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 BUFFER /REPEATER PART AVAILABILITY

Other Parts Discussed in Thread: TXS0102, TCA9617A

Hi,

I am looking for  I2C Buffer/Repeater for 3.4MHz speed. The required  bus voltage is 2.5V. May i know the part availability? 

  • Hello Vishal,
    If you require a repeater (something that buffer's the signal for longer traces), then the cloest option we have is the TXS0102. It features rise time accelerators which help speed transitions up, but it is specced up to 2 Mbps. Our fastest I2C-aimed buffer/translator is the TCA9617A/B which supports FM+ mode (1 MHz).
  • Hi Jonathan,

    Thank you for the information.

    I am also looking for a  "Device" that  should support the following requirement,

    1.The Input of the device should be "Push-Pull/Logical signals(TTL/LVTTL)" compatible

    2.The Output of the device should be an "Open drain" I2C compatible.

    3.The Input & Output lines should be bidirectional with 2.5V compatible.

     

    Kindly suggest us the best suitable part that meets the above requirement.

  • Hello Vishal,

    I am going to assume you also need it to have 2 channels (for SDA and SCL), and relatively low speed (< 2mbps) is ok.

    I'll loop the appropriate application engineer in to help with this.

    I would starting by looking at the TXS0102

  • Yes, I believe that this is the appropriate part for your application.
    -Francis
  • Hi Jonathan,

    We have already selected the part TXS0102 which you have suggested. Now i am looking for a "Device" that is entirely a new requirement. My first post has no relation with my latter one. This is completely for a new design. The speed is of least priority for this requirement.
  • Vishal,

    I understand. The TXS0102 should also meet the new set of requirements you have stated. The TXS0102 is a general purpose translator which supports open drain and push-pull at low to medium frequencies over a large voltage range. As such, it is quite flexible in the use cases it supports.
  • Hi Jonathan,

    Kindly confirm TX0102 supports the below application.

    Buffer Input Logic End – Push-Pull/Totem Pole (TTL/CMOS/LVTTL etc.)

    Buffer Output Logic  End – Open Drain (I2C Compactible)

    Speed Grade – 100KHz or Any

    Input side Voltage level – 2.5V

    Output side Voltage level – 2.5V

    Type – Bidirectional Level translator

    We have connected Port A of TX0102 device to GPIO (Push-Pull/Totem Pole) of PIC Micro-controller which can generate I2C compactible waveform. Other End Port B will be interfaced to I2C Slave device with Open Drain Logic.

  • Hello Vishal,

    THank you for being more clear that you will have 2.5V on both sides of the device. The TXS0102 will support what you are trying to do.

    However, i do have one concern with your approach.
    I2C requires a bi-directional interface in order to accept ACKs from the slave. Unless you can manually tri-state your port to high impedance in order to read the input, you will have communication issues because your master will hold the SDA line high or low, and the slave device will not be able to over come the microcontroller's driver without possible damage to the micro controller (as a result of the slave pulling the line low while the MCU is trying to hold it high, you essentially have a short to ground, which can damage the MCU is the current limit is exceeded.

    Typically, microcontrollers have a hardware module that supports I2C natively, and properly uses an open drain approach to support this.
  • Hi Jonathan,

    Thanks for your suggetion.
    In that case ,Do you have any solution from TI to Interface i2C compactible push-pull driver to Open drain receiver?

    This is a specific requirement where we are not supposed to use Micro-controller I2C Bus.
  • Vishal,

    This is not a device issue. The TXS0102 technically meets your requirements. In most cases where such a conversion is requested, it is not for I2c, but instead for a proprietary 1-directional open-drain bus. Merely translating from push-pull to open drain does not make it I2C compliant. As I had mentioned, bi-directional data on a bus line is a crucial aspect of I2C communication. If you use push-pull on the MCU side, you eliminate the processor's ability to handle incoming data on the SDA line.

    What I'm trying to say is that "I2C Compatible push-pull" is not really a thing. I2C spec mandates open-drain configuration.

    I don't think I understand why you are unable to use the MCU's I2C bus. Perhaps, you would be better served with a bridge that can convert from a push-pull protocol to I2C? For example, there are some SPI to I2C bridge ICs that will handle the conversion from SPI's push-pull interface to a fully I2C compliant signal. TI does not make any such bridge ICs, however NXP and Maxim do have such options.

    If you are wanting to make your push-pull setup work with I2C, then you will need to be able to change the port state from an output to an input on the appropriate bit for ACKs.