• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 34 subscribers
  • Views 572 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.

SN75DP130: Issue of using long cable

masahiko shimizu
Guru 11235 points
Part Number: SN75DP130
Other Parts Discussed in Thread: TDP142

Hi,

My customer has issue.
I have attached file of circuit.

If customer uses 1m cable this circuit, it works well. But if customer uses 2.5m cable, it doesn't work.
Could you check this circuit, please?
Especially, Is it no problem at AUX line?

SN75DP130.pptx

Best regards,
Shimizu

  • 0
    TI__Guru**** 176681 points
    Shimizu-san

    Linear redriver, such as the TDP142 would work better in this design.

    For the source board, does the main lane have AC coupling caps? If so, then AC coupling caps are not needed on the source buffer board.

    How does AUX structure look like on the source board? Typically, only AC caps between source and DP130, and 1M pullup/pulldown is on the AUX_SNK side.

    For the DP130 on the LCD buffer board, please refer to the DP130 in sink implementation app note: www.ti.com/.../slla349.pdf.

    Please check the AUX common mode voltage on both AUX_SRC and AUX_SNK and make sure they are correct.

    Do they have a AUX monitor that can log the link training process between the source and LCD?

    Thanks
    David
  • 0 in reply to David (ASIC) Liu
    Guru 11235 points

    Hi David,

    Thank you for your support.
    I will confirm your question to my customer.
    During waiting of my customer's reply, I confirm about better circuit of this condition.
    Please see the attached file of page 2.
    Is this circuit better?
    This circuit is my idea. I want to confirm correct circuit in this condition.

    4682.SN75DP130.pptx

    Best regards,
    Shimizu

  • 0 in reply to masahiko shimizu
    TI__Guru**** 176681 points
    Shimizu-san

    Your diagram is correct.

    Thanks
    David
  • 0 in reply to David (ASIC) Liu
    TI__Intellectual 1350 points
    David,

    Why you recommend TDP142 rather than DP130? The customer has a chance to replace TDP142 if it can get more good performance. Please let’s know why.

    Best regards,
    Izumi Maruyama
  • 0 in reply to David (ASIC) Liu
    Guru 11235 points

    Hi David,

    Thank you for your support.
    It is very helpful for us.
    I answer your previous question below.

    >Linear redriver, such as the TDP142 would work better in this design.
    ->Could you let me the reason of your recommendation, please?
        I think we shoud adjust EQ level through IIC in using DP130 as sink.
        Is this my idea correct?
        In case of below, Should we use TDP142?
           - we cannot adjust EQ level through IIC at sink side DP130.
           - we can use different cable.

    >For the source board, does the main lane have AC coupling caps? If so, then AC coupling caps are not needed on the source buffer board.
      ->The main lane don't have AC coupling caps on the source board.

    >How does AUX structure look like on the source board?
      ->This showed the ppt which is attached file at previous. 
         No pull up/down and no AC caps.
         It just connect CPU on source board.

    >For the DP130 on the LCD buffer board, please refer to the DP130 in sink implementation app note: www.ti.com/.../slla349.pdf.
      ->Thank you for your recommendation.

    >Please check the AUX common mode voltage on both AUX_SRC and AUX_SNK and make sure they are correct.
      ->My customer is checking this now.

    >Do they have a AUX monitor that can log the link training process between the source and LCD?
    ->Unfortunately, they don't have this data.
       Could you let me know how to monitor this data, please?
        -> Should we monitor AUX line with measuring instrument?
            Or Should we implement software for debug to CPU?

    Best regards,
    Shimizu

  • 0 in reply to masahiko shimizu
    TI__Guru**** 176681 points
    Shimizu-san, Izumi-san

    DP130 is a limited re-driver while DP142 is a linear re-driver. While both limited and linear re-driver has equalizer on the RX, the key difference between limited and linear re-driver is that limited re-driver has the output swing and pre-emphasis control on the TX while linear does not.

    When using a limited re-driver, there could be an inter-op issues because the link training configuration being requested by the sink is based upon the signal path between the limited re-driver and the sink, it does not include the signal path from the source to the limited re-driver.
    For example, when the sink requests pre-emphasis and voltage swing level based upon the output of the limited re-driver, both limited re-driver and source would adjust their pre-emphasis and voltage swing to the same level based upon the request from the sink. While the pre-emphasis and voltage swing may be sufficient for the signal path between limited re-driver and sink, it may not be sufficient for the signal path between the source and limited re-driver. The signal could get either over-equalizer or under-equalized between the source and limited re-driver. Problem gets worse with long cable and if there are multiple limited re-drivers. The way to work around this issue is to disable link training and manually set RX and TX as described in the DP130 app note. The other way to work around this issue is to use DP142 linear re-driver as the entire signal path from the source, through the linear re-driver, to the sink, is seen by the sink.

    Thanks
    David