[sdiy] Octave divider for guitar/bass

[Curtin, Steven D (Steven)]

One thing that would improve the tracking of the comparator is attaching the
output of the envelope follower to the threshold voltage of the comparator.
This way the duty cycle of the output square wave stays more consistent with
the changing amplitude.  Also try attaching the guitar input to a PLL, then
use the square wave output from that with dividers.  The Korg Octave V
worked really well, and I think their "Waveshaper" box was just a PLL.

Steve C

-----------------------------------------------------------------
Steven Curtin  
Agere Systems (formerly
Lucent Technologies Microelectronics)
ph: (732)949-4404   fax: (732)949-6711
http://curtin.emf.org
sdcurtin at agere.com
-----------------------------------------------------------------


> ----------
> From: 	Antti Huovilainen[SMTP:ajhuovil at cc.hut.fi]
> Sent: 	Friday, March 16, 2001 9:30 AM
> To: 	synth-diy at node12b53.a2000.nl
> Subject: 	[sdiy] Octave divider for guitar/bass
> 
> Hi
> 
> After doing some research about MXR BlueBox, I found out that it's not
> really a octave divider (more like random octave + fuzz), so I'm stuck
> with designing my own.
> 
> Are there any schematics (even partial) for guitar octave dividers on the
> net? Has anyone else on the list tried this?
> 
> I'm thinking of something like this:
> 
> in +-> agc -> comparator -> divide by 2/4 -> vca -> mix with dry -> output
>    |                                          ^
>    \-> env detector --------------------------/
> 
> Since vca input is just square wave, I can replace it with a single
> transistor (switch env detector output on/off). The tricky part is
> implementing AGC and the comparator.
> 
> I can't use zero-crossing as a clock for the divider since at first the
> waveform crosses zero four times in one period but later only twice.
> Detection from peaks would work in the beginning but later (about same
> time as extra zero crossings disappear) the peak becomes almost flat with
> a slight dip, so that won't work either. Would some hybrid solution
> work? Say, switching when waveform crosses 2/3 of peak?
> 
> Any thoughts / ideas? Am I on the right track at all?
> 
> Antti
> 
> "If you don't eat yer meat, you can't have any pudding!
>  How can you have any pudding if you don't eat yer meat!?"
>    - Roger Waters
>