I'm trying to estimate the battery life for a system powered by a pair of LiPo batteries, in which the main loads are two BLDC motors. I estimated the battery life by dividing the battery capacity by the average current consumption of the two motors. I estimated the current consumption based on the current value sent by the motor drivers, in a variable named "motor torque current (Iq)". This method should give an overestimation, since I ignore all losses and other components in the system, but the number I'm getting is less than the battery life of the system in practice. I think the problem is coming from the current value I'm using, since it seems like that Iq is not the same as the current consumed by the motors, but I couldn't find any clear answers. Any help would be greatly appreciated.
The details of my system and calculations:
The two batteries are 3s LiPo (11.1 [V]) with a nominal capacity of 2200 mAh each. They are connected in series to get an output voltage of ~24 [V]. The two 330 [W] BLDC motors (nominal voltage 48 [V], nominal and stall currents: 6.5 and 12.5 [A]) and their embedded drivers are directly powered by the batteries. Full specs of the motor+driver module here. The motor drivers are commanded with an embedded computer, also powered by the batteries. I ran a test of about 10 min. and continuously logged the motor currents as returned by their drivers in the variable named "motor torque current (Iq)" in the driver's manual. I calculated the average value of these currents over the whole period, and added them up to get a total average current of 13.9 A. With this, the battery life is:
2.2 [Ah] / 13.9 [A] = 0.16 [h]
...or less than 10 min. But at the end of that testing scenario, the batteries were far from being fully depleted, and could have easily lasted for another 5-6 min. Are there any obvious mistakes in my calculations or procedure? My main guess is that the currents returned by the drivers are somehow different from the actual current consumed by the batteries because of different definitions of "current" in BLDC (as discussed here for example), but I couldn't find any clear information on this.
