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.

BQ78Z100 Spec 7.26 I2C and HDQ: pull down resistor

Other Parts Discussed in Thread: BQ78Z100, TXS0102

BQ78Z100 Spec 7.26 I2C and HDQ Interface I/O Parameter: Rpd=3.3K. Wondering about that, is there really a pull down resistor on Data or Clock port? Isn't that a typo?

  • Hello Jeffrey,
    The I2C clock and data lines have an internal pull-down, as you state.
    I2C spec says the Host Bus should have pullups, 2k Ohms is the norm.

    Page 29 of the bq78z100 datasheet reinforces the function of the internal pulldowns.
    When the gas gauge senses that both lines are low (such as during removal of the pack),
    the device performs auto-offset calibration and then goes into SLEEP mode to conserve power.

    Without the internal pulldowns, the gauge could not identify both lines low to force sleep mode.
  • I also want to know if I can use a 10K pull up resistor instead of what TI recommended 3.3K. Because the other side is TXS0102 Voltage-Level Translator, which low-level input voltage is only 0.15V max..
  • 10K pullup between the BQ and the TXS should be fine.
    As long as the BQ VinH VinL are met by the TXS0102 translator.

    Also see this app note about I2C Speed vs Pullup vs bus capacitance.
     
    How you plan on using the I2C determines how you terminate it.

    Higher speed + high buss capacitance = lower value pullup needed to maintain the speed.

  • Thank you for the reply.

    I understand the pull down resistor for chip sensing the bus line and going into sleep mode. We usually use a external 1M resistor for other chips, like bq20z65 etc. For bq78z100, if a 3.3K internal pull down resistor is always there, how can a bus line be pulled up to high level by a 2K resistor from 3.3V? The level of bus line may be only 3.3/(2+3.3)*3.3V=2.05V when it is pulled up?

  • For VIN HIGH for the BQ, it is 0.7 x Vreg. For Vreg=1.8V that's 1.26V.
    For VIL LOW, 0.3 x Vreg, and for Vreg=1.8V that's 0.54V
    The I2C on BQ78z100 is on a 1.8V engine.
  • I guess the internal pull down resistor is conditionally connected (switched on) to the bus line. I powered up the chip and pulled up the I2C line to 3.3V with 10K resistor, there is no voltage drop observed. Otherwise, it should be 0.8V at the line. Am I right? Thanks!
  • That 3.3K Rpd would be the pull down resistance when the device is actively pulling the I/O low. So, under conditions where no communication is occurring, I wouldn’t expect a voltage drop. There is still a 1M pull-down resistance within the device that should be connected all the time when the device powered.

    thanks
    Onyx
  • Thank you for the reply Onyx! However, I don't understand "That 3.3K Rpd would be the pull down resistance when the device is actively pulling the I/O low". I don't think I2C can work under the condition where the pull down resistance of I/O would be 3.3K. It must be something I misunderstood.

  • Hi Jeffrey,
    There are two pull downs in parallel on the I/O. The weak pull down is 1M Ohm and the strong pull down would be the 3.3k Ohm. The weak pull down is always active, the strong pull down is active when the device pulls the I2C line low.
    thanks
    Onyx