• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 46 subscribers
  • Views 287 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.

DAC38J82: does phase difference between output of DAC38J82 and LMX2594 constant?

Bill Xu
Intellectual 1310 points
Part Number: DAC38J82
Other Parts Discussed in Thread: LMX2594,

Tool/software:

Dear  product line

my customer want to use DAC38J82 and LMX2594(PLL) in their design. they use an 80MHz external clock as input for DAC38J82 and LMX2594. then configure DAC38J82 output to be a sinwave with frequency equal with LMX2594 output. customer's question is whether or not the phase difference between DAC38J82 and LMX2594 is constant at any output frequency? thanks for your supporting.

regards,

Bill

  • 0
    TI__Mastermind 23710 points

    Hi Bill,

    The phase difference will only be constant if the two frequencies match. This does not mean the phase offset will be 0 degrees. Since the DAC will be clocked by the same source as LMX2594, the two frequencies should not drift apart. Does this answer your question? If not, please provide a high level block diagram of what the customer is trying to achieve.

    Thanks, Chase

  • 0 in reply to Chase W
    TI__Intellectual 1310 points

    hi, Chase

    here is feedback from customer.

       "My design is as below:

       Clock source is same (80MHz).

       It distributes to 2 outputs.

       One is input to LMX2594 and the other is input to DAC38J82.

       LMX2594 is used to generate 21MHz clock (for example) and DAC38J82 is used to generate 21MHz single tone waveform (same frequency with LMX2594)

       My question is whether the frequency from LMX2594 output is exactly same with the frequency from DAC38J82 output (same frequency, zero tolerance).

       From the below answer, it seems the frequency from both output are completely same.

       Please double confirm…"

    regards,

    Bill

  • 0 in reply to Bill Xu
    TI__Mastermind 23710 points

    Hi Bill,

    The DAC output will eventually drift due to the NCO resolution if in DDS mode or as a result of the input data stream repeating. There are still questions about whether the DAC is being supplied data by an FPGA or if the DAC is using the NCO in constant input mode (DDS mode). Then comes the question of is the signal coherent or is it an exact 21MHz? In a short span, the frequencies can be exactly the same, over a longer period of time, they likely will not.

    If in the DDS mode, the NCO resolution is the limiting factor. At 80MSPS, the NCO resolution is 80e6/2^48=0.0000286Hz. Next calculate the frequency word as 21e6/0.0000286=734265734265.734 but this is not a perfect integer, so the closest you can get to exactly 21MHz is when the frequency word is 734265734265.0 (round up and down and see which is closer, the rounding does not always guarantee you are closest to desired frequency) -> taking this to frequency by multiplying by NCO resolution, we get 20999999.999991 Hz, which is 0.000009 Hz off. This will slip by 1 cycle in about 1/0.000009=111111.11 seconds, or roughly 30 hours and 52 minutes. 

    If receiving data by the FPGA, the resolution will be limited by the size of the input pattern, the larger the input data pattern, the closer to the desired 21MHz the output will be.

    Thanks, Chase