Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

TUSB1142: TUSB1142 fail USB compliance test TD 7.31 Hot reset failure test

Part Number: TUSB1142
Other Parts Discussed in Thread: HD3SS3220

Tool/software:

Hi TI guru,

We perform USB compliance test on our board with TUSB1142 Redriver. (We used pin strap mode on TUSB1142.)

When we carry out USB compliance test, we fail on TD 7.31 Hot reset failure test with the failure message mention "The PUT without a captive re-timer doesn't transmit at least 16 polling LFPS bursts."

We perform some of the DOE, we remove TUSB1142 and connect it with jumper, by pass TUSB1142 and it able to pass the test.

USB compliance test FAIL with TUSB1142, we notice the electrical Idle duration was 6.528ms, and Rx.Detect ==> Polling LFPS only having 14/12 package.

USB compliance test PASS when we remove TUSB1142 and replace it with jumper (bypass), the electrical Idle duration was 142.024us, and Rx.Detect ==> Polling LFPS having 34/32 package.

What is going wrong here?

Thanks,
Best regards,
Alder

  • Hi Alder,

    This failure is due to the RX detect process of our re-driver consuming some of the LFPS bursts sent by your host in the polling stage. We have seen this before and it should not impact the functionality of the USB port.

    The redriver between the hub and the USB connector may not propagate some of the initial 16 consecutive Polling.LFPS bursts while the redriver is still performing Rx.Detect. This may cause less than 16 consecutive Polling.LFPS bursts to be seen by the USB exerciser. As a result, the TD.7.31 Hot Reset Failure Test (Downstream Port Only) fails. Even though some LFPS bursts are consumed, enough should pass through to successfully exit the polling stage. Please see this waiver request template for more details: Waiver_Request_TUSB1142_TD7_31.docx

    Does the USB port work functionally? If so, you can submit a waiver to the USB-IF using the template included above.

    Best,

    Shane

  • Thanks Shane.

    I have a follow up questions.

    1)

    We have another project used HD3SS3220 and will going for compliance test in a week time.

    We want to know is HD3SS3220 will be the same case here or HD3SS3220 will not see this issue?

    Shorter RX detect time?

    2)

    Although we do not see any functional issue but the team here is a bit reluctant on applying a waiver. Could TI recommend a replacement redriver/mux for TUSB1142 which can achieve our goal?

     

    Thanks,
    Best regards,
    Alder

  • Hi Alder,

    1. The HD3SS3220 is a passive MUX with no re-driving or RX detect functionality. Because of this, you will not have the same issue with the HD3SS3220, (unless there is a separate re-driver in the signal path).

    2. The LFPS consumption is a side effect of our RX detect process. Since all of our USB re-drivers use RX detect when establishing a connection, I don't believe we have a part that would perform better here.

    Are you able to increase the number of LFPS bursts from the USB host to account for the TUSB1142 RX detect process? Its possible that increasing the LFPS bursts will allow enough to pass through for this test.

    Best,

    Shane

  • Thanks Shane.

  • Hello Shane,

    We are thinking to obtain the waiver for TD7-31 however we would like to understand how likely the waiver being approved.

    Does TI successfully obtain the similar waiver for the TUSB1142 and all other redriver products?

    Also, it will be great that if you able to share on TI partner/customer who also do not have issue to obtain the waiver.

    Thanks,

    Best regards,
    Alder

  • Hi Alder,

    Does TI successfully obtain the similar waiver for the TUSB1142 and all other redriver products?

    Yes, we've had customers submit a waiver for this test and get approved. I cannot share which customers have obtained this waiver due to confidentiality concerns.

    Best,

    Shane

  • Thanks Shane