• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 305 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.

TUSB1104: fail at CP10

Tobias Solisch
Prodigy 35 points
Part Number: TUSB1104

Hello everyone,

I have developed a prototype of an active cable that uses the redriver TUSB1104 at both ends. I have different cable lengths from the prototype. With the help of an oscilloscope and a USB 3.2 device, I then look at the eye diagram for a data rate of 5 Gbit/s and 10 Gbit/s. I use the compliance patterns of CP0 and CP9. This works pretty well for all lengths and the eye diagrams look very good. At a certain length, however, the CP10 pattern no longer works and one of the re-drivers then fails or no longer reacts to the lfps ping and the power supply drops. What could be the reason for this?

Best regards,

Tobias

  • 0
    TI__Mastermind 30100 points

    Hi Tobias,

    I hope you're doing well. Have you tried using CP1 and did the cable pass this pattern ok?

    • I am wondering this because CP10 and CP1 are both clock signals. It would be good to know if only the clock signal patterns are causing the cable to fail

    From what I remember, you're making an active cable with two 1104s in series correct?

    Best,

    Shane

  • 0 in reply to Shane Hauser
    Prodigy 35 points

    Hi Shane,

    I am fine thank you for asking:) Hope you are doing well too.

    Yes the cable did pass USB 3.2 Gen1 Compliance test, so yes CP1 works.

    Yes I am using two 1104s in my cable. For all my prototypes I am using the equalization CEQ (at the far end) and if necessary, like the cable reaches a specific length, I also use SSEQ to boost the signal at the near connector. This works pretty good but after a specific length one redriver like shuts down. Are there like any timing requirements? I don't think that the voltage is to high it should be inbetween +-600 mV. 

    Best Regards,

    Tobias

  • 0 in reply to Tobias Solisch
    TI__Mastermind 30100 points

    Hi Tobias,

    I have a few questions to get a better idea

    1. Do all compliance patterns fail at the length where one re-driver shuts down or is it only CP10?

    2. What length and loss (in dB) is the cable that you see the failure on? I'm wondering if the cable is long enough that the RX detect on one 1104 isn't seeing the terminations on the other 1104. You can see in the datasheet that RX detect will shut the 1104 down if id does not see terminations or is disabled:

    Tobias Solisch said:
    At a certain length, however, the CP10 pattern no longer works and one of the re-drivers then fails or no longer reacts to the lfps ping and the power supply drops.

    3. Does the 1104 work, and then fail after a certain time limit? Or does it never turn on at the length you are describing?

    Tobias Solisch said:
    Are there like any timing requirements?

    There are timing requirements in the datasheet, but I don't believe they would cause the 1104 to shut down after it had been working.

    Best,

    Shane

  • 0 in reply to Shane Hauser
    Prodigy 35 points

    Hello Shane,

    I will send you a message with all the informations.

    Best Regards,

    Tobias

  • 0 in reply to Tobias Solisch
    TI__Mastermind 30100 points

    I'll follow up on this in our direct message.

    Thanks,

    Shane

  • 0 in reply to Tobias Solisch
    TI__Mastermind 30100 points

    Hi Tobias,

    I hope you're doing well. From our direct message, I believe you were able to fix the issue with a different EQ setting.

    I'll mark this as resolved for now. Let me know if you have any questions and the thread will re-open.

    Best,

    Shane