• State TI Thinks Resolved
  • Locked Locked
  • Replies 23 replies
  • Answers 4 answers
  • Subscribers 63 subscribers
  • Views 483 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.

BQ34Z100-G1: SOC Jump to 0%

Yoshinobu Kagawa
Expert 2030 points
Part Number: BQ34Z100-G1
Other Parts Discussed in Thread: EV2400

About 5 cases, the SOC suddenly became 0%.
At least one unit has confirmed the following phenomena.
・ When the remaining power was 90%, the power was turned off (the remaining power monitoring IC was always energized), and after 20 minutes the power was turned on and the remaining power was 0%.
・ The voltage of the battery pack was about 4V (about 1.33V for a single cell). The actual remaining capacity seems to be close to 90%.
・ When the battery pack was refreshed (completely discharged → fully charged) while connected to the equipment, it reached 100%.


It seems that the remaining battery level is estimated to be returned by refreshing the battery pack in 2 out of 3 devices that have 0% remaining battery.
There is no user action such as removing the battery pack.
Although it has been operating stably until now, it is the first time that it has stuck to 0% suddenly, but I am confused because it occurred on multiple units. Frequent occurrence by the user is unacceptable.
Is this possible?

As a user's usage, since it is a cycle that it is fully charged after using it for several hours (converted to a used capacity of about 10%), most of it is 100% → 90% (estimated) → 100% → 90% → 100% → 70% ( I sometimes use it for a long time) → I think that the charge and discharge is rather shallow, such as 100%.

I wonder if some part of the battery characteristics could not be learned, or it became 0% because the battery was discharged and the abnormal area of ​​Golden image was used for the first time.

Best regards,
Kagawa

  • 0
    TI__Guru 52115 points

    Hi Kagawa,

    Which kind of battery are you using?  It seems like the battery pack is always connected to the bq34z100-g1, correct?

    Are you able to connect EV2400 to the gauge device and create a bqStuio log?  I would need to check the bq34z100-g1's run-time information to figure out the root cause?  Thanks.  

    Andy

  • 0 in reply to Andy Liu (BMS)
    Expert 2030 points

    Hi Andy-san,

    Customer is using NiMH(3-Cell) battery.
    The bq34z100-G1 is outside the battery pack, but it's always connected.
    Now, we are asking the customer to create the log.

    Best regards,
    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points

    Hi Kagawa,

    Let me know when the log file is ready.  Thanks.

    Andy

  • 0 in reply to Andy Liu (BMS)
    Expert 2030 points

    Hi Andy-san,

    I sent the log file and golden Image files to you.

    Best regards,
    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points

    Hi Kagawa,

    I have checked your log file.  It doesn't have enough data. Bascially,  I cannot see the data before the SOC became 0.  Can you provide a good log file with enough details? 

    Andy

  • 0 in reply to Andy Liu (BMS)
    Expert 2030 points

    Hi Andy-san,

    As a thing that can be done immediately, I tried removing the battery from the target device for a long time, completely removing the charge of the remaining amount monitoring IC etc. and resetting it.
    Even this is stuck at 0%. (I have already sent you a log file.)

    So I change the question.
    If it is not updated as it is 0% instead of why it will be 0%, will there be any progress in the current log etc.?

    Best regards,
    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points

    I see the Update Staus is still 05 in your gg and log file,  have you ever finished the learning cycle?  If the learning cycle is done, the Update Staut should be 06.

    Andy

  • 0 in reply to Andy Liu (BMS)
    Expert 2030 points

    Hi Andy-san,

    Yes. It's in production after the learning cycle,
    This phenomenon is occurring in the market.
    Why is SOC = 0% even when fully charged?

    Best regards,
    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points

    Hi Kagawa,

    I see that you sent me a new log file.  I will check it again and get back to you in one or two days.

    Andy

  • 0 in reply to Andy Liu (BMS)
    TI__Guru 52115 points

    Hi Kagawa,

    Are you able to provide the srec file of the bq34z100-g1 device that has this FCC jump issue?  Thanks.

    Andy

  • 0 in reply to Andy Liu (BMS)
    Expert 2030 points

    Hi Andy-san,

    I sent you the srec file.

    Best regards,
    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    Expert 2030 points

    Hi Andy-san,

    Please tell me the following method with bq34z100-G1.

    ① Do not update Ra0 Table
    ② Limit the Ra0 Table so that it does not change too much (limit the amount of change)

    Best regards,
    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points

    See my comments below.

    ① Do not update Ra0 Table
    The Ra table needs to updated by the gauge so that the impedance of the battery can be tracked correctly. Otherwise, the RSOC repeated by the gauge will not be accurate.

    ② Limit the Ra0 Table so that it does not change too much (limit the amount of change)
    During the update of Ra values, a filtering process is already performed to eliminate unexpected fluctuations in the updated Ra values and therefore the Ra table will not change dramatically.

    Andy

  • 0 in reply to Andy Liu (BMS)
    Expert 2030 points

    Hi Andy-san,

    The customer uses it by discharging a few percent from full charge and then fully charging again.
    The memory effect is likely to occur, but isn't this usage increasing the Ra value?

