• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 45 subscribers
  • Views 533 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

ADC3244: Creating dBFS spectral plots from ADC count data that compare well with those of the data sheet

Roger Backhus
Intellectual 410 points
Part Number: ADC3244

Hello, 

I'm trying to use raw ADC count data to get spectral plots corresponding to those in the data sheet.  Seems simple but my noise floor is about 40 dB off from that of the data sheet plots on a dBFS scale.  So, I terminate the ADC input with a resistive load to ground, sample at 65.536 MSPS ((for a final spectral line resolution of about 3 Hz (took a lot of data)), scale the data by the maximum ADC count value (2^14 - 1 = 16383), do an FFT in MATLAB, take the absolute value of that, convert to dB (20*Log10(fft data) ), offset the plot by (20*Log(1/fs) = -156 dB) to get it to correspond to a Fourier Transform,  So, my spectral noise floor is 40 dB too low.  I'm not seeing a problem here except for the wrong answer.  What am I missing?  Thanks.  

  • 0
    Intellectual 410 points
    It occurred to me that the resolution bandwidth or the spectral line resolution of the data sheet plots may not be as fine as 3 Hz so the noise level would not be as low as mine. 40 dB => a factor of 10000 so the resolution bandwidth of those plots would be about 30 KHz.

    So, never mind. :)