• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 377 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.

PCM1754 noise

S Stone
Prodigy 20 points
Part Number: PCM1754

Hey all, I was on this forum searching for "PCM1754 noise" and everything pointed to layout issues, grounding and decoupling. Although I had this exact issue seen in some posts, my resolution was none of the above.

At the start, the noise was ubiquitous and could be clearly be heard under the (very clear and high quality) audio input... and no hardware changes changed or mitigated this noise floor.

After hours of fooling around with layout and hardware, I simply did not see any influence changes...at all. After stepping back I found my fix was simply software settings and my input  data rate from 16Khz to 48khz completely (100%) resolved by underlying (totally audible) inband  (white) broadband "noise".

If anyone can explain this I'd be happy to learn. :)

b.t.w. Seems that in my travels I also learned that the PCM1754 is rather robust regarding layout...

stay Safe !

  • 0
    TI__Guru 53283 points

    Hi Stone,

    The delta-sigma modulators on these devices create switching noise.  That noise starts to increase at around 50%×fS (the nyquist rate).  The problem with using such a low sampling rate as 16kHz is that the noise starts at 8kHz, which is well within the audible band.

    When you switched to 48kHz, the noise starts at ~24kHz, which is beyond the human hearing spectrum.  

    I have some images of the noise in this cookbook article:

    You can also look at some of our other families, such as the PCM510xA.  THe PCM510xA features an interpolator that will upsample the input data (assuming it is valid) to about 384kSPS, so then the output of band noise does not start until around 190kHz!

    Thanks,

    Paul