[sdiy] NIC-waveshaper progress

[Ian Fritz]

Hi all --

Here's what I've been doing with the new NIC-based waveshaper idea.

As a recap of my previous post, this new waveshaper uses a saturating 
negative resistor in the lower leg of a voltage divider to produce an 
interesting wave folding.  With circuit parameters set appropriately, it 
will take a sawtooth input wave and produce an output wave that can be 
varied with a pot from a sawtooth to a triangle (and beyond, towards an 
inverted sawtooth).

To put this unit under voltage control, I have replaced the "shape" pot 
with a voltage-controlled resistor built from an LM13600 OTA.  I then used 
the other half of the OTA to make a VCA to compensate for the amplitude 
variation that occurs as the shape is varied. Since the amplitude varies 
directly with the variable "shape" resistance, the amplitude leveling is 
very accurate.

Additionally, I looked (on paper only) at what happens when the output wave 
is combined with the input wave.  Since the output wave is shifted by 180 
deg from the input, it turns out that there is an amazing variety of 
waveshapes available if the input is simply either added to or subtracted 
from the shaper's output.

I considered these limiting cases: either sawtooth or triangle drive, 
either adding to or subtracting from the shaper output, and either the 
sawtooth or triangle limits of the shaper.  The resultant waveshapes 
include: square, trapezoid, frequency-doubled sawtooth, triange with tips 
folded down to zero,  triangle chopped in half with the halves shifted 
vertically, and a few more that are too hard to describe.

The next step will be to built a summing circuit to actually produce these 
waveforms.  After that I want to add a second folder to get even more high 
harmonics.  This will probably not be under voltage control but may be 
switchable between the saw and tri limits.  Also I want to include a 
voltage-controlled comparitor to make interesting pulse trains.

This is really shaping up!  (sorry)

   Ian