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.

LMK5C33216: TOD Accuracy

Part Number: LMK5C33216

Tool/software:

Team,

When we using LMK5C33216 we noticed that the TOD accuracy should be up to 7.5ns. Our system design requires ps level TOD accuracy, by any chance is there configuration method to meet ps level TOD accuracy? Or is there any other DPLL could meet this? Thanks.

Best,

Qiang

  • Hi Qiang,

    I will need more details:

    1. To clarify, is this the LMK5C33216 or LMK5C33216A device?
    2. Can you please provide more description of your system design with the LMK5C device? Such as a block diagram, required inputs and outputs...
    3. Please provide your TICS Pro configure file (.tcs).
    4. Are you using IEEE 1588 PTP in your system?

    Regards,

    Jennifer

  • Jennifer,

    • It's LMK5C33216.
    • System design as below:

    • TCS file we are preparing and should update later
    • 1588 or not: Will not use PTP

    Best,

    Qiang

  • Hi Qiang,

    Thank you for sharing the block diagram.

    I will review and get back to you later today or tomorrow.

    Regards,

    Jennifer

  • Jennifer,

    Thank you for support. Also we got more questions of the TOD module:

    Below is some more question from our side:

    1. Does the Freq Monitoring function only set flag or the specific measurement value can be read?

    Is there any equation for the relationship between measurement delay and accuracy;

     

    1. Will TOD counter be locked until next trigger event;

     

    1. What will be the error for the Elapsed time if we use the nth captured ToD value(n>2)

    Elapsed Time = (2nd captured ToD value - 1st captured ToD value) / ToD Clock Rate

    Best,

    Qiang

  • Hi Qiang,

    1. We strongly urge our customers to use the LMK5C33216A version instead of the LMK5C33216. The "A" version is the latest silicon version of the LMK5C series and has improvements made such as better jitter performance and SYSREF alignment. Additionally, the TICS Pro GUI of the A version has additional features added to improve the user development experience.
    2. Please note that we have renamed the ToD counter to the Time Elapsed Counter (TEC) because the LMK5Cxxxxx and LMK5CxxxxxA devices do not support a ToD counter. I see that the LMK5C33216 datasheet was not updated with this rename. I have made note in our document logging system to make this correction to avoid future confusion. From LMK5C33216A datasheet:
    3. The  highest output frequency that can be outputted from VCO3 (which has the highest accuracy time measurement) is 1228.8 MHz. If an output frequency of 1228.8 MHz is used, then the TEC clock period is around 817 ps. Would this meet your ps level requirement? If so, I would need to check internally to make sure this frequency can be used.

    Regards,

    Jennifer

  • Jennifer,

    Thanks for reply and we plan to use LMK5C33216A. 

    And double confirm with you:

    In the 33216A D/S we can see that the TEC clock runs at PLL3 VCO fequency/8, which should 2457.6/8, correct? 

    Best,

    QIang

  • Hi Qiang,

    That is correct. However, I am also in the process of double checking whether the TEC clock can be increased (beyond 2457.6/8) or if it is stuck at 2457.6/8. Let me get back to you tomorrow or early next week.

    Regards,

    Jennifer

  • Hi Qiang,

    I have confirmed with my team that the TEC clock is fixed to the PLL3 VCO frequency /8 or PLL2 frequency / 20.

    Regards,

    Jennifer

  • Thanks Jennifer.

    Best,

    Qiang