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.

LMK05028: Reference Input Monitoring

Part Number: LMK05028

Hi, I have a question regarding the reference input monitoring of the LMK05028 clock synchronization chip.

The data sheet explains that the following input monitors are included to evaluate the validity of the reference clock: Reference Validation Timer, Amplitude Monitor, Missing Pulse Monitor (Late Detect), Runt Pulse Monitor (Early Detect), Frequency Monitoring, and Phase Validation Monitor for 1-PPS Inputs.

My question is: when one of these monitors determines a reference clock is invalid, is there a way to detect which specific monitor caused the invalidation? E.G., If the Frequency Monitor invalidates the input because the reference input does not meet the frequency threshold, is there a way to determine that this was the exact reason?

Thanks,

Thomas

  • Hi Thomas, 

    Is this more for debugging purposes or for active usage? The reason I ask is the chip technically doesn't have anything specific that will tell you exactly what failed and when however in most cases I believe you should be able to gather all the information you need. 

    For debugging purposes, you can simply just enable only 1 of them at a time and provide the reference clock and see which one when/if reference becomes invalid. Then you can be certain that mechanism is the one invalidating your reference. 

    Other things to know - 

    • Validation timer is really the amount of time device needs to meet the other requirements for it to be valid. So validation timer would be used in conjunction with other mechanisms - for example frequency monitoring. For Validation time and frequency monitor to pass - the reference provided needs to be below the invalid ppm value provided for the length of time specified on the validation timer. 
    • 1 pps / phase detector cannot be used in conjunction with frequency monitor. 1 pps is really only for 1 pps signal and wouldn't be used for other frequencies (although I believe the threshold is 2 kHz and below).
    • The frequency monitor and the missing clock monitor might sound similar and both have to do with frequency drift essentially - but one is very fine (frequency monitor) and other is coarse (missing clock). For example if the reference drifts by 200 ppm the frequency monitor will trigger (should the invalid ppm be anything below 200) whereas this will not trigger the missing clock as there aren't any missing/runt pulses. Should the reference disappear all together then technically both would be triggered - however missing clock is a much faster measurement (as it's coarse) and would trigger immediately therefore that would be the invalidating factor. 

    So I think in all cases it should be clear essentially what got triggered that resulted in the invalidation. 

    Please also note that the status readback does provide: 

    1. LOR_MISSCLKx
    2. LOR_FREQx
    3. LOR_AMPx 

    LOR: Loss of Reference. So each flag above should get triggered if that's what causes the invalidation. 

    If there's a specific scenario where you don't think it can be covered by some evaluating please provide the details so I understand it better. 

    Thanks and regards,

    Amin