1
\$\begingroup\$

This question may be too general, but I will include all the details that I can. My end goal is to achieve a super high-resolution spectrogram of any recording I choose to make. By doing some tests I realized that the higher the sample rate the higher the spectrogram resolution which makes sense because with a higher sample rate the same size bin will correspond to a finer time resolution. However, after looking for the highest sample rate ADC I realized that there is another parameter called resolution in the ADC world that corresponds to the resolution of recorded signal power. I supposed that increasing this parameter would also increase spectrogram resolution because there would be less of a color jump from pixel to pixel. I have tried to understand the ins and outs of sigma-delta ADC modules but I can't seem to get a solid grasp because there are so many parameters.

Note* Before I launched into my ADC investigation, I tried going the consumer route and just buying a good audio ADC. I found that most of these consumer devices were overpriced and under specced (probably because most people would like to record sound for the sake of listening to it not looking at it). I did find some promising products like this one from Korg https://www.korg.com/us/products/audio/ds_dac_10r/, but my suspicion is that a higher sample rate and resolution are physically possible and are obtainable for less money. Also, I have minimal EE experience (competent with Arduino projects and general coding).

So my question is this. What ADC parameters should I look for to fulfill my goal? Here are some that I am looking at:

https://www.digikey.com/product-detail/en/texas-instruments/ADS7886SDBVT/296-19307-1-ND/948628

http://www.ti.com/lit/ds/symlink/ads1605.pdf

I realize I am way out of my depth here. I would appreciate any advice or ways of recording the quality of audio that I need without building an overclocked ADC or buying an overpriced sound recorder.

\$\endgroup\$
  • \$\begingroup\$ What kind of signal do you want to measure? Only audio? In that case a higher bandwidth won't give you better results, because there simply are no higher frequency components in the signal (assuming the audio is coming from typical playback device). If you want to measure something else it would be good to know what kind of signal you have in mind. \$\endgroup\$ – jusaca Jul 21 at 8:03
  • \$\begingroup\$ Realistically, you will not achieve the theoretical performance of even a cheap audio ADC without a lot of experience in the design of the surrounding board and circuit. \$\endgroup\$ – Chris Stratton Jul 21 at 13:57
  • \$\begingroup\$ I am converting human voices into spectrograms. \$\endgroup\$ – Frederick King Cunningham Jul 21 at 20:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.