it seems there is a 660kHz noise present when output a 2.1~2.2V voltage

I use a AD5338R to bias a power amplifier. There is two signal tone present around +/- 660kHz offset. I try sevral things to identify the source.

  1. if replace AD5338R by a LDO to bias PA , these two spur dissappeared;
  2. increase the AD5338R output to 2.3V, these two spur dissappeared;
  • 0
    •  Analog Employees 
    on Jan 14, 2020 1:24 AM

    Hi ,

    I have few questions for you to better understand your problem:

    1) Can you share with us the circuit diagram used here together with the component values?

    2) Using the same spectrum analyzer, can you check and compare the response at AD5338R's output at these conditions:

    a) connected to PA; 2.1V

    b) connected to PA; 2.3V

    c) disconnected to PA; 2.1V

    d) disconnected to PA; 2.3V

    3) Are you using the internal reference of the AD5338R? Or is it supplied by an external reference device?

    4) How often do you update your output?

    Cheers,

    Ivan

  • Hi Ivan, Thank you for your answer.

    There is another question on my side, typical cap load is 10nF in DS. is possible use a higher capacitor(1uF) at output intends provide a more clean output. Output remians constant most of the time and seldom update values.

    Best regards,

    Alex

  • 0
    •  Analog Employees 
    on Jun 9, 2020 2:16 AM in reply to alexyang

    Hi ,

    Yes you may, but at expense of slower settling time and some overshoots during transient response.

    I advise you to check all possible scenarios in your application circuit once you increase the output capacitance.

    Cheers,

    Ivan