• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 2 answers
  • Subscribers 19 subscribers
  • Views 321 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.

Original question:

LSF0204: Threshold voltage of channel and pull-up resistor

LSF0204: Why does 3.3 V IO have a voltage drop?

Seunghwan Lee
Prodigy 20 points
Part Number: LSF0204

Hi TI,

Could you please let me know why PCIE_33_RST_N has a voltage drop?

I did the same test with TXS0104 but it didn't observe the below weird waveform.

Please refer to the below schematic/block diagram and waveform.

Schematic of LSF0204

Waveform: Voltage drop w/LSF0204

Waveform: NO Voltage drop w/TXS0104

Block Diagram (Simplified)

Thank you.

Best Regards,
Seunghwan Lee.

  • 0
    Guru 251950 points

    The LSF is a passive switch. The high-level voltage is generated only by the 10 kΩ pull-up resistor. The voltage is lower because some component is pulling the line low (with about 20 kΩ).

  • 0 in reply to Clemens Ladisch
    Prodigy 20 points

    Thanks for your comment.

    1. As I know LSF has FET but I have no idea why it's a passive switch. Do you mean is that acting like a passive switch?

    2. I'm just wondering why it's 20 K ohm.

    3. 'Some component' means Network Interface Card or something?

    Could you please explain in more detail?

    4. Some NICs have an internal pull-up on the PCIE_RESET_N line, but we don't know the value. It was also dropped.

     

  • 0 in reply to Seunghwan Lee
    Guru 251950 points

    1. The FET is a passive switch. When any input is low, the gate-to-source voltage of the FET is large enough, and it connects the pins on both sides.

    2. The 10 kΩ pull-up and the unknown pull-down form a voltage divider. The resulting voltage is about 2 V. (If there are other pull-ups on the line, then the pull-down also is stronger.)

    3. I don't know. I just see the effect.

  • 0 in reply to Clemens Ladisch
    Prodigy 20 points

    I thought you were talking about FET as a passive component. Sorry, I misread a word as 'a passive switch'.

    Then, why TXS0104 does not cause a voltage drop? It also has an internal 10 K pull-up resistor.

  • 0 in reply to Seunghwan Lee
    Guru 251950 points

    here, "passive" means that the I/Os are not buffered.

    I do not know why the TXS behaves differently; the LSF and TXS have the same architecture (the TXS just adds internal pull-ups and edge accelerators).

  • 0 in reply to Clemens Ladisch
    Prodigy 20 points

    Thanks for your answer.

    I've tested without NIC and It was the same result.

    It doesn't have anything to do with NIC.

    So, please ignore the block diagram.

    LSF0204; Added an external 10K pull-up on PCIe_33_RST_N (Still has a large voltage drop (without NIC))

    TXS0104(Internal 10K pull-up); No external 10K pull-up on PCIe_33_RST_N (without NIC)

    Still wondering why LSF has a large voltage drop even though there's no connection on the B side's IO.

  • 0 in reply to Seunghwan Lee
    TI__Mastermind 24030 points

    Hi Seunghwan,

    When you say no connection on the B side's I/O, do you mean the PCIE connector is disconnected as well?

    Note that your schematic seems to indicate the device has no outputs at all, as PCIE_RST and PCIE_33 seems to be both going into the device. Please help confirm for bus contention as the TXS case also seems to be dropping after a longer while, thanks.

    Best Regards,

    Michael.

  • 0 in reply to Michael Ikwuyum
    Prodigy 20 points

    Hi Michael,

    Thank you for your comment.

    No connection on the B side's I/O means the PCIE connector is disconnected.

    There's nothing connected to B-side.

    I wonder why LSF0204 has a voltage drop on B-sides at the first waveform.  

    Because the TXS series does not have a voltage drop. (Please refer to the above capture, RED circle)

    It's okay to be dropping after a longer while.

    Thank you.

    Best Regards,

    Seunghwan Lee.

  • 0 in reply to Seunghwan Lee
    TI__Mastermind 24030 points

    Hi Seunghwan,

    Note that the TXS has edge accelerators which speeds up the edges. Would the host happen to have internal pull-downs? 

    Best Regards,

    Michael.

  • 0 in reply to Michael Ikwuyum
    Prodigy 20 points

    Yes, the host has an internal pull-down.

    But I don't know the specific value.

    I knew the TXS series have an edge accelerator. 

    But I thought the edge accelerator doesn't affect dropping voltage.

    If the level translator has edge accelerators, it influences to the voltage drop?

  • +1 in reply to Seunghwan Lee
    TI__Mastermind 24030 points

    Hi Seunghwan,

    No, the edge accelerators may speed the rising edge while working with the internal 10K of the TXS device. 

    It seems the 10K is stronger than the host's pull-down. However, the LSF do not have pull-ups. You may add pull-ups on A1 to help pull-up, thanks.

    Best Regards,

    Michael.

  • 0 in reply to Michael Ikwuyum
    Prodigy 20 points

    Thank you, Michael.

    It helps a lot!