Suboctave generator - 4013?

[Tony Allgood]

>Anyone use 4013-based sub octave generators in their designs?
>Drawbacks? Limitations?

Actually I have had a problem when using one manufacturer's version of
the IC in the TB3030. I use a comparator to square up the signal from
the 'square' output of the TB3030 VCO and then a further spare CMOS nand
gate to straighten the edges just in case. However some samples appear
not to divide by two. This maybe because of the less than fast rise time
at the clock input (don't believe this bit), or because set-up times are
not being met. Early CMOS 4000 specs are a bit wayward for this. I think
there's a bit in Horowitz and Hill about this very thing. Or maybe there
is a runt pulse being delivered by the comparator as it swings
through... either way I can't get my scope to detect either runt pulse
or rise time error. I tell my customers to use the Motorola part,
MC14013s, which seem to be used in many early 1980s synths and stomp
boxes. I have had no trouble with the Motorola part ever, and
fortunately, they seem to be the most common part.

As a footnote, one customer did suggest to use a 2K2 resistor as the
feedback path, not just a simple wire link. This would always create a
delay in the feedback path compared to the clock rising edge. I have
never tried it to see if it would work.

Regards,

Tony Allgood  Penrith, Cumbria, England

Oakley Modular Synth and TB3030:
www.techrepairs.freeserve.co.uk/projects.htm
My music: www.mp3.com/taklamakan