• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 19 subscribers
  • Views 1265 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.

TXS0108E: Wrong Output

Kuan Wei Chen (Kevin Chen)
Genius 9710 points
Part Number: TXS0108E
Other Parts Discussed in Thread: SN74AXC8T245, , SN74LVC1T45

Dear Team, 

Can you take a look, any reason cause the output wrong here?

The output side's 1.8V signal is wrong, as you can see. The 3.3V signal is good.

The SDIO Clock speed now is 400kHz at the beginning,  later 50MHz

They already try:

1. Tried to add an extra 10k pull high on AM_MMC1_CLK at 3.3V side, but it seems nothing changes 

BR

Kevin

  • 0
    TI__Guru 61940 points
    Kuan,

    The device doesn't guarantee 50 MHz operation. The datasheet mentions that the 3.3 to 1.8V translation min guaranteed max data rate is 60Mbps.
    This is at the rated loading conditions. It seems that the loading conditions at the output is higher than what the TXS can handle.
    Can you remove the 10k series resistors and probe the output to see the response?
  • 0 in reply to ShreyasRao
    TI__Guru 61940 points
    Also, you could consider using the SN74AXC8T245 device for 3.3 to 1.8V. It can support higher data rate ~150Mbps.
    I would not recommend having external pullups or pulldown on the TXS devices.
  • 0 in reply to ShreyasRao
    Prodigy 20 points
    I agree with you, loading is too heavy to TXS. TXS output is good when not wired to EVM board, I also checked TXS drive strength by adding a 6.8k pull down at 1.8V output, high voltage becomes 1.1V, this behavior comply with datasheet description.
    System works fine when using SN74LVC1T45 to replace TXS0108E on CLK signal, but this makes me worried about time violation. Original system has only 1ns time margin in worst case.
    Another question is why EVM board shows such a heavy load, maybe it is partly broken?? suffered ESD overstress??
  • 0 in reply to ShreyasRao
    Prodigy 20 points
    I agree with you, loading is too heavy. TXS output is good when not wiring to EVM board. I've checked TXS drive strength by adding a 6.8k pull down, high voltage becomes 1.1V, this behavior complies with datasheet description.
    System works when using SN74LVC1T45 on CLK signal, but this change makes me worry about timing violation. Original system has very little time margin.
    Another question is why EVM board be such a heavy load. It's partly broken?? Suffered ESD overstress?? We plan to keep both TXS and 1T45 option on sample run and check if heavy load is a common issue
  • 0 in reply to Sunjar CUI
    TI__Guru 61940 points
    Sunjar,

    What timing violation are you worried about the LVC1T45?
    With active drive devices like the LVC1T45, the specs mentioned in the datasheet should hold good with the rated cap loading.
    It seems that the output loading on the EVM is higher than normal.
    Is there external cap or ESD devices? ESD device can present cap loading. It is usually mentioned in the datasheet of the ESD device.