-->
Page 1 of 2

Help with RCWL-0516 avoiding false positive triggers

PostPosted: Thu Nov 12, 2020 1:24 am
by docmattman
I'm using a bare eps8266 (running esphome). It's being powered by a USB wall charger at the moment. I have a RCWL-0516 microwave radar module attached to use for motion detection. It seems when the RCWL-0516 is the only thing connected to the chip, it functions properly (shows motion when it should and doesn't when it shouldn't). However, when I connect addressable LEDs (ws2812b) to the chip as well, the RCWL-0516 basically stops functioning. It will have small windows of time where it might work for a few seconds. But it pretty much always shows motion detected even when there isn't any.

Ultimately what I'm trying to accomplish is a nightlight for my daughter that turns on when it detects motion. So I'm trying to use the RCWL-0516 to detect movement, then turn on the LEDs. As mentioned, the motion detection seems to work fine as far as I can tell when the RCWL-0516 is the only thing connected to the esp8266 chip. Then when I attach the LEDs, it becomes completely unreliable.

Anyone have any suggestions or tips that I can try to get this working?

Re: Help with RCWL-0516 avoiding false positive triggers

PostPosted: Sat Nov 14, 2020 3:58 am
by schufti
with all three components it is important to decouple the supply rail. Connect them directly to the supply, not in chain, and have caps on each device. Shield the back of the RCWL against the esp, mount the LED outside the view of the RCWL.

Re: Help with RCWL-0516 avoiding false positive triggers

PostPosted: Wed Nov 18, 2020 7:20 pm
by StickyBurr
Having the same issue. It might be a manufacturing variability issue or some inband or harmonic interference. I did see markedly better results without an LED strip churning away on the same 5V rail so this suggests it's a power conditioning issue. However you will still get occasional false positives with these gadgets even on pure (battery) power, totally standalone. It's the nature of the design.

For a nightlight I would use a PIR sensor anyway. Works fine with humans (and pets) and never false-triggers. Cheap enough on amazon or alibaba. We have them all over our house. Don't forget the 10k pullup on the signal output. And BTW they don't work with the white freznel lens removed (I tried) and they also don't work behind 6mm plexiglass, despite findings to the contrary elsewhere.

Re: Help with RCWL-0516 avoiding false positive triggers

PostPosted: Thu Nov 19, 2020 2:27 am
by rudy
There is a spot for a resistor on the back of the board that will reduce the sensitivity to motion. Maybe that can also reduce the sensitivity to the leds.

This might be something to try in addition to the above posts.