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.

TXS02612: specification of i/o levels

Part Number: TXS02612

Tool/software:

The datasheet does not tell me the Vih minimum from the B-side with 3.3V supply.

I have a reference design where a XILINX ultrazynq+ interfaces to an SD card. Between the FPGA and the card, I find a TXS02612. The levels shifter has 3.3V supply om the A-side to satisfy the FPGA, and 3.3V on the B-side, that is connected to the SD card.

The boot-loader in the FPGA makes the card switch to 1.8V interface levels. The bootloader is hard coded in the FPGA, and can NOT be changed. This means, that my own design, without the TXS02612 does NOT work, because of the Vih min specification of the FPGA is too high.

Your datasheet for TXS02612 does NOT give me Vih min for 3.3Volts on the B-side. I need this information, to understand why it works.

Please specify the values, and hopefully update the data sheet,

Kind regards, Bo Bjerre

  • In the recommended operating conditions, VIH is specified as VCCI − 0.2 V. (For open-drain signals, the line is pulled to VCCI by the pull-up resistor.)

    If the reference design does not switch the VCCB supply to 1.8 V, then I do not understand how it would work. (The TXS works like an analog switch for voltages between 0 V and VIH.)

  • Hi Bo Bjerre,

    The VIH thresholds is stated here:

    Note the I/O lines have different architecture (switch-based, similar to that mentioned by Clemens) where as the SEL/CLK line has buffering circuitry, hence the different threshold requirements. 

    You may see this FAQ regarding the stringent thresholds for the I/O lines here: e2e.ti.com/.../faq-why-are-the-txs01xx-vih-vil-specifications-so-stringent

    Regards,

    Jack

  • exactly, how I read it. So an SD card, which has "stepped down" to 1.8V and connected to the B-side (with 3.3V supply) should NOT work.

    I must contact TRENZ about their consideration about WHY it works. I will arrange for measurements of the actual signal levels.

    Thanks so far,

    Bo //

  • Hi Bo,

    I see, please comment here on any updates regarding this application. (Any block diagrams would be useful as well). 

    Regards,

    Jack

  • Hello,

    The problem has been solved. Interfacing an SD card needs pull-ups on the CMD and datalines. This is the reason, why the design which uses the TXS02612 works. The levetranslation from 3.3V to 3.3V is not needed, but the build-in pull-ups saves the design.

    I just added pull-up resistors to the design, and now it works.

    Thanx