[sdiy] Waveshaper update

[Ian Fritz]

Hi Michael --

>Your waveshaper sounds really great (just listening to the little demo
>soundbite), and the idea is very clever.

Thanks.  There's a lot of bang-for-the-buck with this one!

>I'm wondering two things:  What happens if
>you drive it with a hard-synced sawtooth wave?  The sound of the Five Pulser
>is kind of reminiscent of the sound of a swept hardsynced oscillator -- it
>would be interesting to see how the two sounds would interact (that is, if 
>the 5P
>would even accept the synced wave as input.

Haven't tried that, but any 0-5V signal will work, as long as it is mostly 
smooth.

>Also, did you ever think about
>(or try) cascading two LM3914's, to make a Ten Pulser circuit?  I wonder 
>if the
>additional pulses would make for a more detailed, more interesting, or more
>harmonically rich sound?

I'm sure that could be done.  The 3914 data sheet gives detailed info on 
cascading.  A tricky detail would be that the voltage swing at the input to 
the comparitor would be reduced by a factor of two (about one volt swing 
instead of two volt swing). You might want to use two separate comparitors 
and AND their output.

>BTW, I imagine that a properly DC level-shifted triangle would also work well
>as an input wave.  Have you tried that?

Yes.  You can get (for example) five pulses going up and five going down.

BTW, I chose to use a 0-5V input with the idea of dedicating the circuit to 
the raw output of a VCO.  This allows the drive levels to be precisely 
set.  Also, the direct connection minimizes problems with triggering during 
the SAW reset, as the reset trace is fastest at that point.  If you want a 
more general input then just set up the input amplifier to map (-5,+5) to 
(0,1).

Another version of the circuit I made fed the 1,3,5 pulses to one 
comparitor and the 2,4 pulses to another.  The second set went through a 
VCA/inverter/summer to produce a voltage controlled bipolar amplitude 
signal, which was then added to the first set.  This gives another timbral 
dimension, but at the expense of quite a bit of extra circuitry.

   Ian