    The following are reports and inquiries from customers.

    Ra Table was returned to the initial value in the two units that were refreshed (completely discharged → fully charged).
    When refreshed, the remaining amount monitoring IC itself will be in a non-energized state and will be reset when it is completely discharged.
    Is it possible that Ra Table became the initial value following this reset?
    (I think that the operation method to occasionally remove the battery pack is within the assumption, so it is unlikely that the data will easily return to the initial value.)

    Or, since the battery is not so deteriorated, the battery was activated during the process of full discharge → full charge, and the result better than the initial value of the original Ra Table (low impedance result) was obtained, but as a result, Is it possible to think that the initial values ​​of Ra Table have been adopted?
    (For example, when the initial value is 40 mΩ and the battery was 60 mΩ during the charging/discharging process, when it is measured as 20 mΩ after complete discharge → full charge, 20 mΩ is discarded and it is imagined to learn as 40 mΩ. I am.)

    Best regards,
    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points

    See my comments below.

    Is it possible that Ra Table became the initial value following this reset?

    [Andy] I don't think so. The Ra Table will NOT be reset to its default even after it goes through a power cycling.  The table will only be updated during discharge.

     

  • 0 in reply to Andy Liu (BMS)
    Expert 2030 points

    Hi Andy-san,

    Regarding Load Mode
    What is the effect of using LoadSelect=0 on equipment that is close to constant power?

    Regarding Ra Filter
    Ra Filter seems to be the maximum value of 1000, but if it is set to 1000, is it good to understand that Ra Table will not be updated?

    Best regards,
    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    Expert 2030 points

    Hi Andy-san,

    Questions from customer.

    1) It's DoDatEOC16384 in the Log I sent you before.
    What is the cause of DoDatEOC=16384?

    2) I would like to understand why I would change the parameters below if the battery is not designed to be isolated.
    Max Res Scale Current status changed from 32000 to 5000
    Min Res Scale Current status changed from 1 to 200

    3) What is the difference between Maximum Qmax Change and Qmax Max Delta %?

    Best regards,
    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points

    See my comments below.

    Regarding Load Mode

    What is the effect of using LoadSelect=0 on equipment that is close to constant power?

    [Andy] You mentioned constant power and I assume that the Load Mode is set to 1.  LoadSelect = 0 means the Average discharge power from the previous cycle is used for simulations. 

    Regarding Ra Filter
    Ra Filter seems to be the maximum value of 1000, but if it is set to 1000, is it good to understand that Ra Table will not be updated?

    [Andy] Yes. If the Ra Filter is set to 1000, the only the old resistance will remain.  I don't think this kind of setting is good since you try to stop the Ra Table from changing.

    Andy 

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points

    1) It's DoDatEOC16384 in the Log I sent you before.
    What is the cause of DoDatEOC=16384?

    [Andy]I need to check the log file again before I provide my answer.

    2) I would like to understand why I would change the parameters below if the battery is not designed to be isolated.
    Max Res Scale Current status changed from 32000 to 5000
    Min Res Scale Current status changed from 1 to 200

    [Andy]These two settings are recommended for non-removable Battery Packs.  Let me check if more details are available.

    3) What is the difference between Maximum Qmax Change and Qmax Max Delta %?

    Max Qmax Change specifies the maximum allowed change in Qmax value during Qmax update. Qmax
    update is disqualified if the change from previous Qmax value is greater than Max Qmax Change.

    Qmax Max Delta is the percentage of DesignCapacity() that limits how much Qmax may grow or shrink during any

    one Qmax update.

    Andy

  • 0 in reply to Andy Liu (BMS)
    Expert 2030 points

    Is there any particular problem if the device shipped with the constant current setting is changed to the constant power setting halfway ?

    Best regards,

    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points

    I assume that you mean the Load Mode has been set to Constant Current originally but it changes to Constant Power by itself.

    I don't think it could happen. Someone must have written to the Data Flash via i2c to change the Load Mode setting. Seal the device so that the Data Flash will not be changed by accident.

  • 0 in reply to Andy Liu (BMS)
    Expert 2030 points

    If the customer intentionally changes the Load Mode during use, will there be any problems?

    How about this ?

    1) It's DoDatEOC16384 in the Log I sent you before.
    What is the cause of DoDatEOC=16384?

    [Andy]I need to check the log file again before I provide my answer.

    Best regards,

    Kagawa

  • 0 in reply to Yoshinobu Kagawa
    TI__Guru 52115 points
    See my comments below.
     
    If the customer intentionally changes the Load Mode during use, will there be any problems?
    [Andy]   The setting should be finalized before a golden image is generated.  I don't think it is good to change it on the fly.
    1) It's DoDatEOC16384 in the Log I sent you before.
    What is the cause of DoDatEOC=16384?
    [Andy]As far as I know, the DoDatEOC is updated when a valid charge termination is detected.  In the log file,  the DoDatEOC is always 16384 because the battery is in the discharge cycle. It should have been updated when the battery is charged to full earlier.
     
    Andy