• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 22 subscribers
  • Views 681 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.

LMX2594: Spurious emissions

jianjun lu
Prodigy 50 points

Part Number: LMX2594
Other Parts Discussed in Thread: LMX2572

The problem with LMX2594 is that there are spurious emissions moving back and forth near the main signal.

The main signal is 4703.75M, clock ref 100M, Phase detection frequency 200M, moving spurious emissions disappear at 4700M. It seems to be related to the integer boundary

Another question:

Regarding the problem of LMX2594 integer boundary spurs. clock ref 100M, Phase detection frequency 200M, integer boundary spurs appear at integer multiples of 100M

Change the Phase detection frequency through the Multiplier Frequency divider and Frequency multiplier of the reference channel.

For example, if the phase detection frequency is changed to 233.33M or 160M or 240M, there will still be integer boundary spurs at integer multiples of 100M.

The integer boundary cannot be eliminated by changing the Phase detection frequency, but changing the reference frequency to 80M can eliminate spurious emissions at integer multiples of 100M, but integer boundary spurs appear at integer multiples of 80M.

My question is why changing the Phase detection frequency using Multiplier cannot eliminate integer boundary spurs, but changing the reference frequency can. Isn't the integer boundary defined by Phase detection frequency?

Also, the reference signal I used is a square wave. Does it have any impact?

  • 0
    TI__Guru** 120300 points

    Hi Lu,

    Are you using a big PLL DEN with 2nd order MASH? 

    Try use 3rd order MASH, then adjust PFD_DLY_SEL to optimize phase noise.

    In your configuration, even if you change the fpd to avoid IBS between fpd and VCO, you still has crosstalk between OSCin and VCO, this spurs frequency is same as IBS. 

    Changing fosc to 80MHz will completely change all the spurs frequencies. 

    To reduce crosstalk due to OSCin, suggest use differential clock input such as LVDS.

  • 0 in reply to Noel Fung
    Prodigy 50 points

    When using the sysref function of lmx2572, the sysrefreq pin inputs a 100M reference clock, and the repeat mode is selected, can RFoutB output a 100M reference clock? Will the output signal maintain phase noise of the 100M reference signal?

  • 0 in reply to jianjun lu
    TI__Guru** 120300 points

    Hi Jianjun,

    In sysref repeater mode, sysref divider block is bypassed, the input sysref signal will be routed to a delay circuit and output. Phase noise of the output will be equal to the input.