running off-grid, you really do want to charge to 100% every day possible to start the night with full batteries. A 'correct' charge curve for LFP batteries is to "bulk charge" once per day to 56.2 volts (3.5 vpc, or whatever the battery mfg recommends), and then allow the system to "float" at 53.3-54.0 volts, which is full, but not over-charging. The inverter itself self-consumes a little energy all the time, so there will be a little current going in and out all afternoon. This is fine for Li, and will not reduce their life.
If you run your system "closed loop" using rj485 communication, the batteries do this internally by themselves and you see the (calculated) SOC%, but you cannot control the end-of-charge since you are presuming the internal battery charge curves set by mfg are optimal. You can set % cutoffs at the bottom end to protect for over-discharge, and charge from grid or generator.
If you run open loop communications, and manage by Voltage, intending to have the inverter-mppt charge controller run the show, then you *should* be able to set your desired bulk and float setpoints to whatever you want... but the end-of-charge "bulk" setpoint needs to be something over 55-56 volts to enable the last little bit of internal cell balancing at low current.
The BUG in the existing 6000xp firmware is that you have to set battery type=lead-acid and manage setpoints by voltage, not SOC% because calculations for SOC in lead-acid are inaccurate for LFP. The current BUG is that system ignores the "float" voltage setpoint when running off-grid open-loop Li batteries using the "lead-acid" battery type, and EQ day=0=EQ time as described in EG4 documentation.... and the voltage sticks at the BULK voltage instead of the FLOAT voltage.
EG4 assures us they are working on this, but they have known about it for several months, and I have not seen a patch release for firmware yet (11/1/2024)
