• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 161 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.

Original question:

BQ25628E: Detecting absent battery

BQ25628E: OCP gets unexpectedly activated

Martin
Prodigy 40 points
Part Number: BQ25628E

Tool/software:

Hi,

I am working on implementing BQ25628E in my device and I'm facing further issues - the device takes many attempts to start as the part is repetitively cut by BATFET for an unknown reason.

  • Device is turned on and runs fine for 1-10 seconds, then the power gets cut by the charger. The reset seems to occur at a random time following the start or sometimes not at all. The power is then restored within ~150 ms.
  • The device repeats the restart 0-10 times and then remains operational as normal. The longer the device has been turned off for, the more restart attempts occur.
  • There is no such issue when powered by USB, with or without the battery.
  • There are no appreciable spikes in the current (I have a 10 mOhm series sense resistor) and the load is ~0.5 A, which is well below the 6 A IBATFET_OCP.
  • There are no spikes or changes in the battery voltage, beyond a small increase in voltage when the load is disconnected by BATFET.
  • Based on the oscilloscope shot below, it seems this is the behaviour described in datasheet section 8.3.9.1.2, "slow, fixed-range OCP". That is a relatively long turn-off with a ~100 ms re-activation timer. At the same time, I am quite confident the current is nowhere near 6A.
  • There is ~400 µF on VSYS and 100 µF on VBAT. I've tested 50-400 µF for VSYS, all yielding identical results.
  • I am unable to get any flag/errors out as the power is cut before I do the I2C readout.
  • I set CHARGE CONTROL 3 to 0b01000000, so peak discharge 3A and external ILIM disabled.

Pink = VBAT, Yellow = VSYS

Thanks in advance for your help

Martin

  • 0
    TI__Mastermind 31500 points

    Hi Martin, 

    Thank you for the detailed explanation of the behavior you are observing. I have a couple follow-up questions. 

    1)What I2C mode (i.e. clock frequency) are you using in your system? 

    2)Are you still able to observe the batfet turnoff/reset in battery only operation when you keep IBAT_PK setting at default 12A? 

    Best Regards,

    Garrett  

  • 0 in reply to Garrett Kreger
    Prodigy 40 points

    Hi Garrett,

    Thanks for your response.

    1) I2C is 400 kHz, 3V3. I've attached an oscilloscope shot of an I2C transaction if that's helpful. I always do readback after writing registers and don't really see issues with the bus.

    2) 12A setting resulted in the same behaviour.

    Best

    Martin

  • 0 in reply to Martin
    TI__Mastermind 31500 points

    Hi Martin, 

    Thank you for your response. 

    Please allow me to refer you to section 8.5.1 Serial Interface in the BQ25628E datasheet. There is a requirement related to I2C communication when using the BQ25628E. 

    When in battery only mode operation if the I2C requirement is not met the BATFET turn off behavior you observe can occur. 

    Additionally, we are currently in the process of releasing a new revision of the datasheet where timing requirements are relaxed compared to the overly conservative requirements listed in the current revision. Therefore please refer to the timing requirements shown below. 

    Best Regards,

    Garrett

  • 0 in reply to Garrett Kreger
    Prodigy 40 points

    Hi Garrett,

    This resolved it, thanks!!

    I think a very unusual issue / behaviour and I would have never solved it without your advice. It might be worth strongly highlighting in the datasheet just how critical the I2C timing is and that it manifests in behaviour seeming unrelated to I2C.

    Best

    Martin

  • 0 in reply to Martin
    TI__Mastermind 31500 points

    Hi Martin, 

    Great to hear the modification resolved the observed behavior and thank you for the feedback on datasheet improvement. 

    Best Regards,

    Garrett