• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 554 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.

TDP142EVM: No image when used with Dell WD19TB Dock

James Hillman
Expert 1436 points
Part Number: TDP142EVM
Other Parts Discussed in Thread: TDP142

Hello, I am currently experiencing an issue with the TDP142EVM when used between a Dell  WD19TB  Dock and a monitor. The host (DisplayPort source) device (either a  Windows PC or Mac) is able to read the EDID of the monitor and detect the screen, but no image is visible and the screens soon goes to sleep after saying that no DisplayPort signal is being detected, any ideas why this might be the case? 

List of Jumper states:

J1: Low

J2: Low

J3: Open

J4: Low

J5: Low

J6: High

J7: High

J11: Open

J12 : DPPWR

J14: Closed

We originally saw the issue on our product but then moved to the evaluation board to isoalte the failure. Our displayport logs suggest link training is failing. 

Thanks,
James

  • +1
    TI__Guru*** 146055 points

    Did you change J4/J5 to see any difference.

  • +1 in reply to Brian Zhou
    Expert 1436 points

    Yes. No change. 

  • 0 in reply to Brian Zhou
    Expert 1436 points

    I tested this and had no change. This is still an issue 

  • 0 in reply to James Hillman
    TI__Guru**** 177301 points

    Hi,

    Attached is the DP142 EVM schematic, 

    1663.HSDC029A_Schematic.pdf

    • Pull J7 low
    • Double check the BOARD_3P3V and make sure it is 3.3V? Or if it possible, provide 3.3V externally instead from the dock itself? 
    • Pull SNOOPENZ high to disable AUX snoop and enable all lanes
    • What EQ values have you tried? Have you tried with a short DP cable between the dock and the EVM?
    • Can you also share the link training log file?

    Thanks

    David

  • 0 in reply to David (ASIC) Liu
    Expert 1436 points
    David (ASIC) Liu said:
    SNOOPENZ

    Disabling snoopenz did the trick!

    #1. Can you help me understand the impact of doing this? Is it just no power saving? I have already done a lot of compatibility testing but my gut feel is doing this shouldn't break an existing compatibility, agree? In my case power comes from a separate source not the DP source or sink, so power is not a problem.

    #2. Also, if you could potentially explain why this would have caused incompatibility then I'd love to learn that answers. Thanks for your help and time.

  • 0 in reply to James Hillman
    TI__Guru**** 177301 points

    Hi,

    1. The impact of disabling SNOOPENZ is only on the total power consumption number, it does not have any functional impact. 

    Base on the snooping the AXU bus traffic, the DP142 would turn off unused lane. For example, if the source and the sink communicates for 2 lanes, then the DP142 would snoop this and turn off lane 2 and lane 3 since they are not being used for power saving.

    2. Potential root cause,

    • If there is noise on the AUX bus, the DP142 could incorrectly turn off the wrong number of lanes. For example, the source and sink want to communicate for 4 lanes, but the DP142 turned off 2 lanes. In this case, the link training will always fail because the sink will never able to get link training to work on the 2 lanes that are being turned off. 
    • You could also have wrong physical connection of the lanes. For example, DP142 lane3 is physically mapped to lane0. If lane3 gets turned off in the 2 lane case, then logically lane 0 data will never received by the sink.

    Thanks

    David

  • 0 in reply to David (ASIC) Liu
    Expert 1436 points
    David (ASIC) Liu said:
    You could also have wrong physical connection of the lanes. For example, DP142 lane3 is physically mapped to lane0. If lane3 gets turned off in the 2 lane case, then logically lane 0 data will never received by the sink

    Most excellent response. How would I check if we did this? I have tested with perhaps 8 displays and 8 sources in most combinations, would it be surprising that so many would work if this error had been made.

  • 0 in reply to James Hillman
    TI__Guru**** 177301 points

    HI,

    I would agree with your assessment. You can probe the TDP142 output and see if the wrong lane gets turned off.

    The other option is to look at the TDP142 power consumption between working and non-working case. For 4 lane DP active mode, the power is around 660mW.

    Thanks

    David