• Locked Locked
  • Replies 2 replies
  • Subscribers 34 subscribers
  • Views 264 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.

DS100DF410: SI failure issue (Reopen at public forum)

Dave Chang
Expert 5475 points
Part Number: DS100DF410
Other Parts Discussed in Thread: USB2ANY

Hello Lucas,

This thread is for Lanner's DS100DF410.

I open a thread at public forum is that Lanner will also follow on this case through E2E with you, thanks.

After trying and error to fine tune the SoC NXP TX pattern through the retimer to measure SI; below are the measuring results.

The NXP TX pattern waveform has been tried with the best configuration, you may see the waveform is better than through retimer one.

Lanner has been fine tuned the retimer amplitude/ de-emphasis by SW, but it seemed no response from the settings.

Furthermore, they’ve consulted SI engineer about the XFI SI measuring issue of TI retimer, it was once fine-tuned by TI US FAE, which is by EEPROM setting via retimer EVB with SigCon tool.

A few questions need to resolve:

  1. Does EEPROM is the MUST with DS100DF410?
  2. Why original TX pattern waveform is better than retimer output? How to fine tune it?

Looking forward your reply soon, thx.

Best Regards,
Dave

  • TI__Expert 5475 points

    I attach some figure here:

    SI measuring setup by differential probe:

    NXP SoC TX pattern:

    After TI's retimers result:

    Schematic: do not reserve for EEPROM:

    TI’s retimer SI is fine tuned by retimer EVB with tool SigCon, and to burn the image settings into EEPROM :

  • in reply to Dave Chang
    TI__Mastermind 36523 points

    Hi Dave,

    Lucas is out of office and can follow up later this week.

    In the meantime, I have a couple comments/questions.

    Dave Chang said:
    Does EEPROM is the MUST with DS100DF410?

    The question is not clear to me, but it sounds like the customer is asking if using EEPROM is a requirement for DS100DF410?  Using EEPROM is not a requirement for DS100DF410.  However, the customer does need some way to configure the retimer.  This can either be an I2C controller in their system, or an EEPROM.

    Please help clarify if I misinterpreted the question.

    Dave Chang said:
    Why original TX pattern waveform is better than retimer output? How to fine tune it?

    It is unusual that the TX amplitude from the retimer is small.  I think the customer should target a larger TX amplitude in order to increase the eye opening.  You mentioned that the customer has tried adjusting the TX amplitude, but has not observed change.

    Can you please share the steps that the customer is taking to configure the retimer TX amplitude?

    If the customer is using SigCon Architect with a USB2ANY to directly configure the retimer, they need to select the TX amplitude in the GUI, then press "apply to channel".

    Thanks,

    Drew