[sdiy] They aren't sawtooths, they're ramps

[Donald Tillman]

   > Date: Wed, 04 Nov 2009 14:31:02 -0800
   > From: Dave Manley <dlmanley at sonic.net>
   > 
   > David G. Dixon wrote:
   > 
   > > Waveforms derived from relaxation oscillators definitely do not
   > > fit the bill, and therefore should not be called "sawtooth",
   > > any more than some random "rounded" waveform should be called
   > > "sine".

Well, we don't use relaxation oscillators (ie., RC charge up and
reset) for electronic music, so it's never going to come up.  I mean,
if you're using a relaxation oscillator you have probably have more
important things to worry about than the word used to describe the
waveform.

For people who do use relaxation oscillators, "sawtooth" is probably a
fine term in their context.

   > I think there's some 'prior art' that may conflict with your definition. 
   > Didn't the original Moog modular oscillators and first gen minimoogs use 
   > a unijunction transistor in the oscillator core?  I think a buffered 
   > version of that is the sawtooth.

Lots of older VCO's used unijunctions to reset the waveform.  That's
okay, they work like schmitt triggers.

   > What does the saw look like on those old Moogs?

Depends on the model and whether you're measuring it at the cap, after
the buffer circuit, after a DC blocking cap, or after the VCF and VCA.

At the cap, it has to be mighty linear for tuning accuracy.

  -- Don

-- 
Don Tillman
Palo Alto, California
don at till.com
http://www.till.com