• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 184 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.

BQ78350-R1A: AFEconfig randomly returns bad data

Jeffrey Wright
Expert 2875 points
Part Number: BQ78350-R1A
Other Parts Discussed in Thread: BQ78350

Hello;

While testing, I ask for AFEconfig (0x59) at each ALERT. Randomly, I observe that the value for CELLBAL1 is invalid, listing adjacent cells being balanced (e.g. 0x03, 0x06, 0x07, 0x0F, etc.). Is this a known issue? If not, how could this be occurring?

Thanks

  • 0
    TI__Guru* 81034 points

    Hi Jeffrey,

    This seems very strange. The BQ78350 should not balance on adjacent cells. Are you sure you are reading the correct AFE register? Have you confirmed that balancing is actually occurring (check with an oscilloscope on the VCx pins)?

    Regards,

    Matt

  • 0 in reply to Matt Sunna
    Expert 2875 points

    Yes, I'm reading CELLBAL1, and it reads correctly 99.9% of the time...but it randomly (seemingly) has bad data...which I've implemented a screen for to block it from the GUI, since it confuses some tech's while testing. This is being observed on a simulator which can simulate imbalanced cells, and it is confirmed that balancing is indeed taking place. But then, every couple minutes, the data is bad from the gauge.

    This appears similar to the issue with the PF bit being incorrect when first reading the OperationStatus() after a permanent fail, where I must read it a few times with delays until it is set....while it could be my driver, I doubt it, since the first read of OperationStatus() correctly has the XDSG and XCHG bits set.  I note that the PF, SS, and SAFE bits act the same way...I must read the register several times before they are set.

    Question: How long should I wait for the gauge to perform all the actions after a permanent fail (i.e. blackbox, etc.)???? I'm just wondering because I put it in ship mode after a PF event (which also requires about 3-4 seconds to respond to the shutdown command), and so am wondering if I'm sending the shutdown command too soon

  • 0 in reply to Jeffrey Wright
    TI__Guru* 81034 points

    Hi Jeffrey,

    I haven't seen this issue reported before. 

    With some bits like PF, XDSG, XCHG, there is a sequence so not all bits are changed simultaneously. When a Shutdown command is sent, the device will finish all of its actions before it tells the AFE to turn off. For example, it will go write lifetime data to data flash as part of the Shutdown sequence.

    Matt