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.

TMDS361B: Compliance Testing

Part Number: TMDS361B

I have a customer using the TMDS361BPAGR to bring in 3 HDMI signals to a single HDMI receiver chip Analog Devices ADV7611.  When he sent his product through HDMI compliance he had one failure when they ran the HDMI Sink Jitter test 8-7 at 27Mhz (480P resolution).  The console passed testing at higher resolutions of 720P and 1080P.  This was done using a “TYPE 2 Cable Emulator” that adds skew into the signal. Although he suspects the issue is at the ADV7611 side of the design, there is potential the TI part is contributing to the failure.  He has had several calls with the compliance lab and Analog Devices on this and there were some questions that came up on the TI part.  Could you please take a look at these questions below?

 

  1. Did the TMDS361B chip go through HDMI compliance testing?
  2. If it was HDMI tested, would a report be available for the test?
  3. If it was HDMI tested, would the evaluation board that was used still be available?  We would like to run a test with the TMDS361B evaluation board connected to the ADV7611 evaluation board using an HDMI cable to compare performance with our PCB design.
  4. Our design has the TMDS361B feed directly into the ADV7611 using traces on the PCB, as shown in the attached image.  Does TI see any issues with using the chip in this manner.

Thank you for your consideration!

 


  • Jarrett

    Unfortunately, we do not have the RX jitter tolerance compliance report for TMDS361B. 

    Can they share their jitter tolerance report with us? If jitter tolerance is passing at higher data rate, I would expect it to pass at lower frequency as well.

    Thanks

    David 

  • I believe that this might be what you are looking for.  Do we have an eval board they can test with?

     
     
     
    8-7:
    TMDS – Jitter Tolerance
    Video Format #1
    Vid. Format:
    720x480p @ 60
           TMDS clock:
    27
    MHz
    D_JITTER = 500kHz, C_JITTER = 10MHz
    1
    PASS/FAIL = (Jitter only on Clock)
    FAIL
    Failed with Type 2 (JAE) cable, DUT could not display the signal
    D_JITTER = 1MHz, C_JITTER = 7MHz
    1
    PASS/FAIL = (Jitter only on Clock)
    FAIL
    Failed with Type 2 (JAE) cable, DUT could not display the signal
    Video Format #2
    Vid. Format:
    1280x720p @ 60
           TMDS clock:
    74.25
    MHz
    D_JITTER = 500kHz, C_JITTER = 10MHz
    0
    PASS/FAIL = (Jitter only on Clock)
    PASS
    D_JITTER = 1MHz, C_JITTER = 7MHz
    0
    PASS/FAIL = (Jitter only on Clock)
    PASS
    Video Format #3
    Vid. Format:
    1920x1080p @ 60
           TMDS clock:
    148.5
    MHz
    D_JITTER = 500kHz, C_JITTER = 10MHz
    0
    PASS/FAIL = (Jitter only on Clock)
    PASS
    D_JITTER = 1MHz, C_JITTER = 7MHz
    0
    PASS/FAIL = (Jitter only on Clock)
    PASS

  • Jarrett

    Are they able to pass the test without the injected jitter (turn the jitter off)?

    Does the sink support 27MHz clock frequency?

    Thanks

    David

  • Jarrett:

        any update?

    Regards,

    Brian

  • Yes, the board supports 27Mhz for 480p60 operation.  When the jitter was applied for this resolution our board could not lock on the TMDS clock.  We have been working on some register setting changes to the ADV7611 using the same values their evaluation kit had when it went to be tested for compliance.  Some registers are related to equalization that would be appropriate for an HDMI cable, while our application has just PCB traces from the TI chip to the ADV7611, so we are not certain if there will be adjustments needed for the equalization values.  We will be shipping the console back to the lab for a 2 hour consulting test to see if we have resolved the issue.  We will also be logging information in case it still fails.

  • Waiting feedback from customer

  • The console was retested on 8/30/19.  It still has a failure at 480P.  The lab comment on this is:

    • Retested Input1 with JAE 27M CE and repeated video failure (unstable/flickering test pattern on display) at 0.6 Tbit Skew of Clock10MHz-Data0.5MHz Jitter Profile

    When I look at section 8-6 of the HDMI 1.4a testing spec, it appears to show that a 0.4Tbit skew would still be acceptable.  I only see the lab tested at 0.6Tbit of skew, so I have sent in a question about the 0.4Tbit.

  • Steve

    Do you see the failure if the jitter is off?

    Thanks

    David

  • David,

    The problem only shows up while jitter is applied for a 480P resolution.  There are no issues when jitter is not applied.  There are also no issues if at 720p or 1080p with or without jitter applied.

  • Steve

    If the problem only shows up when jitter is applied, then I don't believe this is a equalizer issue. 

    The TMDS361B will not clean up the jitter, so we need to take a look at the  ADV7611, to see if it has a CDR or PLL and change the bandwidth to better track and filter out the jitter.

    Thanks

    David

  • Steve:

       any more update?