+- Smart Home Automation Forum (https://www.kincony.com/forum)
+-- Forum: Technical Support (https://www.kincony.com/forum/forumdisplay.php?fid=87)
+--- Forum: N30 (https://www.kincony.com/forum/forumdisplay.php?fid=97)
+--- Thread: N30 (KCS): energy counters of BL0910 #2 and #3 published as 0 kWh at boot (/showthread.php?tid=9860)
N30 (KCS): energy counters of BL0910 #2 and #3 published as 0 kWh at boot - dominique_manas@hotmail.com - 10-09-2026
Hello,
I found an issue with the energy counters on the N30 running KCS firmware (version: v3.27.2), integrated in Home Assistant through MQTT discovery.
What happens:
Each time the N30 restarts (firmware flash, power cut, reboot), the channels of BL0910 #2 and #3 (Energy 2 and Energy 3, all 10 channels each, plus energy_total) publish energy = 0 kWh for a moment, then the real value again a few seconds later.
BL0910 #1 (Energy 1) does not do this. Its channels only go unavailable during the restart and come back with the correct value, never 0.
The counters are not lost inside the N30: after the restart the values continue from where they were (e.g. 258.062 kWh -> 0 -> 258.062 kWh).
Why it is a problem:
Home Assistant treats energy sensors as increasing counters (state_class: total_increasing). A drop to 0 is seen as a meter reset, and the return to the real value is counted as new consumption.
This morning the N30 restarted 3 times after a firmware flash. Each channel of Energy 2 and Energy 3 was counted as consuming its whole counter value 4 times. For example, one channel with 258 kWh got 1032 kWh of phantom consumption added in one morning. I had to correct the statistics by hand.
Request:
Please do not publish energy values until the real counters have been read from the BL0910 chips at boot. Publishing nothing (or keeping the previous retained value) is fine. Publishing 0 is not.
It would also help to know why only chips #2 and #3 behave like this and not chip #1.
Thank you.