• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 148 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.

DS320PR810: test failed issue

Hailiang Zhang
Expert 8155 points
Part Number: DS320PR810
Other Parts Discussed in Thread: SN75LVPE5412, DS160PT801

Hi Team,

My customer is testing DS320PR810 for a PCIe 5.0 cable. With the EQ turned off, the test specifications cannot be met. The curve obtained from the test is as below. And the picture of customer's cable is also attached.

There are some questions here.

  1. Please share the method or test guide for testing PCIE5.0 cables.
  2. What are the parameter indicators that need to be tested?
  3. What is the difference between linear redriver and limiting redriver? Is there any AN you can share with customers? Which one should be chosen for making this type of cable?

Regards,
Hailiang

  • 0
    TI__Genius 14840 points

    Hi Hailiang,

    Hailiang Zhang said:
    With the EQ turned off, the test specifications cannot be met.

    Can you tell us more about what the customer was doing?

    • What does "EQ turned off" mean?  Is EQ set to 0 or did the customer do something else?
    • Why did they want to turn the EQ off for this test? The redriver depends on EQ to improve signal quality by compensating for insertion loss, so if EQ is off then I expect the redriver cannot be helpful and results would not look good.
    • As I understand, the chart is for insertion loss (S_DD[2,1]) and I see the red line is the standard. What exactly is failing? I see that the red line ends at 24 GHz. Between 21 and 24 GHz, it looks like some of the TX/RX curves dip below the red line. Is this the problem?
    Hailiang Zhang said:
    Please share the method or test guide for testing PCIE5.0 cables.

    We don't have recommended test procedures specifically for PCIe active cables. We do have procedures for PCIe systems (motherboards, cards) but they focus on compliance testing and redriver tuning based on functional results from software. These could be difficult or not relevant for cable manufacturers.

    Hailiang Zhang said:
    What are the parameter indicators that need to be tested?

    Not sure about what "needs" testing but I have seen other active cable customers test insertion loss and crosstalk, in simulations as well as from lab measurements.

    Hailiang Zhang said:
    What is the difference between linear redriver and limiting redriver? Is there any AN you can share with customers? Which one should be chosen for making this type of cable?

    Correction: after further reading, a limiting redriver is a nonlinear redriver that has additional features compared to a linear redriver such as deemphasis control, however limiting redrivers are more likely to interfere with the PCIe link training process and are more difficult to use as a result. There is a Precision Labs video that explains the difference here: https://www.ti.com/video/6144123457001 (talks about DisplayPort but the concepts are very similar for PCIe).

    For the latest PCIe generations, TI has linear redrivers (DS320PRxx products), a retimer (DS160PT801, not for Gen 5), and redriver mux/demux (SN75LVPE5412/421, most likely not relevant for active cables). We do not have limiting/nonlinear redrivers for Gen 4 or Gen 5.

    Also, I want to note that this thread is on the E2E public forum. If any information about this case is confidential, I can ask an administrator to move this to the internal forum.

    Best,

    Evan Su