How to troubleshoot OpenDTU instances not receiving data ?

[mathieucarbou]

Hello,

I bought 3x ESP32 NodeMCU boards and 3x NRF24L01+ radio boards on Amazon (links provided).

I broke the usb connector of one ESP32 so I have flashed OpenDTU to the 2 other ones I have.

After having put back the config and flashing https://github.com/tbnobody/OpenDTU/commits/422d489, the 2 OpenDTU are really having issues receiving inverters data. I had the same issue with the first one also: I could only get signal at one specific place at my home, and it was really random. Sometimes I see an inverter, for about 5min, then the DTU is losing it. Same thing for the 2 new boards flashed.

I've tried different combinations with and without the power stabilising module.
And increased the power to max.
Both powered by a 5V 3A adapter.
DTUs are also 3-6m from the inverters, in the garage, without any thick wall.
I have tried with a longer antenna, more directional, but same issue.

I really don't know how to debug this... Any help would be appreciated.

I also don't know how to assert if the problem comes from the ESP32, radio board, antenna, etc.

Thanks for your help!

[smarthausleben]

I also had some trouble to get data from the inverters at the beginning. I could get it fixed by soldering an electrolytic capacitor directly over the power supply pins on the NRF24L01+.

I don’t remember in detail but I guess it worked on gain setting Minimum (-18db) adding a 10 µF capacitor. To get it work on Low (-12db) I had to add an other 100µF capacitor. The higher gain modes (-6 dBm) and (0 dBm) did never work. So power supply for the NRF2401+ was the key factor on my setup.

[smarthausleben]

Adding a 100 uF + changing the adapter and position helped a little, but not by much. It is still random. Sometimes I pick the inverters, sometimes not. I really don't know what to try next. These are supposed to work ~50m inside, I am less than 10m away from the farther away inverter.

I'm wondering if the 3 NRF24L01+ modules I got are really what they are supposed to be...

Just try to get it work in Minimum(-18 dBm) first. Having a small and big capacitor in parallel may improve the situation as well.

As I said, I just got it working with -18dBm and -12dBm using a PoE powered ESP32 PoE.

[mathieucarbou]

Thanks! I will try next if I see some issues. I was able to get things quite stable now since a few minutes with a dedicated 5V 3A adapter for Raspberry, a 100 uF as close as possible to the radio module and the DTU at a right place im the room, using 5 sec polling and maximum for PA level.
If I see other issues I will reduce the PA level and add a 10uF.
Thanks a lot for your help 👍

[smarthausleben]

Adding a 100 uF + changing the ...

We are dealing with high frequency parts too and if you have a 0.1 µF ceramic on hand, try this one in parallel to the 10 µF ceramic.

[mathieucarbou]

wait... you meant: 0.1 uF ceramic + 10 uF electrolytic + 100 uF electrolytic ? Or only 0.1 uF ceramic + 10 uF electrolytic ? Turns out I have plenty of 0.1 uF ceramic... I'm glad to have bought this Elegoo starter kit once lol!

[smarthausleben]

Just add an 0.1 µF ceramic for the higher frequency part.
I would first try it only with 10 µF electrolyte in parallel. Maybe this is enough. If not, then add the 100 µF electrolyte. That’s how I would try to improve it.