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.

BQ24196: I want to know what is the abnormal state? Under what circumstances does BQ24196 enter this state?

Part Number: BQ24196

BQ24196 works in the default mode, I did not turn off the watchdog, and the watchdog reset time is set to 40 seconds. In most cases, the charging can be normally completed, but sometimes, BQ24196 enters an abnormal state, in which the watchdog timer overflow error is set and cannot be automatically recovered.  The watchdog timer overflow error can only be cleared by unplugging the battery, that is to say, the watchdog timer overflow error can only be cleared by powering off BQ24196. I want to know what is the abnormal state?  Under what circumstances does BQ24196 enter this state?

  • Hi,

    Today is a US holiday. We will get back to you tomorrow.

    Regards,

    Jeff

  • Hello, 

    Can you please confirm that the "watchdog timer overflow error" is the watchdog fault (REG09 bit 7=1)? Even with the watchdog fault present the device should still be able to function normally in default mode.

    Referring to section 8.4.1 of the datasheet (page 25), which states any write command over I2C should transition the device from default mode back to host mode and clear the watchdog fault. 

    Regards,

    Garrett

  • Hi Garrett,Thanks for getting back to me.
    Yes, the "watchdog timer overflow error" I'm talking about is the watchdog error (REG09 bit 7=1).
    In my program, whenever the MCU(STM32) is powered on, the BQ24196 register will be configured through 12C. The configured register is as follows:
    BQ24196_WriteReg(REG00, 0x7F);
    BQ24196_WriteReg(REG01, 0x1F);
    BQ24196_WriteReg(REG03, 0x10);
    BQ24196_WriteReg(REG04, 0xB3);
    In this configuration, the charging current is around 300mA (I simulated the pre-charging stage through this configuration to protect the battery).
    After 40 seconds, the register of BQ24196 is reset to default parameter through the watchdog, and normal charging is carried out under the default parameter (charging current above 1A).
    The problem I have now is that sometimes the watchdog error of BQ24196 will not be cleared automatically, and the parameters of the register will not be reset after 40 seconds, so the charging current is always around 300mA, charging is very slow, or it cannot be charged.
    I would like to know what is the problem that the watchdog error cannot be cleared automatically, and how can I solve this problem so that my product can resume normal charging.

  • Hello,

    Thank you for the additional information. The behavior you describe is not expected for this device. The watchdog timer expiring should always reset register settings back to default values.

    Would you be able to provide the register read values for all registers, as well as, input voltage and battery voltage/current when your device is in this abnormal state? This will be helpful to see if there are any other faults besides the watchdog and why the register values aren't resetting as they should. 

    Best Regards,

    Garrett

  • Hello,
    In the abnormal state, the values of all registers are as follows:  
    REG00=0x7F,
    REG01=0x1F,
    REG02=0x60,
    REG03=0x10,
    REG04=0xB3,
    REG05=0x9A,
    REG06=0x03,
    REG07=0x4B,
    REG08=0xAC,
    REG09=0x80,
    REG0A=0x2B,
    Input voltage is 5.15V, battery voltage is 3.3V to 3.8V (this abnormal state has occurred)

  • Hello, 

    Please see my comments below following review of all register values. 

    Based on the status and fault registers even in this abnormal state the device is in fast charge mode with ICHG = 2.048 A. REG08 does indicate your device is in dynamic power management mode (DPM). The device being in either VINDPM or IINDPM is likely why you are seeing lower charge current than expected. 

    According to section 8.3.4.3 of the datasheet on page 21 when a fault, such as watchdog expire, occurs the fault register (REG09) keeps the fault state until the host reads the register. To see the current fault status the host most read REG08 two times consecutively. My best guess is your host is re-writing the registers you set originally following the watchdog fault to transition from default mode back to host mode. Hence why REG00,01,03,04 are not at default values. 

    Could you please look to read REG09 twice in a row during the abnormal state to see if the watchdog fault is cleared on the 2nd read. Also I recommend investigating why your device is in VINDPM/IINDPM by reviewing your settings on REG00 and resistor component at ILIM pin. 

    Best Regards,

    Garrett