• State TI Thinks Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 3 answers
  • Subscribers 64 subscribers
  • Views 464 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.

BQ25790: BQ25790 doesn't detect full-charged when VCHG < VBATT

Hideki Hayashi1
Expert 8390 points
Part Number: BQ25790

Tool/software:

HI team,

Under specific situation, BQ25790 doesn't work as expected.

With full-capacity battery equipped with BQ25790, and plug a USB cable into the system, BQ25790 doesn't detect full-charged.

Example

full-capacity battery: 8.4V

Charging voltage: 8.3V (Reg01h 033Eh)

Judge for full-charged: Read CHG_STAT (Reg 1Ch bit7:5). Under this situation, the bits doesn't become 7h.

Is it expected or not? And please teach me how to detect this situation as full-charged.

Regards,

Hayashi

  • 0
    TI__Intellectual 1110 points

    Hi Hayashi,

    A few thing might be the issue here:

    • The charger regulates that BATP pin. Be sure the comparison is made between VREG and BATP voltage.
    • Ensure the charger is not in BAT_OVP 

    Regards,

    Jared Baxter

  • 0 in reply to Jared Baxter
    TI__Expert 8390 points

    Hi Jared,

    Yes customer is testing at BATP pin. OVP setting is 8.3V x 103% = 8.549V, so it doesn't become higher than VBAT_OVP(rising,min).

    Do you have any other thoughts?

    REgards,

    Hayashi

  • 0 in reply to Hideki Hayashi1
    TI__Intellectual 1110 points

    Hi Hideki,

    Is termination enabled?

    Does VBAT_PRESENT_STAT show VBAT is there?

    What is the CHG_STAT status?

    Regards,

    Jared Baxter

  • 0 in reply to Jared Baxter
    TI__Expert 8390 points

    Hi Jared,

    Jared Baxter said:
    Is termination enabled?

    Reg0Fh bit1=1b: Enable

    Jared Baxter said:
    Does VBAT_PRESENT_STAT show VBAT is there?

    Reg1Bh bit0=1b: VBAT present

    Jared Baxter said:
    What is the CHG_STAT status?

    Reg1Ch bit7-5=100b=4h: so 4h = Taper Charge (CV mode)

    Is there any idea of the root cause?

    Regards,

    Hayashi

  • 0 in reply to Hideki Hayashi1
    TI__Intellectual 1110 points

    Hi Hayashi,

    If the part is in taper charge (CV), the charge current has not dropped low enough to go into termination. The termination current can be adjusted through the  REG09_Termination_Control Register. Also, ensure that you do not have an external load on the battery that could be drawing current and preventing termination.

    Regards,

    Jared Baxter

  • 0 in reply to Jared Baxter
    TI__Expert 8390 points

    Hi Jared,

    REG09_Termination_Control Register: 3h＝120mA

    REG33_IBAT_ADC Register: 0h＝0mA

    REG01_Charge_Voltage_Limit Register: 33Eh＝8.3V

    VBAT =8.4V, so the current never flow into battery. and it doesn't detect stop charging even the IBAT is below REG09 termination_control register.

    Customer noticed that BQ25790 can detect full-charge when the charging, but it can't detect when 8.4V charged battery is equipped.

    Is it correct behavior? I assume BQ25790 doesn't start switching when VBAT > REG01_Charge_Voltage_Limit and no switching doesn't rewrite CHG_STAT register, is it right?

    Regards,

    Hayashi

  • 0 in reply to Hideki Hayashi1
    TI__Guru**** 166626 points

    Hi Hayashi,

    In forward/charge/sink mode, the IBAT ADC channel can only report charge current flowing from SYS to BAT.  If there is a load on SYS that results in reverse current from BAT to SYS, IBAT will report 0.  Also, we do not spec the ADC absolute accuracy.  I have seen the IBAT ADC channel measure up to 50mA inaccurately. 

    If the host reduces the ICHG register value < 1000mA when close to termination, the termination current and IBAT ADC accuracy improve.  Can you ask your customer to try that? Also, when in buck or boost (VBUS not close to VBAT), termination accuracy improves if PFM is disabled.  Alternatively, when in buck-boost (VBUS close to VBAT), termination accuracy improves if PFM is enabled.

    Regards,

    Jeff

        

  • 0 in reply to Jeff F
    TI__Expert 8390 points

    Hi Jeff,

    Customer tried both, but it can't be resolved.

    Jeff F said:
    If the host reduces the ICHG register value < 1000mA when close to termination, the termination current and IBAT ADC accuracy improve.
    Jeff F said:
    when in buck or boost (VBUS not close to VBAT), termination accuracy improves if PFM is disabled. 

    VBUS=5V, VREG=8.3V and PFM_FWD_DIS=0h

    In customer's reporting situation, VBAT is bigger than VREG, so it may produce reverse current to SYS pin, so the register value report 0.

    Basically, the BQ25790 doesn't consider cases that a battery with higher voltage (full-charged) is equipped? 

    If so, customer will modify software to do something in the case. Please let me know the thoughts on this use-case.

    Regards,

    Hayashi

  • 0 in reply to Hideki Hayashi1
    TI__Guru**** 166626 points

    Hi Hayashi,

    If V(BAT) > VREG but not greater than BAT_OVP threshold, charging stops.  If V(BAT) > BAT_OVP threshold, charging stops and a 30mA pull turns on.  How is the battery voltage getting higher than VREG?

    Regards,

    Jeff

  • 0 in reply to Jeff F
    TI__Expert 8390 points

    Hi Jeff,

    There are some cases that "battery voltage getting higher than VREG"

    1. Other battery charger product with different battery charger IC exists.

    2. EE is camera. The full charge voltage at room temperature is normally 8.4V even when being charged by camera (BQ25790), but the specification is such that the battery is charged at 8.3V when it cannot be determined due to concerns about charging deteriorating batteries.

    After removing battery to set the battery, this case also happens.

    3. When a fully charged battery at room temperature is attempted to be charged in a high temperature environment, VREG is set to a lower voltage, so there are cases where a battery with a higher voltage than VREG is inserted.

     

    Have you seen same issue on EVM? Customer said they can reproduce it by EVM.

    The way of confirmation is 

    VREG set to 8.3V -> ENABLE_CHARGE set to OFF -> apply 8.4V on VBAT pin -> ENABLE_CHARGE set to ON -> REG1C CHG_STAT is Taper Charge (CV mode)

    Best regards,

    Hayashi

  • 0 in reply to Hideki Hayashi1
    TI__Guru**** 166626 points

    Hi Hayashi,

    I have not seen this on the EVM.  If charge enabled is toggled again after battery attached, does the IC terminate as expected?

    Regards,

    Jeff