• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 21 subscribers
  • Views 382 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.

LMX2486: RF & IF detector frequency signal divided from oscin: are they phase synchronized?

Jhon Hoo
Prodigy 30 points
Part Number: LMX2486
Other Parts Discussed in Thread: TIDA-01410

In LMX2496, both the RF and IF have a phase detector frequency (pdf)  signal divided from the oscin.

If RF_R_divider == IF_R_divider, the RF and IF parts have the same pdf, but do they have the same phase? 

In the TI tech document  (tidud11.pdf):

TI Designs: TIDA-01410
Phase Synchronization of Multiple PLL Synthesizers Reference Design

section 2.2.1 states:

"One thing that can cause the phase relationships to be different is if there are divides involved. For instance, whenever a signal is divide-by-two there are two possible output phases, which requires the designer to take action. "

For example, if oscin=30MHz, and  RF_R_divider = IF_R_divider = 3, so both the pdf signals are 10MHz, but since they are divided by 3 from 30MHz, so they may have 3 possible phases, and the two pdf signals may not phase synchronized. Consequently, the output RF and IF signal may also not synchronized.

  • 0
    TI__Guru** 110530 points

    Hi Jhan,

    If both VCO frequency is identical and is an integer channel, then RF and IF are synchronized (maybe there exists some phase difference because the structure of RF and IF are different).

  • 0 in reply to Noel Fung
    Prodigy 30 points

    Thanks for your reply, Fung.

    What I am concerned about is the phase difference between the two RF and IF signals.

    If both R_Divider and N_Divider are the same and both the PLLs are integer, resulting in identical RF and IF frequencies, is the phase difference between them also constant? The pdf signals are divided from the oscin signal, so the phase difference between the two pdf signals may be random, right?

  • 0 in reply to Jhon Hoo
    TI__Guru** 110530 points

    Hi Jhon,

    pdf signal is used to phase lock the vco, the ambiguous phase of R divider is therefore not matter. As long as the synthesizers have the same configuration, output frequency is an integer multiple of fosc, no matter what is the R divider value, output of the synthesizers will have the same phase. 

  • 0 in reply to Noel Fung
    Prodigy 30 points

    Hi Fung, please see the figure,

    The first signal (f) is oscin, and R divider is 4, and the divided signal may have 4 phases. Suppose the RF PDF signal is state 1, and the IF PDF signal is state 4, they have 3 oscin clocks delay in time, and I think it is a 270 degree phase difference. Then the RF feedback input signal is synchronized to state 1, and IF feedback input signal is synchronized to state 4, so I think the RF and IF signal have 270 degree phase difference too. Am I right? What's wrong?

  • 0 in reply to Jhon Hoo
    TI__Guru** 110530 points

    Hi Jhon,

    If you have a multi-output clock device with individual divider at it outputs, yes, each output may have different phases. Here we are talking about distributing a clock source to different channels. In fact, the propagation delay of each channel will create uneven phase (or skew) between channels.

    In PLL, the PDF is used to lock the VCO but not distribution. The PDF signal is used to trigger the phase detector and charge pump to correct the VCO frequency. Which edge is used to trigger the phase detector is not important, as the goal is to make VCO = N x PDF. (here should we count the phase ambiguous due to the N counter?). In addition, no one know how many PDF cycle is actually used until the VCO is locked. If the RF PLL uses one more PDF cycle than the IF PLL, will the phase of RF PLL lag behind IF PLL? The answer is not.