• State TI Thinks Resolved
  • Locked Locked
  • Replies 17 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 3536 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.

BQ20Z65-R1: Charge Fault

Chuchen Wang
Expert 3470 points
Part Number: BQ20Z65-R1
Other Parts Discussed in Thread: BQ20Z65

Hi,

My customer is encountering an issue when charging back a battery with bq20z65 after certain period of storage. The OC (Over Charge Capacity) bit was found et showing a charge fault and the Smart charger therefore stopped charging the battery.

Investigation found the cell voltage was reduced alone with circuitry leakage (sleep mode), but the fuel gauge RSOC did not decrease. When battery fuel gauge kept at a higher percentage but the cell capacity actually was lower, the internal counter might start counting up at that higher level and would eventually exceed the limit of FCC + OC Capacity during next charging. That causes a OC fault if there is no any discharge happened in advance. 

For instance, if the cell voltage has dropped below 3.2V, says to be less than 10%, but the fuel gauge RSOC keeps at 30% (which is the level we ship the battery), since OC Capacity setting is 15% of FCC, when the battery is charged around 85%, the “Internal Counter” can reach 115% (30% + 85%) and therefore over the limit of FCC+15%, the OC fault then occurs.

 

The battery summary is as follows and I also attached data flash setting for your reference.

4S2P

Panasonic NCR18650BF cell

14.4V

6500mAh

Over Charge Capacity=900mAh (Charge Control)

 

Questions:

  1. Why RSOC does not update per OCV change?
  2. Is that related to CHGOCV_DIS bit setting? (Configuration: Operation Cfg C)
  3. Is there any data flash setting can cause such issue?

Thanks,

Chuchen

 

U80388 Data Flash R2.xls

 

 

 

  • 0
    TI__Guru* 76230 points
    Hi Chuchen

    Your customer will have to disable the OCV_WGHT function of operation cfg C since it appears they have it currently set. operation cfg flag has to be set to 50 from 70 to clear that. Issue a reset command after clearing the flag to have Ram updated. Then rerun your test. Let me know if that helps
    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    TI__Expert 3470 points

    Hi Onyx,

    Please see customer response:

    I have done the test but cannot verify it.

    My test steps are:

    1. Modified Cfg C from 0x0070 to 0x0050 to disable OCV_WGHT bit

    2. Sent 0x41 to reset bq20z65r1

    3. Discharged the battery to 0% then charged up to 30%.

    4. Discharged the battery on cell side until 12.6V (3.15V/cell), so that RSOC kept at 30%, but cell voltage dropped down to 3.15V (equivalent <5% capacity).

    5. Left the battery overnight and checked next day, RSOC didn’t update, still 30%.

    6. Charge the battery to full (by power supply 16.8V 2.5A), OC bit was found set.

    Thanks,

    Chuchen

  • 0 in reply to Chuchen Wang
    TI__Guru* 76230 points
    Hi Chuchen,
    Did the customer have have log files of this test? Can they attach it here in addition to the gg file. Pls do not save the gg file as an excel file.

    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    TI__Expert 3470 points

    Hi Onyx,

    Please see attached.

    Thanks,

    Chuchen6661.Log.zip

  • 0 in reply to Chuchen Wang
    TI__Guru* 76230 points
    Hi Chuchen,
    I will be at the deep dive for most of this week and will be unable to review your file until friday. I suggest emailing the file to me so i have it in my inbox as a reminder

    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    TI__Guru* 76230 points

    Hi Chuchen

    With multicell gauges, in relaxation mode, the device performs rem cap simulations every 4 hours to account for self discharge. Can you increase your quit current to 80mA, increase dsg current threshold to 150 and charge to 100. Keep sleep current as 30mA. Then repeat the test.

    After you perform the self discharge test, make sure the battery pack is kept idle for > 4 hours without any load and see if a rem cap simulation is performed.

    thanks

    Onyx

  • 0 in reply to Onyx Ahiakwo
    Intellectual 270 points

    Hi Quote,

    I am Jeffrey Ye, one of Chuchen's customer. I modified the setting per your suggestion and repeated the test, RSOC updated after 2 hours rest when performed self discharge test.

