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.

BQ25792:

Part Number: BQ25792
Other Parts Discussed in Thread: TPS25750

When we use your company's bq25792 for circuit design, after the type-C is inserted, the output of bq25792 is abnormal, and the VSYS quickly drops from about 1.2V to 0.

When we use battery power supply, the output voltage of VSYS is the same as the battery voltage.

What causes this situation? The circuit design is as follows, please help we  to analyze it, thank you!

  • Hi,

    With no battery at BAT pin and minimum recommended capacitors on SYS and BAT, the SYS and/or BAT output overshoots at startup, hitting OVP, which causes the converter to momentarily turn off.  If disable charge at startup or add more capacitance to BAT or SYS pin to absorb the overshoot or have a battery at BAT, the shutdown will not occur.

    Regards,

    Jeff

  • Dear Jeff:

     Thank you for your reply!

     When we plug in the battery and then insert VBUS, bq25792 still doesn't work

     The waveforms of SW1 / SW2 / STAT / REGN have been measured under three conditions: USB power supply only / connecting battery and then inserting USB / battery power supply only

      Please refer to the attachment for measurement results

      In addition,because the layout personnel are self-learning AD18, and the first time layout board, layout has many shortcomings, BQ25792 layout as follows, please help to confirm whether bq25792 does not work due

    to layout, thank you!

       BQ25792 waveform_20210125A.xlsx

  • Hi,

    The charger is tripping a fault after 7 tries.  Can you I2C read the fault and status registers as this is happening?  The fault registers will tell you what is tripping but they clear immediately after being read.

    This charger is sensitive to board layout and needs to follow the EVM as close as possible.  I have attached the Atlium files for the EVM.

    BMS034A_nologo.zip

    Regards,

    Jeff

  • Dear Jeff:

       Thank you for your reply!

       I'd like to ask you another question

       With our current circuit design, if there is no problem with layout, can we realize battery charging? We use the type-C interface, but it only uses USB2.0 protocol. Do we need to add a USB controller between type-C and bq25792? If we realize PD fast charging, do we need to add a CC controller?

      With our current layout motherboard, the input current of VBUS is only about 25mA. Is the current supplied by USB too small to cause bq25792 not to work?

    The measured VBUS current waveform is as follows:

     

     I look forward to your reply,Thank you!

  • Hi,

    To use USB TYPE C PD, you will need a PD controller like TPS25750 before the charger.

    Your waveform plot did not post correctly.

    If the charger's input power = VBUS*input current limit is greater than its output power = VSYS*ISYS+VBAT*ICHG, then you should be able to charge.  If not, then ICHG is reduced to feed ISYS.  Input current limit is set to the lower of ILIM pin resistor or IINDPM per the D+/D- pins. 

    The charger has fault and status I2C register that should tell you why are not are charging.  The fault registers clear after they are read.  Can you read those and reply back?

    Regards,

    Jeff 

     

  • Dear Jeff:

       We use the logic analyzer to capture the I2C signal of PMIC. The data read by I2C is basically 0xff. I refer to bq25792 datesheet," register read beyond reg25 (0x25), returns 0xff"

    We don't quite understand this part of the problem, I would like to ask you more,Thank you very much!

    The waveform and log of I2C data are as follows:

     I2C DATA OF PMIC.txt

  • Dear Jeff:

       The I2C waveform captured by logic analyzer in the last email is not be the information  in the PMIC status register.

    Does the status register information of PMIC need to input the address of the status register through the software to read? Do you need to input any

    instructions?

    In addition, for our development board, DC power supply is used to supply 4.89V voltage to REGN today. STAT is still high level, it can't charge the battery.

    Please confirm it,Thank you very much!

    Regards,

    Ellie

  • Hi Ellie,

    The datasheet section 8.3.14 starting on page 51 explains the I2C registers.  In short, you have to send the IC address and the register address that you want to read/write.

    Regards,

    Jeff

  • Dear Jeff:

       According to the specification, the address of bq25792 is 0x6b,the address of REG20_ FAULT_Status_0 Register is 0x20?

     If not, how do I know the address of this register?

      We use the address of 0x20 to write and read,the data is 0000000000.Is the register address wrong? Or is the method we read wrong?

       Looking forward to your reply! Thank you!

      ellie

  • Dear Jeff:

       Can I use I2C to read and write normally when PMIC doesn't work? Now the status of our development board is VBUS=5V.

     After the battery is inserted, VSYS = Vbat, REGN and other signals are low . Can I read PMIC register information normally in this state?

      What does bq25792 need MCU to control through I2C? Is it only necessary to write charging parameters through I2C?

    Is it necessary to write control signal information?

       Looking forward to your reply! Thank you!

      ellie

     

  • Hi Ellie,

    The I2C command requires both the IC address and the register address.  

    The FAULT registers clear after the first read.  So, if you made multiple reads, the 2nd read would read 0 unless another fault occurred between reads.

    Regards,

    Jeff

  • Dear jeff:

        For the application of bq25792, would you like to ask if the IC will write information to the IC before leaving the factory? For example, the F/W & the default parameters of charging current, etc. if the information is not written in the factory, will it not work when it is powered on

    At present, I have disconnected the SYS back-end circuit of our IC to confirm that the bq25792 phenomenon is the same as before. The measured waveform is shown in the attachment. Would you please help me to see if there are other things to confirm?

    The information of the register has not been read out correctly at present

    BQ25792 單板量測波形_20210208A.xlsx

  • Hi,

    The charger doesn't have a processor or firmware.  The charger uses an asynchronous state machine that makes decisions based on I2C registers.  Those registers have default values.  I am assuming that you still have no battery connected. With no battery connected and EVM default settings and components, the BAT pin voltage and the SYS pin voltage overshoots, causing the charger to hit OVP protection, which shuts down the converter.  To prevent that, 

    1. Did you disable charge to BAT pin using the /CE pin (JP17) on the EVM?

    2. With no battery attached, did you also remove the shunt on JP18 (TS Normal) in order to simulate the battery thermistor being removed, as would be the case if no battery is attached?

    3. Alternatively, you can add more capacitance on BAT pin to absorb the overshoot.  The exact value varies depending on your battery regulation voltage setting (1S-4S) and input current limit setting (ILIM_HIZ pin resistor) but you can start with 470uF.

    Regards,

    Jeff