Team,
I had a question regarding the BQ27510-G1 Fuel Gauge.
We have recently had a fuel gauge greatly underestimate battery capacity. This was observed in an event where the fuel gauge made a large jump in ABC (53% to 24% in 10 minutes). I want to note that our current Quit Relax Time is set to 30 seconds, while our Discharge Current Threshold is currently set to 60 mA. At the moment, the only viable explanation for the large jump in ABC seems to be that the fuel gauge had taken an OCV reading possibly during a brief (1 second or less) spike in battery current consumption (LCD + on board radio + motor movement), causing this reading to take place during a brief drop in battery voltage. This OCV reading created an offset which greatly underestimated the remaining battery capacity for the remainder of its use. At the end of this use case, the fuel gauge was later found to be reporting 75% remaining capacity when the battery voltage was already at 4.2V. This fuel gauge clearly should have been reporting 100% ABC.
There is a statement in the application note for the fuel gauge that questions the viability of this assumption. This statement reads (see pg 32 of http://www.ti.com/lit/an/slua458/slua458.pdf ): After 30 minutes in relaxation mode, bq27500 starts checking if the dV/dt < 4 µV/s requirement for OCV readings is satisfied.
Unless my understanding is incorrect, this statement seems to imply that an invalid OCV reading as described by the siutation above is not possible due to the dV/dt < 4uV/s requirement for OCV readings. Would you be able to verify if the scenario described in blue could be explained by an invalid OCV reading and if so, would you be able to reconcile this with the statement found from the fuel gauge app note?
Feel free to let me know if you require any additional information.
Regards,
John