I have 2 wireless digital light switches installed, one each in 2 different rooms. One of them uses 433mhz, and has a capacitive touch sensor(glass surface). The other uses Wifi, and has a tactile type switch. Occasionally, while the lights in the rooms are off, when I switch on/off the light in my kitchen which uses a traditional light switch, the light in either one of those rooms might turn on for a split second, like a short blink. The lights in the rooms are using between 15-30w LED bulbs.
Why does this happen? Is this a product of bad wiring, or something in the likes of EMI? And what can I do to rectify this issue? Thank you.
Oh well, that doesn't sound good. I remembered one more thing. I have a reed switch connected to an ESP8266, which in turn controls a LED strip via mosfets. I also noticed that sometimes while the strip is ON this time, when the kitchen light is being switched on/off, it affects the LED strip occasionally. The strip will turn off and on instead, I assume some kind of EMI affected the reading of the reed switch.
I would only suggest you have filtering on the power supply and any input connections are short with twisted wire connections and pull up or down resistors as needed .
If your kitchen lights are fluorescent , these can generate noise when switching .
I have a fan in my bathroom that sometimes comes on when I operate the lights .
It is a ferrite bead/ choke, connect in series with power supply wires. The switch has 2 wires, so connected in series with them. Similar method to reduce EMI = ferrite on pc cables.
You are right about the kitchen light, they are fluorescent. They are fitted to some really old light fixtures which are a pain to change, which is why they are still fluorescent. So I assume changing them to LEDs would potentially help with the situation. The reed switch with the ESP8266 don't really bother me too much, as I could correct the issue in code. But what's the best way of filtering the light switches? Would it be using ferrite beads like what 79galinakorczak suggested?
I thought an inductive load like a fan would be more resistant to such occurrences? I guess my understanding is too vague on this issue.
The condition of the switch operating your lights might be a factor, plus the age of the fitting .
You could also look at modifying your software , such that an input needs to present for a period of time .
Also keep wiring away from mains leads , and filter your DC supply .
Good luck ! Google might help you with the lights ??
Its possible the wireless light switches aren't designed for real world use on domestic mains - surges and spikes on the mains are a fact of life, even upto the kilovolt range.
As a consequence I'd be worried they aren't safe devices, but some cheap uncertified junk
that's likely to fail as it hasn't been properly designed for the kind of mains transient that
happens all the time.
However it could also be the case that your fluorescent light fitting is faulty, putting much
larger transients out than it should - old fluorescent lights use a large ballast inductor and
capacitive snubbing to reduce the transients it generates - there's also a starter and these
often deteriorate and fail over time (they are easy to replace though).
Got it, I think I'll start off by changing those old fluorescent fittings to LEDs. I've been wanting to get it done for a long time anyway, and maybe it's a good reason I do it once and for all. And from there I will work through the solutions proposed. Many thanks.
I had the same problem with a 120vac LED light flickering on occasionally when the digital switch controlling it was off. A solution that I found, was to place a 100k ohm-1/4 watt resistor across the LED light's 120 volt input. I believe the resistor bleeds off any accumulated charge coming from the switch that may build up across the LED, when it is off. It's been a long time ago that I did this. But it worked to keep the LED off. The power dissipation in the resistor calculates to 144 milliwatts when the light is on.
