• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 21 subscribers
  • Views 1741 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.

TXB0304: INPUT DRIVE REQUIREMENT

Nutan Reddy
Prodigy 30 points
Part Number: TXB0304
Other Parts Discussed in Thread: TXS0104E

In TXB0304 datasheet section 8.3.1 says min input drive requirement is +/-3mA. We are using TXB0304 to level shift the SPI  to HMC7044. In table 3. of HMC7044 datasheet max. supported DC current for SDATA is mentioned as 0.6mA. We are using the Digital IO in CMOS mode. Does this mean HMC7044 cannot drive TXB0304?

We want to use this level converter in various other chips(HMC6300, HMC6301, AD9680 etc) for the SPI. In most cases the drive current capability of SPI is not mentioned in the datasheets. Can you please throw more light on this?

  • 0
    TI__Guru* 97046 points
    Hi Nutan,

    You definitely need at least 3mA of drive current to overpower the TXB0304 outputs (all inputs are also outputs).

    I think the TXS0104E might be a better fit for your application. It uses a more passive translation method that should work well for standard SPI.
  • 0
    Guru 316780 points
    SPI uses only unidirectional signals, so you do not actually need a translator with automatic direction sensing.

    There are many possible unidirectional level shifters, depending on the voltages involved.
  • 0 in reply to Clemens Ladisch
    Prodigy 30 points

    The chips that we are using (HMC7044 and AD9680) both have SDATA pin as a bidirectional interface.  

  • 0 in reply to Emrys Maier
    Prodigy 30 points
    I understand. Couple of points:
    1. In AD9680 eval board TXB0304 has been used and ADI claims it is working. Note that the drive capacity of AD9680 as mentioned in the datasheet is
    LOGIC OUTPUT (SDIO)
    Logic 1 Voltage (IOH = 800 µA)
    Logic 0 Voltage (IOL = 50 µA)
    How do we explain this?

    2. We have an architecture where single SDIO (bidrectional) is routed to various Chips in parallel. If we use TXS0104E since by default TXB0304 are pulled up with a 10K ohm resistor internal to the chip, which effectively when placed in parallel will not be 10K ohm. This might create a problem.

    Any suggestions to solve the issue would be great!
  • 0 in reply to Nutan Reddy
    TI__Guru* 97046 points

    The primary concern with driving TXB0304 is the ability to overdrive the buffer's output. There is a 1kΩ resistor in series with each output that limits the current driving capability of the device.

    For example, if your device is outputting 3.6V (the max value) from the B port, then the B port is driven low by the device at the distant end, the 1kΩ resistor now has a 3.6V drop across it and it produces 3.6mA of current. If the distant end device can only sink, for example, 1mA of current, then the resistor will only have a drop of 1V, and the output at A won't be switched to a low state.

    If this works in your system, it is likely because the output current capability of the driving device is actually larger than what it is rated for on the datasheet (which is very common).

    If you are translating with the TXS0104E, it does use internal 10kΩ pull-ups to achieve logic 'high' voltages, but I'm not sure why that's a concern for you. You mentioned parallelling the TXS0104E... is this device being used also as a multiplexer? Why are there going to be multiple channels in parallel?