On 03/01/2017 01:40 AM, Jim Brain via cctalk wrote:
  On 2/28/2017 2:34 PM, Brent Hilpert wrote:
snippage
  Obviously, given the LM339 issues, this is unworkable
in the end, but
 it did get me to realizing some outputs on the LM339, so thanks to
 both of you.
 I can plainly see the oscillation on the comparator transitions, so it
 is obvious I need a better comparator.  I'll have to source a few
 different comparator options from Digikey to fill my parts box.
 Suggestions are welcome. 
OR  a bit of positive feedback.
   Using the
internal AVR comparator sounds like a better final solution
 (fewer components), but in devising an
 input circuit for the AVR you may be running into the same issue of
 loading a high-impedance source. 
 Though it plays into Tonly's notes about my
analog understanding (or
 lack thereof), I was so disillusioned by the analog uncertainties that
 I initial focused on the AVR comparator.  By correctly biasing the
 comparator, I was able to get much better results tonight, but the
 lack of any ability to add hysteresis into the design hindered my
 success.  At transition, the signal simply bounces badly.  I tried to
 construct a low pass filter (22pF and 5K6), and some variations, but I
 was not able to overcome.  I then tried to deal with it in SW, but I
 have not yet been successful.  I'll try again, but lack of hysteresis
 is a big issue.
 
 Single transistor audio amp before it.
Allison
  Jim