• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 31 subscribers
  • Views 307 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.

Original question:

PCM4204: ADC Full Scale Input Voltage Discrepancy

PCM4204: PCM4204 Questions

Tony Scott
Prodigy 40 points
Part Number: PCM4204

Questions about the PCM4204:

1) The datasheet only provides a typical full scale input voltage of 6Vp-p with no upper or lower limits. Is there any data available on full scale input voltage variations lot-to-lot, or at least the worst case minimum full scale input voltage? 

2) Noise shaping can explain the rising high frequency noise floor above 50kHz in the 192kS/s FFT plots on datasheet page 12. What explains the steeply rising (about 40dB/octave) low frequency noise floor below 70Hz in the 192k plots?  It's not present in the 48k or 96k plots.

3) Curious about the chip architecture - if the modulator reference voltage is 3V, why is the common mode voltage 2.5V (1/2 Vcc) instead of 1.5V (1/2 Vref)?

Thanks!

- Tony

  • 0
    TI__Guru 62935 points

    Hi Tony,

    I can help w/ #1 and #3 but will need Zak to jump in on #2.

    #1.)  You're correct that the PCM4204 does not include a min/max input range.  However, you'll note that the THD and other performance specifications are taken with a -0.5dBFS input range and therefore it gives some confidence that the minimum anticipated minimum full-scale input wouldn't be much below -0.5dBFS or the clipping would corrupt the performance metrics significantly.

    #2.)  I think this is a # of bins issue w/ the FFT but would need Zak to verify.

    #3.)  The reference voltage is 3V, which allows for the 6Vpp full-scale input range.  In order to avoid clipping and linearity issues as the input signals approach the negative rail it makes more sense to bias the internal circuitry to 1/2 supply instead of shifting it lower and having to worry about swing-to-rail issues mentioned before as inputs need to approach the negative rail.