• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 33 subscribers
  • Views 286 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DS90UB914Q-Q1: Camera frame broken/bouncing happen during 8kV contact discharge ESD.

David Trinh
Intellectual 290 points
Part Number: DS90UB914Q-Q1

Dear Team,

I have a question related to Power Over Coax with DS90UB914 SerDes chips.

We have a camera input ECU using TI DS90UB913/4 SerDes pair with PoC. 

Our problem is that we are observing random camera frame bouncing/jumping during 8kV contact discharge ESD.  The problem occurs in below scenarios

-  8kV ESD Discharge at board's metal case (any location)

-  8kV ESD Discharge at camera metal case (any location; camera is 3 meter far from ECU)

-  8kV ESD Discharge at board power input's GND wire.

Our question is: why is this happening and what should we do to fix this.

(So far, our conclusion for this is because the ESD discharge cause ground bouncing and hence causing the ground potential difference between ECU and Camera which then cause interruption in the SerDes link and frame broken as a result. )

We look forward to your kind feedback regarding the mentioned issue.

Thank you and best regards.

David

  • 0
    TI__Expert 5480 points

    Hi David, 

    I will need some time to look into this. I will get back to you next week. 

    Thanks, 
    Sally 

  • 0 in reply to Sally Cheung95
    TI__Expert 5480 points

    Hi David, 

    From a hardware standpoint, make sure you are following the layout guidelines in the datasheet, using ESD protection (ESD diodes etc.), and making sure you have multiple grounding points.  Good grounding is really important when dealing with ESD issues.  From a software standpoint you can increase the parity error threshold, registers 0x18 and 0x19.  This will result in more pixel errors but may help defend against losing lock.

     Also plastic casing typically works better than metal casing when protecting against ESD, which you can look into changing.  I would suggest the case also be grounded at multiple locations as well. 

    Thanks, 

    Sally 

  • 0 in reply to Sally Cheung95
    Intellectual 290 points

    Dear Sally,

    Thanks for your kind reply.

    Firstly, our board is using a wide ground ring running around the whole PCB edge and this ring is clipped by the metal case at both sides. I assume this grounding meets your guide.

    Secondly, I will change the registers as your suggestion and get back to you after I finish the test.

    Moreover, we are in near production phase and changing to plastic case seems to be impossible.

    Lastly, we have been testing the Hardware intensively without any hard failure. Only the random camera frame bouncing issue is happening (2,3 bounces/broken frames per 10 shots). 

    If you have any other guides, I would highly appreciate it.

    Cheers,

    David

  • 0 in reply to David Trinh
    TI__Expert 5480 points

    Hi David, 

    let me look into this more. i will come back with a reply in a few days. 

  • 0 in reply to Sally Cheung95
    TI__Expert 5480 points

    Hi David, 

    Did that software suggestion improve anything? I am still discussing to see if there are any other last minute changes you can make since your project is close to production. 

    Thanks, 

    Sally 

  • 0 in reply to Sally Cheung95
    Intellectual 290 points

    Dear Sally,

    Huge thanks again for the kind supports.

    We are still struggling with the issue. I have tried your software tweak and it seems like nothing improved (Still some random bouncing from time to time)

    So far, the only scenario that could prevent the bounce was to cover the whole ECU with a plastic cover (which won't be available in mass production) to prevent the discharge from happening to the metal case. 

    We have tried everything and the "always true" symptom is if ESD discharge happens at any ground point (ECU side/camera side/power supply side) then bouncing will occur. (We also observed some bouncing at 4kV discharges as well).

    Cheers,

    David

  • 0 in reply to David Trinh
    TI__Expert 5480 points

    Hi David, 

    If the software suggestion doesn't work for you, unfortunately there are only physical/hardware changes that we can suggest. This includes making sure all components properly grounded, using a case that optimizes the space to minimize the air gap, making sure your design is in line with your datasheet layout guidelines, using different casing, ensuring there are multiple ground points. 

    Thanks, 
    Sally