Other Parts Discussed in Thread: GPCCHEM
Dear TI support team,
I have an application that uses the bq34z100-G1 for fuel gauging a LiFePO4 battery pack.
The proper CHEM ID has been programmed and a successful learning cycle has been completed.
However, the current draw of the pack under real world operating conditions vary from the current values used for the learning cycle.
They are much greater and there is often not much of a rest period.
I'm noticing that the fuel gauge tends to drop from 10% down to 0% very quickly at the End of Discharge.
I know LiFePo4 has a very flat discharge curve followed by the sharp drop off.
Also, on the subsequent charge cycle, the fuel gauge will report 100% too quickly.
Would it makes sense to perform a learning cycle using the current draw conditions that are more in tune with the real world values?
I'm hoping this might help to prevent sudden jumps down to 0% and the charge showing 100% prior to the pack actually being fully charged.
Thanks much and best regards,
Kevin