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.

LM98714: LM98714 temperature and jitter problems

Part Number: LM98714

I have two problems with the chip.

With some chips there is a greater jitter on the clock line to the image sensor and with others I have error detections due to temperature fluctuations below 15 degrees Celsius.

I can rule out cold soldering points relatively well in the cases because I have re-soldered the chip, but the error remains. I tested this with cold spray.
Already I have six boards with these errors. Five of them come from a production series. So the chips come from a roll.

Are you familiar with these problems?

Is the chip defective in these cases?

Do you know what can cause this?

  • Hi Steffen,

    Thanks for your question. Let me contact the expert for this part. You can expect a response by 1/29/20.

    Best,

    Grant

  • Hi Steffen,

    It is unlikely an LM98714 issue since multiple parts shows same issue, to my knowledge that's no history of clock jitter at low temp .

    Seems to me a system signal integrity issue, maybe the load changes at low temp

    My first suggestion is to check fist how clean input clock (INCLK) is.  Do you see a bigger than spec CLKIN to CLKOUT propagation delay variation?

    Temperature will affect the propagation delay, but delay will meet spec. Another suggestion for this issue is to change coarse pixel phase alignment 

    Regards,

    Costin

     

  • Hi Costin,

    I tested a few things.

    I have shown the input and output clock with an infinitely long afterglow time on the oscilloscope and good circuit board
     compared with bad circuit board. Unfortunately I can not find anything about propagation delay in the data sheet.

    Here are a few pictures at room temperature and cooled with cold spray.
    With cooled lm98714 there were phase jumps in the output frequency.

    Bad PCB with cooling (Input CLK yellow output CLK green)

    Bad PCB with room temperature (Input CLK yellow output CLK green)

    Good PCB with cooling (Input CLK yellow output CLK green)

    Good PCB with room temperature (Input CLK yellow output CLK green)


    Furthermore I tried different configurations of the lm98714. I tried to change coarse pixel phase alignment, I changed the polarity and the duty cycle of the output clock.
    Unfortunately, always with the same result as in the bad pictures.

    I also disconnected the CIS module. This did not result in any change either.


    Next I'm going to put the chip down and solder a new one onto it. If the error persists, I have suspected the board as the cause of the error. We have already built about 200 boards, but so far we have seven boards with this defect that we know of.

  • The re-soldering of the chip showed no improvement. It was only by soldering down and populating another chip that the error pattern was eliminated. Now I have no more problems with the jitter.
    But that's actually not a solution.