As promised (to nobody....), here the results from my first charge/discharge cycles.
At around 50% state of charge, I measure between 35 milliOhm and 55 milliOhm internal resistance per cell. This from observing voltage drops of 60 to 90 mV when a discharge current of about 0.8 A is switched on, for the whole 13s6p battery pack.
At the knee at ~42.5V, internal resistance was measured once at 100 and once at 112 milliOhm per cell.
At 39V, I measured 130 milliOhm per cell.
The pack seems to have the full advertised charge capacity. I actually do measure about 20.5 Ah for both charge and discharge, using the range between 3.0V and 4.2V per cell.
Update: Clearly visible in the bottom graph is the voltage difference between charge and discharge.
Stored energy is 1008 resp. 1010 Wh charging, and 993 Wh for the discharge. The loss between energy in and energy out is smaller than I expected.
I do not know how precise my clamp-on current meter is. I do know that the displayed current depends on the clamping pressure, with about 5% increase when forcing the clamp shut manually, in addition to the spring force. All measurements are for the clamp forced shut.
The initial charge curves, voltage and charge vs time in minutes, charging with a nominal 2A 54.6V charger, actual charging current during the constant current phase measured at about 1.85A. Close to 54.6V the charger tapers the current off, as it should.
Voltage vs charge for
a) the initial discharge from 43.5V (as delivered) to 39V
a) charge from 39V to 54.6V
b) discharge from 54.6V to 39V
c) 2nd charge from 39V to 54.3 V:
d) 2nd discharge from 54.4V to 39V (
added):
The second charge curve pretty much exactly follows the first one. The wiggle near full charge (top left; the charge curves are plotted backwards, essentially plotting voltage vs 'charge left to go until full charge') is due to interrupting the charging overnight, and resuming in the morning; there was a 0.4V voltage offset at that point.
The second discharge shows 0.55Ah less charge and 60Wh less energy.