• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 22 subscribers
  • Views 1201 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.

LMX2572: Walking Spurs issue

Terry Liu70
Intellectual 500 points

Part Number: LMX2572

Hi Ti Experts,

On my own board, I am experiencing abnormal spur issue with LMX2572., conditions are,

1, OSC in = PFD = 100MHz, Single Ended, VCXO[CVHD950] sin-wave output.

2, 2572 output freq = 1200MHz on output A, INT mode, internal VCO = 4800M

unstable phase noise near 10MHz offset were noticed when measuring with SSA. When looking at 1210MHz on spectrum analyzer, it shows a batch of walking spurs. Here are video records.

I also measured frequency at 2400MHz, similar results as 1200M. However, when set pll as VCO output of 4800MHz, phase noise is stable and no walking spurs on spectrum analyzer. Base on above measurement, it seems rf divider contributes the walking spur, but this only occurs near 1200/2400MHz frequency range, no such abnormal issue at 600MHz.

I wonder if there is any furthure effors i can try on this issue.

Thanks in advance!

T.L.

  • 0
    TI__Guru** 119830 points

    Hi Terry,

    I don't know where these walking spurs come from, we never see this problem in our EVM.

    So you have problems with div/2, div/4 but not div/8 with VCO at 4800MHz.

    Do you have similar problem with other VCO frequencies?

    What is OUTB_MUX setting? 

  • 0 in reply to Noel Fung
    Intellectual 500 points

    Hi Noel Fung,

    Thanks a lot for your kind advice. With respect to your questions, OUTB was disabled and MUX was set to 'Not Used'.

    I did some more measurements and i think i am getting closer to the root cause,

    I stried more frequencies (1300M, 1500, 1100M, 2300M, 2500M, 4500M, 4800M ...) no such spur issue at near 10M offset range and 5052 shows stable phase curve. However the spectrum analyzer can always capture  walking spurs at differents power levels at  near 1210M and 2410M (escepilly higher spur level at 2410M) range. Thus, i suspect the walking spurs could be sourced from lab WiFi/BlueTooth signals in the air and 'catached'  by PLL RF chain. ..

    Thanks,

    T.L.

  • 0 in reply to Terry Liu70
    TI__Mastermind 28031 points

    T.L.,

    Have you eliminated the fractional circuitry as a cause?

    1.  Does modulator order impact the magnitude or the speed at which these spurs walk?

    2.  If you are using a large inexact fraction (i.e. 10000000/50000001 instead of 1/5), try simplifying the fraction. 

    3.  If you could shut down the fraction altogether and go to a pure integer channel and select integer mode, then you can be confident it is not related to this part.

    If it's not related to the fractional circuitry, then I would suspect either the power supply or input reference.

    For the reference, if the amplitude is very weak, maybe it could have some impact.  

    For the power supply, try a different supply if possible.

    Also, try lowering the charge pump gain and see if it improves the spurs or not.  If it's power supply getting to the VCO it might not help, but if it's PLL related or input reference related, then it might have some impact.  This is not a solution proposal, but for diagnosic purposes.

    Regards,
    Dean