• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 33 subscribers
  • Views 439 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.

PCA9543A: Problems with PCA9543A implementation

Kirk Festing
Prodigy 70 points
Part Number: PCA9543A
Other Parts Discussed in Thread: LM75B, TCA9517

Dear TI,

We are using both outputs of the PCA9543A to connect to the same I2C bus segment on a removeable module.  The module contains a small number of I2C devices, including a temperature sensor.

We adopted this topology so that the devices could still be accessed if the connector pins of the module ever were damaged.  We could use output 1 of the mux rather than output 0 to read from the devices.  It works, but only if both mux outputs are enabled rather than just one.

We expected that the disabled mux output would appear to the bus as high-impedance and have no effect on I2C operation, but the SDA signal exhibits a 4 MHz oscillation that turns the data into gibberish when the LM75B temperature sensor is accessed.  This disappears if both mux outputs are enabled.

The bus segment going to the PCA9543A is driven by a TCA9517 device.

Our short-term solution is simply to enable both mux outputs when accessing the module, but we want to know what is causing the strange behavior when only one output is enabled.  Thanks!

                                                                                                                          -Kirk

  • 0
    TI__Guru 62920 points
    Hi Kirk,

    I'm having trouble understanding your set-up. When you say you have both PCA9543A outputs connecting to the same I2C bus segment, do you mean that SD0 is shorted to SD1 and SC0 is shorted to SC1? Or do you mean that both SD0/SC0 and SD1/SC1 are used to communicate with the common segment on SDA/SCL?

    Do you have pull-up resistors populated on each output? (This is required for proper functioning of the switch, since the way the device is architected only allows for propagation of "low" levels through it while "high" levels are pulled up externally to each segment's VCC voltage.)

    Would it be possible to show us an oscilloscope screenshot of the oscillation on SDA? I'm curious what seems to initiate it, whether it is a sinusoidal or square-wave oscillation, and what the amplitude might be.

    Regards,
    Max
  • 0 in reply to Miguel Robertson
    Prodigy 70 points
    Max,

    Thank you for responding. I realize that I posted my question on a forum, but can I send you a system diagram privately?

    -Kirk
  • 0 in reply to Kirk Festing
    TI__Guru 62920 points
    Kirk,

    Sure, I can send an email to the address saved in your myTI account to start the conversation. I will likely need to loop in a colleague, though, since I will be out of the office for the remainder of this week.

    Regards,
    Max
  • 0 in reply to Miguel Robertson
    Prodigy 70 points
    A 120 pF capacitor from SDA to GND eliminated the 4 MHz oscillation.