• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 224 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

I2C Buffers for MUX Control

AlexWalton42
Expert 6045 points
Other Parts Discussed in Thread: TCA9548A-Q1, TS3A5017-Q1, SN74LVC2G66-Q1, TCA9548A, PCA9518

Hello,

Customer was looking at the TCA9548A-Q1 for:

"I was trying to find a buffer with an enable such that I could set up an array of buffers and use them as a simple to control mux. I was getting concerned about the number of the buffers warning that one side could not have pullups on it or be connected in parallel to another buffer.  The task I have is to test a device with 4 masters and simulate a I2C device on the other end. The lead wants them to be isolated from one another to avoid cross talk as there are shared addresses."

I am having trouble understanding the customer's design. Before I call the customer to learn more, I'd like to have some background knowledge. I couldn't find a similar application TI.com. Is anyone familiar with this design?

Any help would be greatly appreciated.

Best,

Alex

  • 0
    Guru 276740 points

    These restriction are needed because active buffers must be able to detect in which direction the signals flow.

    Passive switches/muxes like the TCA9548A-Q1 do not have this problem.

    It might be also be possible to use plain analog switches/muxes like the SN74LVC2G66-Q1 or TS3A5017-Q1, but those must be controlled with separate GPIO signals.

  • 0
    TI__Guru 58030 points

    Hey Alex,

    Just to be clear, an I2C buffer redrives a signal and is typically used to separate capacitive loading and sometimes weak I2C drivers. A device like TCA9548A is an I2C switch and does not 'buffer' (redrive) the signal. I2C switches/MUXs typically are used for doing multiple level translations or separate I2C addresses to prevent address conflicts. 

    Like Clemen's  pointed out, the I2C switches are less tricky to use with less restrictions/rules. It seems like a passive pass FET device like TCA9548A would be the right choice to use. The difference between a dedicated I2C switch/mux and a generic MUX/switch is the I2C switch acts as an I2C target device with its own state machine so it does not require a select line. 

    -Bobby

  • 0 in reply to Clemens Ladisch
    TI__Expert 6045 points

    Clemens, 

    Thank you for the response! 

    If possible, could you help clear up my confusions on my reply to Bobby's post?

    Best,

    Alex

  • +1 in reply to AlexWalton42
    TI__Guru 58030 points

    It looks like your follow response didn't make it onto the thread for some reason. 

    As an update: We ended up discussing this thread offline and the matter seems to be closed. Current suggest device is the PCA9518 until additional information is gathered.

    -Bobby