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.

BQ27510EVM: Learning cycle fails (QEN bit always low )

Part Number: BQ27510EVM
Other Parts Discussed in Thread: BQ27510, BQSTUDIO

Dear All,

I tried to operate the learning cycle for bq27510 using the bq27510EVM and bqstudio, I followed these steps:

  1. Send IT_ENABLE cmd. 
  2. Send RESET cmd.

NOTE: the QEN bit in control status is still low.

  1. Let the battery discharge with rate C/10 until it is empty.
  2. Let the battery rest for 6 hr.
  3. Send IT_ENABLE cmd. 
  4. Send RESET cmd.

NOTE: the QEN bit in control status is also still low.

start charging (C/2) till the current drop to zero and the voltage raise to charger voltage but the VOK bit in control status is also still low.

  • Waiting for any replies for this problem

  • when I send REST cmd, DSG bit in flag register set, not the RUP_DIS bit in control status register as mentioned in SLUA903 3.2.1 (is that normal?)

    in the same document 3.2 mentions enabling IT after emptying the battery but in 5 Conclusion mention emptying the batter then enabling IT (which one is the correct?)

  • Hello Khaled,

    It looks like the charge section of the learning cycle was unsuccessful, and the gauge failed to learn Qmax. Unfortunately, it is very hard for me to figure out what happened without a log file. Can you log another attempt at the learning cycle (just charging from empty after resting should be okay) and share it? Could you also share the configuration/values you set for charge termination especially?

    QEN should be forced to 1 after sending the IT_ENABLE command. You should also avoid sending the reset command too much, as I think that is what caused the failure here. Once the battery is discharged and rested for 5-6 hours, then you can send IT_ENABLE and do not send reset. Confirm that the bits are okay, and then start to charge. When you charge, look for FC to set, this indicates a valid full charge termination.  

    As for the other questions, I don't see why the DSG bit would go high after sending a reset command. DSG should only go high when the gauge sees a discharge is occurring (negative current). Again, I think you should avoid sending reset commands during the learning cycle. As for the discrepancy with IT_ENABLE, it has to do with the different purposes for the discharges. The first discharge is simply to empty the battery to start the process, so we disable IT so that it doesn't update impedance during this. The second discharge is to learn impedance, so we want to have IT_ENABLE on for this one. 

    thanks,

    Alex M.

  • I follow these steps:

    1. discharge the battery to empty 
    2. battery relaxed  for 6 hours
    3. send command IT_ENABLE  (Data Memory -> Gas Gauging -> State -> IT Enable still equal 01)
    4. plugin the charger
    5. unplug the charger when FC bit in flags register set 
    6. let the battery relax for 3 hours

    NOTE: VOK bit and QEN bit was low and never been high so far

    I guess the issue may be at the begging of the charging procedure because the VOK bit was low. IT Enable in data memory change from 00 to 01 after sending IT_ENABLE command but the  QEN  in control staus register was always LOW, I don't know why??

     LearnCycle.log

  • Hello Khaled,

    Thanks for providing the logs. I think the QEN issue is causing the problems like you suggest. To be clear, when you send IT_ENABLE you see VOK go high but QEN has never once gone high? You aren't resetting after sending the command either? That is definitely strange. To confirm, can you do this procedure?

    1. Reset everything
    2. export .GG file (before)
    3. Start log
    4. discharge and relax >5Hr
    5. send IT_ENABLE
    6. Charge again
    7. Allow relax again
    8. End Log
    9. Export .GG (after)

    Sorry to ask for more data, but I just don't see why QEN should stay low. You may also want to try while in UNSEAL FULL ACCESS. I don't think that should do anything here but it is worth trying if possible. 

    Also, you may want to address this issue:

    From your log, sorry the labels aren't great. Why is there a spike in voltage near the end that is nearly 5V? I see that your charge terminates okay, but if the battery is actually reaching 5V, there is a chance for damage or even fire. Is it just something to do with the charger's termination?

    Anyways, best of luck.

    Thanks,

    Alex M.