We have been utilizing a Rsense value of 1mOhm in our product for an extended period. However, upon further investigation, we found that the power-off discharge current, registering at 800 uA, was not being detected with this Rsense setting.
As a solution, we transitioned to a 20mOhm Rsense, which effectively addressed the issue, enabling coulomb counting with an acceptable error margin. Despite this improvement, the BQ Studio does not display the instantaneous current due to it being below the of "1 mA".
Subsequently, we attempted to optimize further, reducing the power-off discharge current to a range of 400-550 uA. This was achieved by increasing one of the system's pull-up resistances and activating the gauge's sleep mode. Despite these adjustments, upon prolonged product inactivity, we observed no reduction in capacity.
To ensure that we did not regress into a dead band scenario, we conducted tests by adjusting the CC Gain and Capacity Gain by a factor of 8, confirming an instantaneous current of approximately 4 mA, thus verifying the ADC's sensitivity to low currents. We also experimented with a higher resistance (Rsns=30mOhm), albeit without success.
We have exhaustively attempted various configurations to resolve this issue:
- Modification of Deadband to "0"
- Adjustment of CC offset within the range of 0-200
- Alteration of Coulomb Counter offset samples from 64 to 150
- Exploration of CC Auto Offset functionality (Disabled and Enabled)
Each of the above configurations was tested individually, with the product left inactive for a day to monitor any capacity reduction. However, no deduction in capacity was observed.
We are seeking guidance on potential adjustments or solutions to enable the coulomb counter to accurately read such low discharge currents. Any insights or suggestions would be greatly appreciated.
