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.

TXS0104E: TXS0104E Issue

Part Number: TXS0104E
Other Parts Discussed in Thread: TXS0102

Hi Team,

We penetrated our TXS0104E for customer to work as 1.8V to 5V I2C level-shift. Below is customer's schematic:

Below is the waveform from both 1.8V side and 5V side, data and CLK waveform:

C1(yellow curve): 1.8V side data, C2(pink curve),:1.8V side CLKC3(Blue curve): 5V side CLK, C4(Green curve),:5V side data

As I highlighted in the waveform, the 5V data signal seems affected a lot from the 5V SCK falling and rising edge. But in customer's PCB, the 5V data and SCK pcb trace is very short and no connect between each other. Can expert help analysis why the 5V data waveform seems corss-talked with 5V SCK? How to avoid or solve this cross-talk?

Thanks a lot for your support on this urgent issue!

Best regards,

Sulyn

  • Hi Sulyn,
    I have asked our translation expert to take a look at this. He should get back with you in the morning.
  • Hi Emrys,

    Thanks for the support! As I will go to customer side for support, it's better we can have some solution or root cause analysis, that will be great helps.

    Best regards,

    Sulyn

  • Hi Sulyn,

    couple of points here on the schematic.

    1) I do not see bypass caps on the supply ports. Having bypass caps close to the supply pins are recommended to remove the noise and also helps in providing instantaneous current(prevents voltage drooping).
    2) notice that the 2 ports are unused. consider switching to TXs0102 device
    3) I recommend to remove the external pullup resistors on the ports. Please read the app note on effects of external pullup resistors here.
    www.ti.com/.../scea054a.pdf

    on the crosstalk issue, there will be coupling between channels due to switching on other channels. In order to minimize it, having isolation on the layout is recommended. from the signal scopeshot shown here, I do not believe that this could cause data corruption on the downstream device as the peak to peak signal variation is not crossing the threshold.