) is a definite winner here (far less power dissipation) – ThreePhaseEel Mar 4 '18 at 14:35
1

The Lutron LUT-MLC basically does exactly this, I believe, and I should have one shortly to experiment with.

Another solution is to add more bulbs to the circuit. My switch with one bulb does this - as with two. But four goes off correctly.

  • Yes! Not least, it is equipment listed for the purpose. – Harper Mar 4 '18 at 17:59
0

This to me sounds like a unbalance in the electrical system. Off is a complete break of power no indicator lamp should be allowing voltage of any kind to bleed through. If there is voltage coming back through the neutral there is a defect in said switch or system in general. Also if power is bleeding through the neutral than any other lamp on the same circuit would have a glow as well. I Have Led lamps and dimmers with indicators and they are all working as should. no glow when off no resistor. However some led Diodes will glow a little but turn off soon after. You might want to check if you have Receptacles on the same circuit with lamps plugged in you may have reverse polarity at the receptacle. Great topic and could be a bit complicated with all this technology Good luck.

  • By definition an indicator lamp is using power and it is that power that causes LED’s to glow, your indicator may be LED compatible or use the neutral where the OP’s is not causing the issue, it is not an imbalance it is the way some switches are designed when the switch is open now there is a voltage across the contacts and that s is what lights older illuminated switches, newer ones draw a small enough amount not to cause the glow or use hot to neutral to power the indicator so power is not drawn through the led’s. – Ed Beal Jun 10 at 15:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.