• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 106 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ27421-G1: observed the fuel gauge value drifts over time

P Shashank Shetty
Prodigy 10 points
Part Number: BQ27421-G1

Tool/software:

Hi,
               We need support on BQ27421-G1. We are using this fuel gauge ic in one of our products which is in its final testing/pre-production state!  and we have observed the value drifts over time.
We observed in 2 boards, probably 3months+ without any battery changes
Board 1:battery which was 3.60V shows 80% and after the power cycle and reload of the golden file it shows 59%
Board2: battery which was 3.45V shows 50%. and after the power cycle and reload of the golden file it shows 2%
 
In code we update the golden file only if we detect a complete power cycle where the parameters inside BQ27421 goes back to defauts.
We check full_charge_cap_mah and see if it's below 1500(the default value is 1300ish). If so, we update the chip with the golden file. It's not ideal to update the golden file in every controller power cycle as the controller can restart multiple times a day.
The reading is not reliable! 
Are we missing some procedure here?
  • 0
    TI__Mastermind 24931 points

    Hello, 

    Are you saving the learned parameters onto your host MCU as the battery is cycled? Otherwise when you reprogram the initial golden image onto the gauge, you are programming the parameters for a brand new battery not a deeply cycled one, so it will take a cycle for the accuracy to converge as the gauge relearns the parameters specific for the battery. 

    Regards, 

    Jonny.  

  • 0 in reply to Jonny
    Prodigy 10 points

    Hello,

    We are currently not saving the learned parameters to our host MCU, Do we have to save it to our host MCU or does it happen within the fuel gauge?

    Why should we save it to MCU?

    we have a provision for charging the battery onboard in our PCB so it never goes out of power (we only reload the golden file onto the fuel gauge if we have a power-on reset)

    anyways we don't see the "learning" part happening thus we are getting non-reliable readings of SOC

    please let us know if you need more info 

  • 0 in reply to P Shashank Shetty
    TI__Mastermind 24931 points

    Hello, 

    You are not, for example seeing Qmax Cell 0 update, or any of the resistance values?

    Regards, 

    Jonny.