+ Post New Thread
Results 1 to 6 of 6
  1. #1
    Member level 3
    Points: 1,806, Level: 9

    Join Date
    Dec 2005
    Posts
    65
    Helped
    1 / 1
    Points
    1,806
    Level
    9

    What's wrong in CIC filter spectrum?

    I have designed sigma-delta followed by CIC decimator ,
    and run FFT at CIC filter output.
    The result is shown in attachment file.
    what's wrong in the spectrum??
    thank you!
    -------------------------
    spec:
    fin=1.5625 Hz
    fs=51200 Hz
    decimator ration=256

    •   AltAdvertisment

        
       

  2. #2
    Member level 3
    Points: 1,806, Level: 9

    Join Date
    Dec 2005
    Posts
    65
    Helped
    1 / 1
    Points
    1,806
    Level
    9

    Re: What's wrong in CIC filter spectrum?

    No one knows the like "tone" around the notch frequency??
    Thank you!



    •   AltAdvertisment

        
       

  3. #3
    Super Moderator
    Points: 255,029, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    44,441
    Helped
    13537 / 13537
    Points
    255,029
    Level
    100

    Re: What's wrong in CIC filter spectrum?

    I don't understand, if this is a serious question. The output sampling rate is 200 Hz. How can you calculate spectral components above 200 Hz? The FFT is supposed to be cyclic with 200 Hz, this is simply the image of the low frequency output. Due to the logarithmic scale (not appropriate to FTT, actually), it looks somewhat distorted.



    •   AltAdvertisment

        
       

  4. #4
    Member level 3
    Points: 1,806, Level: 9

    Join Date
    Dec 2005
    Posts
    65
    Helped
    1 / 1
    Points
    1,806
    Level
    9

    Re: What's wrong in CIC filter spectrum?

    Quote Originally Posted by FvM
    I don't understand, if this is a serious question. The output sampling rate is 200 Hz. How can you calculate spectral components above 200 Hz? The FFT is supposed to be cyclic with 200 Hz, this is simply the image of the low frequency output. Due to the logarithmic scale (not appropriate to FTT, actually), it looks somewhat distorted.
    Does the question be caused by "sampling theory"??
    If it is caused by sampling theory,
    how can I measure the SNR at digital output?
    Thank you!



  5. #5
    Junior Member level 3
    Points: 864, Level: 6

    Join Date
    May 2009
    Location
    China
    Posts
    27
    Helped
    1 / 1
    Points
    864
    Level
    6

    What's wrong in CIC filter spectrum?

    It doesn't like fft nor the image of the low frequency output~~~but i don't know the answer



    •   AltAdvertisment

        
       

  6. #6
    Super Moderator
    Points: 255,029, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    44,441
    Helped
    13537 / 13537
    Points
    255,029
    Level
    100

    What's wrong in CIC filter spectrum?

    I don't see a problem at all.

    The CIC filter has an output sampling rate of 200 Hz. Thus the meaningful spectral range in SNR measurement is 0 to 100 Hz.
    Above 100 Hz, you're simply displaying a non-existing signal in your above spectrum chart.



--[[ ]]--