• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 34 subscribers
  • Views 101 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.

LMH1239: Regarding LMH1239 Output Behavior

Kyohei
Genius 3051 points

Part Number: LMH1239

Hi team,

Our customer is currently evaluating the LMH1239, and we would appreciate it if you could review the following inquiry.
--------------------------------------------------
We understand that, unlike previous SDI equalizers, the LMH1239 requires monitoring the CDR_Locked register, and if the device is not locked, it is necessary to perform an ENABLE pin override.
However, when switching frequently between HD-SDI and 3G-SDI formats, there are cases where, even though the CDR_Locked register indicates a locked state, the video output is not displayed correctly at SDI_OUT± (thru output) and OUT0±.
We attempted to fully dump the register data for both the working (OK) and non-working (NG) cases using a 1080/60i input, but we could not clearly identify the differences.
The only differences appear in the registers highlighted in yellow; however, these registers are not described in the programming guide.
At this time, all registers are configured with their default settings, except for input switching and disabling OUT1±.
Could you please advise whether there may be any incorrect usage, missing configuration, or additional considerations when operating the LMH1239 under these conditions?
----------------------------------------
image.png
 
Best regards,
Kyohei
  • 0
    TI__Guru**** 192051 points

    Kyohei-san

    Any chance you can share your schematic for review first?

    Thanks

    David

  • 0 in reply to David (ASIC) Liu
    Genius 3051 points

    Hi David-san,

    Thank you for your support.

    I have sent the customer’s circuit diagram and register configuration information via private message, so please review them.
    Best regards,
    Kyohei
  • 0 in reply to Kyohei
    TI__Guru**** 192051 points

    Kyohei-san

    Thanks for sending the schematic. They have the optional 470nF external loop filter capacitor connected between LF+ and LF for reduced CDR loop bandwidth settings. But LOOP_BW_SEL pin looks to be floating which means reduced CDR loop bandwidth is not required. So there is a mis-match between the two, in this case these pins should be left floating.

    Looking at the register map, register 0x06 reads a value of 0x36, so the CDR is locked. The data rate also read 1.5G, which matches with the expected HD-SDI data rate. 

    • Does this happen all the time or once in a while?
    • Also when they say video output is not displayed correctly, what does it show? Blank screen or correct video with some flickers?
    • When not displayed correctly, can they probe the OUT0+/- with a scope? Can they share the scope waveform?
    • Do they see the same issue when using our EVM?

    Thanks

    David

  • 0 in reply to David (ASIC) Liu
    Genius 3051 points

    Hi David-san,

    Thank you for your support.

    We have confirmed the details with our customer and would like to provide the following feedback.

    David (ASIC) Liu said:
    So there is a mis-match between the two, in this case these pins should be left floating.
    We evaluated the device with the external loop filter capacitor removed; however, the observed behavior did not change.
    David (ASIC) Liu said:
    Does this happen all the time or once in a while?
    The occurrence rate is relatively high. When switching between 1080 60i and 1080 59i, the issue occurs approximately once every five attempts.
    David (ASIC) Liu said:
    Also when they say video output is not displayed correctly, what does it show? Blank screen or correct video with some flickers?
    When the image does not display correctly, the signal itself is present, but it is not recognized as a valid video image.
    David (ASIC) Liu said:
    When not displayed correctly, can they probe the OUT0+/- with a scope? Can they share the scope waveform?
    We will send the waveforms for the 1080 60i OK case and NG case via private message. Since we were not able to prepare a differential probe, the measurements were taken on only one side of the signal.
    From reviewing the waveforms, we noticed that in the NG case, the signal interval appears to be shorter.
    David (ASIC) Liu said:
    Do they see the same issue when using our EVM?
    At present, we do not have an EVM available. If there are any other verification methods or checks you would recommend, we would appreciate your guidance.

    Best regards,

    Kyohei

  • 0 in reply to David (ASIC) Liu
    TI__Guru**** 192051 points

    Kyohei-san

    Can you please reach out to our local TI office and ask them to request LMH1239 EVM for you? I would like to confirm if the same issue can be observed on the EVM itself. 

    I will also wait for your scope waveform.

    Thanks

    David