• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 10 subscribers
  • Views 1427 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TRF3722: Reference Clock and PFD Feedthrough

Jon Watkins71
Prodigy 40 points
Part Number: TRF3722

I am currently utilizing the TRF3722 in a transmitter design.  While testing with the TRF3722EVM eval board (100MHz ref from a microwave source, 25 MHz PFD, and 2800 MHz output), I noticed what seemed like the reference frequency feeding through to the modulated output.  See attached picture.

A second test on the eval board showed that with the DAC input disabled, the PFD frequency is also seeming to bleed through.  See second attached picture.

The output of the TRF3722 is fed through a low pass filter to reduce the 2nd and 3rd harmonics, but I still have an issue of the reference clock and PFD feedthrough, which can cause "close-in" mixed spur content at the output of an upstream mixer.

Is there a way that I can reduce this feedthrough?

Thank you in advance.

Jon

Reference Feedthrough

DAC/Modulator disabled

  • 0
    TI__Genius 9695 points
    Hi Jon,

    We are looking into this.

    Thanks,
    Eben.
  • 0
    TI__Mastermind 18065 points

    Jon:

    The high order PFD spurs are present; however, I do not see them quite as high.  I measured on the LO output shown below.  The largest high order spur is the -150 MHz offset at about -83 dBm.  If I measure on the RF output with no BB input signal the level drops to -95 dBm.

    There are a few knobs that impact PFD spur performance.  Care should be taken as there is usually a correllation to phase noise whereas when you improve one, the other degrades.  The loop filter is one of the first areas to explore.  I recommend to use the fractional loop filter from the TI EVM.  You can adjust the charge pump voltage to effectively change the loop filter response.  I was able to see some PFD spur reduction with lower CP, but this does impact phase noise performance.

    There are a couple of other knobs that may provide some improvement: ISOURCE_TRIM and TX_DIV_BIAS.  I saw marginal changes in PFD spurs with modification of these parameters.

    If your frequencies are always even multiples of the PFD, you can opt to switch to integer mode where the high order PFD spurs are not present.

    --RJH