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.

BQ25890: DPM Issue - Maximum Input Current Detected

Part Number: BQ25890

Dear experts:

If the input voltage VBUS stays below DPM threshold for a while. REG13 IDPM_LIM will fall down to minimum value 100mA. When the input voltage ramps up again. Sometimes we can find "Maximum Input Current Detected" in REG14 bit 6.

When it happened, IDPM_LIM would be locked at 100mA, unless the battery is depleted, or we issue a system reset.

Anyway, we duplicate this problem by using power supply. Maybe the ramp-up speed is too fast. I just want to confirm, this scenario is correct. IDPM_LIM will be locked at 100mA after "Maximum Input Current Detected".

Thanks for replying!

  • Hi Charles, 

    Thank you for reaching out via E2E. The behavior you are observing is due to the input current optimizer (ICO) feature described in section 9.2.4 on page 22 of the BQ25890 datasheet. 

    When your VBUS input is measured below the VINDPM threshold the IC interprets that the adapter is being overloaded and therefore sets the input current limit to the minimum value in an attempt to get VBUS to rise back to a normal level.  You can overwrite the 100mA limit by disabling the ICO (set ICO_EN = 0 in REG02[4]) and then setting a new input current limit in REG00. 

    If you do not want the input current limit being set to 100mA due to the ICO algorithm you are free to have ICO disabled at all times and control IINLIM via your MCU host. 

    Please let me know if you have any further questions. 

    Best Regards,

    Garrett 

  • Dear Garrett:

    Do you mean, the IDPM_LIM value would not go back after ICO changes it? Or the ramp-up speed is too fast to response for ICO?

    The nominal input is 12V and the DMP threshold is set to 6V. There is a Schottky diode and some MOSFETs for Power Mux circuits at the DC input. When the DC voltage drops to 6V, the VBUS of bq25890 is below 6V actually.

    The input voltage goes back immediately after touching DPM threshold. It seems to be fine. Only if the voltage stays below DPM threshold for a few seconds (maybe 3 seconds or longer), it happens. So, is there any active time for ICO?

    Any engineering change is not necessary. The threshold is far away minimum input (10V - defined by customers) after all. Our customers just did a boring test and found it. If ICO algorithm is the root cause, then it’s not a big issue. We just need an explanation.

    Appreciated

    Charles Lai

  • Dear Garrett:

    The only thing we need to know is that, "Why the current limit would not go back to original settings?".

    1.  The current limit would not go back after ICO changes it.
    2. “Maximum Input Current Detected” is found.

    DPM controll should be dynamic and automatic.

    Best regards,

    Charles Lai

  • Hi Charles, 

    Firstly in general the IINDPM setting is not dynamic. Other than the ICO ability to change it after D+/D- detection initially sets IINDPM it does not change. For example if ICO is disabled D+/D- detection is the only event which will change IINDPM register setting.

    VBUS voltage being below VINDPM level for a few seconds is long enough for the ICO to interpret that the current IINDPM setting is crashing the adapter, hence it reduces the setting. 

    1) Correct after the ICO reduces the input current limit it will not automatically go back to initial setting when VBUS voltage rises to 12V. To increase the current limit you can force the ICO algorithm to run again after VBUS has recovered by setting REG09[7] = 1 or as I previously mentioned disable the ICO and manually set IINDPM. 

    2) The "Maximum input current detected" status bit is just an indicator that the ICO algorithm has finished and set an input current limit.  

    Best Regards,

    Garrett 

  • Dear Garrett:
    Sorry! I was confused by the wording of DPM ("Dynamic" Power Management). Your answer is really helpful to us. I think, the explanation would be enough for our customers.

    Thanks a lot!

    Charles Lai