I am designing a guitar pedal to do pitch shifting, and I have most of it built and working reasonably well, but there one final point which could make it really great.
The audio signal is used to trigger a comparator, but the problem I have is that low-level harmonics cause spurious triggering that spoil the rest of the circuit's operation.
What I'd really like is a circuit that would totally ignore signals in the crossover region, in the range of about 4.45 to 4.55V, or perhaps a little narrower (the circuit operates from +9V, not bipolar). Hopefully this would cut out the annoying harmonics, at least 90% of the time.
I have attached a picture to show what I mean better. I don't much care what happens in the x-over region; it could either switch directly between the two levels, or decay gradually as shown by the dots, whatever.
Obviously a pair of anti-parallel diodes could normally do this job, but the signal levels are too small in this case (very limited headroom). I've also tried SIMing a complementary pair of transistors biased just below cutoff, but it just doesn't seem accurate enough.
Any suggestions?
The audio signal is used to trigger a comparator, but the problem I have is that low-level harmonics cause spurious triggering that spoil the rest of the circuit's operation.
What I'd really like is a circuit that would totally ignore signals in the crossover region, in the range of about 4.45 to 4.55V, or perhaps a little narrower (the circuit operates from +9V, not bipolar). Hopefully this would cut out the annoying harmonics, at least 90% of the time.
I have attached a picture to show what I mean better. I don't much care what happens in the x-over region; it could either switch directly between the two levels, or decay gradually as shown by the dots, whatever.
Obviously a pair of anti-parallel diodes could normally do this job, but the signal levels are too small in this case (very limited headroom). I've also tried SIMing a complementary pair of transistors biased just below cutoff, but it just doesn't seem accurate enough.
Any suggestions?
Attachments
-
25 KB Views: 26