My system regularly reports a higher State of Charge than the battery charging would indicate. The drift is about 1% / day. When the FNDC/Victron indicate 100%, there is still charging to do to reach CPM.gtarolli wrote: ↑Mon Feb 01, 2021 9:38 am I am not 100% convinced that return amps measures net, I have mine set up at 32 amps (400 ah batteries) and still it doesn't trigger sometimes even when I've completed absorb. It doesn't bother me too much as its just a notification, it doesn't really control anything. What is disturbing is when SOC jumps to 100% from the low 90's due to the net amps reaching 0, and I can still see the batteries are not full. I don't have charge termination control enabled, so the system stays in bulk mode and continues charging until Absorb is done. The net amps on the status page seems to be derated by the charge factor/efficiency, but the net amps on the meters page does not. It seems as if the non-derated net amps is used?
I'm convinced the Corrected Net Battery figure is not precise. If for no other reason, Puekert's Constant is fixed for Lead Acid batteries on the FNDC. Changing Puekert's on the Victron to the LiFePO4 constant doesn't seem to make a material difference, however. This may be an artifact of my generally low discharge rate.
Peukert's not withstanding, Corrected Net Battery remains elusive to rationalize, so I don't really bother. This is another reason I choose the battery physics at full charge as my benchmark.
For me, using the 56.4V termination is like clockwork. I did some experimenting when I set up the Battery Charged parameters. I found that, at full charge, with the Charged Return Amps (Tail Amps in Victron-speak) set to the Phi specified 2% of C20 (9A in my case), the instantaneous charge current dives asymtotically to zero amps within the 1 minute Charged Time. I didn't care for that behavior, so "juked" the Charged Return Amps, setting it as 24A. Now, with the Charged Voltage = 64V, when the charging current hits 24A, the Charged Time counter starts. At the end of 1 minute, the charge current declines to 9A, and since Charged Time is met, the Charged Parameters Met flag is set and charging terminates.
That's as "dialed-in" as I can get it, and it works for me.