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.

SN74CB3T3306: Can we use this device as I2C level shifter ?

Guru 19775 points
Part Number: SN74CB3T3306

Hi Team,

Please allow me to confirm whether SN74CB3T3306 could be used as a I2C level shifter.

Customer is planning to use this device as following, can this use case supported for this device ?

* I understand that TCA9517 would best match for this application, however, please tell us if SN74CB3T3306 could be used or not.

Best Regards,

Kawai

  • Hello Patrick-san,

    Thank you for your support.
    I look forward to hearing from you soon.

    I believe that if the device is used with VCCA=5.0V and VCCB=3.3V, there is a current path from VCCA to VCCB. So, I guess High voltage will be limited to around 4.15V and this will appear on both sides(Master and Slave side).

    Best Regards,
    Kawai
  • Hello Kawai-san,

    Thank you for your patience. Based on my preliminary measurements, you should be able to use this part for I2C. The bandwidth is 100MHz well above the max of an I2C clock even when operated in ultra fast mode.

    The measurements I have done so far were to assess if the signal could be pulled low from either side for bidirectional communication, as well as see what the voltage levels would be if both sides (A and B) see an open drain.

    The basic test setup I used is illustrated below. In case 1, both inputs 1A and 1B are left floating and were pulled up to 5V and 3.3V respectively. In case 2, 0V was applied to pin 1A, and it was observed that pin 1B was pulled down to ~0V. For case 3, 0V was applied to pin 1B, and it was observed that 1A was pulled down to ~0V. We expect similar behavior from channel 2 corresponding to pins 2A and 2B.

  • Hi Patrick-san,

    My apologies for my delay. Thank you for the test.

    I understood as below and this device would be able to be used as I2C level shifter.

    I have two more questions.

    Q). Did you pull down output enable (Low active) pin ?

    Q). Do you have any reason why you had chosen 300 ohm for pull up resistor ? I believe 2k to 5k is nominally used in I2C.

    Best Regards,

    Kawai

  • Hello Kawai-san,

    A). Yes pin 1 (low active enable for switch 1) was set to 0V during these tests.

    A). I knew that a pull up was used for I2C communication, but I did not realize that 2k - 5k was the standard.  As the pullup is essentially forming a voltage divider with an internal mosfet impedance,  similar behavior should still be observed with these values, as the impedance of the open drain mosfet should be greater than 10 times the pull-up values. 

  • Hi Patrick-san,

    Thanks for the answers and your support.

    Best Regards,
    Kawai