After reading all that posts (thanks HeyImAlex!) and lot of other discussions on net, I’ve just finished STABLE WORKING prototype of ESP8266 and RWCL-0516 device.
But first let me share my experience with RWCL-0516 in short.
Actually I have 4 of it working in my house, and all of them works perfect, the oldest one for over 2 years now. And I’m really happy with the doppler radars instead of PIR’s - the sensitivity, working distance and shape of the area, and reliability is far beyond any PIR I’ve tested.
But my current devices are build on nrf24l01+ as transport, with atmega328 MCU and MySensors framework. Although nrf24l01 uses 2,4 Ghz, it is completely different transmission than WiFi, with very narrow band and low power emission.
Few weeks ago I’ve decided to build two more devices for my ceiling lighting on stairways, but this time using WeMos D1 mini clones with ESPHome, instead of Arduino with MySensors.
For the beginning there was a disaster! There were so much of false positive signals that is was completely useless. So I’ve started to search over internet, and found mostly lot of info regarding power filtering.
Long story short - nothing worked good. Tried pi filters, decoupling capacitors, ferrites, powering from 3,3V, 5V, 12V, not solving the problem. Because the problems has nothing to do with power! Now I’m powering RWCL-0516 directly from 3,3V GPIO pin (to turn it off when I don’t need it) without ANY additional elements, and it works all fine.
So I totally agree that the real problem with ESP and RWCL-0516 is the radio interferences. Once focused on radio, I’ve started to get more and more promising effects.
Now, after few more days of designing and tests, I’ve come to the final, working solution.
First off all, I’ve modified ESP antenna, replacing it with standard dipole (5 min. of work).
The second part is the physical placement of the two modules. When you look at radio propagation of dipole antenna, you will see that least gain is straight in ilne of dipole. So I’ve placed RWCL in line with antenna “arms”, in distance of about 10 cm.
That completely eliminated false positives inducted by WiFi transmission.
More of that, the dipole antenna added between 5 to 10 dB to WiFi signal of ESP!
So I’m hardly testing this setup now, before designing and printing final enclosure but it looks very stable so far.
Hope my experience will help.
PS: I’ve always using additional 1 Mohm R-GAIN resistor on RWCL board.