I didn't top balance my packs before installing

[Saladin]

I read somewhere that it is balancing in that time.

Interesting...perhaps, but that'd be a VERY aggressive balance current and the fact it is only active at high SOC makes me think it's a resistor array which are not that. It might be both. A setpoint triggered two stage charger with a secondary balance circuit.

[Andy]

@Tinbum I presume has been running his more than me without troubles, so I shall push on. I've only had mine on a few nights as I was unhappy with the wire that came with the Anderson connectors and the cheap amazon fuse. It was getting hotter and the other day was 78º. I've made my own 10mm wire and got a proper fuse in on a new lynx power in board. No point skimping early I've found!

Can anyone recommend some cheap controllable DC relays? I can command the current before breaking them so there should be no arc damage.

[MrPablo]

My understanding of the Pylontech systems is that the coloumb counting is not fantastically accurate (like many BMS incorporated shunts).
As a result, the SOC can drift overtime, so the Pylontech batteries fudge the numbers a little.
When charging, as the SOC reaches 89% the BMS will stop incrementing the SOC until a certain voltage threshold is met.
That makes sense as the cells will be heading into the knee portion of the voltage curve, where voltage is a more reliable indicator of SOC.
Once the voltage threshold is met, the reported SOC will rapidly increment or jump upwards.

The same is seen at the low end, with the SOC pausing at 11 or 12%, dropping only once the voltage is past a low threshold.

The pausing can be more prominent when there are multiple batteries, or if charging current is low.

[Andy]

Elteks are probably diode protected outputs because they're designed to be paralleled.

You are right, they have OR-ing protection. (google = diode)

https://www.tempestns.com/batteriesdcpower/