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.

BQ27530-G1: Query on Fuel Gauge Learning cycle in BQSTUDIO(VOK and RUPDIS bit setting issue)

Part Number: BQ27530-G1
Other Parts Discussed in Thread: BQSTUDIO

Hi TI Team,

We have started learning cycle for BQ27530 in BQStudio.

As per TI's learning cycle reference document(https://www.ti.com/lit/an/slua903/slua903.pdf), at the end of initial discharge and relaxation period RUPDIS should be cleared. But, when we tried RUPDIS bit is not cleared.

At the same time, once after initiating charging VOK bit also was not set

I have attached the logs for reference. Please suggest us what could be the reason for this issue

Discharge_and_relaxation_log.csv

Warm Regards,

Mythili B

  • Hi Mythili,

    Is it possible to receive the .gg file of this gauge?

    From what I can see from the log file, the gauge is entering sleep mode at 31542.6 seconds, making the voltage almost drop to 0V. The .gg file will allow us to see what thresholds are causing this to enter sleep mode.

    Regards,

    Anthony Baldino

  • Hi Anthony,

    I have attached .gg file for your reference

    7028.GG_file.zip

    Warm Regards,

    Mythili B

  • Hi Mythili,

    Thank you for sending the .gg file.

    Regarding the log file, can you please confirm if there something occurring that would make the voltage reading go to 0V such as the cells being disconnected?

    Regards,

    Anthony Baldino

  • Hi Anthony,

    Nowhere I have seen 0V in log file. Only current it is 0mA because it is in relaxation time

    Fuel Gauge is entering into sleep mode during relaxation time because Sleep current threshold is set as 10mA.

    We have cleared OPCONFIG[SLEEP] bit after that fuel gauge has not entered into sleep mode, Is it ok to proceed with clearing the SLEEP bit?

    While calibration after clearing SLEEP bit, VOK and RUPDIS bits are updating properly

    Warm Regards,

    Mythili B

  • Hi Mythili,

    Below is the voltage graph of the sent learning cycle where the voltage is dropping to around 5mV:

    During this time, there is still communication occurring between the gauge and the host, so we do not believe that this is being caused by the gauge is entering sleep mode. Can you please tell us if your battery pack has any attached protections on it? This discharge cycle could be triggering one of them and sending the battery cell into this state.

    Regards,

    Anthony Baldino

  • Hi Anthony,

    Yes, battery has over discharge protection and it will not discharge beyond 3V

    Still I2C communication is happening because fuel gauge is getting power from battery charger and we have disabled charging since it is in relaxation period

    In learning cycle process, we have cleared OPCONFIG[SLEEP] bit after that fuel gauge has not entered into sleep mode

     

    We have done Discharge-Relaxation-Charge-Relaxation-Discharge cycle. At the final discharge cycle, RDIS bit is set in between the discharging process and second QMAX value is not updated. But when we proceeded with another charging cycle QMAX value updated and UPDATE_STATUS also updated to 0x02

    So, Can we consider that the learning cycle is completed. Please suggest us what could be the reason for QMAX update failure after discharge cycle.

    I have attached logs and GG files for your reference.

    Learning_Cycle_20-01-2024.zip

    Warm Regards,

    Mythili B

  • Hi Mythili,

    Is there a way to disable the over discharge protection on the battery for the learning cycle?

    The voltage drop being caused by this protection is most likely what is causing the problems with this learning cycle, since the OCV reading during relax (when the read voltage is 0V) will not be correct. This will also effect the algorithm and the gauging measurements will not be accurate if used after the learning cycle.

    Regards,

    Anthony Baldino

  • Hi Anthony,

    Thanks for your response, now QMAX has updated properly and the UPDATE_STATUS is also updated to 0x02.

    As per the reference document, once UPDATE_STATUS is updated to 0x02 IT_ENABLE will be disabled. But in our application, IT_ENABLE is not disabled automatically

    How to proceed further for generating the golden image?

    Will QMAX be updated even after flashing Golden file into the fuel gauge?

    Please provide your inputs on the above Queries

    Warm Regards,

    Mythili B

  • Hi Mythili,

    When the Update Status is changed to 0x02, how long does it take for IT_ENABLE to be disabled?

    How to proceed further for generating the golden image?

    To generate the golden image file (.srec), please go to the bqStudio Golden Image tab and click Create Image:

    Will QMAX be updated even after flashing Golden file into the fuel gauge?

    Yes, Qmax will be updated after flashing the golden image if IT_ENABLE is set. More information about how this is calculated during application is below:

    https://e2echina.ti.com/cfs-file/__key/telligent-evolution-components-attachments/00-24-00-00-00-00-02-91/Theory-and-Implementation-of-Impedance-Track-Battery-Fuel_2D00_Gauging-Algorithm.pdf 

    Regards,

    Anthony Baldino

  • Hi Anthony,

    Once after Update Status changed to 0x02, IT_ENABLE is not disabled at all, Do we need to disable it manually before generating golden file?

    Once after flashing the golden image to another fuel gauge, IT_ENABLE should be enabled or disabled before sealing the flash?

    Will Qmax be updated even after sealing the flash?

    Also, please let us know why the difference between Full Charge Capacity True and Full Charge Capacity Smooth is more sometimes and exactly the values are matching sometimes, Is this an expected one?

    Warm Regards,

    Mythili B

  • Hi Mythili,

    When the Update Status is changed to 0x02, this should disable IT_ENABLE automatically.

    Once the golden Image is flashed, IT_ENABLE should be enabled before sealing the flash.

    Qmax will be updated after sealing the flash.

    The difference between FCC True and FCC Smoothing is whether the [SmoothEn] bit is set in OpConfigC. When there are drastic temperature changes or changes in current rate, there can be jumps in the calculation of the FCC. When SmoothEn is set, the gauge will account for these jumps and calculate accordingly. FCC Smoothing is the calculated FCC value with SmoothEn set, while FCC True is the FCC value without SmoothEn set.

    Regards,

    Anthony Baldino