• State
  • +1 person also asked this people also asked this
  • Locked Locked
  • Replies 7 replies
  • Subscribers 64 subscribers
  • Views 223 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.

BQ41Z50: GPIO Mapping to ITStatus() causes high power consumption

Pat Kusbel
Prodigy 110 points
Part Number: BQ41Z50

Tool/software:

I'm trying to use the Flag Map to map ITStatus() to LEDCNTL_A.   As soon as I map ITStatus(), the current jumps up to ~800mA.   Some combinations of registers partially work, such as BatteryStatus(), and I can get the GPIO to reflect the DSG bit for BatteryStatus().   Setting Flag Map 1 to 0x8261 correctly drives the LEDCNTL_A gpio high.   But as soon as ITStatus() is selected as the source (0x8267), the current of the BMS jumps to 800mA. It seems to change also when the Polarity is toggled.  Almost like the wrong pins are being driven....  The document does say it is driving Pin 20, 21, and 22 in the Flag Map Setup section, but that can't be write as those pins are the VC pins for the cells.


Looks like a firmware bug with regards to Flag Mapping.   And a ton of documentation errors.   We need the FLAG_MAP to work, we are blocked on this at this time and can't ship until we have a solution for this.

  • 0
    TI__Mastermind 27321 points

    Hello,

    Can you please share the FW version you are currently working with? 
    If you share your gg file, it will display this info on the top left corner of the file.

  • 0 in reply to Jose Couso
    Prodigy 110 points 
    *
    * Device Number 4150
    * Firmware Version 0.01
    * Build Number 12
    * Order Number 0
    *


    Here's the GG file, this currently has the Flag Map turned off until this is resolved.    The Flag map we are trying to use is as above, 0x8267.   

    acprs_settings.gg.csv

  • 0 in reply to Pat Kusbel
    TI__Mastermind 27321 points

    Hello,

    The issue you are facing is a rare case. We have not seen this occur.

    Maybe we need more info. 

    Can you please share a picture of your current setup? 

    Can you please share more details about this issue that can help us to understand a bit more? 

  • 0 in reply to Jose Couso
    Prodigy 110 points

    Sure thing.  Note that I also duplicated this on the BQ41Z50 EVM.   Here's a picture of my setup with the EVM, the setup is a 4 cell pack connected to the BQ41Z60 EVM with a Keysight N6705C with an N6781 SMU measuring the current.   I'm doing low-side measurement so the current source is in the ground path between the ground lead of the battery pack and the ground lead of the EVM.  

    Load the GG file I gave earlier, and set the Chem ID to 0x2068 (that's the cells I'm using).

    Remove jumper J5 on the EVM so the LED doesn't turn on when the GPIO toggles so we don't see the current from the LED.  Observe the current through the pack to the EVM: for me it's ~100 -200 uA +/- with a 4 cell pack and the EVM 2400 hooked up.

    Then set Flag Map Setup 1 to 0xA207 as shown which maps ITStatus() RESET bit to LED_CNTLA.  Note that this is Open Drain configuration so no current is being sourced from the GPIO, so the GPIO is NOT the source of the increased current.  

    Observe the current, it jumps up to 800uA. 

    Now turn off the LED (set FLAG_EN to 0).  Observe the current does NOT decrease. It's stuck in this state forever.  This will break with ANY GPIO selected as the output GPIO from the flag map register, not just LED_CNTLA.  

    Now, with FLAG_EN still off, select RESET command to reset the BQ41Z50 and observe the current goes back to normal.  

    So basically enabling any flag map to any GPIO (including DISP) will cause the current to jump up by a factor 8x.  It stays stuck this way even if the flag map is disabled.   The only way to recover is disable the flag map and then reset the BQ41Z50 after the flag map is disabled.

     Here's my setup



  • 0 in reply to Pat Kusbel
    TI__Mastermind 27321 points

    Hello,

    A few questions. 
    1- Are you using RevA of bq41z50? 

    2- Please remove the J2-Jumper if installed. 
    If you programmed a FW while J2 installed, it can cause issues with the board. This is explained in Users Guide. 





    I would recommend ordering the Rev.B board which is available to order. Or replacing the IC. 

  • 0 in reply to Jose Couso
    Prodigy 110 points

    J2 is not on.


    THis is NOT an EVM issue, my real design does this as well, that pin is not even connected.   This happens when Flag Map is set to LED_CNTRL_A/b/c/disp.  This is an issue almost certainly with software.

    I was just duplicating this on the EVM to show that this is a software issue.   Please read the above that I wrote carefully.   It's almost certainly software since resetting the chip can recover it (software reset).

    The board is not damaged, the chip is not damage.  Please duplicate this yourself, it's very easy on your EVM.   Just enable the flag map, measure current and observe it increase.  I gave step by step details to duplicate this above on the EVM.

  • 0 in reply to Pat Kusbel
    TI__Mastermind 27321 points

    We are looking at this. Please allow some time.

    In the meantime, please try clearing FLAG_OD if set, the polarity for it is backwards in bq41z50 build 12.
    Clearing it makes the IO do Hi-z vs driven hi. 

    Note that we will be releasing build-13 soon with these fixes.