So I have 5 Tuya repeaters, based on the recommendation of @aaiyar . For the most part they are working great, however I have 1 that kept falling off(using generic repeater driver w/ presence check). I've tried re-pairing it to no avail. As it turns out I've installed about 5 wall outlets with built in USB charging, and the repeater that kept falling off was powered via the outlet's USB A port via being directly inserted in the outlet's USB port. Chasing this theory that the interference was being caused by the outlet itself, I added a 3 foot cable to distance the Tuya repeater from the outlet's circuitry, still no joy. Next step was to move the repeater to another regular outlet without built in USB ports, this time using an assumed "high quality" Blackberry tiny charger block(the squarish black ones) to power the repeater versus using the power from the outlet with integrated USB charging ports. So the repeater has not fallen off in a week, when before it was every day or 2. The new location for the repeater is less than 5 foot from its original location.
So for you engineer types out there, is this actually possible that a badly designed charger circuit could interfere with Zigbee?
Thanks and a great thought but before my return period was up when I purchased the outlets I validated the claimed output, both 2.4amp and the high power USB PD outputs. Output checked out as specified
It is quite feasible that the USB output has high frequency noise. Is it from a reputable manufacturer that would conduct FCC testing?
Here is a 2012 reference showing an Apple charger. Notice how many components are labeled "snubber" or "filter" or even "shield". Any shortcuts in designing a flyback power supply could leave you with high-frequency transients.
I had one usb charger that would kill a powerline network. Took some fault finding for that one. Put it down to excessive high frequency noise, and threw it away!
The reference to 70kHz is the frequency of the transistor turning on and off (for that particular design). Each time that it turns on, current ramps up in the flyback transformer. When the transistor turns off, the current turns off suddenly. Below is an image that I stole from a quick search. It illustrates the transient spikes and oscillations that can occur (which are much faster than the circuit switching frequency).
FCC testing is fun, in that it teaches you that everything is an antenna ... especially when you don't want it to be. It has been 20 years since I have gone through it, but chasing a source of stray radio frequency radiation is the still the stuff of nightmares. UL generally only cares if something will cut/shock/burn you. They are indifferent to product functionality.
