• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 33 subscribers
  • Views 429 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.

DS92LV2412 lock drop out

John Merrill
Intellectual 420 points
Other Parts Discussed in Thread: DS92LV2412

My DS92LV2412 is on a pcb to which a flex circuit from my LCD is attached.  When I test the output signals with the lcd detached all signals look normal (clk is 10.6MHz and LOCK is a steady High).

When I attach the LCD the LOCK pin is occasionally being pullled low.

I am under the impression that the LOCK pin is only affected by the input data stream being validated.  What could i apply to the outputs of the DS92LV2412 to cause the LOCK pin to be driven low?

  • 0
    TI__Mastermind 20820 points

    Hi John,

    Somebody is looking into this and will respond to this shortly.

    Regards,

    Mike

  • 0 in reply to Michael Peffers
    TI__Guru* 93670 points

    Hi John,

    Hope all is well.

    I am thinking out loud here but potentially there would be some potential root causes to this issue which we have to run perhaps some experiment to make sure we are on the right track:

    1). LCD panel draws large amount of current(800 to 1000mA) and in the in rush current could be even higher. Perhaps the switching noise due to LCD could be effecting the DS92LV2412 performance. If possible please isolate LCD panel power from the device power to see if we are on the right track.

    2). I am not sure if you are following the supply filtering especially PLL supply per data sheet. We need to use pie filter type filtering per data sheet and use large cap shunt across small cap as well.

    3). Another potential but less likely scenario could be some type of cross coupling between the parallel and the serial side. 

    In any case, the LCD panel seems to be injecting noise into the de-serializer to the point the signal integrity is sacrificed and we lose lock(or framing to declare lock).

    Regards,,nasser

  • 0 in reply to Nasser Mohammadi
    Intellectual 420 points

    Hi Nasser,

    I believe that issue 1 is the most likely at this point but to answer your questions:

    1. The LCD construction is 3 part, each circuit being routed through its own flex circuit.  These circuits are LED Back light, Resistive touch screen, and video signals.  The Back light and TS circuits are galvanically isolated from the Video circuits. The only point they meet at is at the power source (battery  or well regulated supply)  I ran an experiment to see if the TS or Back light circuitry had any affect on the operation of the DS92LV2412 with and without the LCD video signal FPC connected.  With the video FPC disconnected from the PCB the back light and touch screen circuits do not effect the lock status of the 2412.  With the video FPC connected the 2412 loses lock (some cases are much worse or better than others).

    2. I am following the filter circuit very closely, in fact i have more bulk capacitance on the front end of the filter than listed on the data sheet.  I have a 10uF and a 0.1uF cap in parallel in front of every ferrite and a 0.1uF cap after every ferrite on both the 3.3V and 1.8V rails. 

    3. I have different revisions of the LCD and some work and other do not.  This leads me again to something inherent in the display causing the issue and nothing i am doing in the display.