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: Data speed restriction

Part Number: TXS0108E
Other Parts Discussed in Thread: TXU0304

Hi team,

My customer is using TXS0108E and it is used to transfer SPI data. But the data speed is restricted to 600k .

Do u have any advice on this case? The SCH is as below.

  • This schematic is incomplete. I'd guess that there are long traces, connectors, or cables that result in a large capacitive load.

    For SPI, you should use a buffered translator like the TXU0304.

  • Hi Clemens,

    Sorry for mistaking the application. TXS0108 is used to transfer the logic between the bi-directional signal of SD card. They don't have any connector and the traces are not vey long. But they still find that the rate is districted under 600k. 

    TXU0304 is  directional and can't used in this u have any better recommendation of higher rate signal products?

  • In theory, the TXS should support SD card signals.

    Please show an oscilloscope trace.

  • We test with an oscilloscope, the signal edge of the microcontroller is very good, after the conversion IC connected to the SD card signal edge slowed down, resulting in the signal frequency is adjusted to high read card failure, the frequency is reduced to within 1M on normal

    In the waveform below, the yellow is the waveform of 1.8V level at the input of the conversion chip, and the blue is the waveform of 3.3V level connected to the SD card after the conversion chip.

  • The output signal edges are not straight. This shows that the capacitance is larger than 70 pF.

    What is the total trace length, from the TXS to the SD card? Please show all components connected to these traces.

  • Hi Clemens,

    Add more components for your reference. Thanks!

  • Hey Zoey,

    Is it possible to zoom in on the I/O waveforms that you have captured above? (There seems to be a slow rising RC curve but its hard to tell from this scale). As Clemens' mentioned, this could be an effect from having too long of total trace length. Could you please confirm the length of the PCB traces on the left hand side of the device? 

    Can this anomaly be seen on other channels and/ or devices?