• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 64 subscribers
  • Views 152 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.

BQ25792: Charge current set in the register and measured charge current

Masashi Mineda
Expert 1215 points
Part Number: BQ25792
Other Parts Discussed in Thread: TPS25751,

Tool/software:

Dear Sirs and Madams,

We are debugging using a combination of BQ25792 and TPS25751.

The configuration is as follows:

- Battery : 4-cell lithium-ion battery

- Input voltage: 12V (PDO of USB PD is 12V/2.25A)

The charging current of the BQ25792 is set to 1100mA.

A low resistor (3mΩ) was connected to the low side of the battery and the voltage across the resistor was measured with a digital multimeter to obtain the current value, which was 1248mA.

When both ends of the resistor were measured with the MCU's current sense function, the MCU measured 1334mA.

We can see that the battery voltage is being charged at a constant current from 13V to 16.4V, but the charging current is higher, such as 1248mA (113%) or 1334mA(121%), than the 1100mA set in the register.

Is it possible for an error of nearly 20% to occur in the charging current during boost operation like this?

Regards,

Masashi

  • 0
    TI__Guru**** 167196 points

    Hi Masashi,

    That error is too high.  What does the charger's ADC IBAT measurement report?  If it reports the same, the only known current accuracy issue is switching noise into the current sense amplifier at the SYS pin.  This switching noise can be filtered using the 0.1uF capacitor from SYS to GND but only if that capacitor is placed on top of the IC, with no vias to connect to SYS and GND pins.  This means the SWx traces to the inductor must be routed under the IC and then back up to the inductor using a different layer.  The datasheet's example layout is shown below:

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Expert 1215 points

    Hello Jeff,

    The board layout was created based on the data sheet.

    The board is also a four-layer board.

    We checked the value of ICHG (REG33) and the charging IC used during charging had ICHG = 0, while the response of the charging IC not used during charging was ICHG = 65535.

    Are these values ​​correct?

    Regards,

    Masashi

  • +1 in reply to Masashi Mineda
    TI__Guru**** 167196 points

    Hi Masashi,

    The layout looks good.  My only recommendation would be to move the 0.1uF PMID and SYS caps directly on top of the IC.  The ADC will not read correctly unless the EN_ADC bit in REG0x0E.  If the charge currents are still not correct, can you reply back with the status and fault registers in REG0x1B through REG0x27?

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Expert 1215 points

    Hello Jeff,

    When charging the battery with constant current, the low side sense of the battery was displayed as 1334mA on the MCU and 1248mA on the digital multimeter, and when the IBAT_ADC register was read, it was 70mA (0x46).

    REG03 (Charge Current Limit) : 1100mA (0x6E)

    ADC resolution : 15bits

    EN_IBAT(REG14[5]) : 'H'

    The values ​​of each register are as follows. No particular faults or other problems have occurred.

    --

    REG1B (Charger Status0) : 0x8F

    REG1C (Charger Status1) : 0x7F (CHG_STAT : 0x3, VBUS_STAT : 0x8, BC1.2_DONE_STAT : 0x0)

    REG1D (Charger Status2) : 0x01 (VBAT_PRESENT_STAT : 0x1)

    REG1E (Charger Status3) : 0x00

    REG1F (Charger Status4) : 0x00

    REG20 (Fault Status0) : 0x00

    REG21 (Fault Status1) : 0x00

    --

    Regards,

    Masashi

  • 0 in reply to Masashi Mineda
    TI__Guru**** 167196 points

    Hi Masashi,

    To confirm, REG0x2E[7]=1 to enable ADC?  Also, is the charge current being regulated to the value per the value in ICHG registers REG0x03-04?  The status registers are reporting  IINDPM

    So, charge current should be reduced from the ICHG registers' value, due to input  power (current) not being high enough to supply both charge current and system load current.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Expert 1215 points

    Hello Jeff,

    We found that the correct ICHG value can be read from the BQ25792 when the connection to the TPS25751 I2C master is disconnected.

    In other words, there was no problem with the path between the BQ25792 and the battery.

    This is likely due to the TPS25751 accessing some kind of register, so we will continue to investigate.

    I always appreciate your support.

    Regards,

    Masashi