• State TI Thinks Resolved
  • Replies 11 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 177 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

TCA6424A: TCA6424A I2C NACK

?? ?
Prodigy 10 points
Part Number: TCA6424A

Tool/software:

Dear support team,

I'm trying to get a TCA6424A to work. Sadly, it doesn't reply at all to I2C commands. Another slave on the same bus works without any problems. I would appreciate if you could point me at the thing I'm missing here...

This is a pic of the scope screen:

As you can see, the address byte doesn't get ACK'd.

This is the schematic:

I noticed that there was a similar issue six years ago.It was caused by layout error. I have confirmed mine is correct.

I tryed to remove the another slave on the same bus but it still doesn't work。

As I'm running out of ideas, I'd really appreciate some help here!

Thank you!

Best regards

  • 0
    Guru 297090 points

    The waveform looks OK.

    Check the voltages at the supply and control pins.

  • 0
    TI__Mastermind 37293 points

    Hi ?? ?,

    Are you able to take oscilloscope captures of the SDA/SCL lines? 

    What are the PU resistor values of the SDA/SCL lines? 

    Regards,

    Tyler

  • 0 in reply to Clemens Ladisch
    Prodigy 10 points

    Thank you for your reply，

    The voltages at the supply pin is 1.8V seems normal，RESET pin is 1.8V,  and INT pin is 1.67V.

  • 0 in reply to Tyler Townsend
    Prodigy 10 points

    Thank you for your reply，

    The PU resistor values of the SDA/SCL lines is 4.7K and it connected VCCI/VCCP.

    The waveform seems normal.

  • 0 in reply to ?? ?
    TI__Mastermind 37293 points

    Hi,

    With ADDR = 1, the target address is correct. 

    This is a read command since 8th bit = 1. 

    Is this read command being issued at power up? If there is not command byte written to the device, issuing a read command will NACK since the pointer byte has not been set (the TCA6424A does not know which register to read from upon power up). 

    First write the command byte to set the pointer byte, then issue a read command. This should ACK. 

    Example in figure 8-9. 

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    Prodigy 10 points

    Hi,Tyler

    We've tried to write to the device,but it is still nack. 

  • 0 in reply to ?? ?
    TI__Mastermind 37293 points

    Where is this scope being measured from? Can we scope the SDA/SCL directly at the IC? 

    Have we checked that VCC is reaching the IC? 

    GND'ing connection good? 

    Same with Address pins, are the voltages measured on a DMM as expected? 

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    Prodigy 10 points

    HI Tyler

    I have checked all the points you mentioned, and everything seems to be in order.

    And I noticed a detail in the datasheet as shown in the figure below. Do we need to control VCCP for a power reset cycle?

  • 0 in reply to ?? ?
    TI__Mastermind 37293 points

    VCCP does control the power reset cycle yes. What is happening to VCCP when you are writing I2C commands? 

    Both VCCI and VCCP should be powered up for this device to work properly. VCCI is the I2C supply, VCCP is the p-port side voltage, VCCP is also connected to the PoR. 

  • 0 in reply to Tyler Townsend
    Prodigy 10 points

    VCCP and VCCI are connected together and are powered by a constant 1.8V supply. When writing to the I2C, the VCCP remains unchanged, as it is provided by the internal PMIC of the SoC. If the voltage drops below 0.2V, the system will fail to operate.

  • 0 in reply to ?? ?
    TI__Mastermind 37293 points

    This sounds like correct operation of the device. 

    If VCCP falls below the operation voltage of 1.65V, then behavior is not characterized for the device. 

    Has this issue been repeated on multiple TCA6424A's?