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.

BQ40Z50-R2: Hard to send back a right device name

Part Number: BQ40Z50-R2
Other Parts Discussed in Thread: BQ40Z50

Tool/software:

Dear Expert

+Issue description

‒Customer device connect the charger first, then connect the battery which is in shut down mode. Customer device report issue because the DeviceName from battery is not right.

  • The issue only happened on the BMS with BQ40Z50, test some samples that around 50% have this issue.
  • The issue only happen when the battery wakes up from shutdown mode to active mode.

‒When connect the battery again, is OK.

‒When wake up from sleep to active mode, it’s OK.

‒When connect battery first and then connect the charger, it’s OK.

+Possible root cause

‒The DeviceName is not right during initialization from wake up.

+Test (Test 9 times at BMS level.)

‒Test step:

  • Wake up the BMS from shut down mode and start to read the DeviceName continuously at the same time.

‒Result

  • The DeviceName is not right at the beginning of wake up, but it’s OK after around 140ms later.
  • ...
  • ‒Compare the IC spec found the sequence of data access is different, and we ran some test sequences and determined that BMS with BQ20Z95 takes 1.68s to initialize and the BMS with BQ40Z50 0.2s, It may be the key to the issue, if the BQ20Z95 has time to "stabilize" and communicate the Device Name with SMBus

    ‒Also, we suspect there might be some noise from power supply/charger during initialization when every item is plugged in together

    ‒The parameters for the IC is not suitable?

  • Hello Gabriel,

    Yes, the gauge will need some time to initialize when waking up from shutdown. You should poll the INIT flag to know when the gauge is done initializing. The INIT flag is in the BatteryStatus() register.

    Regards,

    Adrian

  • Hello Adrian,

    The thing is, in this case, there is a resistor on one pin of the batteries that is used by the customer device to send the battery is plugged in. As soon as the device senses the resistor it will start to ask for DeviceName continuously every ms. This means that it does not wait for the battery to have [INIT] flage set.

    thanks

    Best regards,

    João

  • Hello Joao,

    I see, well the device will need time to initialize so initial data received back from the device before this initialization is complete should be disregarded.

    Regards,

    Adrian

  • Hello Adrian,

    I understand what you are saying. And if it was a new development we would have done just that. The situation is that this product/device has been on the market for 10 years and with the BQ20Z95 we never faced such issue. And it impossible for my customer to update all their units on the field to work with the new battery (and the BQ40Z50).

    Is there anything we can do on the BQ side? We know with a longer initialization like in the 20Z95 the issue does not appear...

    thanks

    Regards,

    João

  • Hello Joao,

    I don't see any solution on the gauge side. This is how the gauge is designed to behave.

    Regards,

    Adrian

  • Hello Adrian,

    We have been reviewing internally your feedback and took the time to perform some additional testing, always comparing BQ40Z50 and BQ20Z95.

    First, we counted the time (from wake up from shutdown mode) to get the [INIT] flag to 1:

    - BQ20Z95: 2.606s

    - BQ40Z50: 1.077s

    This means that if our issue was just because of the INIT was not raised to 1 then we should have had the same issue with BQ20Z95, which is not the case.

    Then, we also counted the time (from wake up from shutdown mode) to get a good answer to the questions "what's the DeviceName":

    - BQ20Z95: 0.733s

    - BQ40Z50: 0.172s

    After that, we looked at what was replied by both BQ:

    - BQ20Z95: 

          - 1st trial - 17x in total: 15x answer 0000000000000000 then 1x FFFFFFFFFFFFFFFF then 1x Correct DeviceName

          - 2nd trial - 18x in total: 15x answer 0000000000000000 then 2x FFFFFFFFFFFFFFFF then 1x Correct DeviceName

    - BQ40Z50:

         - 1st trial - 2x in total: 1x FFFFFFFFFFFFFFFF then 1x Correct DeviceName

         - 2nd trial - 3x in total: 1x0000000000000000 then 1x0000B4B4B4B4B4B4 then 1x Correct DeviceName

         - 3rd to 5th - 3x in total: 1x0000000000000000 then 1x0000B4B4B4B4B4B4 then 1x Correct DeviceName

    As a reminder, the customer never had issues with BQ20Z95, so it means the replies with 0's and F's does not affect the identification of the device.

    Now what seems odd, is the fact that the BQ40Z50 send B4's when asked about DeviceName. And that is what our customer saw when the battery was detected as Invalid.

    Do you have an idea of what those B4's are returned by the 40Z50? It looks like our cause for issue, but can it be corrected?

    thanks a lot in advance for your patience and feedback.

    Best regards,

    João

  • Hello Joao,

    The B4 is the data return from the gauge when the data is invalid. 

    Regards,

    Adrian

  • Hello Adrian,

    Well noted and thank you for the answer.

    In our case we believe this is what is creating trouble on the customer side. Any chances this can be cancelled on the BQ and just have like in the BQ20Z95 empty answers?

    thanks

    Regards,

    João

  • Hello Joao,

    Unfortunately this will require a change to the FW. This is unlikely to happen as our FW is constantly busy supporting previous requests and creating new FW for our newer gauges.

    I noticed you are using the R2 version of the BQ40z50 FW, have you tried testing the R5 version to see if this behaves the same?

    Regards,

    Adrian

  • Hello Adrian,

    So, we finally managed to upgrade our BQ40Z50R2 to R5 and flashed a R5 compatible golden file in it. This morning we rna the same test and we got the following result:

    BQ answer correct DeviceName after 3 request.
    1. 0000000000000000
    2. 0000B40000000000
    3. Correct device name

    As you cna see on step 2 we still have a B4 appearing. So it seems we are stuck on the same step. Any other ideas? 

    thanks, I really appreciate your help.

    Happy holidays!

    João

  • Hello Joao,

    Unfortunately I do not have any other ideas as this is simply how the gauge FW behaves. I personally can not change that.

    Regards,

    Adrian