• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 3 answers
  • Subscribers 21 subscribers
  • Views 532 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: LMX2594 integer boundary spurious problem

Martin Ho
Intellectual 545 points
Part Number: LMX2594


Hi,

   I built a LMX2594 synthesize as following, the reference signal comes from a 2.5GHz BAW oscillator following a programmable divider :

   We could choose different reference frequency for different output frequency to reduce the integer boundary spurious. However, for the 7500MHz nearby output, which is triple of 2.5G BAW oscillator, it seems useless to choose non integer reference frequency.

   For example, I tried following settings. All got similar spurious level which is not acceptable. It seems there are other mechanism dominate the spurious generation. What that could be ? How to reduce spurious level in this condition ? Thanks.

Martin

  • 0
    TI__Guru** 109330 points

    Hi Martin,

    In your first configuration, what spurs frequencies did you get? I actually don't except there is spurs with this configuration.

    second configuration, for sure there will be 500kHz and 1MHz spurs.

    The third configuration is not a valid configuration because Doubler and Multiplier cannot be used at the same time.

  • 0 in reply to Noel Fung
    Intellectual 545 points

    Hi Noel,

         Thanks your feedback.  I will do more test about the first configuration and get back to you.

        Besides, could you advise the optimum setting for the below parameters in this configuration. Is there any rule about this ? What I did is trying. 

    Martin

  • 0 in reply to Martin Ho
    TI__Guru** 109330 points

    Hi Martin,

    VCO Calibration: these register settings should match with the fpd and fosc. You can mouse-over to the configuration box to see the description in the left window.

    VCO Assist: Keep these settings unless you want to use full assist. See datasheet section 7.3.6.1 for details.

    MASH and Phase Sync: see datasheet Table 2 for the suggested value for PFD_DLY_SEL. We can choose different MASH_ORDER to optimize spurs. 

  • 0 in reply to Noel Fung
    Intellectual 545 points

    Hi Noel,

        Thanks for your feedback. We will try to optimize configuration 1 and get back the result to you.

    Martin

  • 0 in reply to Martin Ho
    Intellectual 545 points

    Hi Noel,

      MASH and Phase Sync: see datasheet Table 2 for the suggested value for PFD_DLY_SEL. We can choose different MASH_ORDER to optimize spurs. 

    =>  If two mash order comply with Table 2 for some frequency,  Is there any rule like " choose larger / smaller mash order " to get better spur ? I saw "biger mash order could spread spur more effectively" in some other thread . Does this mean we should always choose the larger mash order ?

    Martin

  • 0 in reply to Martin Ho
    TI__Guru** 109330 points

    Hi Martin,

    in general, higher order mash returns better spurs but phase noise may get higher. For LMX2594, so far 3rd order mash is my preference. 

  • 0 in reply to Noel Fung
    Intellectual 545 points

    Hi Noel,

       The synthersizer is used for generating a QAM modulated signal. What we care is the EVM. For EVM, we should concern the sum of spurs and phase noise, right ? In this case, what is your prefernce ? higher or lower order mash ? Thanks.

    Martin  

  • 0 in reply to Martin Ho
    TI__Guru** 109330 points

    Hi Matin,

    Right, EVM takes phase noise and spurs into account. A good start for LMX2594 is 3rd order mash.