• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 34 subscribers
  • Views 356 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.

PCA9534A: The software cannot read the device through IIC and has abnormal waveform issues

Tonyx.wu
Genius 4445 points

Part Number: PCA9534A

Using TI's expansion chip, model: PCA9534ARGTR

schematic diagram:

Problem phenomenon:

1. The software cannot read the device through IIC;

2. In the schematic diagram above, before the IIC inlet of R18 R19 (remove the resistance), measure the IIC waveform and confirm it;

However, with the addition of R18 and R19 (after the expansion chip is connected), the IIC waveform is abnormal and the device cannot be read;

  • 0
    TI__Guru 55808 points

    Hi Tony,

    I am looking at the scope capture and notice the write condition sent to address 0111000 = 0x38h, the ACK bit seems to be a bit odd. With the EEPROM device disconnected, have you been able to communicate with the PCA9534A standalone?

    Is it possible to set the SGM4574 level translator to high-z, remove 0-ohm resistors to separate EEPROM (BL24C64A) device from the bus so that only the PCA9535A is connected? Then re-run tests to see if communication can occur with only the IO expander. If communication begins to work, then slowly add back I2C devices onto the bus to see what makes the system fail. It could be that the added input capacitance from these devices is causing a bit error, but I would think we would see that result with the device NACK'ing.

    What are the oscillations occurring after the 9th clock cycle? Is this just noise? Are there pull-up resistors always present on SDA/SCL lines? 

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    Guru 320330 points

    The I²C signals are very noisy. The SGM4574 has edge accelerators. These continuing oscillations are caused by ringing on the I²C lines that is falsely detected as the beginning of a signal edge and triggers the edge accelerators.

    You should better shield the SDA/SCL lines from each other (and from other digital signals). Also, the total bus capacitance (trace capacitance and input capacitance of all connected I²C devices) might be too high for the SGM4574.