    Can you please explain how the setting affects remaining capacity simulation? We usually use TI default setting for Quit Current=10mA; Sleep Current=10mA; Dsg Current=100mA and Chg Current=50mA.

    Thank you!

  • 0 in reply to jeffrey ye
    Intellectual 270 points
    Sorry, above message is for Onyx Ahiakwo.
  • 0 in reply to jeffrey ye
    TI__Guru* 76230 points

    Hi Jeffrey,

    I believe your quit current in the files sent was lower than the sleep current, so the gauge never got into relaxation mode before entering sleep and that caused the simulations to not occur.

    thanks

    Onyx

  • 0 in reply to Onyx Ahiakwo
    Intellectual 270 points

    Hi Onyx,

    Thank you for the reply!

    Do you mean if the gauge enters relaxation mode prior to sleep mode, then the simulations during sleeping can occur?

    So, as long as set quit current > sleep current will be ok? Or set Chg Relax Time and Dsg Relax Time shorter than Bus Low Time (Power) will insure the gauge entering relaxation mode prior to sleep mode?

    Moreover, in my setting, both quit current and sleep current are 10mA. Is 10mA threshold too low to be recognized by gauge?

    Thanks

    Jeffrey

  • 0 in reply to jeffrey ye
    TI__Guru* 76230 points
    hI Jeffrey,
    Yes it is better to enter relax mode prior to getting into sleep mode. I am not sure i follow what you mean about bus low time.10 mA is not too low to be recognized by the gauge. The resolution of the gauge is 1mA
    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Intellectual 270 points

    Hi Onyx,

    I meant if Bus Low time is 5s (this is what I set), then fuel gauge shall go into sleep mode after 5s no matter the current has been lower than 10mA earlier. And since Dsg Relax Time is 1s, the fuel gauge shall always get into relax mode firs, right?

    I tried to set Quit current at 20mA and keep sleep current at 10mA, but fuel gauge didn't update after self discharge. What is the lowest threshold it should be? Last time you suggested Quit current 80mA and sleep current 30mA, it looks a bit higher for our application.

    Thanks!

    Jeffery 

  • 0 in reply to jeffrey ye
    TI__Guru* 76230 points
    Hi Jeffrey
    A recommended value for quit current should be c/20. Sleep can be kept at 10mA. if the gauge didn't update, it means a condition (likely current spike) caused the gauge to exit relax mode. it is important that the current stays below the quit current for that entire 4 hours for a rem cap simulation to occur.
    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Intellectual 270 points

    Hi Onyx,

    Usually we set the taper current to C/20, will fuel gauge get into relax before charge termination if quit current is C/20 too?

    The scenario of fuel gauge without update happened during storage (didn't connect with any device), looks had no chance to exit relax mode.

    Thanks

    Jeffrey

  • 0 in reply to jeffrey ye
    TI__Guru* 76230 points
    hi Jeffrey,

    taper current should be greater than charge and discharge current threshold, and these two should be greater than quit current.

    thanks
    Onyx
  • 0 in reply to Onyx Ahiakwo
    Intellectual 270 points
    Hi Onyx,

    Does it make sense that beside the current threshold, the fuel gauge will be into Relax prior to Sleep surely as long as Chg Relax Time or Dsg Relax Time is shorter than sleep mode delay time?

    Datasheet didn't specify the delay time for sleep mode when fuel gauge detected battery current <10mA sleep current.
    Is Bus Low Time also (a delay time) for fuel gauge going into sleep mode when chg/dsg current< Sleep Current ?


    Thank you

    Jeffrey
  • 0 in reply to jeffrey ye
    Intellectual 270 points

    Hi Onyx,

    I changed the setting as suggested: Dsg Current Thre=150; Chg Current Thre=100; Quit Current=80; Sleep Current=30 and test many times, but not every time fuel gauge got updated after self-discharge.

    If I keep the setting without change but increase Over Charge Capacity threshold to 30% FCC, then OC fault will not happen because the shipping capacity is less than 30%, the fuel gauge counter will never exceed 130% in initial charging after long storage.

    Please advise.

    Thank you

    Jeffrey