Hi all,

Should any one need, the fix mentioned in this page actually works.

Thanks,

No, actually I use the RPi just as power source. The gateway connects to the hub over WiFi.

The RPi is good (the original one). I will try to move to an independent power source just to test.

Thanks for the reply again.

Guys,

Just quick update to help others that may find this topic relevant: problem is solved. Problem was the step-up (probably low quality). Sensor is working well and with very low battery usage (no change in 5 days).

Thanks for the comments.

Problem solved: I was using as --my-mqtt-client-id the same name as my previous broker.

Thanks for your help all!

I haven’t moved the sensor to a weatherproof case (still figuring out the new design), but I connected it to an external power source.

The circuit is operating perfectly.

If humidity or something alike had damaged it, it should be not working, right?

Although I think it is important to use the weatherproof case, I feel something else is happening here.

@skywatch: where I live in BR temperature ranges between 5-36C. I can see some fluctuations in battery V when temp changes, but usually around 0.05V

Thanks!

Thanks for the hint @mfalkvidd

After much research (mainly here and here), it does look like RFM69+RPI+Mysensors 2.3.x don't like each other very much.

However, I updated my RPI code to the latest version (at RPI, move to MySensors directory, then "git pull", then make+install gateway again) and the NACK problem was completely solved. I also had many NACK during presentation, which all went away. Problem solved and it seems that the radio is performing faster (probably less time waiting ACK).

I also tried to address the low RSSI at gateway (again, it seems a recurring problem for RFM69+RPI) using this code, but it didn't "make".

Would you have any insight on the RSSI issue as well? It seems that you participated in some part of the discussion (here :wink: ).

Thanks!

Hi all,

Should anyone face the same problem, I found the root cause: brownout threshold. I burned a new bootloader (Optiboot 8.0) without such trigger and the node has been working with used batteries (~2.7V) since May.

Hope this helps.