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.

CDCM61001: Clock input Requirements

Part Number: CDCM61001

Good day,

I am using a CDCM61001 to generate the 200MHz clock on a board. The reference i am using is a ABM3B-25.000MHZ-10-1-U-T. Attached you will find a schematic of the clock generation section.

The output clock I am measuring appear to be less stable than I would expect by jumping in frequency by 130kHz and the first thing I am suspecting is my reference clock which is rated 10ppm. I do not fully understand how the shunt capacitance C1 should be calculated and I have been unsuccessful to find any practical documentation that explains this design choice. I am also not 100% sure if the XO I have chosen is acceptable.

Would it be possible to explain to me how to choose a suitable reference oscillator and shunt capacitor based on the requirements of the CDCM61001? Any other insight as to why my clock isn't as stable as expected would also be appreciated.

Thank you.

Regards,

Charl

Clock Generation.pdf

  • Hi Charl,

    Can you remove C1 and measure again? How are you measuring the crystal stability?
    The crystal's rated load capacitance matches the on-chip load capacitance, thus no additional load capacitance is necessary.

    The crystal you selected looks OK. You could use Equation 2 from the datasheet to estimate the frequency error of the crystal due to a load capacitance mismatch given the crystal's motional capacitance.

    Kind regards,
    Lane
  • Hi Lane,

    Thank you very much for your response and clarifying that the load capacitance of the crystal needs to mach that of the on-chip load capacitance.

    I have removed the C1 capacitor and saw no noticeable change in the stability. I am doing my measurements with a differential probe across R7 which isn't populated since we are using LVPECL. Attached are screenshots of the clock frequency measurement as well as jitter measurement.

    I tried using the equation from the datasheet, but unfortunately I can't find the crystal's motional capacitance anywhere on its datasheet. 

    Regards,

    Charl 

      

  • Good day,

    Have you been able to further look into my response to your questions?

    Any further assistance would be greatly appreciated.

    Regards,
    Charl

  • Hi Charl,

    It is recommended to measure stability using a frequency counter or a phase noise analyzer rather than an oscilloscope. These instruments can offer more precise frequency measurements. To measure accuracy down to the ppm level, I would recommend frequency counter with at least 1 Hz resolution.

    You could measure OSC_OUT to check the stability of the reference. The crystal you selected is +/-10ppm, so I would expect an OSC_OUT frequency counter measurement of 24.999750 MHz to 25.000250 MHz (depending on which crystal sample is tested). You may need to provide an precise external reference clock input to the frequency counter/phase noise analyzer for best results.

    After that, you could measure OUTP/N. The crystal is +/-10ppm, so I would expect an OUTP/N output frequency measurement of 199.998000 MHz to 200.002000 MHz (depending on which crystal sample is tested).

    Kind regards,
    Lane