• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 559 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.

DAC5675A spurs

Michael Kobayashi
Prodigy 80 points
Other Parts Discussed in Thread: DAC5675A

Hello,

I am working on a design that uses the DAC5675A as a DDS output at 20MHz. We are feeding the DAC with a 50MHz clock from a TCXO. It is unfavorable as it is close to the Nyquist, but we could not use the PLL in the FPGA to bring the sample frequency up without degrading the phase noise. Phase noise and SFDR is important in our application.

At 20MHz output, we see a clean 20MHz that we can filter out the aliases as expected with good phase noise. When we try to change the output to 20.1MHz or any other frequency, we start to see very low frequency spurs at the DAC output. The spurs are about 120Hz away from the main carrier, and they look to be aliasing with something. We basically see a picket fence of spurs very close in to the carrier, maybe more than 10 on each side.

I noticed that the data sheet lists SFDR at 100MHz clock, but then the output is at 19.9MHz. Is there a reason why the 20MHz region may be specially susceptible to noise/spurs, etc.?

Any ideas or recommendations on debugging this problem?

  • 0
    TI__Mastermind 20541 points
    Hi Michael,

    I am looking into the problem will respond back to you early next week.

    Regards,
    Neeraj Gill
  • 0 in reply to Neeraj Gill
    Prodigy 10 points
    To give a bit more information, the spurs we are seeing are actually coming from the digital lines and exist there as a function of our DDS. But for some reason they are also on showing up on the DAC output.

    What are some common problems that would cause such crossover in the DAC?
    Any suggestions on how determine if the coupling is happening within the chip or external to it, such as on the pcb?
  • 0 in reply to Kameron Larsen1
    TI__Mastermind 20541 points

    Hi,

    It sounds like noise from your power supply is coupling into your digital lines/clock signal.

    I would recommend using a cleaner power supply. And also recommend to use independent 3.3V analog and digital supplies to provide higher performance.

    Regards,

    Neeraj Gill

    HSCC

  • 0 in reply to Neeraj Gill
    Prodigy 80 points

    Thanks Neeraj,

    We were able to get some of the spurs down by shortening the ground loop at the output of the balun to bring the signals to single-ended. We also were able to reduce common-mode noise by double-terminating the output.

    We actually pass this signal into a PLO to up-convert to X-Band, but we still see some spurs at X-Band. We're still looking into it. The 20MHz tone is brought up to 8400MHz by a multiplier with N about 420.