• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 19 subscribers
  • Views 864 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.

TXS0108E: output is easy to have noise

Howard Zou
Genius 12365 points
Part Number: TXS0108E
Other Parts Discussed in Thread: TMS320F28034

Hi,

my customer is using TXS0108E to translate 3.3V SPI to 5V SPI.

3.3V is connected to TMS320F28034, 5V is connected to an ADC AD7705.

We tried to measure the CLK signal with scope.

And we find that the input (A6 pin) is rather stable, but B6 pin is easy to be influenced as shown below.

When B6 pin should be high, we touch the pin with multimeter, and there will be noise on the waveform (red rectangle part).

But we also touch A6 pin with multimeter, there is no noise.

And to ensure it's not related to ADC, we have disconnected ADC to TXS0108E, and the waveform is still easy to have noise.

  • 0
    TI__Mastermind 44775 points

    Hey Howard,

    Just to confirm this "noise" that is happening when you touch the measured pin with a DMM?

  • 0 in reply to Dylan Hubbard
    TI__Genius 12365 points

    Yes, 

    When we touch the pin with a DMM, and the waveform will not be clean.

    Although we haven't captured the noise big enough to make logic high to low or low to high, there will be communication error when touch the pin with DMM.

    If we don't touch the pin with DMM, there will be no communication error.

    The same happens to DIN and DOUT pin, not only the clock pin.

  • 0 in reply to Howard Zou
    Guru 284830 points

    Can you zoom in on the noise?

    Are there long traces, cables, or connectors on the 5 V side?

    When the signal is high, the TXS does not drive the output, and there are only the pull-up resistors (4 kΩ || 10 kΩ = 2.9 kΩ). I suspect that outside noise is coupled through the DMM leads acting as an antenna.

    Why are there external 10 kΩ resistors in addition to the TXS's internal ones?

  • 0 in reply to Clemens Ladisch
    TI__Genius 12365 points

    Clemens,

    it's just that their original SPI without level shifter have pull up resistor on SPI lines.

    I guess it doesn't matter if they have the pull up resistor connected, right?

    The trace on 5V side is only 20mm.

    Do you have any idea how to make the output side of TXS0108E resistant to the noise?

    Or do we have any device pin to pin compatible with TXS0108E that is more resistant?

  • 0 in reply to Howard Zou
    TI__Mastermind 44775 points

    Hey Howard,

    I'm not really sure why this is an issue as you won't be probing this lines with a DMM in normal operation. Its plausible that this DMM introduces this noise, but even so it doesn't seem to be that drastic. Though, its hard to tell since all of the important info has been cut out of the screenshot like the X and Y scaling, BW, etc.

  • 0 in reply to Dylan Hubbard
    TI__Genius 12365 points

    Dylan,

    thanks. 

    We are probing the lines with DMM because:

    there will be a unpluggable analog signal applying to ADC AD7705, they tried to plug and unplug the signal, and the signal on SPI line behaves the same as we probe the lines with DMM, and there will be communication error.

    And we tried to remove the analog signal and AD7705, and then we probe the lines with DMM, the waveform is still the same.

    That's why we got the waveform of SPI when probing the lines with DMM.

    So need your help on what information needed to help debug the issue, so we need a zoomed noise waveform with X and Y label scaling, bandwidth of the scope. What else do we need to do?

  • 0 in reply to Howard Zou
    TI__Mastermind 44775 points

    Hi Howard,

    The amplitude of that noise looks small, seeing the Y scale will help me verify. This doesn't look like it should cause issues with the digital communication since it likely stays above the required VIH(min) of the following device.