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.

Reference clock lock between two signal generators



Hi,

I am reading the application report 'High-Speed, Analog-to-Digital Converter Basics' (SLAA510).

Can you explain to me that why should I have to use a reference clock like 10MHz to lock the clock between two signal generators or between signal generator and spectrum analyzer, just as below?

Thanks for your kind help!

 

 

  • Yarn,

    For coherent sampling of tones you will need to have the clock and signal source locked. This will guarantee that the sampling clock and signal source are coherent (no windowing required in the FFT). Non-coherent sampling without windowing causes a lot of power leaking into adjacent FFT bins and results in a not so good spectrum of the input tone.

    It is not necessary to lock them if windowing can be applied to the captured data to limit the spectral leakage. This will give you a good response where sources cannot be locked together.

    Ken.
  • Hi, Can:

    In oder to prevent power leaking, I think that the input signal frequency should meet Fin = M*Fs/N.
    If the signal generator and sampling clock are not coherent, but meets the formula above, I think that there is no power leaking in FFT. Right?

    Look forward to your kind guidance!
    Yarn.
  • You are correct, however without the 2 sources being phase locked there is no guarantee that the Fin and Fs are the expected frequencies. There may be enough drift in each of the sources to cause coherency problems. Locking the sources guarantees the sources will be coherent at the desired frequencies.

    Ken